This document provides a survey of object detection methods in visual sensor networks (VSNs). It discusses two main categories of object detection: 1) conventional methods using background subtraction to detect moving objects, and 2) methods involving camera nodes to perform preprocessing and send only useful information like bounding boxes or features. The document analyzes and compares specific methods within each category, discussing their advantages and limitations. It concludes that involving camera nodes in preprocessing can reduce data transmission costs and improve energy efficiency in VSNs, but open challenges remain regarding performance metrics and accommodating different application needs.
A Study of Motion Detection Method for Smart Home SystemAM Publications
Motion detection surveillance technology give ease for time-consuming reviewing process that a normal video
surveillance system offers. By using motion detection, it save the monitoring time and cost. It has gained a lot of interests
over the past few years. In this paper, a proposed motion detection surveillance system, through the study and evaluation
of currently available different methods. The proposed system is efficient and convenient for both office and home uses as
a smart home security system technology.
Analytical framework for optimized feature extraction for upgrading occupancy...IJECEIAES
The adoption of the occupancy sensors has become an inevitable in commercial and non-commercial security devices, owing to their proficiency in the energy management. It has been found that the usages of conventional sensors is shrouded with operational problems, hence the use of the Doppler radar offers better mitigation of such problems. However, the usage of Doppler radar towards occupancy sensing in existing system is found to be very much in infancy stage. Moreover, the performance of monitoring using Doppler radar is yet to be improved more. Therefore, this paper introduces a simplified framework for enriching the event sensing performance by efficient selection of minimal robust attributes using Doppler radar. Adoption of analytical methodology has been carried out to find that different machine learning approaches could be further used for improving the accuracy performance for the feature that has been extracted in the proposed system of occuancy system.
A Study of Motion Detection Method for Smart Home SystemAM Publications
Motion detection surveillance technology give ease for time-consuming reviewing process that a normal video
surveillance system offers. By using motion detection, it save the monitoring time and cost. It has gained a lot of interests
over the past few years. In this paper, a proposed motion detection surveillance system, through the study and evaluation
of currently available different methods. The proposed system is efficient and convenient for both office and home uses as
a smart home security system technology.
Analytical framework for optimized feature extraction for upgrading occupancy...IJECEIAES
The adoption of the occupancy sensors has become an inevitable in commercial and non-commercial security devices, owing to their proficiency in the energy management. It has been found that the usages of conventional sensors is shrouded with operational problems, hence the use of the Doppler radar offers better mitigation of such problems. However, the usage of Doppler radar towards occupancy sensing in existing system is found to be very much in infancy stage. Moreover, the performance of monitoring using Doppler radar is yet to be improved more. Therefore, this paper introduces a simplified framework for enriching the event sensing performance by efficient selection of minimal robust attributes using Doppler radar. Adoption of analytical methodology has been carried out to find that different machine learning approaches could be further used for improving the accuracy performance for the feature that has been extracted in the proposed system of occuancy system.
Framework for Contextual Outlier Identification using Multivariate Analysis a...IJECEIAES
Majority of the existing commercial application for video surveillance system only captures the event frames where the accuracy level of captures is too poor. We reviewed the existing system to find that at present there is no such research technique that offers contextual-based scene identification of outliers. Therefore, we presented a framework that uses unsupervised learning approach to perform precise identification of outliers for a given video frames concerning the contextual information of the scene. The proposed system uses matrix decomposition method using multivariate analysis to maintain an equilibrium better faster response time and higher accuracy of the abnormal event/object detection as an outlier. Using an analytical methodology, the proposed system blocking operation followed by sparsity to perform detection. The study outcome shows that proposed system offers an increasing level of accuracy in contrast to the existing system with faster response time.
Applying edge density based region growing with frame difference for detectin...eSAT 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
SAR Image Classification by Multilayer Back Propagation Neural NetworkIJMTST Journal
A novel descriptive feature extraction method of Discrete Fourier transform and neural network classifier for classification of Synthetic Aperture Radar (SAR) images is proposed. The classification process has the following stages (1) Image Segmentation using statistical Region Merging (SRM) (2) Polar transform and Feature extraction using Discrete Fourier Transform (3) Neural Network classification using back propagation. This is generally the first step in image analysis. Segmentation subdivides an image into its constituent parts or objects. The level to which this subdivision is carried depends on the problem being solved. The image segmentation in this study is performed using Statistical Region Merging proposed Richard Nock and Frank Nielsen. The key idea of the Statistical Region Merging model is to formulate image segmentation as an inference problem. Here the merging procedure is based on the theorem. Feature vectors as the input for the neural network. Polar transform is applied to segmented SAR image. The rotation problem under the Cartesian coordinates becomes the translation problem under the polar coordinates.
ABSTRACT Feature extraction plays a vital role in the analysis and interpretation of remotely sensed data. The two important components of Feature extraction are Image enhancement and information extraction. Image enhancement techniques help in improving the visibility of any portion or feature of the image. Information extraction techniques help in obtaining the statistical information about any particular feature or portion of the image. This presented work focuses on the various feature extraction techniques and area of optical character recognition is a particularly important in Image processing. Keywords— Image character recognition, Methods for Feature Extraction, Basic Gabor Filter, IDA, and PCA.
REGISTRATION TECHNOLOGIES and THEIR CLASSIFICATION IN AUGMENTED REALITY THE K...IJCSEA Journal
The registration in augmented reality is process which merges virtual objects generated by computer with real world image caught by camera. This paper describes the knowledge-based registration, computer vision-based registration and tracker-based registration technology. This paper mainly focused on trackerbased
registration technology in augmented reality. Also described method in tracker- based technology, problem and solution.
Face Recognition Based Intelligent Door Control Systemijtsrd
This paper presents the intelligent door control system based on face detection and recognition. This system can avoid the need to control by persons with the use of keys, security cards, password or pattern to open the door. The main objective is to develop a simple and fast recognition system for personal identification and face recognition to provide the security system. Face is a complex multidimensional structure and needs good computing techniques for recognition. The system is composed of two main parts face recognition and automatic door access control. It needs to detect the face before recognizing the face of the person. In face detection step, Viola Jones face detection algorithm is applied to detect the human face. Face recognition is implemented by using the Principal Component Analysis PCA and Neural Network. Image processing toolbox which is in MATLAB 2013a is used for the recognition process in this research. The PIC microcontroller is used to automatic door access control system by programming MikroC language. The door is opened automatically for the known person according to the result of verification in the MATLAB. On the other hand, the door remains closed for the unknown person. San San Naing | Thiri Oo Kywe | Ni Ni San Hlaing ""Face Recognition Based Intelligent Door Control System"" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-4 , June 2019, URL: https://www.ijtsrd.com/papers/ijtsrd23893.pdf
Paper URL: https://www.ijtsrd.com/engineering/electrical-engineering/23893/face-recognition-based-intelligent-door-control-system/san-san-naing
Interactive Full-Body Motion Capture Using Infrared Sensor Network ijcga
Traditional motion capture (mocap) has been well-studied in visual science for the last decades. However the field is mostly about capturing precise animation to be used in specific applications after intensive post processing such as studying biomechanics or rigging models in movies. These data sets are normally captured in complex laboratory environments with sophisticated equipment thus making motion capture a
field that is mostly exclusive to professional animators. In addition, obtrusive sensors must be attached to actors and calibrated within the capturing system, resulting in limited and unnatural motion. In recent year the rise of computer vision and interactive entertainment opened the gate for a different type of motion capture which focuses on producing optical markerless or mechanical sensorless motion capture. Furthermore a wide array of low-cost device are released that are easy to use for less mission critical applications. This paper describes a new technique of using multiple infrared devices to process data from multiple infrared sensors to enhance the flexibility and accuracy of the markerless mocap using commodity
devices such as Kinect. The method involves analyzing each individual sensor data, decompose and rebuild
them into a uniformed skeleton across all sensors. We then assign criteria to define the confidence level of
captured signal from sensor. Each sensor operates on its own process and communicates through MPI.
