Distance measurement is part of various robotic applications. There exist many methods for this purpose. In this paper, we introduce a new method to measure the distance from a digital camera to an arbitrary object by using its pose (X,Y pixel coordination and the angel of the camera). The method uses a pre-data that stores all the information about the relation between the pose and the distance of an object to the camera. This process designed for a robot that is a part of a robotic team participating in RoboCup KSL competition
This paper presents an improved edge detection algorithm for facial and remotely sensed images using
vector order statistics. The developed algorithm processes coloured images directly without been converted
to grey scale. A number of the existing algorithms converts the coloured images into grey scale before
detection of edges. But this process leads to inaccurate precision of recognized edges, thus producing false
and broken edges in the output edge map. Facial and remotely sensed images consist of curved edge lines
which have to be detected continuously to prevent broken edges. In order to deal with this, a collection of
pixel approach is introduced with a view to minimizing the false and broken edges that exists in the
generated output edge map of facial and remotely sensed images.
Stereo Correspondence Algorithms for Robotic Applications Under Ideal And Non...CSCJournals
The use of visual information in real time applications such as in robotic pick, navigation, obstacle avoidance etc. has been widely used in many sectors for enabling them to interact with its environment. Robotics require computationally simpler and easy to implement stereo vision algorithms that will provide reliable and accurate results under real time constraint. Stereo vision is a less expensive, passive sensing technique, for inferring the three dimensional position of objects from two or more simultaneous views of a scene and there is no interference with other sensing devices if multiple robots are present in the same environment. Stereo correspondence aims at finding matching points in the stereo image pair based on Lambertian criteria to obtain disparity. The correspondence algorithm will provide high resolution disparity maps of the scene by comparing two views of the scene under the study. By using the principle of triangulation and with the help of camera parameters, depth information can be extracted from this disparity .Since the focus is on real-time application, only the local stereo correspondence algorithms are considered. A comparative study based on error and computational costs are done between two area based algorithms. Evaluation of Sum of absolute Difference algorithm, which is less computationally expensive, suitable for ideal lightening condition and a more accurate adaptive binary support window algorithm that can handle of non-ideal lighting conditions are taken for this study. To simplify the correspondence search, rectified stereo image pairs are used as inputs.
An algorithm to quantify the swelling by reconstructing 3D model of the face with stereo images is presented. We
analyzed the primary problems in computational stereo, which include correspondence and depth calculation. Work has been carried out to determine suitable methods for depth estimation and standardizing volume estimations. Finally we designed software for reconstructing 3D images from 2D stereo images, which was built on Matlab and Visual C++. Utilizing
techniques from multi-view geometry, a 3D model of the face was constructed and refined. An explicit analysis of the stereo
disparity calculation methods and filter elimination disparity estimation for increasing reliability of the disparity map was
used. Minimizing variability in position by using more precise positioning techniques and resources will increase the accuracy of this technique and is a focus for future work
This document is an assignment submission that discusses and compares various image processing commands and edge detection techniques in MATLAB. It provides examples and descriptions of the 'imshow', 'geoshow', and 'mapshow' commands for displaying images and their differences. It also examines the 'edge' command for edge detection and compares the Sobel, Prewitt, Roberts, Laplacian of Gaussian, and Canny edge detection methods. Finally, it defines what a world file is, its structure and extensions, and how to read and write world files in MATLAB using the 'worldfileread' and 'worldfilewrite' commands.
International Journal of Engineering Research and Development (IJERD)IJERD Editor
journal publishing, how to publish research paper, Call For research paper, international journal, publishing a paper, IJERD, journal of science and technology, how to get a research paper published, publishing a paper, publishing of journal, publishing of research paper, reserach and review articles, IJERD Journal, How to publish your research paper, publish research paper, open access engineering journal, Engineering journal, Mathemetics journal, Physics journal, Chemistry journal, Computer Engineering, Computer Science journal, how to submit your paper, peer reviw journal, indexed journal, reserach and review articles, engineering journal, www.ijerd.com, research journals,
yahoo journals, bing journals, International Journal of Engineering Research and Development, google journals, hard copy of journal
The document discusses image processing and provides information on several key topics:
1. Image processing can be grouped into compression, preprocessing, and analysis. Preprocessing improves image quality by reducing noise and enhancing edges. Analysis extracts numeric or graphical information for tasks like classification.
2. Images are 2D matrices of intensity values represented by pixels. Common digital formats include grayscale, RGB, and RGBA. Higher bit depths allow more intensity levels to be represented.
3. Basic measurements of images include spatial resolution in pixels per unit, bit depth determining representable intensity levels, and factors like saturation and noise.
Paper 58 disparity-of_stereo_images_by_self_adaptive_algorithmMDABDULMANNANMONDAL
This document summarizes a research paper that proposes a new stereo matching algorithm called Self Adaptive Algorithm (SAA) to efficiently compute stereo correspondence or disparity maps from stereo images. SAA aims to improve matching speed by reducing the search zone and avoiding false matches through an adaptive search approach. It dynamically selects the search range based on previous matching results, reducing the range by 50% with each iteration. Experimental results on standard stereo datasets show that SAA outperforms other methods in terms of speed while maintaining accuracy, with processing speeds of 535 fps and 377 fps for different image pairs. SAA reduces computational time by 70.53-99.93% compared to other state-of-the-art methods.
Image segmentation techniques
More information on this research can be found in:
Hussein, Rania, Frederic D. McKenzie. “Identifying Ambiguous Prostate Gland Contours from Histology Using Capsule Shape Information and Least Squares Curve Fitting.” The International Journal of Computer Assisted Radiology and Surgery ( IJCARS), Volume 2 Numbers 3-4, pp. 143-150, December 2007.
This paper presents an improved edge detection algorithm for facial and remotely sensed images using
vector order statistics. The developed algorithm processes coloured images directly without been converted
to grey scale. A number of the existing algorithms converts the coloured images into grey scale before
detection of edges. But this process leads to inaccurate precision of recognized edges, thus producing false
and broken edges in the output edge map. Facial and remotely sensed images consist of curved edge lines
which have to be detected continuously to prevent broken edges. In order to deal with this, a collection of
pixel approach is introduced with a view to minimizing the false and broken edges that exists in the
generated output edge map of facial and remotely sensed images.
Stereo Correspondence Algorithms for Robotic Applications Under Ideal And Non...CSCJournals
The use of visual information in real time applications such as in robotic pick, navigation, obstacle avoidance etc. has been widely used in many sectors for enabling them to interact with its environment. Robotics require computationally simpler and easy to implement stereo vision algorithms that will provide reliable and accurate results under real time constraint. Stereo vision is a less expensive, passive sensing technique, for inferring the three dimensional position of objects from two or more simultaneous views of a scene and there is no interference with other sensing devices if multiple robots are present in the same environment. Stereo correspondence aims at finding matching points in the stereo image pair based on Lambertian criteria to obtain disparity. The correspondence algorithm will provide high resolution disparity maps of the scene by comparing two views of the scene under the study. By using the principle of triangulation and with the help of camera parameters, depth information can be extracted from this disparity .Since the focus is on real-time application, only the local stereo correspondence algorithms are considered. A comparative study based on error and computational costs are done between two area based algorithms. Evaluation of Sum of absolute Difference algorithm, which is less computationally expensive, suitable for ideal lightening condition and a more accurate adaptive binary support window algorithm that can handle of non-ideal lighting conditions are taken for this study. To simplify the correspondence search, rectified stereo image pairs are used as inputs.
An algorithm to quantify the swelling by reconstructing 3D model of the face with stereo images is presented. We
analyzed the primary problems in computational stereo, which include correspondence and depth calculation. Work has been carried out to determine suitable methods for depth estimation and standardizing volume estimations. Finally we designed software for reconstructing 3D images from 2D stereo images, which was built on Matlab and Visual C++. Utilizing
techniques from multi-view geometry, a 3D model of the face was constructed and refined. An explicit analysis of the stereo
disparity calculation methods and filter elimination disparity estimation for increasing reliability of the disparity map was
used. Minimizing variability in position by using more precise positioning techniques and resources will increase the accuracy of this technique and is a focus for future work
This document is an assignment submission that discusses and compares various image processing commands and edge detection techniques in MATLAB. It provides examples and descriptions of the 'imshow', 'geoshow', and 'mapshow' commands for displaying images and their differences. It also examines the 'edge' command for edge detection and compares the Sobel, Prewitt, Roberts, Laplacian of Gaussian, and Canny edge detection methods. Finally, it defines what a world file is, its structure and extensions, and how to read and write world files in MATLAB using the 'worldfileread' and 'worldfilewrite' commands.
International Journal of Engineering Research and Development (IJERD)IJERD Editor
journal publishing, how to publish research paper, Call For research paper, international journal, publishing a paper, IJERD, journal of science and technology, how to get a research paper published, publishing a paper, publishing of journal, publishing of research paper, reserach and review articles, IJERD Journal, How to publish your research paper, publish research paper, open access engineering journal, Engineering journal, Mathemetics journal, Physics journal, Chemistry journal, Computer Engineering, Computer Science journal, how to submit your paper, peer reviw journal, indexed journal, reserach and review articles, engineering journal, www.ijerd.com, research journals,
yahoo journals, bing journals, International Journal of Engineering Research and Development, google journals, hard copy of journal
The document discusses image processing and provides information on several key topics:
1. Image processing can be grouped into compression, preprocessing, and analysis. Preprocessing improves image quality by reducing noise and enhancing edges. Analysis extracts numeric or graphical information for tasks like classification.
2. Images are 2D matrices of intensity values represented by pixels. Common digital formats include grayscale, RGB, and RGBA. Higher bit depths allow more intensity levels to be represented.
3. Basic measurements of images include spatial resolution in pixels per unit, bit depth determining representable intensity levels, and factors like saturation and noise.
Paper 58 disparity-of_stereo_images_by_self_adaptive_algorithmMDABDULMANNANMONDAL
This document summarizes a research paper that proposes a new stereo matching algorithm called Self Adaptive Algorithm (SAA) to efficiently compute stereo correspondence or disparity maps from stereo images. SAA aims to improve matching speed by reducing the search zone and avoiding false matches through an adaptive search approach. It dynamically selects the search range based on previous matching results, reducing the range by 50% with each iteration. Experimental results on standard stereo datasets show that SAA outperforms other methods in terms of speed while maintaining accuracy, with processing speeds of 535 fps and 377 fps for different image pairs. SAA reduces computational time by 70.53-99.93% compared to other state-of-the-art methods.
