This document summarizes a research paper that presents a hybrid approach for detecting and localizing color text in natural scene images. The approach uses both region-based and connected component-based methods. In the preprocessing stage, a text region detector is used to detect text regions and generate candidate text components. A conditional random field model combines unary component properties and binary contextual relationships to filter non-text components. Finally, neighboring text components are grouped into text lines or words using a learning-based energy minimization method. The paper is evaluated on a natural scene image dataset and shows improvements over existing methods.
Text detection and recognition in scene images or natural images has applications in computer
vision systems like registration number plate detection, automatic traffic sign detection, image retrieval
and help for visually impaired people. Scene text, however, has complicated background, blur image,
partly occluded text, variations in font-styles, image noise and ranging illumination. Hence scene text
recognition could be a difficult computer vision problem. In this paper connected component method
is used to extract the text from background. In this work, horizontal and vertical projection profiles,
geometric properties of text, image binirization and gap filling method are used to extract the text from
scene images. Then histogram based threshold is applied to separate text background of the images.
Finally text is extracted from images.
A Texture Based Methodology for Text Region Extraction from Low Resolution Na...CSCJournals
Automated systems for understanding display boards are finding many applications useful in guiding tourists, assisting visually challenged and also in providing location aware information. Such systems require an automated method to detect and extract text prior to further image analysis. In this paper, a methodology to detect and extract text regions from low resolution natural scene images is presented. The proposed work is texture based and uses DCT based high pass filter to remove constant background. The texture features are then obtained on every 50x50 block of the processed image and potential text blocks are identified using newly defined discriminant functions. Further, the detected text blocks are merged and refined to extract text regions. The proposed method is robust and achieves a detection rate of 96.6% on a variety of 100 low resolution natural scene images each of size 240x320.
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.
Comparative analysis of c99 and topictiling text segmentation algorithmseSAT Journals
Abstract In this paper, the work done includes the extraction of information from image datasets which contain natural text. The difficulty level of segmenting natural text from an image is too high and so precision is the most important factor to be kept in mind. To minimize the error rates, error filtration technique is provided, as filtration is adopted while doing image segmentation basically text segmentation present in images. Furthermore, a comparative analysis of two different text segmentation algorithms namely C99 and TopicTiling on image documents is presented. To assess how well each algorithm works, each was applied on different datasets and results were compared. The work done also proves the efficiency of TopicTiling over C99. Index Terms: Text Segmentation, text extraction, image documents,C99 and TopicTiling.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
TEXT DETECTION AND EXTRACTION FROM VIDEOS USING ANN BASED NETWORKijscai
With fast intensification of existing multimedia documents and mounting demand for information indexing and retrieval, much endeavor has been done on extracting the text from images and videos. The prime intention of the projected system is to spot and haul out the scene text from video. Extracting the scene text from video is demanding due to complex background, varying font size, different style, lower resolution and blurring, position, viewing angle and so on. In this paper we put forward a hybrid method where the two most well-liked text extraction techniques i.e. region based method and connected component (CC) based method comes together. Initially the video is split into frames and key frames obtained. Text region indicator (TRI) is being developed to compute the text prevailing confidence and
candidate region by performing binarization. Artificial Neural network (ANN) is used as the classifier and Optical Character Recognition (OCR) is used for character verification. Text is grouped by constructing the minimum spanning tree with the use of bounding box distance.
Text detection and recognition in scene images or natural images has applications in computer
vision systems like registration number plate detection, automatic traffic sign detection, image retrieval
and help for visually impaired people. Scene text, however, has complicated background, blur image,
partly occluded text, variations in font-styles, image noise and ranging illumination. Hence scene text
recognition could be a difficult computer vision problem. In this paper connected component method
is used to extract the text from background. In this work, horizontal and vertical projection profiles,
geometric properties of text, image binirization and gap filling method are used to extract the text from
scene images. Then histogram based threshold is applied to separate text background of the images.
Finally text is extracted from images.