Our method emphasizes on the need of minimum calculation overhead for better real time performance
while being able to maintain good scalability.
An optimized framework for detection and tracking of video objects in challen...ijma
Segmentation and tracking are two important aspects in visual surveillance systems. Many barriers such as
cluttered background, camera movements, and occlusion make the robust detection and tracking a difficult
problem, especially in case of multiple moving objects. Object detection in the presence of camera noise
and with variable or unfavourable luminance conditions is still an active area of research. This paper
propose a framework which can effectively detect the moving objects and track them despite of occlusion
and a priori knowledge of objects in the scene. The segmentation step uses a robust threshold decision
algorithm which uses a multi-background model. The video object tracking is able to track multiple objects
along with their trajectories based on Continuous Energy Minimization. In this work, an effective
formulation of multi-target tracking as minimization of a continuous energy is combined with multibackground
registration. Apart from the recent approaches, it focus on making use of an energy that
corresponds to a more complete representation of the problem, rather than one that is amenable to global
optimization. Besides the image evidence, the energy function considers physical constraints, such as target
dynamics, mutual exclusion, and track persistence. The proposed tracking framework is able to track
multiple objects despite of occlusions under dynamic background conditions.
Interactive full body motion capture using infrared sensor networkijcga
Traditional motion capture (mocap) has been
well
-
stud
ied in visual science for
the last decades
. However
the fie
ld is mostly about capturing
precise animation to be used in
specific
application
s
after
intensive
post
processing such as studying biomechanics or rigging models in movies. These data set
s are normally
captured in complex laboratory environments with
sophisticated
equipment thus making motion capture a
field that is mostly exclusive to professional animators.
In
addition
, obtrusive sensors must be attached to
actors and calibrated within t
he capturing system, resulting in limited and unnatural motion.
In recent year
the rise of computer vision and interactive entertainment opened the gate for a different type of motion
capture which focuses on producing
optical
marker
less
or mechanical sens
orless
motion capture.
Furtherm
ore a wide array of low
-
cost
device are released that are easy to use
for less mission critical
applications
.
This paper
describe
s
a new technique of using multiple infrared devices to process data from
multiple infrared sensors to enhance the flexibility and accuracy of the markerless mocap
using commodity
devices such as Kinect
. The method involves analyzing each individual sensor
data, decompose and rebuild
them into a uniformed skeleton across all sensors. We then assign criteria to define the confidence level of
captured signal from
sensor. Each sensor operates on its own process and communicates through MPI.
Our method emphasize
s on the need of minimum calculation overhead for better real time performance
while being able to maintain good scalability
International Journal of Engineering Research and DevelopmentIJERD Editor
Electrical, Electronics and Computer Engineering,
Information Engineering and Technology,
Mechanical, Industrial and Manufacturing Engineering,
Automation and Mechatronics Engineering,
Material and Chemical Engineering,
Civil and Architecture Engineering,
Biotechnology and Bio Engineering,
Environmental Engineering,
Petroleum and Mining Engineering,
Marine and Agriculture engineering,
Aerospace Engineering.
Threshold based filtering technique for efficient moving object detection and...eSAT Journals
Abstract Detection and tracking of moving objects are an important research area in a video surveillance application. Object tracking is
used in several applications such as video compression, surveillance, robot technology and so on. Recently many researches has
been developed for video object detection, however the object detection accuracy and background object detection in the video
frames are still poses demanding issues. In this paper, a novel framework called Threshold Filtered Video Object Detection and
Tracking (TFVODT) is designed for effective detection and tracking of moving objects. TFVODT framework initially takes video
file as input, and then video frames are segmented using Median Filter-based Enhanced Laplacian Thresholding for improving
the video quality by reducing mean square error. Next, Color Histogram-based Particle Filter is applied to the segmented objects
in TFVODT framework for video object tracking. The Color Histogram-based Particle Filter measures the likelihood function,
particle posterior and particle prior function based on the Bayes Sequential Estimation model for improving the object tracking
accuracy. Finally, the objects detection is performed with help of Improvisation of Enhanced Laplacian Threshold (IELT) to
enhance video object detection accuracy and to recognize background moving object detection. The proposed TFVODT
framework using video images obtained from Internet Archive 501(c) (3) for conducting experiment and comparison is made with
the existing object detection techniques. Experimental evaluation of TFVODT framework is done with the performance metrics
such as object segmentation accuracy, Peak Signal to Noise Ratio, object tracking accuracy, Mean Square Error and object
detection accuracy of moving video object frames. Experimental analysis shows that the TFVODT framework is able to improve
the video object detection accuracy by 18% and reduces the Peak Signal to Noise Ratio by 23 % when compared to the state-ofthe-
art works.
Keywords: Object segmentation, Object tracking, Object Detection, Enhanced Laplacian Thresholding, Median
Filter, Color Histogram-based Particle Filter
Objects detection and tracking using fast principle component purist and kalm...IJECEIAES
The detection and tracking of moving objects attracted a lot of concern because of the vast computer vision applications. This paper proposes a new algorithm based on several methods for identifying, detecting, and tracking an object in order to develop an effective and efficient system in several applications. This algorithm has three main parts: the first part for background modeling and foreground extraction, the second part for smoothing, filtering and detecting moving objects within the video frame and the last part includes tracking and prediction of detected objects. In this proposed work, a new algorithm to detect moving objects from video data is designed by the Fast Principle Component Purist (FPCP). Then we used an optimal filter that performs well to reduce noise through the median filter. The Fast Regionconvolution neural networks (Fast- RCNN) is used to add smoothness to the spatial identification of objects and their areas. Then the detected object is tracked by Kalman Filter. Experimental results show that our algorithm adapts to different situations and outperforms many existing algorithms.
Framework for Contextual Outlier Identification using Multivariate Analysis a...IJECEIAES
Majority of the existing commercial application for video surveillance system only captures the event frames where the accuracy level of captures is too poor. We reviewed the existing system to find that at present there is no such research technique that offers contextual-based scene identification of outliers. Therefore, we presented a framework that uses unsupervised learning approach to perform precise identification of outliers for a given video frames concerning the contextual information of the scene. The proposed system uses matrix decomposition method using multivariate analysis to maintain an equilibrium better faster response time and higher accuracy of the abnormal event/object detection as an outlier. Using an analytical methodology, the proposed system blocking operation followed by sparsity to perform detection. The study outcome shows that proposed system offers an increasing level of accuracy in contrast to the existing system with faster response time.
Applying edge density based region growing with frame difference for detectin...eSAT 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
SAR Image Classification by Multilayer Back Propagation Neural NetworkIJMTST Journal
A novel descriptive feature extraction method of Discrete Fourier transform and neural network classifier for classification of Synthetic Aperture Radar (SAR) images is proposed. The classification process has the following stages (1) Image Segmentation using statistical Region Merging (SRM) (2) Polar transform and Feature extraction using Discrete Fourier Transform (3) Neural Network classification using back propagation. This is generally the first step in image analysis. Segmentation subdivides an image into its constituent parts or objects. The level to which this subdivision is carried depends on the problem being solved. The image segmentation in this study is performed using Statistical Region Merging proposed Richard Nock and Frank Nielsen. The key idea of the Statistical Region Merging model is to formulate image segmentation as an inference problem. Here the merging procedure is based on the theorem. Feature vectors as the input for the neural network. Polar transform is applied to segmented SAR image. The rotation problem under the Cartesian coordinates becomes the translation problem under the polar coordinates.
ABSTRACT Feature extraction plays a vital role in the analysis and interpretation of remotely sensed data. The two important components of Feature extraction are Image enhancement and information extraction. Image enhancement techniques help in improving the visibility of any portion or feature of the image. Information extraction techniques help in obtaining the statistical information about any particular feature or portion of the image. This presented work focuses on the various feature extraction techniques and area of optical character recognition is a particularly important in Image processing. Keywords— Image character recognition, Methods for Feature Extraction, Basic Gabor Filter, IDA, and PCA.