Image segmentation techniques
More information on this research can be found in:
Hussein, Rania, Frederic D. McKenzie. “Identifying Ambiguous Prostate Gland Contours from Histology Using Capsule Shape Information and Least Squares Curve Fitting.” The International Journal of Computer Assisted Radiology and Surgery ( IJCARS), Volume 2 Numbers 3-4, pp. 143-150, December 2007.
An Enhanced Computer Vision Based Hand Movement Capturing System with Stereo ...CSCJournals
This framework is a hand movement capturing method which could be done in three different depth levels. The algorithm has the capability of capturing and identifying when the hand is moving up, down, right and left. From these captured movements four signals could be generated. Moreover, when these hand movements are done, 15cm-75cm, 75cm-100cm, 100cm- 200cm from the camera (3 depth levels), twelve different signals could be generated. These generated signals could be used for applications such as game controlling (gaming).The existing method uses an object area based method for depth analysis. The results of the proposed work shows it has high accuracy compared to the existing method when tested for depth analysis.
This document describes a new 3D Areal Image (3D-AI) sensor that can measure the heights of hundreds of micro-scale bumps or other interconnect features simultaneously. The sensor uses phase imaging to measure the height of each feature in its field of view in parallel. Initial tests show it can accurately measure features from a few microns like copper nails to tens of microns like solder bumps used in chip packaging. The sensor provides both 3D metrology for height and coplanarity inspection as well as 2D defect detection in a single tool, offering high throughput needed for production. Measurements of micro bumps, C4 bumps, and copper nails demonstrate its capability and correlation to other metrology tools
This document presents a system for measuring doors and windows using images taken on a mobile device. The system allows a user to take a photo of a door or window, select its corners, and then measure dimensions. It removes the existing object from the image and overlays a 3D model of a replacement. Tests show it can measure with an average error of 0.5% across different viewing angles, distances, and image resolutions. The system allows measuring inaccessible objects and visualizing replacements.
This paper discusses techniques for digital image processing, including noise reduction, edge detection, and histogram equalization. Noise reduction techniques discussed include mean, Gaussian, and median filters to remove salt and pepper noise and Gaussian noise. Edge detection algorithms like Sobel and Laplacian are introduced to reduce image data while preserving object boundaries. Histogram equalization is used for image enhancement by spreading pixel values across the full intensity range for increased contrast. The goal is recognizing objects in images through these preprocessing steps.
EFFECTIVE INTEREST REGION ESTIMATION MODEL TO REPRESENT CORNERS FOR IMAGE sipij
One of the most important steps to describe local features is to estimate the interest region around the feature location to achieve the invariance against different image transformation. The pixels inside the interest region are used to build the descriptor, to represent a feature. Estimating the interest region
around a corner location is a fundamental step to describe the corner feature. But the process is challenging under different image conditions. Most of the corner detectors derive appropriate scales to estimate the region to build descriptors. In our approach, we have proposed a new local maxima-based
interest region detection method. This region estimation method can be used to build descriptors to represent corners. We have performed a comparative analysis to match the feature points using recent corner detectors and the result shows that our method achieves better precision and recall results than
existing methods.
At the end of this lesson, you should be able to;
describe Connected Components and Contours in image segmentation.
discuss region based segmentation method.
discuss Region Growing segmentation technique.
discuss Morphological Watersheds segmentation.
discuss Model Based Segmentation.
discuss Motion Segmentation.
implement connected components, flood fill, watershed, template matching and frame difference techniques.
formulate possible mechanisms to propose segmentation methods to solve problems.
Depth Estimation from Defocused Images: a SurveyIJAAS Team
An important step in 3D data generation is the generation of depth map. Depth map is a black and white image which has exactly the same size of the original captured 2D image that indicates the relative distance of each pixel from the observer to the objects in the real world. This paper presents a survey of Depth Perception from Defocused or blurs images as well as image from motion. The change of distance of the object from the camera has direct relation with the amount of blurring of object in the image. The amount of blurring will be calculated with a comparison in front of the camera directly and can be seen with the changes at gray level around the edges of objects.
Enhancing Security and Privacy Issue in Airport by Biometric based Iris Recog...idescitation
Few years ago a self service has been predominant way of passenger at airport.
For the passenger that is a very enjoyable and comfort situation because it keeps control
over all process during their complete journey. For airport and for airlines is also very
interesting evolution because self service allows increasing capacity of airport without any
significant extra investment. However success of self service induces one potential risk. That
is of lack of human contact between airline operator and passenger, there is a problem in
identifying a passenger. This is definitely the problem for immigrations forcibly. This
potential risk of the industry is needed to be addressed and biometrics definitely can solve
this kind of problem. Nowadays biometric is considered to be the most important and
reliable method for personal identification. Iris recognition is considered as most personal
identification.
Region-based image segmentation partitions an image into regions based on pixel properties like homogeneity and spatial proximity. The key region-based methods are thresholding, clustering, region growing, and split-and-merge. Region growing works by aggregating neighboring pixels with similar attributes into regions starting from seed pixels. Split-and-merge first over-segments an image and then refines the segmentation by splitting regions with high variance and merging similar adjacent regions. Region-based segmentation is used for tasks like object recognition, image compression, and medical imaging.
Blind detection of image manipulation @ PoliMiGiorgio Sironi
The document discusses various techniques for the blind detection of image manipulation without the use of digital watermarks. It outlines pixel-based, format-based, camera-based, physics-based, and geometric-based approaches. It focuses on the use of projective geometry tools and geometric-based techniques like analyzing the assumptions of manual text selection and rectification to known fonts or objects to detect tampering. Key steps involve finding keypoints with SIFT feature detection, matching keypoint pairs with RANSAC, and comparing rectified images to reference samples to judge manipulation.
This document provides an overview of the application of remote sensing and geographical information systems in civil engineering. It discusses key concepts such as image interpretation, data preprocessing, feature extraction, image classification, and accuracy assessment. The document aims to explain how remote sensing and GIS techniques can be used to extract useful information from imagery and geospatial data for civil engineering applications.
Goal location prediction based on deep learning using RGB-D camerajournalBEEI
In the navigation system, the desired destination position plays an essential role since the path planning algorithms takes a current location and goal location as inputs as well as the map of the surrounding environment. The generated path from path planning algorithm is used to guide a user to his final destination. This paper presents a proposed algorithm based on RGB-D camera to predict the goal coordinates in 2D occupancy grid map for visually impaired people navigation system. In recent years, deep learning methods have been used in many object detection tasks. So, the object detection method based on convolution neural network method is adopted in the proposed algorithm. The measuring distance between the current position of a sensor and the detected object depends on the depth data that is acquired from RGB-D camera. Both of the object detected coordinates and depth data has been integrated to get an accurate goal location in a 2D map. This proposed algorithm has been tested on various real-time scenarios. The experiments results indicate to the effectiveness of the proposed algorithm.
This document discusses various techniques for image segmentation. It begins by defining image segmentation as dividing an image into constituent regions or objects based on visual characteristics. There are two main categories of segmentation techniques: edge-based techniques which detect discontinuities, and region-based techniques which partition images into regions of similarity. Popular region-based techniques include region growing, region splitting and merging, and watershed transformation. Edge-based techniques detect edges using methods like edge detection. The document provides an overview of these segmentation techniques and their applications in image analysis tasks.
Lec11: Active Contour and Level Set for Medical Image SegmentationUlaş Bağcı
ActiveContour(Snake) • LevelSet
• Applications
Enhancement, Noise Reduction, and Signal Processing • MedicalImageRegistration • MedicalImageSegmentation • MedicalImageVisualization • Machine Learning in Medical Imaging • Shape Modeling/Analysis of Medical Images Deep Learning in Radiology Fuzzy Connectivity (FC) – Affinity functions • Absolute FC • Relative FC (and Iterative Relative FC) • Successful example applications of FC in medical imaging • Segmentation of Airway and Airway Walls using RFC based method Energyfunctional – Data and Smoothness terms • GraphCut – Min cut – Max Flow • ApplicationsinRadiologyImages
The flow of baseline estimation using a single omnidirectional cameraTELKOMNIKA JOURNAL
1. The document describes a method for estimating the baseline of a single omnidirectional camera using optical flow tracking of points on an object.
2. As the camera is moved horizontally, tracking points on an object in panoramic images produces coordinate shifts that are saved and represented as graphs.
3. Analyzing the graphs allows determining the equation that estimates the baseline flow and coefficients of the equation.
The document summarizes a novel approach for multisensor biometric fusion of face and palmprint images using wavelet decomposition and SIFT features for person authentication. Face and palmprint images are decomposed using wavelets and fused to create an enhanced fused image. SIFT features are extracted from the fused image and used for matching based on a monotonic-decreasing graph approach. Experimental results on a 150 person database show the proposed fusion method achieves 98.19% accuracy, outperforming individual face and palmprint recognition.
International journal of signal and image processing issues vol 2015 - no 1...sophiabelthome
This document reviews commonly used calibration patterns for camera calibration and image rectification. It discusses traditional 2D and 3D patterns using points, lines or geometric shapes. Structured light patterns using diffractive optical elements are also presented. Extraction of pattern data is important and can be done through intensity-based subpixel detection or edge detection techniques. Accuracy is evaluated using metrics like root mean square error. Image rectification transforms distorted images into rectilinear images by modeling and removing lens distortion.
Face Pose Classification Method using Image Structural Similarity Indexidescitation
This document proposes a new method for classifying face pose using structural similarity index (SSIM). SSIM is used to measure similarity between a test facial image and images in a database with known poses. The test image is assigned the pose of the database image with the highest SSIM value. Experimental results on the Pointing'04 database show the method can accurately classify poses when many training images are available. Classification confidence decreases when fewer training images are used, as poses may not be directly represented. The method could be useful for applications like driver monitoring that require pose authentication.