A Texture Based Methodology for Text Region Extraction from Low Resolution Na...CSCJournals
Automated systems for understanding display boards are finding many applications useful in guiding tourists, assisting visually challenged and also in providing location aware information. Such systems require an automated method to detect and extract text prior to further image analysis. In this paper, a methodology to detect and extract text regions from low resolution natural scene images is presented. The proposed work is texture based and uses DCT based high pass filter to remove constant background. The texture features are then obtained on every 50x50 block of the processed image and potential text blocks are identified using newly defined discriminant functions. Further, the detected text blocks are merged and refined to extract text regions. The proposed method is robust and achieves a detection rate of 96.6% on a variety of 100 low resolution natural scene images each of size 240x320.
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.
Comparative analysis of c99 and topictiling text segmentation algorithmseSAT Journals
Abstract In this paper, the work done includes the extraction of information from image datasets which contain natural text. The difficulty level of segmenting natural text from an image is too high and so precision is the most important factor to be kept in mind. To minimize the error rates, error filtration technique is provided, as filtration is adopted while doing image segmentation basically text segmentation present in images. Furthermore, a comparative analysis of two different text segmentation algorithms namely C99 and TopicTiling on image documents is presented. To assess how well each algorithm works, each was applied on different datasets and results were compared. The work done also proves the efficiency of TopicTiling over C99. Index Terms: Text Segmentation, text extraction, image documents,C99 and TopicTiling.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
TEXT DETECTION AND EXTRACTION FROM VIDEOS USING ANN BASED NETWORKijscai
With fast intensification of existing multimedia documents and mounting demand for information indexing and retrieval, much endeavor has been done on extracting the text from images and videos. The prime intention of the projected system is to spot and haul out the scene text from video. Extracting the scene text from video is demanding due to complex background, varying font size, different style, lower resolution and blurring, position, viewing angle and so on. In this paper we put forward a hybrid method where the two most well-liked text extraction techniques i.e. region based method and connected component (CC) based method comes together. Initially the video is split into frames and key frames obtained. Text region indicator (TRI) is being developed to compute the text prevailing confidence and
candidate region by performing binarization. Artificial Neural network (ANN) is used as the classifier and Optical Character Recognition (OCR) is used for character verification. Text is grouped by constructing the minimum spanning tree with the use of bounding box distance.
Texture features based text extraction from images using DWT and K-means clus...Divya Gera
Text extraction from different kind of images document, caption and scene text images. Discret wavelet transform was used to exract horizontal, vertical and diagonal features and k-means clustering was used to cluster the features into text and background cluster. For simple images k = 2 worked i.e. text and backgroud cluster while for complex images k=3 was used i.e. text cluster, complex background ad simple background.
A Survey On Thresholding Operators of Text Extraction In VideosCSCJournals
ideo indexing is an important problem that has interested by the communities of visual information in image processing. The detection and extraction of scene and caption text from unconstrained, general purpose video is an important research problem in the context of content-based retrieval and summarization. In this paper, the technique presented is for detection text from frames video. Finding the textual contents in images is a challenging and promising research area in information technology. Consequently, text detection and recognition in multimedia had become one of the most important fields in computer vision due to its valuable uses in a variety of recent technical applications. The work in this paper consists using morphological operations for extract text appearing in the video frames. The proposed scheme well as preprocessing to differentiate among where it as the high similarity between text and background information. Experimental results show that the resultant image is the image with only text. The evaluated criteria are applied with the image result and one obtained bay different operator.
Improved wolf algorithm on document images detection using optimum mean techn...journalBEEI
Detection text from handwriting in historical documents provides high-level features for the challenging problem of handwriting recognition. Such handwriting often contains noise, faint or incomplete strokes, strokes with gaps, and competing lines when embedded in a table or form, making it unsuitable for local line following algorithms or associated binarization schemes. In this paper, a proposed method based on the optimum threshold value and namely as the Optimum Mean method was presented. Besides, Wolf method unsuccessful in order to detect the thin text in the non-uniform input image. However, the proposed method was suggested to overcome the Wolf method problem by suggesting a maximum threshold value using optimum mean. Based on the calculation, the proposed method obtained a higher F-measure (74.53), PSNR (14.77) and lowest NRM (0.11) compared to the Wolf method. In conclusion, the proposed method successful and effective to solve the wolf problem by producing a high-quality output image.