REGISTRATION TECHNOLOGIES and THEIR CLASSIFICATION IN AUGMENTED REALITY THE K...IJCSEA Journal
The registration in augmented reality is process which merges virtual objects generated by computer with real world image caught by camera. This paper describes the knowledge-based registration, computer vision-based registration and tracker-based registration technology. This paper mainly focused on trackerbased
registration technology in augmented reality. Also described method in tracker- based technology, problem and solution.
Face Recognition Based Intelligent Door Control Systemijtsrd
This paper presents the intelligent door control system based on face detection and recognition. This system can avoid the need to control by persons with the use of keys, security cards, password or pattern to open the door. The main objective is to develop a simple and fast recognition system for personal identification and face recognition to provide the security system. Face is a complex multidimensional structure and needs good computing techniques for recognition. The system is composed of two main parts face recognition and automatic door access control. It needs to detect the face before recognizing the face of the person. In face detection step, Viola Jones face detection algorithm is applied to detect the human face. Face recognition is implemented by using the Principal Component Analysis PCA and Neural Network. Image processing toolbox which is in MATLAB 2013a is used for the recognition process in this research. The PIC microcontroller is used to automatic door access control system by programming MikroC language. The door is opened automatically for the known person according to the result of verification in the MATLAB. On the other hand, the door remains closed for the unknown person. San San Naing | Thiri Oo Kywe | Ni Ni San Hlaing ""Face Recognition Based Intelligent Door Control System"" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-4 , June 2019, URL: https://www.ijtsrd.com/papers/ijtsrd23893.pdf
Paper URL: https://www.ijtsrd.com/engineering/electrical-engineering/23893/face-recognition-based-intelligent-door-control-system/san-san-naing
Interactive Full-Body Motion Capture Using Infrared Sensor Network ijcga
Traditional motion capture (mocap) has been well-studied in visual science for the last decades. However the field is mostly about capturing precise animation to be used in specific applications after intensive post processing such as studying biomechanics or rigging models in movies. These data sets are normally captured in complex laboratory environments with sophisticated equipment thus making motion capture a
field that is mostly exclusive to professional animators. In addition, obtrusive sensors must be attached to actors and calibrated within the capturing system, resulting in limited and unnatural motion. In recent year the rise of computer vision and interactive entertainment opened the gate for a different type of motion capture which focuses on producing optical markerless or mechanical sensorless motion capture. Furthermore a wide array of low-cost device are released that are easy to use for less mission critical applications. This paper describes a new technique of using multiple infrared devices to process data from multiple infrared sensors to enhance the flexibility and accuracy of the markerless mocap using commodity
devices such as Kinect. The method involves analyzing each individual sensor data, decompose and rebuild
them into a uniformed skeleton across all sensors. We then assign criteria to define the confidence level of
captured signal from sensor. Each sensor operates on its own process and communicates through MPI.
Our method emphasizes on the need of minimum calculation overhead for better real time performance
while being able to maintain good scalability.
An optimized framework for detection and tracking of video objects in challen...ijma
Segmentation and tracking are two important aspects in visual surveillance systems. Many barriers such as
cluttered background, camera movements, and occlusion make the robust detection and tracking a difficult
problem, especially in case of multiple moving objects. Object detection in the presence of camera noise
and with variable or unfavourable luminance conditions is still an active area of research. This paper
propose a framework which can effectively detect the moving objects and track them despite of occlusion
and a priori knowledge of objects in the scene. The segmentation step uses a robust threshold decision
algorithm which uses a multi-background model. The video object tracking is able to track multiple objects
along with their trajectories based on Continuous Energy Minimization. In this work, an effective
formulation of multi-target tracking as minimization of a continuous energy is combined with multibackground
registration. Apart from the recent approaches, it focus on making use of an energy that
corresponds to a more complete representation of the problem, rather than one that is amenable to global
optimization. Besides the image evidence, the energy function considers physical constraints, such as target
dynamics, mutual exclusion, and track persistence. The proposed tracking framework is able to track
multiple objects despite of occlusions under dynamic background conditions.
Interactive full body motion capture using infrared sensor networkijcga
Traditional motion capture (mocap) has been
well
-
stud
ied in visual science for
the last decades
. However
the fie
ld is mostly about capturing
precise animation to be used in
specific
application
s
after
intensive
post
processing such as studying biomechanics or rigging models in movies. These data set
s are normally
captured in complex laboratory environments with
sophisticated
equipment thus making motion capture a
field that is mostly exclusive to professional animators.
In
addition
, obtrusive sensors must be attached to
actors and calibrated within t
he capturing system, resulting in limited and unnatural motion.
In recent year
the rise of computer vision and interactive entertainment opened the gate for a different type of motion
capture which focuses on producing
optical
marker
less
or mechanical sens
orless
motion capture.
Furtherm
ore a wide array of low
-
cost
device are released that are easy to use
for less mission critical
applications
.
This paper
describe
s
a new technique of using multiple infrared devices to process data from
multiple infrared sensors to enhance the flexibility and accuracy of the markerless mocap
using commodity
devices such as Kinect
. The method involves analyzing each individual sensor
data, decompose and rebuild
them into a uniformed skeleton across all sensors. We then assign criteria to define the confidence level of
captured signal from
sensor. Each sensor operates on its own process and communicates through MPI.
Our method emphasize
s on the need of minimum calculation overhead for better real time performance
while being able to maintain good scalability
International Journal of Engineering Research and DevelopmentIJERD Editor
Electrical, Electronics and Computer Engineering,
Information Engineering and Technology,
Mechanical, Industrial and Manufacturing Engineering,
Automation and Mechatronics Engineering,
Material and Chemical Engineering,
Civil and Architecture Engineering,
Biotechnology and Bio Engineering,
Environmental Engineering,
Petroleum and Mining Engineering,
Marine and Agriculture engineering,
Aerospace Engineering.
Threshold based filtering technique for efficient moving object detection and...eSAT Journals
Abstract Detection and tracking of moving objects are an important research area in a video surveillance application. Object tracking is
used in several applications such as video compression, surveillance, robot technology and so on. Recently many researches has
been developed for video object detection, however the object detection accuracy and background object detection in the video
frames are still poses demanding issues. In this paper, a novel framework called Threshold Filtered Video Object Detection and
Tracking (TFVODT) is designed for effective detection and tracking of moving objects. TFVODT framework initially takes video
file as input, and then video frames are segmented using Median Filter-based Enhanced Laplacian Thresholding for improving
the video quality by reducing mean square error. Next, Color Histogram-based Particle Filter is applied to the segmented objects
in TFVODT framework for video object tracking. The Color Histogram-based Particle Filter measures the likelihood function,
particle posterior and particle prior function based on the Bayes Sequential Estimation model for improving the object tracking
accuracy. Finally, the objects detection is performed with help of Improvisation of Enhanced Laplacian Threshold (IELT) to
enhance video object detection accuracy and to recognize background moving object detection. The proposed TFVODT
framework using video images obtained from Internet Archive 501(c) (3) for conducting experiment and comparison is made with
the existing object detection techniques. Experimental evaluation of TFVODT framework is done with the performance metrics
such as object segmentation accuracy, Peak Signal to Noise Ratio, object tracking accuracy, Mean Square Error and object
detection accuracy of moving video object frames. Experimental analysis shows that the TFVODT framework is able to improve
the video object detection accuracy by 18% and reduces the Peak Signal to Noise Ratio by 23 % when compared to the state-ofthe-
art works.