Gesture Recognition Review: A Survey of Various Gesture Recognition AlgorithmsIJRES Journal
This document reviews various algorithms for gesture recognition from images and video. It discusses approaches such as pixel-by-pixel comparison, edge detection, orientation histograms, thinning, hidden Markov models, color space segmentation using YUV and tracking using CAMSHIFT, naive Bayes classification, 3D hand modeling, appearance-based modeling using eigenvectors, finite state machines, and particle filtering using condensation. It evaluates these methods and concludes that combining YUV segmentation, CAMSHIFT tracking and hidden Markov modeling provides an effective approach for hand detection and gesture recognition.
This document summarizes a research paper on gesture recognition techniques for controlling mouse events without physically touching a mouse. The paper presents a technique using color detection and tracking of colored caps on fingers. By analyzing the number and positions of color regions in camera frames, various mouse gestures can be recognized, such as left click, right click, drag, etc. An algorithm was implemented in MATLAB using color space conversion from RGB to YCbCr to track hand gestures. Experimental results showed high recognition rates for common mouse events like cursor movement, clicking, and dragging. The technique provides an accessible way for people with disabilities to control computing devices through natural hand gestures.
Las didácticas contemporáneas se refieren a metodologías de enseñanza que promueven la participación activa de estudiantes y profesores (1). Se diferencian de las didácticas tradicionales en que combinan enseñanza y aprendizaje de manera interactiva (2). Su objetivo principal es lograr el desarrollo integral de los estudiantes a nivel cognitivo, práctico y emocional (3).
This document provides a SWOT analysis and product evaluation of McDonald's original cheeseburger. It identifies strengths like being the closest McDonald's for miles and having multiple entrances/exits, but also weaknesses like outdated facilities and loud environment. Opportunities include renovating to seem more modern. Threats are other updated fast food restaurants. The cheeseburger is analyzed as an iconic affordable product, though quality could be improved. Customers are satisfied by the low price point. Packaging and positioning have remained consistent over time.
An Enhanced Computer Vision Based Hand Movement Capturing System with Stereo ...CSCJournals
This framework is a hand movement capturing method which could be done in three different depth levels. The algorithm has the capability of capturing and identifying when the hand is moving up, down, right and left. From these captured movements four signals could be generated. Moreover, when these hand movements are done, 15cm-75cm, 75cm-100cm, 100cm- 200cm from the camera (3 depth levels), twelve different signals could be generated. These generated signals could be used for applications such as game controlling (gaming).The existing method uses an object area based method for depth analysis. The results of the proposed work shows it has high accuracy compared to the existing method when tested for depth analysis.
This document describes a new 3D Areal Image (3D-AI) sensor that can measure the heights of hundreds of micro-scale bumps or other interconnect features simultaneously. The sensor uses phase imaging to measure the height of each feature in its field of view in parallel. Initial tests show it can accurately measure features from a few microns like copper nails to tens of microns like solder bumps used in chip packaging. The sensor provides both 3D metrology for height and coplanarity inspection as well as 2D defect detection in a single tool, offering high throughput needed for production. Measurements of micro bumps, C4 bumps, and copper nails demonstrate its capability and correlation to other metrology tools
This document presents a system for measuring doors and windows using images taken on a mobile device. The system allows a user to take a photo of a door or window, select its corners, and then measure dimensions. It removes the existing object from the image and overlays a 3D model of a replacement. Tests show it can measure with an average error of 0.5% across different viewing angles, distances, and image resolutions. The system allows measuring inaccessible objects and visualizing replacements.
This paper discusses techniques for digital image processing, including noise reduction, edge detection, and histogram equalization. Noise reduction techniques discussed include mean, Gaussian, and median filters to remove salt and pepper noise and Gaussian noise. Edge detection algorithms like Sobel and Laplacian are introduced to reduce image data while preserving object boundaries. Histogram equalization is used for image enhancement by spreading pixel values across the full intensity range for increased contrast. The goal is recognizing objects in images through these preprocessing steps.
EFFECTIVE INTEREST REGION ESTIMATION MODEL TO REPRESENT CORNERS FOR IMAGE sipij
One of the most important steps to describe local features is to estimate the interest region around the feature location to achieve the invariance against different image transformation. The pixels inside the interest region are used to build the descriptor, to represent a feature. Estimating the interest region
around a corner location is a fundamental step to describe the corner feature. But the process is challenging under different image conditions. Most of the corner detectors derive appropriate scales to estimate the region to build descriptors. In our approach, we have proposed a new local maxima-based
interest region detection method. This region estimation method can be used to build descriptors to represent corners. We have performed a comparative analysis to match the feature points using recent corner detectors and the result shows that our method achieves better precision and recall results than
existing methods.
At the end of this lesson, you should be able to;
describe Connected Components and Contours in image segmentation.
discuss region based segmentation method.
discuss Region Growing segmentation technique.
discuss Morphological Watersheds segmentation.
discuss Model Based Segmentation.
discuss Motion Segmentation.
implement connected components, flood fill, watershed, template matching and frame difference techniques.
formulate possible mechanisms to propose segmentation methods to solve problems.
Depth Estimation from Defocused Images: a SurveyIJAAS Team
An important step in 3D data generation is the generation of depth map. Depth map is a black and white image which has exactly the same size of the original captured 2D image that indicates the relative distance of each pixel from the observer to the objects in the real world. This paper presents a survey of Depth Perception from Defocused or blurs images as well as image from motion. The change of distance of the object from the camera has direct relation with the amount of blurring of object in the image. The amount of blurring will be calculated with a comparison in front of the camera directly and can be seen with the changes at gray level around the edges of objects.
Enhancing Security and Privacy Issue in Airport by Biometric based Iris Recog...idescitation
Few years ago a self service has been predominant way of passenger at airport.
For the passenger that is a very enjoyable and comfort situation because it keeps control
over all process during their complete journey. For airport and for airlines is also very
interesting evolution because self service allows increasing capacity of airport without any
significant extra investment. However success of self service induces one potential risk. That
is of lack of human contact between airline operator and passenger, there is a problem in
identifying a passenger. This is definitely the problem for immigrations forcibly. This
potential risk of the industry is needed to be addressed and biometrics definitely can solve
this kind of problem. Nowadays biometric is considered to be the most important and
reliable method for personal identification. Iris recognition is considered as most personal
identification.
Region-based image segmentation partitions an image into regions based on pixel properties like homogeneity and spatial proximity. The key region-based methods are thresholding, clustering, region growing, and split-and-merge. Region growing works by aggregating neighboring pixels with similar attributes into regions starting from seed pixels. Split-and-merge first over-segments an image and then refines the segmentation by splitting regions with high variance and merging similar adjacent regions. Region-based segmentation is used for tasks like object recognition, image compression, and medical imaging.
Blind detection of image manipulation @ PoliMiGiorgio Sironi
The document discusses various techniques for the blind detection of image manipulation without the use of digital watermarks. It outlines pixel-based, format-based, camera-based, physics-based, and geometric-based approaches. It focuses on the use of projective geometry tools and geometric-based techniques like analyzing the assumptions of manual text selection and rectification to known fonts or objects to detect tampering. Key steps involve finding keypoints with SIFT feature detection, matching keypoint pairs with RANSAC, and comparing rectified images to reference samples to judge manipulation.
This document provides an overview of the application of remote sensing and geographical information systems in civil engineering. It discusses key concepts such as image interpretation, data preprocessing, feature extraction, image classification, and accuracy assessment. The document aims to explain how remote sensing and GIS techniques can be used to extract useful information from imagery and geospatial data for civil engineering applications.
Goal location prediction based on deep learning using RGB-D camerajournalBEEI
In the navigation system, the desired destination position plays an essential role since the path planning algorithms takes a current location and goal location as inputs as well as the map of the surrounding environment. The generated path from path planning algorithm is used to guide a user to his final destination. This paper presents a proposed algorithm based on RGB-D camera to predict the goal coordinates in 2D occupancy grid map for visually impaired people navigation system. In recent years, deep learning methods have been used in many object detection tasks. So, the object detection method based on convolution neural network method is adopted in the proposed algorithm. The measuring distance between the current position of a sensor and the detected object depends on the depth data that is acquired from RGB-D camera. Both of the object detected coordinates and depth data has been integrated to get an accurate goal location in a 2D map. This proposed algorithm has been tested on various real-time scenarios. The experiments results indicate to the effectiveness of the proposed algorithm.
This document discusses various techniques for image segmentation. It begins by defining image segmentation as dividing an image into constituent regions or objects based on visual characteristics. There are two main categories of segmentation techniques: edge-based techniques which detect discontinuities, and region-based techniques which partition images into regions of similarity. Popular region-based techniques include region growing, region splitting and merging, and watershed transformation. Edge-based techniques detect edges using methods like edge detection. The document provides an overview of these segmentation techniques and their applications in image analysis tasks.
Lec11: Active Contour and Level Set for Medical Image SegmentationUlaş Bağcı
ActiveContour(Snake) • LevelSet
• Applications
Enhancement, Noise Reduction, and Signal Processing • MedicalImageRegistration • MedicalImageSegmentation • MedicalImageVisualization • Machine Learning in Medical Imaging • Shape Modeling/Analysis of Medical Images Deep Learning in Radiology Fuzzy Connectivity (FC) – Affinity functions • Absolute FC • Relative FC (and Iterative Relative FC) • Successful example applications of FC in medical imaging • Segmentation of Airway and Airway Walls using RFC based method Energyfunctional – Data and Smoothness terms • GraphCut – Min cut – Max Flow • ApplicationsinRadiologyImages
The flow of baseline estimation using a single omnidirectional cameraTELKOMNIKA JOURNAL
1. The document describes a method for estimating the baseline of a single omnidirectional camera using optical flow tracking of points on an object.
2. As the camera is moved horizontally, tracking points on an object in panoramic images produces coordinate shifts that are saved and represented as graphs.
3. Analyzing the graphs allows determining the equation that estimates the baseline flow and coefficients of the equation.
The document summarizes a novel approach for multisensor biometric fusion of face and palmprint images using wavelet decomposition and SIFT features for person authentication. Face and palmprint images are decomposed using wavelets and fused to create an enhanced fused image. SIFT features are extracted from the fused image and used for matching based on a monotonic-decreasing graph approach. Experimental results on a 150 person database show the proposed fusion method achieves 98.19% accuracy, outperforming individual face and palmprint recognition.