Data reduction techniques for high dimensional biological dataeSAT Journals
Abstract
High dimensional biological datasets in recent years has been growing rapidly. Extracting the knowledge and analyzing highdimensional
biological data is one the key challenges in which variety and veracity are the two distinct characteristics. The
question that arises now is, how to perform dimensionality reduction for this heterogeneous data and how to develop a high
performance platform to efficiently analyze high dimensional biological data and how to find the useful things from this data. To
deeply discuss this issue, this paper begins with a brief introduction to data analytics available for biological data, followed by
the discussions of big data analytics and then a survey on various data reduction methods for biological data. We propose a dense
clustering algorithm for standard high dimensional biological data.
Keywords: Big Data Analytics, Dimensionality Reduction
Anatomical Survey Based Feature Vector for Text Pattern DetectionIJEACS
The vital objective of artificial intelligence is to discover and understand the human competences, one of which is the capability to distinguish several text objects within one or more images exhibited on any canvas including prints, videos or electronic displays. Multimedia data has increased rapidly in past years. Textual information present in multimedia contains important information about the image/video content. However it needs to technologically verify the commonly used human intelligence of detecting and differentiating the text within an image, for computers. Hence in this paper feature set based on anatomical study of human text detection system is proposed.
Text documents clustering using modified multi-verse optimizerIJECEIAES
In this study, a multi-verse optimizer (MVO) is utilised for the text document clus- tering (TDC) problem. TDC is treated as a discrete optimization problem, and an objective function based on the Euclidean distance is applied as similarity measure. TDC is tackled by the division of the documents into clusters; documents belonging to the same cluster are similar, whereas those belonging to different clusters are dissimilar. MVO, which is a recent metaheuristic optimization algorithm established for continuous optimization problems, can intelligently navigate different areas in the search space and search deeply in each area using a particular learning mechanism. The proposed algorithm is called MVOTDC, and it adopts the convergence behaviour of MVO operators to deal with discrete, rather than continuous, optimization problems. For evaluating MVOTDC, a comprehensive comparative study is conducted on six text document datasets with various numbers of documents and clusters. The quality of the final results is assessed using precision, recall, F-measure, entropy accuracy, and purity measures. Experimental results reveal that the proposed method performs competitively in comparison with state-of-the-art algorithms. Statistical analysis is also conducted and shows that MVOTDC can produce significant results in comparison with three well-established methods.
RECOGNITION OF HANDWRITTEN MEITEI MAYEK SCRIPT BASED ON TEXTURE FEATURE kevig
Recognition of Manipuri Script called Meitei Mayek is still in the infant stage due to its complex structure. In this paper, an attempt has been made to develop an offline Meitei Mayek handwritten character recognition model by exploiting the texture feature, Local Binary Pattern (LBP). The system has been developed and evaluated on a large dataset consisting of 3,780 characters which are collected from different people of varying age group. The highest recognition rate achieved by the proposed method is 93.33% using Support Vector Machine (SVM). So, the contribution of this paper is bi-fold: firstly, a collection of a large handwritten corpus of Meitei Mayek Script and secondly developing character recognition model on the collected dataset.
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
Texture features based text extraction from images using DWT and K-means clus...Divya Gera
Text extraction from different kind of images document, caption and scene text images. Discret wavelet transform was used to exract horizontal, vertical and diagonal features and k-means clustering was used to cluster the features into text and background cluster. For simple images k = 2 worked i.e. text and backgroud cluster while for complex images k=3 was used i.e. text cluster, complex background ad simple background.
A Survey On Thresholding Operators of Text Extraction In VideosCSCJournals
ideo indexing is an important problem that has interested by the communities of visual information in image processing. The detection and extraction of scene and caption text from unconstrained, general purpose video is an important research problem in the context of content-based retrieval and summarization. In this paper, the technique presented is for detection text from frames video. Finding the textual contents in images is a challenging and promising research area in information technology. Consequently, text detection and recognition in multimedia had become one of the most important fields in computer vision due to its valuable uses in a variety of recent technical applications. The work in this paper consists using morphological operations for extract text appearing in the video frames. The proposed scheme well as preprocessing to differentiate among where it as the high similarity between text and background information. Experimental results show that the resultant image is the image with only text. The evaluated criteria are applied with the image result and one obtained bay different operator.