Keywords: Object segmentation, Object tracking, Object Detection, Enhanced Laplacian Thresholding, Median
Filter, Color Histogram-based Particle Filter
Objects detection and tracking using fast principle component purist and kalm...IJECEIAES
The detection and tracking of moving objects attracted a lot of concern because of the vast computer vision applications. This paper proposes a new algorithm based on several methods for identifying, detecting, and tracking an object in order to develop an effective and efficient system in several applications. This algorithm has three main parts: the first part for background modeling and foreground extraction, the second part for smoothing, filtering and detecting moving objects within the video frame and the last part includes tracking and prediction of detected objects. In this proposed work, a new algorithm to detect moving objects from video data is designed by the Fast Principle Component Purist (FPCP). Then we used an optimal filter that performs well to reduce noise through the median filter. The Fast Regionconvolution neural networks (Fast- RCNN) is used to add smoothness to the spatial identification of objects and their areas. Then the detected object is tracked by Kalman Filter. Experimental results show that our algorithm adapts to different situations and outperforms many existing algorithms.
An Analysis of Various Deep Learning Algorithms for Image Processingvivatechijri
Various applications of image processing has given it a wider scope when it comes to data analysis.
Various Machine Learning Algorithms provide a powerful environment for training modules effectively to
identify various entities of images and segment the same accordingly. Rather one can observe that though the
image classifiers like the Support Vector Machines (SVM) or Random Forest Algorithms do justice to the task,
deep learning algorithms like the Artificial Neural Networks (ANN) and its subordinates, the very well-known
and extremely powerful Algorithm Convolution Neural Networks (CNN) can provide a new dimension to the
image processing domain. It has way higher accuracy and computational power for classifying images further
and segregating their various entities as individual components of the image working region. Major focus will
be on the Region Convolution Neural Networks (R-CNN) algorithm and how well it provides the pixel-level
segmentation further using its better successors like the Fast-Faster and Mask R-CNN versions.
Online video-based abnormal detection using highly motion techniques and stat...TELKOMNIKA JOURNAL
At the essence of video surveillance, there are abnormal detection approaches, which have been proven to be substantially effective in detecting abnormal incidents without prior knowledge about these incidents. Based on the state-of-the-art research, it is evident that there is a trade-off between frame processing time and detection accuracy in abnormal detection approaches. Therefore, the primary challenge is to balance this trade-off suitably by utilizing few, but very descriptive features to fulfill online performance while maintaining a high accuracy rate. In this study, we propose a new framework, which achieves the balancing between detection accuracy and video processing time by employing two efficient motion techniques, specifically, foreground and optical flow energy. Moreover, we use different statistical analysis measures of motion features to get robust inference method to distinguish abnormal behavior incident from normal ones. The performance of this framework has been extensively evaluated in terms of the detection accuracy, the area under the curve (AUC) and frame processing time. Simulation results and comparisons with ten relevant online and non-online frameworks demonstrate that our framework efficiently achieves superior performance to those frameworks, in which it presents high values for the accuracy while attaining simultaneously low values for the processing time.
Machine learning based augmented reality for improved learning application th...IJECEIAES
Detection of objects and their location in an image are important elements of current research in computer vision. In May 2020, Meta released its state-ofthe-art object-detection model based on a transformer architecture called detection transformer (DETR). There are several object-detection models such as region-based convolutional neural network (R-CNN), you only look once (YOLO) and single shot detectors (SSD), but none have used a transformer to accomplish this task. These models mentioned earlier, use all sorts of hyperparameters and layers. However, the advantages of using a transformer pattern make the architecture simple and easy to implement. In this paper, we determine the name of a chemical experiment through two steps: firstly, by building a DETR model, trained on a customized dataset, and then integrate it into an augmented reality mobile application. By detecting the objects used during the realization of an experiment, we can predict the name of the experiment using a multi-class classification approach. The combination of various computer vision techniques with augmented reality is indeed promising and offers a better user experience.
Automated Traffic sign board classification system is one of the key technologies of Intelligent
Transportation Systems (ITS). Traffic Surveillance System is being more and important with improving
urban scale and increasing number of vehicles. This Paper presents an intelligent sign board
classification method based on blob analysis in traffic surveillance. Processing is done by three main
steps: moving object segmentation, blob analysis, and classifying. A Sign board is modelled as a
rectangular patch and classified via blob analysis. By processing the blob of sign boards, the meaningful
features are extracted. Tracking moving targets is achieved by comparing the extracted features with
training data. After classifying the sign boards the system will intimate to user in the form of alarms,
sound waves. The experimental results show that the proposed system can provide real-time and useful
information for traffic surveillance.
Satellite Image Classification with Deep Learning Surveyijtsrd
Satellite imagery is important for many applications including disaster response, law enforcement and environmental monitoring etc. These applications require the manual identification of objects in the imagery. Because the geographic area to be covered is very large and the analysts available to conduct the searches are few, thus an automation is required. Yet traditional object detection and classification algorithms are too inaccurate and unreliable to solve the problem. Deep learning is a part of broader family of machine learning methods that have shown promise for the automation of such tasks. It has achieved success in image understanding by means that of convolutional neural networks. The problem of object and facility recognition in satellite imagery is considered. The system consists of an ensemble of convolutional neural networks and additional neural networks that integrate satellite metadata with image features. Roshni Rajendran | Liji Samuel ""Satellite Image Classification with Deep Learning: Survey"" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-2 , February 2020,
URL: https://www.ijtsrd.com/papers/ijtsrd30031.pdf
Paper Url : https://www.ijtsrd.com/engineering/computer-engineering/30031/satellite-image-classification-with-deep-learning-survey/roshni-rajendran
Deep-learning based single object tracker for night surveillance IJECEIAES
Tracking an object in night surveillance video is a challenging task as the quality of the captured image is normally poor with low brightness and contrast. The task becomes harder for a small object as fewer features are apparent. Traditional approach is based on improving the image quality before tracking is performed. In this paper, a single object tracking algorithm based on deep-learning approach is proposed to exploit its outstanding capability of modelling object’s appearance even during night. The algorithm uses pre-trained convolutional neural networks coupled with fully connected layers, which are trained online during the tracking so that it is able to cater for appearance changes as the object moves around. Various learning hyperparameters for the optimization function, learning rate and ratio of training samples are tested to find optimal setup for tracking in night scenarios. Fourteen night surveillance videos are collected for validation purpose, which are captured from three viewing angles. The results show that the best accuracy is obtained by using Adam optimizer with learning rate of 0.00075 and sampling ratio of 2:1 for positive and negative training data. This algorithm is suitable to be implemented in higher level surveillance applications such as abnormal behavioral recognition.
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.
Object gripping algorithm for robotic assistance by means of deep leaning IJECEIAES
This paper exposes the use of recent deep learning techniques in the state of the art, little addressed in robotic applications, where a new algorithm based on Faster R-CNN and CNN regression is exposed. The machine vision systems implemented, tend to require multiple stages to locate an object and allow a robot to take it, increasing the noise in the system and the processing times. The convolutional networks based on regions allow one to solve this problem, it is used for it two convolutional architectures, one for classification and location of three types of objects and one to determine the grip angle for a robotic gripper. Under the establish virtual environment, the grip algorithm works up to 5 frames per second with a 100% object classification, and with the implementation of the Faster R-CNN, it allows obtain 100% accuracy in the classifications of the test database, and over a 97% of average precision locating the generated boxes in each element, gripping successfully the objects.
REGISTRATION TECHNOLOGIES and THEIR CLASSIFICATION IN AUGMENTED REALITY THE K...IJCSEA Journal
The registration in augmented reality is process which merges virtual objects generated by computer with
real world image caught by camera. This paper describes the knowledge-based registration, computer
vision-based registration and tracker-based registration technology. This paper mainly focused on trackerbased registration technology in augmented reality. Also described method in tracker- based technology,
problem and solution.