International journal of signal and image processing issues vol 2015 - no 1...sophiabelthome
This document reviews commonly used calibration patterns for camera calibration and image rectification. It discusses traditional 2D and 3D patterns using points, lines or geometric shapes. Structured light patterns using diffractive optical elements are also presented. Extraction of pattern data is important and can be done through intensity-based subpixel detection or edge detection techniques. Accuracy is evaluated using metrics like root mean square error. Image rectification transforms distorted images into rectilinear images by modeling and removing lens distortion.
Face Pose Classification Method using Image Structural Similarity Indexidescitation
This document proposes a new method for classifying face pose using structural similarity index (SSIM). SSIM is used to measure similarity between a test facial image and images in a database with known poses. The test image is assigned the pose of the database image with the highest SSIM value. Experimental results on the Pointing'04 database show the method can accurately classify poses when many training images are available. Classification confidence decreases when fewer training images are used, as poses may not be directly represented. The method could be useful for applications like driver monitoring that require pose authentication.
Gesture Recognition Review: A Survey of Various Gesture Recognition AlgorithmsIJRES Journal
This document reviews various algorithms for gesture recognition from images and video. It discusses approaches such as pixel-by-pixel comparison, edge detection, orientation histograms, thinning, hidden Markov models, color space segmentation using YUV and tracking using CAMSHIFT, naive Bayes classification, 3D hand modeling, appearance-based modeling using eigenvectors, finite state machines, and particle filtering using condensation. It evaluates these methods and concludes that combining YUV segmentation, CAMSHIFT tracking and hidden Markov modeling provides an effective approach for hand detection and gesture recognition.
This document summarizes a research paper on gesture recognition techniques for controlling mouse events without physically touching a mouse. The paper presents a technique using color detection and tracking of colored caps on fingers. By analyzing the number and positions of color regions in camera frames, various mouse gestures can be recognized, such as left click, right click, drag, etc. An algorithm was implemented in MATLAB using color space conversion from RGB to YCbCr to track hand gestures. Experimental results showed high recognition rates for common mouse events like cursor movement, clicking, and dragging. The technique provides an accessible way for people with disabilities to control computing devices through natural hand gestures.
Las didácticas contemporáneas se refieren a metodologías de enseñanza que promueven la participación activa de estudiantes y profesores (1). Se diferencian de las didácticas tradicionales en que combinan enseñanza y aprendizaje de manera interactiva (2). Su objetivo principal es lograr el desarrollo integral de los estudiantes a nivel cognitivo, práctico y emocional (3).
This document provides a SWOT analysis and product evaluation of McDonald's original cheeseburger. It identifies strengths like being the closest McDonald's for miles and having multiple entrances/exits, but also weaknesses like outdated facilities and loud environment. Opportunities include renovating to seem more modern. Threats are other updated fast food restaurants. The cheeseburger is analyzed as an iconic affordable product, though quality could be improved. Customers are satisfied by the low price point. Packaging and positioning have remained consistent over time.
The skeletal system has several major functions including support, movement, protection of organs, storage of calcium, and regulation. Bones are connected by ligaments at joints which allow different ranges of motion like hinge, pivot, and ball and socket. The skeletal system works with other systems like the muscular system which contracts muscles to move bones. Diseases can affect the bones like scoliosis causing spine curvature and osteoporosis making bones fragile.
Photography session with Elissa Viramontes. Wanderlust: a strong desire to travel. Lost in a dream this model finds herself in a desert state where she is left alone only with the strong desire to escape her surrounding and come into a place where she is free.
SlideShare es un sitio web que permite subir y compartir presentaciones de PowerPoint, documentos PDF y archivos de Word de manera pública o privada. Permite administrar una cuenta para archivar y publicar presentaciones de manera fácil para compartir con colegas desde cualquier computadora a través de una página web, aunque no admite animaciones, sonido o imágenes interactivas.
Fareed Ebrahim is a South African national currently working as a Private Banking Suite Head at Standard Bank. He has over 20 years of experience in various roles at Standard Bank including Branch Manager, Private Banker, and positions in branch operations and customer service. He holds qualifications including a matric certificate, SAIM Programme in Business Management, and B.Com Accounting. His duties have included ensuring excellent customer service, compliance with relevant laws, health and safety, risk management, and people management responsibilities as a Branch Manager.
The document is a catalogue for the 2012 Graduating Exhibition at Langara College's Fine Arts Department. It features 36 artworks from the graduating class of 2012, including paintings, prints, sculptures, and ceramics. The chair of the Fine Arts department writes in the introduction that the students have spent years developing their skills and creating thoughtful and challenging artworks. The exhibition and catalogue showcase the achievements of the graduating class as they move forward in their artistic careers.
Este proyecto pedagógico propone prevenir el maltrato infantil en los estudiantes de 4o grado del Colegio Berlín a través del uso de las TIC. El proyecto involucra a los estudiantes, padres y docentes, y consiste en charlas, videos y foros para crear conciencia sobre este problema. El objetivo general es promover los derechos de los niños para eliminar todas las formas de maltrato, y los objetivos específicos incluyen determinar las causas del maltrato en la comunidad, diseñar estrateg
Este proyecto pedagógico de dos meses utiliza las TIC para que los estudiantes investiguen diferentes hábitats y aprendan sobre las plantas y animales que viven en ellos. Los objetivos son desarrollar habilidades con las TIC y usar el minicomputador para obtener y analizar información sobre hábitats. Las actividades incluyen visitas de campo y el uso de páginas web para responder la pregunta "Si un camello viviera en un bosque tropical ¿conservaría la joroba?".
This document is a listing of the name "Antonis Silva Digital Design" repeated multiple times with no other context or information provided. It does not contain enough meaningful details to generate an informative summary.
Dr. Cassandra Wiley has hired private investigator Ian Kane to look into the disappearance of her friend and colleague Dr. Gail Honeywell, whose death was ruled a suicide by police but whose body was never found. Cassandra meets with Kane at Morgan's Bar in a run-down district of New Orleans to discuss the case. Kane agrees to take the case and find out what really happened to Gail, warning Cassandra that the truth may be ugly.
Enhanced Algorithm for Obstacle Detection and Avoidance Using a Hybrid of Pla...IOSR Journals
This document presents an enhanced algorithm for obstacle detection and avoidance using a hybrid of plane-to-plane homography, image segmentation, corner detection, and edge detection techniques. The algorithm aims to improve upon previous methods by eliminating false positives, reducing unreliable corners and broken edges, providing depth perception without planar assumptions, and requiring less processing power. The key components of the algorithm include plane-to-plane homography, image segmentation, Canny edge detection, Harris corner detection, and the RANSAC sampling method for system analysis. Test results on sample images show the algorithm can accurately detect obstacles based on texture differences while reducing noise from ground plane textures.
Matching algorithm performance analysis for autocalibration method of stereo ...TELKOMNIKA JOURNAL
Stereo vision is one of the interesting research topics in the computer vision field. Two cameras are used to generate a disparity map, resulting in the depth estimation. Camera calibration is the most important step in stereo vision. The calibration step is used to generate an intrinsic parameter of each camera to get a better disparity map. In general, the calibration process is done manually by using a chessboard pattern, but this process is an exhausting task. Self-calibration is an important ability required to overcome this problem. Self-calibration required a robust and good matching algorithm to find the key feature between images as reference. The purpose of this paper is to analyze the performance of three matching algorithms for the autocalibration process. The matching algorithms used in this research are SIFT, SURF, and ORB. The result shows that SIFT performs better than other methods.
This document describes a laser distance measurement system using a webcam. It consists of a laser transmitter and webcam receiver. The laser pulse is reflected off an object and received by the webcam. Software calculates the distance based on the time of flight. The system achieves high accuracy of ±3cm. It calibrates the system using test measurements to determine the relationship between pixel location of the laser dot and actual distance. This allows accurate distance measurements within a few percent of error out to over 2 meters. Potential improvements discussed are using a laser line instead of dot for more data points and a green laser for better visibility.
This document presents a methodology for real-time object tracking using a webcam. It combines Prewitt edge detection for object detection and Kalman filtering for tracking. Prewitt edge detection is used to detect the edges of the moving object in each video frame. Then, Kalman filtering is used to track the detected object across subsequent frames by predicting its location. Experiments show the approach can efficiently track objects under deformation, occlusion, and can track multiple objects simultaneously. The combination of Prewitt edge detection and Kalman filtering provides an effective method for real-time object tracking.
Leader Follower Formation Control of Ground Vehicles Using Dynamic Pixel Coun...ijma
This paper deals with leader-follower formations of non-holonomic mobile robots, introducing a formation
control strategy based on pixel counts using a commercial grade electro optics camera. Localization of the
leader for motions along line of sight as well as the obliquely inclined directions are considered based on
pixel variation of the images by referencing to two arbitrarily designated positions in the image frames.
Based on an established relationship between the displacement of the camera movement along the viewing
direction and the difference in pixel counts between reference points in the images, the range and the angle
estimate between the follower camera and the leader is calculated. The Inverse Perspective Transform is
used to account for non linear relationship between the height of vehicle in a forward facing image and its
distance from the camera. The formulation is validated with experiments.
Nowadays crowd analysis, essential factor about decision management of brand strategy, is not a controllable field by individuals. Therefore a technology, software products is needed. In this paper we focused on what we have done about crowd analysis and examination the problem of human detection with fish-eye lenses cameras. In order to identify human density, one of the machine learning algorithm, which is Haar Classification algorithm, is used to distinguish human body under different conditions. First, motion analysis is used to search for meaningful data, and then the desired object is detected by the trained classifier. Significant data has been sent to the end user via socket programming and human density analysis is presented.
Nowadays crowd analysis, essential factor about decision management of brand strategy, is not a
controllable field by individuals. Therefore a technology, software products is needed. In this paper we
focused on what we have done about crowd analysis and examination the problem of human detection with
fish-eye lenses cameras. In order to identify human density, one of the machine learning algorithm, which
is Haar Classification algorithm, is used to distinguish human body under different conditions. First,
motion analysis is used to search for meaningful data, and then the desired object is detected by the trained
classifier. Significant data has been sent to the end user via socket programming and human density
analysis is presented.