Improved wolf algorithm on document images detection using optimum mean techn...journalBEEI
Detection text from handwriting in historical documents provides high-level features for the challenging problem of handwriting recognition. Such handwriting often contains noise, faint or incomplete strokes, strokes with gaps, and competing lines when embedded in a table or form, making it unsuitable for local line following algorithms or associated binarization schemes. In this paper, a proposed method based on the optimum threshold value and namely as the Optimum Mean method was presented. Besides, Wolf method unsuccessful in order to detect the thin text in the non-uniform input image. However, the proposed method was suggested to overcome the Wolf method problem by suggesting a maximum threshold value using optimum mean. Based on the calculation, the proposed method obtained a higher F-measure (74.53), PSNR (14.77) and lowest NRM (0.11) compared to the Wolf method. In conclusion, the proposed method successful and effective to solve the wolf problem by producing a high-quality output image.
Data reduction techniques for high dimensional biological dataeSAT Journals
Abstract
High dimensional biological datasets in recent years has been growing rapidly. Extracting the knowledge and analyzing highdimensional
biological data is one the key challenges in which variety and veracity are the two distinct characteristics. The
question that arises now is, how to perform dimensionality reduction for this heterogeneous data and how to develop a high
performance platform to efficiently analyze high dimensional biological data and how to find the useful things from this data. To
deeply discuss this issue, this paper begins with a brief introduction to data analytics available for biological data, followed by
the discussions of big data analytics and then a survey on various data reduction methods for biological data. We propose a dense
clustering algorithm for standard high dimensional biological data.
Keywords: Big Data Analytics, Dimensionality Reduction
Anatomical Survey Based Feature Vector for Text Pattern DetectionIJEACS
The vital objective of artificial intelligence is to discover and understand the human competences, one of which is the capability to distinguish several text objects within one or more images exhibited on any canvas including prints, videos or electronic displays. Multimedia data has increased rapidly in past years. Textual information present in multimedia contains important information about the image/video content. However it needs to technologically verify the commonly used human intelligence of detecting and differentiating the text within an image, for computers. Hence in this paper feature set based on anatomical study of human text detection system is proposed.
Text documents clustering using modified multi-verse optimizerIJECEIAES
In this study, a multi-verse optimizer (MVO) is utilised for the text document clus- tering (TDC) problem. TDC is treated as a discrete optimization problem, and an objective function based on the Euclidean distance is applied as similarity measure. TDC is tackled by the division of the documents into clusters; documents belonging to the same cluster are similar, whereas those belonging to different clusters are dissimilar. MVO, which is a recent metaheuristic optimization algorithm established for continuous optimization problems, can intelligently navigate different areas in the search space and search deeply in each area using a particular learning mechanism. The proposed algorithm is called MVOTDC, and it adopts the convergence behaviour of MVO operators to deal with discrete, rather than continuous, optimization problems. For evaluating MVOTDC, a comprehensive comparative study is conducted on six text document datasets with various numbers of documents and clusters. The quality of the final results is assessed using precision, recall, F-measure, entropy accuracy, and purity measures. Experimental results reveal that the proposed method performs competitively in comparison with state-of-the-art algorithms. Statistical analysis is also conducted and shows that MVOTDC can produce significant results in comparison with three well-established methods.
RECOGNITION OF HANDWRITTEN MEITEI MAYEK SCRIPT BASED ON TEXTURE FEATURE kevig
Recognition of Manipuri Script called Meitei Mayek is still in the infant stage due to its complex structure. In this paper, an attempt has been made to develop an offline Meitei Mayek handwritten character recognition model by exploiting the texture feature, Local Binary Pattern (LBP). The system has been developed and evaluated on a large dataset consisting of 3,780 characters which are collected from different people of varying age group. The highest recognition rate achieved by the proposed method is 93.33% using Support Vector Machine (SVM). So, the contribution of this paper is bi-fold: firstly, a collection of a large handwritten corpus of Meitei Mayek Script and secondly developing character recognition model on the collected dataset.