Real Time Detection of Moving Object Based on Fpgaiosrjce
IOSR Journal of Electronics and Communication Engineering(IOSR-JECE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of electronics and communication engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in electronics and communication engineering. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Similar to A SURVEY ON OBJECT DETECTION METHODS IN VISUAL SENSOR NETWORKS (20)
Saudi Arabia stands as a titan in the global energy landscape, renowned for its abundant oil and gas resources. It's the largest exporter of petroleum and holds some of the world's most significant reserves. Let's delve into the top 10 oil and gas projects shaping Saudi Arabia's energy future in 2024.
We have compiled the most important slides from each speaker's presentation. This year’s compilation, available for free, captures the key insights and contributions shared during the DfMAy 2024 conference.
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdffxintegritypublishin
Advancements in technology unveil a myriad of electrical and electronic breakthroughs geared towards efficiently harnessing limited resources to meet human energy demands. The optimization of hybrid solar PV panels and pumped hydro energy supply systems plays a pivotal role in utilizing natural resources effectively. This initiative not only benefits humanity but also fosters environmental sustainability. The study investigated the design optimization of these hybrid systems, focusing on understanding solar radiation patterns, identifying geographical influences on solar radiation, formulating a mathematical model for system optimization, and determining the optimal configuration of PV panels and pumped hydro storage. Through a comparative analysis approach and eight weeks of data collection, the study addressed key research questions related to solar radiation patterns and optimal system design. The findings highlighted regions with heightened solar radiation levels, showcasing substantial potential for power generation and emphasizing the system's efficiency. Optimizing system design significantly boosted power generation, promoted renewable energy utilization, and enhanced energy storage capacity. The study underscored the benefits of optimizing hybrid solar PV panels and pumped hydro energy supply systems for sustainable energy usage. Optimizing the design of solar PV panels and pumped hydro energy supply systems as examined across diverse climatic conditions in a developing country, not only enhances power generation but also improves the integration of renewable energy sources and boosts energy storage capacities, particularly beneficial for less economically prosperous regions. Additionally, the study provides valuable insights for advancing energy research in economically viable areas. Recommendations included conducting site-specific assessments, utilizing advanced modeling tools, implementing regular maintenance protocols, and enhancing communication among system components.
Cosmetic shop management system project report.pdfKamal Acharya
Buying new cosmetic products is difficult. It can even be scary for those who have sensitive skin and are prone to skin trouble. The information needed to alleviate this problem is on the back of each product, but it's thought to interpret those ingredient lists unless you have a background in chemistry.
Instead of buying and hoping for the best, we can use data science to help us predict which products may be good fits for us. It includes various function programs to do the above mentioned tasks.
Data file handling has been effectively used in the program.
The automated cosmetic shop management system should deal with the automation of general workflow and administration process of the shop. The main processes of the system focus on customer's request where the system is able to search the most appropriate products and deliver it to the customers. It should help the employees to quickly identify the list of cosmetic product that have reached the minimum quantity and also keep a track of expired date for each cosmetic product. It should help the employees to find the rack number in which the product is placed.It is also Faster and more efficient way.
6th International Conference on Machine Learning & Applications (CMLA 2024)ClaraZara1
6th International Conference on Machine Learning & Applications (CMLA 2024) will provide an excellent international forum for sharing knowledge and results in theory, methodology and applications of on Machine Learning & Applications.
Understanding Inductive Bias in Machine LearningSUTEJAS
This presentation explores the concept of inductive bias in machine learning. It explains how algorithms come with built-in assumptions and preferences that guide the learning process. You'll learn about the different types of inductive bias and how they can impact the performance and generalizability of machine learning models.
The presentation also covers the positive and negative aspects of inductive bias, along with strategies for mitigating potential drawbacks. We'll explore examples of how bias manifests in algorithms like neural networks and decision trees.
By understanding inductive bias, you can gain valuable insights into how machine learning models work and make informed decisions when building and deploying them.
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...ssuser7dcef0
Power plants release a large amount of water vapor into the
atmosphere through the stack. The flue gas can be a potential
source for obtaining much needed cooling water for a power
plant. If a power plant could recover and reuse a portion of this
moisture, it could reduce its total cooling water intake
requirement. One of the most practical way to recover water
from flue gas is to use a condensing heat exchanger. The power
plant could also recover latent heat due to condensation as well
as sensible heat due to lowering the flue gas exit temperature.
Additionally, harmful acids released from the stack can be
reduced in a condensing heat exchanger by acid condensation. reduced in a condensing heat exchanger by acid condensation.
Condensation of vapors in flue gas is a complicated
phenomenon since heat and mass transfer of water vapor and
various acids simultaneously occur in the presence of noncondensable
gases such as nitrogen and oxygen. Design of a
condenser depends on the knowledge and understanding of the
heat and mass transfer processes. A computer program for
numerical simulations of water (H2O) and sulfuric acid (H2SO4)
condensation in a flue gas condensing heat exchanger was
developed using MATLAB. Governing equations based on
mass and energy balances for the system were derived to
predict variables such as flue gas exit temperature, cooling
water outlet temperature, mole fraction and condensation rates
of water and sulfuric acid vapors. The equations were solved
using an iterative solution technique with calculations of heat
and mass transfer coefficients and physical properties.
Student information management system project report ii.pdfKamal Acharya
Our project explains about the student management. This project mainly explains the various actions related to student details. This project shows some ease in adding, editing and deleting the student details. It also provides a less time consuming process for viewing, adding, editing and deleting the marks of the students.
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Dr.Costas Sachpazis
Terzaghi's soil bearing capacity theory, developed by Karl Terzaghi, is a fundamental principle in geotechnical engineering used to determine the bearing capacity of shallow foundations. This theory provides a method to calculate the ultimate bearing capacity of soil, which is the maximum load per unit area that the soil can support without undergoing shear failure. The Calculation HTML Code included.
A SURVEY ON OBJECT DETECTION METHODS IN VISUAL SENSOR NETWORKS
1. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
DOI:10.5121/ijassn.2016.6201 1
A SURVEY ON OBJECT DETECTION METHODS IN
VISUAL SENSOR NETWORKS
Shamim Yousefi and Samad Najjar Ghabel
Faculty of Electrical and Computer Engineering, University of Tabriz, 29th Bahman
Boulevard, Tabriz, Iran, Islamic Republic of
ABSTRACT
Object detection is one of the major challenges in visual sensor networks (VSNs) which is set up in the
monitoring applications. Many approaches proposed to solve the object detection problem in VSNs,
considering diverse metrics such as reliability, energy consumption, detection accuracy and being real-
time. In this paper, a survey on the object detection methods in visual sensor networks is presented for the
first time. Furthermore, this paper classified the methods precisely. Two main object detection categories
in VSNs that explored in this paper are conventional object detection methods and object detection
approaches with the camera nodes involvement. To be more precise, presented survey promotes an
overview of recent object detection methods' literature with their performance evaluation. Also, this
research is challenging and the object detection issue in the visual sensor networks is open caused by
differences in estimations and performance metrics. Therefore, the survey concludes with open research
challenges.
KEYWORDS
Detection Accuracy, Energy Consumption, Object Detection, Object Recognition, Visual Sensor Networks
1. INTRODUCTION
Recent advances in low-power and resource constraining self-organizing sensor nodes have led to
the development of visual sensor networks (VSNs) [1, 2, 3]. The camera nodes that constitute the
visual sensor networks are able to capture the multimedia data from the monitoring area in the
form of conventional or infrared images and video streaming. Therefore, VSNs can lead to the
development of monitoring and surveillance applications such as traffic control systems [4, 5],
person locator services [6], industrial process control systems, seismic sensing and hazardous
environment exploration [7] automated assistance for the elderly and family monitoring,
biomedical health monitoring [8, 9] and virtual reality [10].
On most of the expressed applications, object detection and recognition are one of the major
challenging issues in the visual sensor networks [11]. Two principal object detection categories in
visual sensor networks that explored in this paper are conventional object detection methods and
object detection approaches with the camera nodes involvement. The conventional object
detection approaches in VSNs used various types of the background subtraction techniques [12]
to recognize the difference between the background and foreground images. In such method, if
the background subtraction result exceeds a pre-determined threshold, the camera node will
detect moving objects within the monitoring area and sends the captured/difference image to the
base station for recognition operations [13, 14]. In visual sensor networks, the data
communication cost is usually much higher than the image processing cost [15]. Therefore, the
2. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
2
conventional object detection approaches are not suitable for monitoring and surveillance
applications.