A NOVEL APPROACH TO SMOOTHING ON 3D STRUCTURED ADAPTIVE MESH OF THE KINECT-BA...csandit
3-dimensional object modelling of real world objects in steady state by means of multiple point
cloud (pcl) depth scans taken by using sensing camera and application of smoothing algorithm
are suggested in this study. Polygon structure, which is constituted by coordinates of point
cloud (x,y,z) corresponding to the position of 3D model in space and obtained by nodal points
and connection of these points by means of triangulation, is utilized for the demonstration of 3D
models. Gaussian smoothing and developed methods are applied to the mesh consisting of
merge of these polygons, and a new mesh simplification and augmentation algorithm are
suggested for the over the 3D modelling. Mesh consisting of merge of polygons can be
demonstrated in a more packed, smooth and fluent way. In this study is shown that applied the
triangulation and smoothing method for 3D modelling, perform to a fast and robust mesh
structures compared to existing methods therewithal no remeshing is necessary for refinement
and reduction.
An Accurate Scheme for Distance Measurement using an Ordinary Webcam IJECEIAES
Nowadays, image processing has become one of the widely used computer aided science. Two major branches of this scientific field are image enhancement and machine vision. Machine vision has many applications and demands in robotic and defense industries. Detecting distance of objects is one of the extensive research in the defense industry and robotic industries that a lot of annual projects have been involved in this issue both inside and outside the country. So, in this paper, an accurate algorithm is presented for measuring the distance of the objects from a camera. In this method, a laser transmitter is used alongside a regular webcam. The laser light is transmitted to the desired object and then the distance of the object is calculated using image processing methods and mathematical and geometric relations. The performance of the proposed algorithm was evaluated using MATLAB software. The accuracy rate of distance detection is up to 99.62%. The results also has shown that the presented algorithms make the obstacle distance measurement more reliable. Finally, the performance of the proposed algorithm was compared with other methods from different literatures.
An Accurate Scheme for Distance Measurement using an Ordinary Webcam Yayah Zakaria
Nowadays, image processing has become one of the widely used computer aided science. Two major branches of this scientific field are image enhancement and machine vision. Machine vision has many applications and demands in robotic and defense industries. Detecting distance of objects is
one of the extensive research in the defense industry and robotic industries that a lot of annual projects have been involved in this issue both inside and outside the country. So, in this paper, an accurate algorithm is presented for measuring the distance of the objects from a camera. In this method, a laser
transmitter is used alongside a regular webcam. The laser light is transmitted to the desired object and then the distance of the object is calculated using image processing methods and mathematical and geometric relations. The performance of the proposed algorithm was evaluated using MATLAB software. The accuracy rate of distance detection is up to 99.62%. The results
also has shown that the presented algorithms make the obstacle distance measurement more reliable. Finally, the performance of the proposed algorithm was compared with other methods from different literatures.
Robotic navigation algorithm with machine vision IJECEIAES
In the field of robotics, it is essential to know the work area in which the agent is going to develop, for that reason, different methods of mapping and spatial location have been developed for different applications. In this article, a machine vision algorithm is proposed, which is responsible for identifying objects of interest within a work area and determining the polar coordinates to which they are related to the observer, applicable either with a fixed camera or in a mobile agent such as the one presented in this document. The developed algorithm was evaluated in two situations, determining the position of six objects in total around the mobile agent. These results were compared with the real position of each of the objects, reaching a high level of accuracy with an average error of 1.3271% in the distance and 2.8998% in the angle.
A NOVEL APPROACH TO SMOOTHING ON 3D STRUCTURED ADAPTIVE MESH OF THE KINECT-BA...cscpconf
3-dimensional object modelling of real world objects in steady state by means of multiple point cloud (pcl) depth scans taken by using sensing camera and application of smoothing algorithm
are suggested in this study. Polygon structure, which is constituted by coordinates of point cloud (x,y,z) corresponding to the position of 3D model in space and obtained by nodal points and connection of these points by means of triangulation, is utilized for the demonstration of 3D models. Gaussian smoothing and developed methods are applied to the mesh consisting of merge of these polygons, and a new mesh simplification and augmentation algorithm are suggested for the over the 3D modelling. Mesh consisting of merge of polygons can be demonstrated in a more packed, smooth and fluent way. In this study is shown that applied the triangulation and smoothing method for 3D modelling, perform to a fast and robust mesh structures compared to existing methods therewithal no remeshing is necessary for refinement and reduction.
This document proposes and evaluates several deep learning models for unsupervised monocular depth estimation. It begins with background on depth estimation methods and a literature review of recent work. Four depth estimation architectures are then described: EfficientNet-B7, EfficientNet-B3, DenseNet121, and DenseNet161. These models use an encoder-decoder structure with skip connections. An unsupervised loss function is adopted that combines appearance matching, disparity smoothness, and left-right consistency losses. The models are trained on the KITTI dataset and evaluated using standard KITTI metrics, showing improved performance over baseline methods using less training data and lower input resolution.
Fuzzy-proportional-integral-derivative-based controller for object tracking i...IJECEIAES
This paper aims at designing and implementing an intelligent controller for the orientation control of a two-wheeled mobile robot. The controller is designed in LabVIEW and based on analyzed image parameters from cameras. The image program calculates the distance and angle from the camera to the object. The fuzzy controller will get these parameters as crisp input data and send the calculated velocity as crisp output data to the right and left wheel motor for the robot tracking the target object. The results show that the controller gives a fast response and high reliability and quickly carries out data recovery from system faults. The system also works well in the uncertainties of process variables and without mathematical modeling.
This document describes a system for visualizing deviations on arbitrary surfaces for quality assurance and process control. The system combines 3D measurement devices like terrestrial laser scanners with a projector. This allows detected deviations between a reference model (like CAD data) and measured data to be projected directly onto the measured object. The key steps are determining the relative orientation of the measurement device and projector, computing an inspection map of deviations, and projecting this map. A novel approach called "virtual targets" is presented for establishing the relative orientation without using physical targets when laser scanning. Sample applications for quality inspection and object alignment are demonstrated.
AUTOMATED MANAGEMENT OF POTHOLE RELATED DISASTERS USING IMAGE PROCESSING AND ...ijcsit
Potholes though seem inconsequential, may cause accidents resulting in loss of human life. In this paper, we present an automated system to efficiently manage the potholes in a ward by deploying geotagging and image processing techniques that overcomes the drawbacks associated with the existing
survey-oriented systems. Image processing is used for identification of target pothole regions in the 2D
images using edge detection and morphological image processing operations. A method is developed to
accurately estimate the dimensions of the potholes from their images, analyze their area and depth,estimate the quantity of filling material required and therefore enabling pothole attendance on a priority basis. This will further enable the government official to have a fully automated system for e f f e c t i v e l y ma n a g i ng pothole related disasters.
This document summarizes an analysis of iris recognition based on false acceptance rate (FAR) and false rejection rate (FRR) using the Hough transform. It first provides an overview of iris recognition and its typical stages: image acquisition, localization/segmentation, normalization, feature extraction, and pattern matching. It then describes existing methods used in each stage, including the Hough transform and rubber sheet model for localization and normalization. The proposed methodology applies Canny edge detection, Hough transform for boundary detection, normalization with the rubber sheet model, and calculates metrics like mean squared error, root mean squared error, signal-to-noise ratio, and root signal-to-noise ratio to evaluate the accuracy of iris recognition using FAR
This document discusses a system for capturing both the shape and material properties of physical objects, especially those that exhibit subsurface scattering effects. The system uses coded light patterns projected from multiple inexpensive projectors and captured with digital cameras. Preliminary results show this approach can estimate both shape and subsurface scattering properties by separating direct and indirect light paths based on the projected patterns. The goal is to create an inexpensive capture system that produces models accurate enough for computer graphics rendering.
Stereo matching based on absolute differences for multiple objects detectionTELKOMNIKA JOURNAL
This article presents a new algorithm for object detection using stereo camera system. The problem to get an accurate object detion using stereo camera is the imprecise of matching process between two scenes with the same viewpoint. Hence, this article aims to reduce the incorrect matching pixel with four stages. This new algorithm is the combination of continuous process of matching cost computation, aggregation, optimization and filtering. The first stage is matching cost computation to acquire preliminary result using an absolute differences method. Then the second stage known as aggregation step uses a guided filter with fixed window support size. After that, the optimization stage uses winner-takes-all (WTA) approach which selects the smallest matching differences value and normalized it to the disparity level. The last stage in the framework uses a bilateral filter. It is effectively further decrease the error on the disparity map which contains information of object detection and locations. The proposed work produces low errors (i.e., 12.11% and 14.01% nonocc and all errors) based on the KITTI dataset and capable to perform much better compared with before the proposed framework and competitive with some newly available methods.
Image fusion using nsct denoising and target extraction for visual surveillanceeSAT Publishing House
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
Similar to Distance Estimation to Image Objects Using Adapted Scale (20)
A review on techniques and modelling methodologies used for checking electrom...nooriasukmaningtyas
The proper function of the integrated circuit (IC) in an inhibiting electromagnetic environment has always been a serious concern throughout the decades of revolution in the world of electronics, from disjunct devices to today’s integrated circuit technology, where billions of transistors are combined on a single chip. The automotive industry and smart vehicles in particular, are confronting design issues such as being prone to electromagnetic interference (EMI). Electronic control devices calculate incorrect outputs because of EMI and sensors give misleading values which can prove fatal in case of automotives. In this paper, the authors have non exhaustively tried to review research work concerned with the investigation of EMI in ICs and prediction of this EMI using various modelling methodologies and measurement setups.