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
Sender Authentication with Transmission Power Adjustment Method Using RSSI in...IJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
Low Cost Self-assistive Voice Controlled Technology for Disabled PeopleIJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
Combating Bit Losses in Computer Networks using Modified Luby Transform CodeIJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
Load balancing in Content Delivery Networks in Novel Distributed EquilibriumIJMER
In today’s world’s to provide service to netizen’s with good availability of data, content
delivery networks (CDNs) must balance requests between servers while assigning clients to closet
servers. In this paper, we describe a new CDN design that associates artificial load-aware coordinates
with clients and data servers and uses them to direct content requests to cached data. This approach
helps achieve good accuracy and service when request workloads and resource availability in the CDN
are dynamic. A deployment and evaluation of our system on Planet Lab demonstrates how it achieves low
request times with high cache hit ratios when compared to other CDN approaches.
Analysis and Improved Operation of PEBB Based 5-Level Voltage Source Convert...IJMER
The paper presents the power-electronic devices are increasing in several applications, and
power-electronic building blocks (PEBBs) are a strategic concept to increase the reliability of the
power-electronic converters and to minimize their cost. Magnetic elements, such as zigzag
transformers, phase-shifted transformers (PST), or zero-sequence blocking transformers (ZSBT), are
used to interconnect the PEBBs. In this paper, by using 5-level voltage source converter the operation
of multi-pulse converters will be analyzed, describing the harmonic cancellation and minimization
techniques that could be used in these multi-pulse converters, focusing on the power-electronics flexible
ac transmission systems devices installed at the NYPA Marcy substation. In order to improve the
dynamic response of this system, the use of selective harmonic elimination modulation is proposed and
implemented
Complex test pattern generation for high speed fault diagnosis in Embedded SRAMIJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
Influence choice of the injection nodes of energy source on on-line losses of...IJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
A Comparative Study on Privacy Preserving Datamining TechniquesIJMER
Privacy protection is very important in the recent years for the reason of increasing in the
ability to store data. In particular, recent advances in the data mining field have lead to increased
concerns about privacy. Data in its original form, however, typically contains sensitive information about
individuals, and publishing such data will violate individual privacy. The current practice in data
publishing based on that what type of data can be released and use of that data. Recently, PPDM has
received immersed attention in research communities, and many approaches have been proposed for
different data publishing scenarios. In this comparative study we will systematically summarize and
evaluate different approaches for PPDM, study the challenges ,differences and requirements that
distinguish PPDM from other related problems, and propose future research directions
Prediction of groundwater quality in Selected Locations in Imo StateIJMER
The prediction of groundwater quality in selected locations was carried out in Owerri-West
L.G.A. of Imo State. The Physical, chemical and biological parameters of groundwater samples from
Nekede (Ward A), Ihiagwa (Ward B), Eziobodo (Ward C), Obinze (Ward D) and Avu (Ward E) were
analysed using the Atomic Absorption Spectrophotometer (AAS). A total of three replicates of fifteen
different borehole water samples were collected based on distances from closest potential sources of
contamination. All parameters were detected up to 61m from pollution source and most of them
increased in concentration during the periods, pointing to infiltrations from storm water. The results
for Iron, pH and TVC decreased as distance increases while for nitrate and BOD increased as distance
increases. Results also showed that most of the boreholes were polluted and not suitable for human
consumption without adequate treatment, Regular monitoring of groundwater quality, abolishment of
unhealthy waste disposal practices and introduction of modern techniques are recommended.
Deformation Analysis of a Triangular Mild Steel Plate Using CST as Finite Ele...IJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
International Journal of Modern Engineering Research (IJMER) covers all the fields of engineering and science: Electrical Engineering, Mechanical Engineering, Civil Engineering, Chemical Engineering, Computer Engineering, Agricultural Engineering, Aerospace Engineering, Thermodynamics, Structural Engineering, Control Engineering, Robotics, Mechatronics, Fluid Mechanics, Nanotechnology, Simulators, Web-based Learning, Remote Laboratories, Engineering Design Methods, Education Research, Students' Satisfaction and Motivation, Global Projects, and Assessment…. And many more.