For reducing data transmission cost, some works involves the camera nodes to perform
preprocessing after background subtraction and transmit only the bounding box of the objects to
the network [16]. In these methods, if there are some non-object pixels between the objects in the
foreground image, they are sent to the base station, too. As a result, this type of methods will be
suitable on applications for detecting a single object. Some other works involve the camera nodes
to perform preprocessing tasks, after background subtraction and send only the features or key-
points of the objects to the base station [17, 18, 19]. The overall taxonomy of the object detection
methods in visual sensor networks are shown in Figure 1.
Figure 1: Overall Taxonomy of the Object Detection Methods in VSNs
In the recent papers, various approaches for object detection in visual sensor networks are
presented, but there is a deficiency of well-defined grouping of them based on applications'
requirements. So, this survey presents taxonomy of object detection methods and discusses each
approach under the appropriate category. The rest of the paper is organized as following: Section
2 provides a more detailed survey about the object detection methods in visual sensor networks
and categorizes the approaches. In Section 3, reviewed methods are compared and the paper is
concluded based on VSNs' applications.
2. TAXONOMY OF OBJECT DETECTION APPROACHES IN VSNS
Two main object detection categories in visual sensor networks that are examined in this section
are conventional object detection methods and object detection methods with the camera nodes
involvement. The conventional object detection approaches in VSNs used various types of
background subtraction techniques [12] to recognize the moving objects in the camera node's
field-of-view. In such methods, camera nods send the captured/difference image that include
objects to the base station [13, 14]. While object detection methods with the camera nodes
involvement perform preprocessing tasks after background subtraction and send only the
bounding box of the objects or the useful information of them into the network [16, 17, 18, 19].
2.1. CONVENTIONAL OBJECT DETECTION METHODS
In most of the traditional visual sensor networks' applications, background subtraction techniques
[12] are one of the common approaches to detect the presence of moving objects. These
approaches are based on the difference between the background and foreground images [20]. At
first, each camera node captures a reference image from its field-of view without any moving
objects, which called background image [21]. After saving the background image, the camera
nodes periodically capture the foreground image [22] from its field-of view.
3. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
3
In background subtraction approaches for object detection in VSNs, each camera node subtracts
the background and foreground images. If the difference between the background and foreground
images exceeds a pre-determined threshold, the camera node will detect moving objects within its
monitoring area and sends the foreground/difference image to the base station for recognition and
classification operations [13, 14].
2.1.1 OBJECT DETECTION USING THE BACKGROUND SUBTRACTION AND IMAGE
RECONSTRUCTION
In most of the machine vision applications [23], detecting the moving objects in the camera
nodes' field-of view is one of the very important issues. To reduce the energy consumption of the
network, Kenchannavar et al. [13] suggested that the camera nodes must process the captured
image, and then send the useful information to the base station.
As shown in Figure 2, each camera node has a background image and periodically captures the
images from its monitoring area. The moving objects are detected using a background subtraction
technique with some pre-determined threshold values. If the background subtraction result
exceeds the pre-determined threshold value, the camera node will detect moving objects within
the monitoring area and calculates the energy that required for subtraction. Then, the difference
image is sent to the base station using TCP/IP protocol. The foreground image is reconstructed
using the background image, difference image and the median filter [24] at the server and the
energy consumed during reconstruction and transmission is calculated.
Figure 2: Block Diagram for the Camera node and server [13]
The results of authors' implementations show that the visual sensor network's bandwidth [25] in
the cases that the captured images have been processed is efficient as compared to without
processing the images, which in turn increase the network's lifetime [26]. However, in
subtraction/reconstruction approaches, non-object pixels with zero value (black) are injected into
the network, when the background subtraction result is sent to the base station. Therefore, the
transmission energy of the network is increased and its performance will not be acceptable.
Furthermore, background subtraction-based object detection methods are sensitive to
environmental factors like luminance [18].
4. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
4
2.1.2 MULTI-OBJECT DETECTION USING THE HAAR-LIKE FEATURES
To accelerate multi-object detection in VSNs with low training data, Vaidehi et al. [14] have been
proposed a method using Haar-like features [27] and Joint-Boosting algorithm [28]. The
foundation of work is that each camera node has a background image and periodically captures
the foreground images from its field-of-view. If the difference between the background and
foreground images exceeds a pre-determined threshold, foreground image is sent to the base
station for extra recognition tasks.
As shown in Figure 3, the foreground images received are treated as the inputs to the multi-object
detection system in the base station. In first step of the multi-object detection approach, the
integral images [29] are defined to compute the Haar-like features fast. The integral image is a
matrix in which its parts include sum of all the pixels in the left-upper part of the foreground
image. The rectangle areas and their Haar-like features are directly obtained in the integral image.
Haar-like features are the input of the decision tree classifiers. In the next step, after a strong
classifier is trained, it can be applied to the foreground image's regions for detecting the
object/non-object sub-window.
Figure 3: Multi-object Detection System Architecture [14]
The simulation results of the author's implementation proved that classifier requires low training
data by using Haar-like features, which in turn accelerate multi-object detection. Furthermore, the
paper shows that detection system recognizes all instances of the objects regardless of their scale
and location. However, further investigation states non-object pixels are injected into the visual
sensor network, when all the foreground image's pixels are sent to the base station. Therefore, the
VSNs' lifetime is decreased and proved inefficiency of the multi-object detection method.
2.2. OBJECT DETECTION METHODS WITH THE CAMERA NODES INVOLVEMENT
It is worth mentioning that calculating the difference between the background and foreground
images are the basis task of all object detection methods in visual sensor networks. However,
analyzing the conventional object detection methods shows using only the background
subtraction techniques increases the network's energy consumption, which in turn decrease the
lifetime of the camera nodes. To reduce data transmission cost in VSNs, some of the existence
approaches involves the camera nodes to perform various preprocessing tasks after background
subtraction and transmit only the useful information such as bounding box, features or key-points
of the objects to the base station [16, 17, 18, 19].
5. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
5
2.2.1 EXTRACTING BOUNDING BOX OF THE OBJECTS
To detect objects using minimum hardware and improve the detection costs, Pham et al. [16]
have suggested an efficient approach for extracting the bounding box of the objects in camera
nodes. In bounding box detection method, each stable camera node captures a background image
and converts to gray-scale one. Then, the camera nodes periodically capture foreground images
from own field-of-view and convert to gray-scale ones, too. The moving objects are detected
using a background subtraction technique with some pre-determined threshold values. If the
difference between background and foreground images exceeds the pre-determined threshold, the
result of background subtraction that is a black-and-white image with only values of 0 and 1, will
be treated as the input of the bounding box extraction step. As shown in Figure 4, bounding box
extraction step involves row and column scans to detect whether the number of consecutive
differences is greater than a pre-determined threshold or not (difference threshold for objects'
length and width). During row scans, the first pixel location of the first threshold consecutive
differences and the last pixel location of the last threshold consecutive differences are recorded as
the row edges of bounding box. Similarly, in column scans, the first pixel location of the first
threshold consecutive differences and the last pixel location of the last threshold consecutive
differences have been recorded as the column edges of bounding box. The extracted rows and
columns' edges are used for determining bounding box of the objects in the colored foreground
image.
Figure 4: Extracting the Bounding Box of Objects: (A) Black-and-white Subtraction Result, (B) Rows and
Columns Scan, (C) Bounding Box of the Objects [16]
It is worth mentioning that the noises in the image cause finding some false targets in the most of
the object detection approaches [30]. The results of the authors' implementation show that images'
noises are usually in small groups of consecutive pixels and the approach discards them.