Electric vehicle and photovoltaic advanced roles in enhancing the financial p...IJECEIAES
Climate change's impact on the planet forced the United Nations and governments to promote green energies and electric transportation. The deployments of photovoltaic (PV) and electric vehicle (EV) systems gained stronger momentum due to their numerous advantages over fossil fuel types. The advantages go beyond sustainability to reach financial support and stability. The work in this paper introduces the hybrid system between PV and EV to support industrial and commercial plants. This paper covers the theoretical framework of the proposed hybrid system including the required equation to complete the cost analysis when PV and EV are present. In addition, the proposed design diagram which sets the priorities and requirements of the system is presented. The proposed approach allows setup to advance their power stability, especially during power outages. The presented information supports researchers and plant owners to complete the necessary analysis while promoting the deployment of clean energy. The result of a case study that represents a dairy milk farmer supports the theoretical works and highlights its advanced benefits to existing plants. The short return on investment of the proposed approach supports the paper's novelty approach for the sustainable electrical system. In addition, the proposed system allows for an isolated power setup without the need for a transmission line which enhances the safety of the electrical network
Literature Review Basics and Understanding Reference Management.pptxDr Ramhari Poudyal
Three-day training on academic research focuses on analytical tools at United Technical College, supported by the University Grant Commission, Nepal. 24-26 May 2024
Using recycled concrete aggregates (RCA) for pavements is crucial to achieving sustainability. Implementing RCA for new pavement can minimize carbon footprint, conserve natural resources, reduce harmful emissions, and lower life cycle costs. Compared to natural aggregate (NA), RCA pavement has fewer comprehensive studies and sustainability assessments.
Comparative analysis between traditional aquaponics and reconstructed aquapon...bijceesjournal
The aquaponic system of planting is a method that does not require soil usage. It is a method that only needs water, fish, lava rocks (a substitute for soil), and plants. Aquaponic systems are sustainable and environmentally friendly. Its use not only helps to plant in small spaces but also helps reduce artificial chemical use and minimizes excess water use, as aquaponics consumes 90% less water than soil-based gardening. The study applied a descriptive and experimental design to assess and compare conventional and reconstructed aquaponic methods for reproducing tomatoes. The researchers created an observation checklist to determine the significant factors of the study. The study aims to determine the significant difference between traditional aquaponics and reconstructed aquaponics systems propagating tomatoes in terms of height, weight, girth, and number of fruits. The reconstructed aquaponics system’s higher growth yield results in a much more nourished crop than the traditional aquaponics system. It is superior in its number of fruits, height, weight, and girth measurement. Moreover, the reconstructed aquaponics system is proven to eliminate all the hindrances present in the traditional aquaponics system, which are overcrowding of fish, algae growth, pest problems, contaminated water, and dead fish.
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Sinan KOZAK
Sinan from the Delivery Hero mobile infrastructure engineering team shares a deep dive into performance acceleration with Gradle build cache optimizations. Sinan shares their journey into solving complex build-cache problems that affect Gradle builds. By understanding the challenges and solutions found in our journey, we aim to demonstrate the possibilities for faster builds. The case study reveals how overlapping outputs and cache misconfigurations led to significant increases in build times, especially as the project scaled up with numerous modules using Paparazzi tests. The journey from diagnosing to defeating cache issues offers invaluable lessons on maintaining cache integrity without sacrificing functionality.
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...University of Maribor
Slides from talk presenting:
Aleš Zamuda: Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapter and Networking.
Presentation at IcETRAN 2024 session:
"Inter-Society Networking Panel GRSS/MTT-S/CIS
Panel Session: Promoting Connection and Cooperation"
IEEE Slovenia GRSS
IEEE Serbia and Montenegro MTT-S
IEEE Slovenia CIS
11TH INTERNATIONAL CONFERENCE ON ELECTRICAL, ELECTRONIC AND COMPUTING ENGINEERING
3-6 June 2024, Niš, Serbia
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Distance Estimation to Image Objects Using Adapted Scale
1. The International Journal Of Engineering And Science (IJES)
|| Volume || 6 || Issue || 1 || Pages || PP 39-50|| 2017 ||
ISSN (e): 2319 – 1813 ISSN (p): 2319 – 1805
www.theijes.com The IJES Page 39
Distance Estimation to Image Objects Using Adapted Scale
M. Zuckerman(1)
Intel Corp., Israel
E. Kolberg(2)*
Bar-Ilan University, Israel
--------------------------------------------------------ABSTRACT--------------------------------------------------
Distance measurement is part of various robotic applications. There exist many methods for this purpose. In this
paper, we introduce a new method to measure the distance from a digital camera to an arbitrary object by using
its pose (X,Y pixel coordination and the angel of the camera). The method uses a pre-data that stores all the
information about the relation between the pose and the distance of an object to the camera. This process
designed for a robot that is a part of a robotic team participating in RoboCup KSL competition.
---------------------------------------------------------------------------------------------------------------------------------------
Date of Submission: 12 January 2017 Date of Accepted: 05 February 2017
--------------------------------------------------------------------------------------------------------------------------------------
I. INTRODUCTION
Distance measurement from a camera to an arbitrary object is widely needed. We use the fact that an object
looks larger as it gets closer to the camera. That means that the same object when close to the camera will have
more pixels representing it than if it was located further. It allowed us to develop a robust mapping method for
measuring a distance to an object.
Among the common methods used to yield distance measurement from image processing we find Pinhole cam-
era model [1, 2], Stereo vision [3, 4] and object volume [5, 6]. In [2] distance measurements are based on foot-
print size. In [1] the distance to an object is derived using the laws of perspective and depends on the road geo-
metry and the point of contact between the vehicle and the road. This method is used in order to estimate a dis-
tance to a car or a truck that are an order of magnitude bigger than the objects we deal with. Measuring a
distance of 90 meters, the error is 10%. The object size and the dependency of the road geometry using a pin-
hole camera is therefore less adequate in our case. In [2] the method is again deals with vehicles and focus on
the detection of preceding vehicles. It takes an area in an image bounded by lane marks, (taking advantage that
lane marks are at 45 when they appear in the image), and execute a logical-AND operation with the binary road
image with no lane marks. It also takes advantage of the non symmetrical shape of a vehicle in order to verify
the vehicle footprint. These two last methods while adequate for vehicle identification are less useful for the
case described in this paper. In [3] the distance measurement is based on stereoscopic pictures. It depends on
horizontally aligned cameras, taking pictures at the same time, parallel axis of the two cameras. As the distance
is longer, the error grows in proportion to the squared distance. In the case we describe here, the robot moves so
a static infrastructure is not relevant here. In addition, we need to reduce the error even in long distances. In [4]
there is a comparison between camera and human depth accuracy. They concluded that the focal length affected
the depth resolution remarkably. Wide-angle lens deteriorated the focal resolution as well. The calibration errors
added to the overall error. In [5] the authors describe a method for measure distance to thin and high pole like
objects. Each pole needs three alignment measurements for each placement. While this is fine for a pole posi-
tioned in a single static position, it is less adequate for our case. In [6] the method for distance calculation has
two steps. First, calculating an interpolation function based on the height and the horizontal angle of the camera.
Second, using this function to calculate the distance of the object from the camera. This method relies upon the
camera height and horizontal angle. The paper shows results of distances up to 1m. It is based on interpolation,
which in our case might increase the error. Using laser like the one presented in [7] integrated with a camera can
increase accuracy on a static system as described there. Besides the fact that it is forbidden for use in the compe-
tition, it will require substantial design and additional hardware and software for using it on a walking robot. In
[8] plate size of car number is used to estimate distance to a car. While this method might be adequate for
distance from a car, it proved to be less useful in our case. The method described in [9], deals with a flat board
of size 750mm x 750 mm, with target points on it. This target is too big for our case. The focal length and cam-
era's lens diameter along with object base size were used to derive the distance to an object. The limitations of
this method include relevancy for exact camera model in combination with particular lens, and target should be
in optical axes.
In this paper we will present the volume solution for deriving a distance to an object. The Mapping method is
based on the idea of the volume solution.
2. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 40
II. VOLUME SOLUTION
When dealing with known objects, it will be useful to measure the object size with different distances, as it ap-
pears on the image. As close the object to the camera as bigger it appears. The object, which appears on an
image, is called reflected object. This information needs to be recorded and saved as a function [5] [6] . We took
measurements of distance versus radius of a ball with a radius of 5 cm. We shot images out of Logitech C905
camera with 1280x720 pixels. Figure 1 presents a graph of the measurements we took. The graph looks similar
to the one presented by Sant’Ann et. al [7] . In our example, the software program identified the ball. Since the
ball was a known object, the program could use pre-stored data about ball diameter, ball color, and ball center.
While these parameters are known it is possible to use a scale tool that measures the different radius sizes as
appear on the image.
Figure 1: Distance Vs Radios
From figure 1 we see that the reflected object distance is an exponential function of the number of its radius
pixels. This means that when the ball is far from the camera from a certain distance, it will have relatively less
changes in number of pixels. On the other hand, the sensitivity will increase as the object is getting close to the
camera, since the reflected object radius will have substantially less pixels in comparison to a closer distance
when the ball is close to the camera.
From figure 1, we can see that there is a correlation between the number of pixels of the object radius in the
image and the distance of the object from the camera in the real world.
In this particular graph, the correlation is presented by the function 𝑦 = 54181𝑥−1.164
, when y stands for the
distance and x stands for the number of pixels. Next step is to determine the accuracy of the correlation. It will
be presented by the error between the real distance and the computed one.
III. VOLUME SOLUTION ACCURACY
The function accuracy depends on the following factors.
3.1 Camera's angle aimed to an object
Objects in image might look different from various angles and different lighting conditions. One common me-
thod of identifying objects is first saving an image as a binary picture and then count the white pixels of an ob-
ject like the radius of a ball or the width of a goal in a soccer game. The accuracy of the function will depend on
the way the object is presented. Figure 2 presents the error ratings obtained when we tested the distance from
the camera to a 10cm diameter orange ball on a green field.
Table 1 presents the average distance error of short and long distance areas in the graph.