Fatigue Performance in Grinding and Turning: An OverviewIJMER
This paper analysis the influence of Abrasive Flow Machining (AFM), Turning and Grinding on
fatigue performance of Fe250. Surface condition has a strong effect on fatigue life, and that most surfaces
produced by conventional manufacturing operations such as machining and forging have poor fatigue
behavior than polished surfaces commonly used for laboratory specimens. It is found that the surfaces
produced with different machining process and having the same surface roughness having different fatigue
performances. High –cycle fatigue data was obtained for Fe 250 using three types of machining process
viz, AFM, Turning and Grinding .S-N curve is plotted for the samples obtained with all the three process. It
was found that the samples produced with AFM having the highest fatigue life.
An Hybrid Learning Approach using Particle Intelligence Dynamics and Bacteri...IJMER
The foraging behavior of E. Coli is used for optimization problems. This paper is based on a
hybrid method that combines particle swarm optimization and bacterial foraging (BF) algorithm for
solution of optimization results. We applied this proposed algorithm on different closed loop transfer
functions and the performance of the system using time response for the optimum value of PID
parameters is studied with incorporating PSO method on mutation, crossover, step sizes, and chemotactic
of the bacteria during the foraging. The bacterial foraging particle swarm optimization (BFPSO)
algorithm is applied to tune the PID controller of type 2, 3 and 4 systems with consideration of minimum
peak overshoot and steady state error objective function. The performance of the time response is
evaluated for the designed PID controller as the integral of time weighted squared error. The results
illustrate that the proposed approach is more efficient and provides better results as compared to the
conventional PSO algorithm.
NEW ONTOLOGY RETRIEVAL IMAGE METHOD IN 5K COREL IMAGESijcax
Semantic annotation of images is an important research topic on both image understanding and database
or web image search. Image annotation is a technique to choosing appropriate labels for images with
extracting effective and hidden feature in pictures. In the feature extraction step of proposed method, we
present a model, which combined effective features of visual topics (global features over an image) and
regional contexts (relationship between the regions in Image and each other regions images) to automatic
image annotation.In the annotation step of proposed method, we create a new ontology (base on WordNet
ontology) for the semantic relationships between tags in the classification and improving semantic gap
exist in the automatic image annotation.Experiments result on the 5k Corel dataset show the proposed
method of image annotation in addition to reducing the complexity of the classification, increased accuracy
compared to the another methods.
NEW ONTOLOGY RETRIEVAL IMAGE METHOD IN 5K COREL IMAGESijcax
Semantic annotation of images is an important research topic on both image understanding and database
or web image search. Image annotation is a technique to choosing appropriate labels for images with
extracting effective and hidden feature in pictures. In the feature extraction step of proposed method, we
present a model, which combined effective features of visual topics (global features over an image) and
regional contexts (relationship between the regions in Image and each other regions images) to automatic
image annotation.In the annotation step of proposed method, we create a new ontology (base on WordNet
ontology) for the semantic relationships between tags in the classification and improving semantic gap
exist in the automatic image annotation.Experiments result on the 5k Corel dataset show the proposed
method of image annotation in addition to reducing the complexity of the classification, increased accuracy
compared to the another methods
NEW ONTOLOGY RETRIEVAL IMAGE METHOD IN 5K COREL IMAGESijcax
Semantic annotation of images is an important research topic on both image understanding and database
or web image search. Image annotation is a technique to choosing appropriate labels for images with
extracting effective and hidden feature in pictures. In the feature extraction step of proposed method, we
present a model, which combined effective features of visual topics (global features over an image) and
regional contexts (relationship between the regions in Image and each other regions images) to automatic
image annotation.In the annotation step of proposed method, we create a new ontology (base on WordNet
ontology) for the semantic relationships between tags in the classification and improving semantic gap
exist in the automatic image annotation.Experiments result on the 5k Corel dataset show the proposed
method of image annotation in addition to reducing the complexity of the classification, increased accuracy
compared to the another methods.