Furthermore, the paper proves that the transmission costs in visual sensor networks are more than
processing ones. So, preprocessing images in the camera nodes and sending only the bounding
box of the objects to the base station increase the network lifetime, significantly. However,
further investigation shows extracting each object's bounding box eliminates non-object pixels in
the overall bounding box and decreases the injected traffic into the VSNs. In other words,
bounding box extraction approach can be suitable for only single object detection. Furthermore,
in the human detection applications, extracting the bounding box of each face [31] and sending it
to the base station is adequate to satisfy the recognition tasks requirements.
To further improvement in visual sensor networks' lifetime, some works have been proposed
various low-complexity face detection methods in camera nodes for detecting the existence faces
in the extracted bounding box. Yousefi et al. [32] have suggested an energy-aware multi-object
method that works based on extracting the bounding box of the objects and Boosting-based face
detection algorithm. The simulation results demonstrated that face detection method injects low
volume of traffic into the network and saves camera nodes energy. However, the complexity of
the Boosting-based face detection algorithm depends on the size of the input boxes and non-
object pixels as the detected objects increases the size of the input boxes, which in turn raise the
6. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
6
processing complexity in the camera nodes. In the other hands, face detection methods are
suitable only for the VSNs' applications with the aim of human recognition.
2.2.2 EXTRACTING OBJECTS' FEATURES OR KEY-POINTS
Some object detection approaches in visual sensor networks involve the camera nodes to perform
preprocessing tasks, after background subtraction and sent only the features or key-points of the
objects to the base station [17, 18, 19]. The qualities of the detected objects are accelerated by
using the features or key-points extraction methods, especially in the applications that the
distance between the camera node and objects is changing, continuously. Furthermore, extracting
the features or key-points of the objects and injecting them into the networks decrease the
processing and transmission costs, these in turn improve the performance of the visual sensor
networks.
2.2.2.1 OBJECT DETECTION METHODS BASED OF BINARY ROBUST INVARIANT SCALABLE
KEY-POINTS
To maximize the quality of reconstructed pixel-domain representation under limited resources
such as bandwidth and processing power, some approaches have been suggested that camera
nodes extract the main features required for object recognition using Binary Robust Invariant
Scalable Key-points (BRISK) and send them to the base station for extra recognition analysis.
Object detection approach based on BRISK processes the foreground image to recognize a
number of salient key-points that correspond to very different pixels of the underlying image.
Finally, descriptors detected from the foreground image are matched with a set of descriptors
extracted from a database of reference images. Therefore, a ranked list with the most relevant
results is returned.
To use the BRISK for detecting objects, Redondi et al. [17] have extended Binary Robust
Independent Elementary Features (BRIEF) designing to fix it against scale and rotation
transformations. As shown in Figure 5, the camera nodes are responsible for capturing images,
performing key-point detection tasks and finally, transmitting the descriptors to the base station.
The base station performs object recognition leveraging the descriptors received from the camera
node. The relay nodes only perform information communication and routing tasks.
Figure 5: Architecture of Object Detection by BRISK [17]
Paper's simulation results prove that the processing time depends on the images resolution and the
number of highlight key-points. Furthermore, the paper shows object detection accuracy and data
7. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
7
transmission duration is accelerated by using the BRISK to detect the objects. However, using the
local features such as BRISK for object detection in visual sensor networks has not a perfect
performance to predict and formulate the lost data.
To solve the control problem and balance the processing loads in a visual sensor network, some
works have used the temporal correlation in video sequences [19]. The basis of the work is
distributing the processing load by allocating sub-areas of the images to the camera nodes.
Therefore, the threshold and cut-point are estimated for each image, and then the optimal values
of parameters via autoregressive models are predicted. The analytical results of the paper show
that prediction-based methods reach the detection threshold and cut-points in a persuasive
performance. Furthermore, achieving low computational complexity make them a group of
convenient ways to control and balance the processing load on the local feature detection in
VSNs. However, further investigations prove that prediction-based methods will not have an
acceptable performance in visual sensor networks' monitoring and surveillance applications.
2.2.2.2 OBJECT DETECTION METHODS BASED ON ADAPTIVE GAUSSIAN MIXTURE MODEL
To eliminate the influence of environmental factors such as brightness variations, some works
[18] have proposed object detection methods in visual sensor networks based on adaptive
Gaussian mixture model. As shown in Figure 6, the first step of object detection in the camera
nodes is frame reconstruction, which reduces the image size by parting it into blocks with a pre-
determined size. Then, average color value of each image block is computed in a RGB color
space and replaced for the corresponding block. In the second step, background modeling is
performed. The static pixels may only be modeled by a Gaussian member, while other pixels,
which are non-static, should be modeled by multiple Gaussian mixture components. Finally, to
detect whether a pixel matches a component of the Gaussian, sort the Gaussian components, and
then compare them one by one with the corresponding pixel. The pixels are detected as
background pert if matches a component of the Gaussian and foreground one, otherwise.
Figure 6: A. Foreground Image, B. Frame Reconstruction, C. Object Detection in Reconstructed Image and
D. Object Detection in Foreground Image.
Simulation results of the paper show adaptive Gaussian mixture model minimizes the processing
cost and the influence of environmental factors. So, it is suitable for many detection-based
applications in VSNs. However, further investigations prove the faces information is adequate for
detecting the objects, which are humans. Therefore, it is proved transmitting the non-face
information of the humans into the network reduces its lifetime.
3. CONCLUSION AND OPEN ISSUES
In this paper, the importance of object detection and recognition in visual sensor networks was
discussed, which is set up in the monitoring and surveillance applications. Many different
approaches have been proposed to solve the object detection problem in VSNs, considering
8. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
8
diverse metrics such as reliability, energy consumption, object detection accuracy and being real-
time. In this paper, a survey on the object detection methods in visual sensor networks is
presented for the first time. Two principal object detection categories in VSNs that explored in
this paper were conventional object detection methods and object detection approaches with the
camera nodes involvement. To be more precise, presented survey promoted an overview of recent
object detection methods' literature with their performance evaluation.
Table 1 illustrates a summary of approaches discussed in this paper utilizing purpose and their
cons and pros. It clearly indicates object detection approaches with the camera nodes involvement
increase the preprocessing and transmission costs partially, which in turn accelerate lifetime of
the camera nodes. Furthermore, table 1 shows that sending the face's information of the objects to
the base station is adequate, when the objects are humans.
One of the important open research issues for object detection-based applications in visual sensor
networks include decreasing the transmission energy of the networks and raising their lifetime.
This purpose is achieved by increasing low cost preprocessing tasks in camera nodes and sending
only the useful information (such as only the faces' information of the humans) to the base
station. Another interesting issue for object detection approaches in visual sensor networks is
energy efficient missing objects recovery, when the camera nodes fail. Therefore, fault tolerant
object detection methods must be designed for VSNs to maximize the detection accuracy with
minimum hardware.
Table 1 Classification of Object Detection Approaches
Methods
Purpose of
Methods
Advantage(s) Disadvantage(s)
Object Detection
using
background
subtraction
techniques [13,
33]
Detecting the
moving object
bandwidth usage
reduction
Influencing of
environmental factors,
High energy consumption
Object detection
using Haar-like
features [14]
Detecting objects
without size
limitation
high-precision
object detection,
speed acceleration
High transmission energy
Extracting the
bounding box of
the objects [16]
High speed object
detection with
minimum hardware
Increasing residual
energy of the
network
Not perfect in reducing
the transmission costs
Extracting the
face's
information of
the objects [32,
34]
increasing
preprocessing tasks
in camera nodes
Decreasing injected
traffic into the
network, Increasing
network' lifetime
high processing and
transmission costs
Object detection
using BRISK
[17, 35]
Maximizing the
quality of pixel-
domain display by
limited resources
Optimizing
processing time,
Increasing
detection accuracy
Inefficiencies in the
prediction of lost
information
Object detection
by adaptive
Gaussian mixture
model [18]
Decreasing costs,
Eliminating
environmental
factors
Reliable object
detection
Sending objects'
information to the base
station instead of faces'
one
9. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
9
REFERENCES
[1] Soro, S., Heinzelman, W. "A survey of visual sensor networks." Advances in Multimedia 2009
(2009).