Figure 2: Error as function of distance orange
y = 54181x-1.16
R² = 0.987
0
500
1000
0 200 400 600 800 1000
Distancecm
Radios pix
Distance vs Radios pixcels from 0
– 800 cm 1280X720
23%
0%
7%7%
11%
7%
2%1%2%6%
9%
5%4%4%5%4%6%4%4%4%7%4%5%6%6%5%5%4%4%
0%
5%
10%
15%
20%
25%
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
ERROR%
DISTANCE CM
Error as function of distance
3. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 41
Table 1: Average error values of distance from a camera to a 10 cm dia. ball, calculated from B/W image
The values presented in table 1 are reasonable for our purpose of a soccer game since they are close enough to
the real world. When we tried the same technique on the identification of the goal, it was not as robust. One rea-
son to this phenomenon is that the ball looks round from almost any angle and light conditions. However, the
goal might look different from different angles. The gap between the goal's poles will look different when the
camera looks at the goal from different angles. The shape of the goal in these cases might look as a part of a
parallelogram, or a triangle or a trapezoid, etc.
It leads to a conclusion that volume solution by counting white pixels is valid only for a perfect rounded shape
or a simple shape like an equilateral triangle. In case of different objects, the error might increase substantially.
3.2 Light conditions
Different light conditions might change objects' edges as appears in the reflected image. With brighter illumina-
tion, white objects like white lines or white poles may look larger in the reflected image than they are in reality.
Shadows or hiding a part of an object might also considered as different light conditions.
3.2 Filter type
The type of filter will have also an impact on the accuracy of the distance. In some cases, HSV will be used in
others, RGB, etc. The filter types of edge detection will also have an impact on the accuracy.
3.3. Approximation function. Tinnachote and Pimprasen [5], suggest using a polynomial approximation func-
tion in order to calculate the distance, in a case of exponential function. We found that while it improves the
results, it still was not proper to our purposes. We came out with the idea of further dividing the graph into
smaller sections, which makes it more delicate for better approximation. In this specific example, we divided the
graph into three sections. In each section, we used a different type of approximation function:
1. For ball distance of 0 to 55 cm – linear function as presented in figure 3.
2. For ball distance of 60 to 150 cm – 2nd
order polynomial function as presented in figure 4
3. For ball distance of 160 cm and above – exponential function as presented in figure 5
The resulted distance error is presented in figure 6. The error decreased. The average distance error of the di-
vided function was 3% compared to 6% average distance error presented in figure 1.
Figure 3: linear function of section 1
Figure 4: 2nd order polynomial function of section 2
Distance Error values
Short distance 0-55 cm 7%
Long distance 60-150 cm 5%
4. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 42
Figure 5: exponential function of section 3
Figure6: Error as function of distance: average error of 3%
3.4 Color/filter calibration
Before analyzing an image, it is translated from a colored image into a B/W image. One of the tools used for this
purpose is HSV/RGB threshold. With a proper threshold, it is possible to identify object types. Notwithstanding,
it is somewhat problematic. For instance, if we allow for a low threshold, the object will be seen larger com-
pared to the real object. Increasing or decreasing the value of the HSV threshold, will change the object's area.
IV. NEW METHOD'S IDEA
Since distances to objects are important values as inputs to a localization system in general and for RoboCup
soccer game in particular, it was required to develop a reliable distance measuring method that is independent as
much as possible of the factors described above.
After consideration of various options like using more expensive cameras, developing more complex algorithm,
etc., we decided upon an algorithm that take as an input the location (x,y) of the object in 2-D Cartesian coordi-
nate system. The (x,y) coordinates is the only needed input to the algorithm. In addition, the algorithm requires
building in advance, a lookup table that transfers these coordinates into distances. Using such a lookup table
makes it unnecessary to use an approximation function or count white pixels. Actually, the algorithm is inde-
pendent of the binary picture. The only input needed is the (X, Y) coordinates of the object.
In order to implement the lookup it is necessary to find a correlation between any point coordinates (X, Y) and
the distance to the point. For this purpose, we chose to use a scale.
After creating the scale, it is easy to build the lookup table. Then it is straightforward to implement the algo-
rithm that deduce the distance to a point with its Cartesian coordinates. Next we will describe the two scales we
built during the development process.
V. SCALE VERSION 1.0 (Y DEPENDENT )
The scale is made of black and white stripes. We developed a specific software in order to identify the black and
white scale. We added numbers on the scale, which correlates to the distance between the camera and the object.
8765431 2 1211109
10 [cm]
0 - 120 [cm]
Figure 7: Scale 1.0 with is 10[cm] stripes
5. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 43
VI. THE LOOKUP TABLE
The lookup table saves the output information coming from the software in the preparation test mode. The Algo-
rithm of creating to the lookup table appears in table 2.
First, variables and constants are initialized (Lines 1-10). The camera is placed in the middle of first stripe.
Then, the algorithm count the current stripe pixels' number (Lines 12-16). Each stripes pixels count is saved
(Line 18). In addition the algorithm takes care of saving the accumulated pixel counts (Line 19). J presents the
stripe number. Since the scale reflected image looks like a trapezoid, the middle stripe's x value changes as y
value advances to next stripe. Figure 8 presents the geometric scheme of this state.
Here α is the left side slope and
2
.
y
y
x
1st
stripe
2nd
stripe
( )y tg
First stripe’s
edge center
Second stripe’s
edge center
Third stripe’s
edge center
Figure 8: x values change
Table 2: Creation of lookup table algorithm
1 : _ ( , ) :
2 :
3 :
4 :
5 :
6 : ' 0
7 : '
8 : ( , )
A lg o r ith m L o o k u p T a b le N T
p la c e th e c a m e r a in fr o n t o f m id d le s c a l e 's fir s t s tr ip e
y = 0
c o u n t = 0
j = 0
x o f s tr ip e s lo w e r le ft c o r n e r
x x v a lu e o f fir s t s tr ip e s c e n te r
r e a d c o lo r x y
1 ,
2 ,
3 , 2 , 3 , 1
9 : _ ( , )
1 0 : s e c 0
1 1 :
1 2 : s e c
1 3 :
1 4 : ( , )
1 5 : ( ) _
1 6 :
1 7 :
1 8 :
1 9 :
2 0 :
2 1 : s e c s e c 1 0
2 2 : 0
2 3 :
2 4 :
j
j
j j j
c o lo r te m p c o lo r x y
w h ile j N d o
T
y
r e a d c o lo r x y
if c o lo r y c o lo r te m p
c o u n t
e ls e
T c o u n t
T T T
j
c o u n t
e n d if
x x y t
( )
2 5 :
2 6 :
g
e n d w h ile
r e tu r n T
6. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 44
Figure 9: the measurements'' environment
The result is an array similar to the one presented in table 3. We see that as further the stripe from the camera,
the less pixels it has.
Table 3: Lookup table of pixel numbers as a function of the distance
The red line presents the measurement points along the y-axis. The green areas present the field and the yellow
and gray areas presents the area outside the count range. The table (like table 3), can be used to correlate number
of pixels to a point (x,y) further from the camera in the field and then derive the distance to this point. In order
to calculate a distance from the camera to a certain point we will usually have to perform interpolation. Let us
take an example. Suppose we need to know the distance to point (320, 75). We will look at the y-axes value,
which is 75 in this case. From table 3 we see that the appropriate region end-points in terms of accumulated pix-
el values are 72 and 81. The related distance value lies between 40 and 50 cm. The distance value is calculated
by interpolation:
40 +
75−72
81−72
∗ 10 = 43.3333 cm. In general the distance function will be:
1 0
1
p ix
y y i
d ista n ce cm y i
y i y i
Where y(i) is the lower end point and y(i+1) is the higher end point in the region. ypix is the pixel's y value of the
desired point.
VII. MAPPING SOLUTION ACCURACY
The method we described use the scale solely. When the lookup table is complete, there is no need to process
the binary image. The only input data needed is the (x,y) object location.
Figure 10 presents the error as a function of the distance.
Table 4:Lookup table of pixel numbers as a function of the distance and stripe width type.
0 10 20 30 40 50 60 70 CM
0 20 15 14 13 10 9 8 #PIX
0 20 35 49 62 72 81 89 ACC
7. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 45
Figure 10: error as a function of distance
There are areas in figure 10 that show higher error compared to other areas. In order to have a distinction among
the different areas we divided the area into three distance ranges: short range, middle range and long range. Ta-
ble 4 presents the average error ratings for each range. The error in the long distance range is higher by more
than three times compared to the short distance range and more than double compared to the middle distance
range. Clearly, there is a need to adapt the method such that the error will be satisfactory for all ranges.
Table 5: The average error as a function of the distance ranges
Range Distance Error
Short 80-190[cm] 2.68%
Middle 200-300[cm] 4.95%
Long 310-410[cm] 11.6%
Looking closely at figure 9 and table 3 reveals that in the long distance there is a gap, which is realized by a
jump in the pixels count. It means that the algorithm miscount the long or far distance. For example, if the real
distance was 30 [cm] the algorithm calculates it as a distance of 10 [cm] due to the jump in pixels count. We
will call this error as Jump error meaning miscount pixels error.
In order to take care about this problem we will design a new scale that solves the issue of Jump error. The new
scale will take into account the different pixels count of stripes relative to their distance from the camera. After
testing several options, we concluded that the following measures of the different ranges on the scale would be
adequate: stripes' width of be 5 cm, 10 cm, and 20 cm for short, middle, and long distances accordingly.
VIII. SCALE VERSION 2.0
Based on the above discussion the 2nd
version of the scale looks like the one presented in figure 11.
Figure 11: Scale 2.0 with different stripe's size
IX. THE LOOKUP TABLE ALGORITHM FOR SCALE 2.0:
The new lookup table will contain the information of each stripe width. The digit 0 will designate the stripes
with width of 5 cm. The digit 1 will designate the stripes with width of 10 cm, and the digit 2 will designate the
stripes with width of 20 cm. Thus, the lookup table might look as the one presented in table 5.
Reading the new ribbon requires the use of a new algorithm. This new algorithm will be similar to the one pre-
sented in table 2 with additional part for the information about different stipe sizes (5, 10, or 20 cm). Actually,
since the ribbon structure is known, it is easy to manually add the last row of the lookup table.
0
5
10
15
20
25
80
100
120
140
160
180
200
220
240
260
280
300
320
340
360
380
400
ERROR%
DISTANCE [CM]
Error as function of distance
8. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 46
Table 6 presents the algorithm for scale 2.0. The additional inputs compared to the algorithm of table 2 are n1
and n2, which presents the first stripe number with width of 10 cm and 20 cm respectively.