NEW ONTOLOGY RETRIEVAL IMAGE METHOD IN 5K COREL IMAGESijcax
Semantic annotation of images is an important research topic on both image understanding and database or web image search. Image annotation is a technique to choosing appropriate labels for images with
extracting effective and hidden feature in pictures. In the feature extraction step of proposed method, we present a model, which combined effective features of visual topics (global features over an image) and regional contexts (relationship between the regions in Image and each other regions images) to automatic
image annotation. In the annotation step of proposed method, we create a new ontology (base on WordNet ontology) for the semantic relationships between tags in the classification and improving semantic gap exist in the automatic image annotation. Experiments result on the 5k Corel dataset show the proposed
method of image annotation in addition to reducing the complexity of the classification, increased accuracy compared to the another methods.
NEW ONTOLOGY RETRIEVAL IMAGE METHOD IN 5K COREL IMAGESijcax
Semantic annotation of images is an important research topic on both image understanding and database
or web image search. Image annotation is a technique to choosing appropriate labels for images with
extracting effective and hidden feature in pictures. In the feature extraction step of proposed method, we
present a model, which combined effective features of visual topics (global features over an image) and
regional contexts (relationship between the regions in Image and each other regions images) to automatic
image annotation.In the annotation step of proposed method, we create a new ontology (base on WordNet
ontology) for the semantic relationships between tags in the classification and improving semantic gap
exist in the automatic image annotation.Experiments result on the 5k Corel dataset show the proposed
method of image annotation in addition to reducing the complexity of the classification, increased accuracy
compared to the another methods.
NEW ONTOLOGY RETRIEVAL IMAGE METHOD IN 5K COREL IMAGESijcax
Semantic annotation of images is an important research topic on both image understanding and databaseor web image search. Image annotation is a technique to choosing appropriate labels for images with extracting effective and hidden feature in pictures. In the feature extraction step of proposed method, we
present a model, which combined effective features of visual topics (global features over an image) and regional contexts (relationship between the regions in Image and each other regions images) to automatic image annotation.In the nnotation step of proposed method, we create a new ontology (base on WordNet ontology) for the semantic relationships between tags in the classification and improving semantic gap exist in the automatic image
annotation.Experiments result on the 5k Corel dataset show the proposed method of image annotation in addition to reducing the complexity of the classification, increased accuracy compared to the another methods.
NEW ONTOLOGY RETRIEVAL IMAGE METHOD IN 5K COREL IMAGESijcax
Semantic annotation of images is an important research topic on both image understanding and database or web image search. Image annotation is a technique to choosing appropriate labels for images with
extracting effective and hidden feature in pictures. In the feature extraction step of proposed method, we
present a model, which combined effective features of visual topics (global features over an image) and
regional contexts (relationship between the regions in Image and each other regions images) to automatic image annotation.In the annotation step of proposed method, we create a new ontology (base on WordNet ontology) for the semantic relationships between tags in the classification and improving semantic gap exist in the automatic image annotation.Experiments result on the 5k Corel dataset show the proposed
method of image annotation in addition to reducing the complexity of the classification, increased accuracy
compared to the another methods.
Facial image retrieval on semantic features using adaptive mean genetic algor...TELKOMNIKA JOURNAL
The emergence of larger databases has made image retrieval techniques an essential component and has led to the development of more efficient image retrieval systems. Retrieval can either be content or text-based. In this paper, the focus is on the content-based image retrieval from the FGNET database. Input query images are subjected to several processing techniques in the database before computing the squared Euclidean distance (SED) between them. The images with the shortest Euclidean distance are considered as a match and are retrieved. The processing techniques involve the application of the median modified Weiner filter (MMWF), extraction of the low-level features using histogram-oriented gradients (HOG), discrete wavelet transform (DWT), GIST, and Local tetra pattern (LTrP). Finally, the features are selected using Adaptive Mean Genetic Algorithm (AMGA). In this study, the average PSNR value obtained after applying the Wiener filter was 45.29. The performance of the AMGA was evaluated based on its precision, F-measure, and recall, and the obtained average values were respectively 0.75, 0.692, and 0.66. The performance matrix of the AMGA was compared to those of particle swarm optimization algorithm (PSO) and genetic algorithm (GA) and found to perform better; thus, proving its efficiency.