[2] Akyildiz, I.F., Melodia, T. and Chowdhury, K.R. "A survey on wireless multimedia sensor
networks." Computer networks 51.4 (2007): 921-960.
[3] Akyildiz, I.F., Melodia, T. and Chowdury, K.R. "Wireless multimedia sensor networks: A survey."
Wireless Communications, IEEE 14.6 (2007): 32-39.
[4] Tavli, Bulent, et al. "A survey of visual sensor network platforms." Multimedia Tools and
Applications 60.3 (2012): 689-726.
[5] Misra, Satyajayant, Martin Reisslein, and Guoliang Xue. "A survey of multimedia streaming in
wireless sensor networks." Communications Surveys & Tutorials, IEEE 10.4 (2008): 18-39.
[6] Feller, Steven D., et al. "Tracking and imaging humans on heterogeneous infrared sensor arrays for
law enforcement applications." AeroSense 2002. International Society for Optics and Photonics,
2002.
[7] Wener-Allen, G., et al. "Deploying a wireless sensor network on an active volcano, Data-Driven
Applications in Sensor Networks (Special Issue)." IEEE Internet Computing 2 (2006): 18-25.
[8] Gao, Tia, et al. "Vital signs monitoring and patient tracking over a wireless network." Engineering in
Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference of
the. IEEE, 2006.
[9] Lorincz, Konrad, et al. "Sensor networks for emergency response: challenges and opportunities."
Pervasive Computing, IEEE 3.4 (2004): 16-23.
[10] Charfi, Youssef, Naoki Wakamiya, and Masayuki Murata. "Challenging issues in visual sensor
networks." Wireless Communications, IEEE 16.2 (2009): 44-49.
[11] Hu, Weiming, et al. "A survey on visual surveillance of object motion and behaviors." Systems,
Man, and Cybernetics, Part C: Applications and Reviews, IEEE Transactions on 34.3 (2004): 334-
352.
[12] Piccardi, Massimo. "Background subtraction techniques: a review." In Systems, man and
cybernetics, 2004 IEEE international conference on, vol. 4, pp. 3099-3104. IEEE, 2004.
[13] Kenchannavar, Harish H., Sushma S. Kudtarkar, and U. P. Kulkarni. "Energy Efficient Data
Processing In Visual Sensor Network." International Journal of CS & IT (2010).
[14] Vaidehi, V., et al. "Multiclass object detection system in imaging sensor network using haar-like
features and joint-boosting algorithm." Recent Trends in Information Technology (ICRTIT), 2011
International Conference on. IEEE, 2011.
[15] Canclini, A., et al. "Object recognition in visual sensor networks based on compression and
transmission of binary local features."
[16] Pham, Duc Minh, and Syed Mahfuzul Aziz. "Object extraction scheme and protocol for energy
efficient image communication over Wireless Sensor Networks." Computer Networks 57.15 (2013):
2949-2960.
[17] Redondi, Alessandro, Luca Baroffio, Antonio Canclini, Matteo Cesana, and M. Tagliasacchi. "A
visual sensor network for object recognition: Testbed realization." In Proc. of International
Conference on Digital Signal Processing (DSP). 2013.
[18] Wang, Yong, Dianhong Wang, and Wu Fang. "Automatic node selection and target tracking in
wireless camera sensor networks." Computers & Electrical Engineering 40, no. 2 (2014): 484-493.
[19] Eriksson, Emil, György Dán, and Viktoria Fodor. "Prediction-based load control and balancing for
feature extraction in visual sensor networks." Acoustics, Speech and Signal Processing (ICASSP),
2014 IEEE International Conference on. IEEE, 2014.
[20] Sobral, Andrews, and Antoine Vacavant. "A comprehensive review of background subtraction
algorithms evaluated with synthetic and real videos."Computer Vision and Image Understanding 122
(2014): 4-21.
[21] Connell, Jonathan H. "Method and apparatus for maintaining a background image model in a
background subtraction system using accumulated motion." U.S. Patent No. 8,630,442. 14 Jan. 2014.
[22] Szwoch, Grzegorz. "Extraction of stable foreground image regions for unattended luggage
detection." Multimedia Tools and Applications (2014): 1-26.
[23] Sonka, Milan, Vaclav Hlavac, and Roger Boyle. Image processing, analysis, and machine vision.
Cengage Learning, 2014.
10. International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 6, No.2, April 2016
10
[24] Wang, Zhou, and David Zhang. "Progressive switching median filter for the removal of impulse
noise from highly corrupted images." Circuits and Systems II: Analog and Digital Signal Processing,
IEEE Transactions on 46.1 (1999): 78-80.
[25] Urgaonkar, Bhuvan, and Prashant Shenoy. "Sharc: Managing CPU and network bandwidth in shared
clusters." Parallel and Distributed Systems, IEEE Transactions on 15.1 (2004): 2-17.
[26] Cardei, Mihaela, and Ding-Zhu Du. "Improving wireless sensor network lifetime through power
aware organization." Wireless Networks 11.3 (2005): 333-340.
[27] Mita, Takeshi, Toshimitsu Kaneko, and Osamu Hori. "Joint haar-like features for face detection."
Computer Vision, 2005. ICCV 2005. Tenth IEEE International Conference on. Vol. 2. IEEE, 2005.
[28] Mita, Takeshi, Toshimitsu Kaneko, and Osamu Hori. "Joint haar-like features for face detection."
Computer Vision, 2005. ICCV 2005. Tenth IEEE International Conference on. Vol. 2. IEEE, 2005.
[29] Lindsay, J. B., J. M. H. Cockburn, and H. A. J. Russell. "An integral image approach to performing
multi-scale topographic position analysis." Geomorphology (2015).
[30] Netravali, Arun. Digital pictures: representation and compression. Springer Science & Business
Media, 2013.
[31] Wechsler, Harry, et al., eds. Face recognition: From theory to applications. Vol. 163. Springer
Science & Business Media, 2012.
[32] Yousefi, SH., Aghdasi S.H. "Energy Aware Multi-object Detection Method in Visual Sensor
Networks", 5th International Conference on Computer and Knowledge Engineering (ICCKE), 2015.
[33] Tavli, Bulent, et al. "A survey of visual sensor network platforms." Multimedia Tools and
Applications 60.3 (2012): 689-726.
[34] Yildiz, Huseyin Ugur, et al. "Maximizing Wireless Sensor Network lifetime by
communication/computation energy optimization of non-repudiation security service: Node level
versus network level strategies." Ad Hoc Networks 37 (2016): 301-323.
[35] Li, Wei, et al. "Multiple feature points representation in target localization of wireless visual sensor
networks." Journal of Network and Computer Applications 57 (2015): 119-128.
Authors
Shamim Yousefi received the B.S. degree in Information Technology Engineering and
M.S. degree in Computer Engineering (Software) from University of Tabriz, Tabriz, Iran
in 2013 and 2015, respectively. Currently, she is working as a researcher at the Wireless
Ad hoc and Sensor networks research Laboratory (WASL) in University of Tabriz. Her
current interests include light weight methods for object detection and recognition in
visual sensor networks.
Samad Najjar Ghabel is a lecturer at University of Mohaghegh Ardebili, Ardebil, Iran.
He received the B.S. degree in Computer Engineering (Software) from University of
Mohaghegh Ardebili, Ardebil, Iran and M.S. degree in Computer Engineering
(Software) from University of Tabriz, Tabriz, Iran in 2013 and 2015, respectively. His
main interests are Visual sensor networks, Computer Networks, networks security,
developing and modeling software.