Table 6: Creation of lookup table 2.0 algorithm
1 2
1 : _ _ 2 .0 ( , , , ) :
2 :
3 :
4 :
5 :
6 : ' 0
7 : '
8 :
A lg o rith m L o o k u p T a b le N T n n
p la c e th e c a m e ra in fro n t o f m id d le sc a l e 's first strip e
y = 0
c o u n t = 0
j = 0
x o f strip e s lo w e r le ft c o rn e r
x x v a lu e o f first strip e s c e n te r
re a d
1,
2 ,
3 , 2 , 3 , 1
1
4 ,
2
( , )
9 : _ ( , )
1 0 : se c 0
1 1 :
1 2 : se c
1 3 :
1 4 : ( , )
1 5 : ( ) _
1 6 :
1 7 :
1 8 :
1 9 :
2 0 :
2 1 : 0
2 2 :
2 3 :
2 4 :
j
j
j j j
j
c o lo r x y
c o lo r te m p c o lo r x y
w h ile j N d o
T
y
re a d c o lo r x y
if c o lo r y c o lo r te m p
c o u n t
e lse
T c o u n t
T T T
if j n
T
e lse
if j n
T
4 ,
4 ,
1
2 5 :
2 6 : 2
2 7 :
2 8 :
2 9 :
3 0 : se c se c 1 0
3 1 : 0
3 2 :
3 3 : ( )
3 4 :
3 5 :
j
j
e lse
T
e n d if
e n d if
j
c o u n t
e n d if
x x y tg
e n d w h ile
re tu rn T
The part that takes care of writing the stripe width index (0,1,2) is presented in lines 20-28 of the algorithm.
When the lookup table is complete, a distance to a certain point (x,y) is calculated using the algorithm in table 7.
9. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 47
Table 7: Distance calculation algorithm
1 : _ _ 2 .0 ( , , , ) :
2 : [1] [2 ] [3]
3 :
4 :
5 :
6 : _ 5
7 : ( (3, ))
8 : (4, ) 0
9 : [1]
1 0 : _ 5
1 1 : (4, ) 1
1 2 :
y
A lg o rith m D ista n ce C a lcu la tio n x y T D IS
co u n t co u n t = co u n t = 0
in d ex = 0
typ e = 0
j = 1
la st w id th
w h ile y T j
if T j
co u n t
la st w id th
else if T j
co
[2 ]
1 3 : _ 1 0
1 4 :
1 5 : [3]
1 6 : _ 2 0
1 7 :
1 8 :
1 9 :
2 0 :
2 1 : 5 [0 ] 1 0 [1]
(3, 1)
2 0 [3] _
(3, ) (3, 1)
2 2 :
y
y
u n t
la st w id th
else
co u n t
la st w id th
en d if
en d if
j
en d w h ile
D IS co u n t co u n t
y T j
co u n t la st w id th
T j T j
retu rn D IS
For example if we need the distance to a point with y=85, using the data in table 5, we will get:
8 5 8 1
5 5 1 0 2 1 0 5 0
8 9 8 1
cm
X. THE ACCURACY OF SCALE VERSION 2.0
Figure 12 presents the error of the distance value for distances up to 410 cm, using the scale version2. Table 8
presents the average error in short, middle and long distance ranges of the two scale versions.
Figure 12: Error as function as distance in Scale ver 2.0
0
2
4
6
8
80
100
120
140
160
180
200
220
240
260
280
300
320
340
360
380
400
ERROR%
DISTANCE [CM]
Error as function of distance
10. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 48
Figure 13: x-axis width measurement using Scale version 1.0
Table 8: error as a function of distance for the two scale versions
Range Distance Error
ver 1.0
Error ver
2.0
Short 80-190[cm] 2.68% 2.11%
Middle 200-300[cm] 4.95% 1.69%
Long 310-410[cm] 11.6% 3.81%
It is clear from table 8, that using scale version 2, significantly improved the accuracy of the distance measure-
ment.For example at the maximum distance presented in table 8, the error will be about 410 cm*3.81%=15.6cm.
Since the robot feet size is more than 20 cm this is accurate enough for all practical purposes.
XI. ADDING THE X-AXIS INTO CALCULATIONS
Up until now, we used only the y-axis for calculation of the distance. Although the x-axis value has less effect
on the distance, it is still good practice to use its value for better distance estimation.
We can calculate the x-axis value using the scale version 1.0. The number of pixels inside each cell doesn’t
change much, meaning that the number of pixels in the middle cell are more or less equal to those on the edge
cell, as can be seen in Table 9.
This fact led us to an additional design. We sketched two vertical white lines, with a distance of 6m between
them as presented in figure 15. Adding the horizontal lines creates a trapezoid of the 2-D field.
Then, we took measurements at every 20[cm] of the y-axis, starting from the camera position (figure 13). In
each slice that created by the scale 1.0, we counted the number of pixels entering to 10cm segments, starting
from one edge of the trapezoid and finshe in the other edge (figure 14).
Table 9 shows the data we got using Logitech C905 camera. Each column in Table 9 presents a different slice of
y-axis with header yend/ybegin (Top and bottom of the cell [figuer 7] in terms of pixels) [the image produced by
this camera is an inverse image].
Each row presents different segment of 10[cm] (width x-axis). The numbers in Table 9 cell represents the
11. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 49
Figure 14:number of pixels presents 10cm segments in y-axis in avg
X- In terms of cm | Y-In terms of pixels
*The data presented in table 9 might be different for a different
camera, but the calculation will be the same.
Table 9: presents the y-axis slices with the average pixels count for each 10cm segment in x-axis in each slice
number of pixel entering to 10[cm] of width. For example in 100[cm] from the right side of the picture. In high
of 240[pixel] (the second column) enter 63 pixels.
We can see from table 9 that each column standard deviation is in most cases less than 5% and in the worst case
is less than 6.5%. We can conclude that it is safe to use the average number of pixels for distance calculations.
Next, we will explain how to receive the distance from xi to xmiddle. For that, let us define xmiddle as the middle
point value of the horizontal line (figure 15) with yi as its y-axis value. xmiddle is known for every y. First , we
recorded the two left corners of the trapezoid ((x1,y1) and (x2,y2)) (This is true for the right side as well).
Suppose that we are interested in the distance to the point (xi,yi). We know that the distance of any point (x3 , yi )
on the white line to the parallel point (xmiddle ,yi) on the middle line equals to 3[M].
The algorithm will find the x3 using the method proposed in the papers [8] and [9].
It is easy to prove that the x-axis value (xwl) of a point on the left vertical white line, with yi as its y-axis value,
can be calculated to be:
1
1
( )
i
w l i
y y
x f y x
m
Where m is the left white line slope ((y1-y2)/(x1-x2)).
We define two more variables: pix1 and pix2 as follows:
1 3
2
m iddle
w l m iddle
pix x x
pix x x
The algorithm will compute the x-axis distance (DISx) with metric units according to the following equation:
1
2
3[ ]x
p ix
D IS m
p ix
This gives the X distance from the center to the desired point (xi,yi)
12. Distance Estimation to Image Objects Using Adapted Scale
www.theijes.com The IJES Page 50
Figure 15: Field configuration for calculating x-axis value of a certain point
XII. RETRIEVE DISTANCE
The last step required to compute the real distance is by use the right triangle hypotenuse formula:
2 2
( ) ( )x y
distance D IS D IS
XIII. SUMMARY
This paper demonstrates an easy and efficient method to calculate a distance from a camera to an arbitrary ob-
ject in the playing field of a KSL competition in the frame of RoboCup soccer league. In fact, this method is
more accurate and faster than volume solution, in other words, it reducing the error mistake in each step and
have no obligation to count the white pixel in the binary threshold picture.
The base of this idea relies on initial setup, which includes adapting the pixel numbers into the distance. At first,
there was an exponential function, as we proceeded, we decided to divide the graph into small sections, achiev-
ing a better result. Apparently, the problem with long distance remained. Another obstacle was different objects
on the field, different shapes are acting differently, what made us create different function for each shape (Goal,
Ball, White strip, etc.)
The database was prepared by using a suitable scale and merged all of the different functions into one database.
Then, an algorithm with only (x,y) coordination was created. These solved both problems. Additionally, this
method can be adapted to different scenarios and different cameras.
REFERENCES
[1]. G. P. Stein, O. Mano and A. Shashua, "Vision-based ACC with a Single Camera: Bounds on Range and Range Rate Accuracy",
Proc. Intelligent Vehicles Symposium, pp. 120-125, 2003.
[2]. Y.C. Kuo,N.S. Pai and Y.N. Li, "Vision-based Vehicle Detection for a Driver Assistance System," Computers and Mathematics
with Applications (61) ,pp. 2096–2100, 2011.
[3]. J. Mrovlje and D. Vrančić, "Distance Measuring Based on Stereoscopic Pictures", 9th Int. PhD Workshop on Systems and Control:
Young Generation Viewpoint, Slovenia, 2008.
[4]. M. Kyto, M. Nuutinen, and P. Oittinen , "Method for Measuring Stereo Camera Depth Accuracy Based on Stereoscopic Vision",
Proc. SPIE 7864, Three dimensional imaging, Interaction and Measurement, 2011.
[5]. C. Tinnachote and K. Pimprasan, "Distance Measurment from Digital Photograpg Using 3rd Order Polynomial Equation", Proc.
ACRS ,2012..
[6]. M. Mirzabaki, "Depth Detection Through Interpolation Functions", The 12th int. conf. WSCG, pp. 105-108, 2004..
[7]. S.V.F Barreto R E.S. Anna, and M. Feitosa, "A Method for Image Processing and Distance Measuring Based on Laser Distance
Triangulation", IEEE 20th int. conf. on Electronics, Circuits and Systems, pp. 695-698, 2013.
[8]. Y. Lim and H. Lim, "A Method for Measurement of Distances using License Plate Detection", Advanced Science and Technology
Letters, pp. 13-16, 2014.
[9]. A. de La Bourdonnaye, R. Doskocil, V. Krivanek, and A. Stefek, "Practical Experience with Distance Measurement Bsed on Single
Visual Camera," Advances in Military Technology, (7) 2, pp. 49-56, 2012