K2 Algorithm-based Text Detection with An Adaptive Classifier ThresholdCSCJournals
In natural scene images, text detection is a challenging study area for dissimilar content-based image analysis tasks. In this paper, a Bayesian network scores are used to classify candidate character regions by computing posterior probabilities. The posterior probabilities are used to define an adaptive threshold to detect text in scene images with accuracy. Therefore, candidate character regions are extracted through maximally stable extremal region. K2 algorithm-based Bayesian network scores are learned by evaluating dependencies amongst features of a given candidate character region. Bayesian logistic regression classifier is trained to compute posterior probabilities to define an adaptive classifier threshold. The candidate character regions below from adaptive classifier threshold are discarded as non-character regions. Finally, text regions are detected with the use of effective text localization scheme based on geometric features. The entire system is evaluated on the ICDAR 2013 competition database. Experimental results produce competitive performance (precision, recall and harmonic mean) with the recently published algorithms.
Scene Text Detection of Curved Text Using Gradiant Vector Flow MethodIJTET Journal
Abstract--Text detection and recognition is a hot topic for researchers in the field of image processing and multimedia. Content based Image Retrieval (CBIR) community fills the semantic gap between low-level and high-level features. For text detection and extraction that achieve reasonable accuracy for multi-oriented text and natural scene text (camera images), several methods have been developed. In general most of the methods use classifier and large number of training samples to improve the accuracy in text detection. In general, connected components are used to tackle the multi-orientation problem. The connected component analysis based features with classifier training, work well for achieving better accuracy when the images are highly contrast. However, when the same methods used directly for text detection in video it results in disconnections, loss of shapes etc, because of low contrast and complex background. For such cases, deciding geometrical features of the components and classifier is not that easy. To overcome from this problem the proposed research uses Gradiant Vector Flow and Grouping based Method for Arbitrarily Oriented Scene text Detection method. The GVF of edge pixels in the Sobel edge map of the input frame is explored to identify the dominant edge pixels which represent text components. The method extracts dominant pixel’s edge components corresponding to the Sobel edge map, which is called Text Candidates (TC) of the text lines. Experimental results on different datasets including text data that is oriented arbitrary, non-horizontal text data also horizontal text data, Hua’s data and ICDAR-03 data (Camera images) show that the proposed method outperforms existing methods.
A Survey On Thresholding Operators of Text Extraction In VideosCSCJournals
Video indexing is an important problem that has interested by the communities of visual information in image processing. The detection and extraction of scene and caption text from unconstrained, general purpose video is an important research problem in the context of content-based retrieval and summarization. In this paper, the technique presented is for detection text from frames video. Finding the textual contents in images is a challenging and promising research area in information technology. Consequently, text detection and recognition in multimedia had become one of the most important fields in computer vision due to its valuable uses in a variety of recent technical applications. The work in this paper consists using morphological operations for extract text appearing in the video frames. The proposed scheme well as preprocessing to differentiate among where it as the high similarity between text and background information. Experimental results show that the resultant image is the image with only text. The evaluated criteria are applied with the image result and one obtained bay different operator.
Alinteri Journal of Agriculture Sciences aims to create an environment for researchers to introduce, share, read, and discuss recent scientific progress. We adopt the policy of providing open access to readers who may be interested in recent developments. Alinteri Journal of Agriculture Sciences is being published online biannually as of 2007. The journal is an open access, international, double-blind peer-reviewed journal publishing research articles, Invited reviews, short communications, and letters to the Editor in the fields of agriculture, fisheries, veterinary, biology, and closely related disciplines.
Customized mask region based convolutional neural networks for un-uniformed ...IJECEIAES
In image scene, text contains high-level of important information that helps to analyze and consider the particular environment. In this paper, we adapt image mask and original identification of the mask region based convolutional neural networks (R-CNN) to allow recognition at 3 levels such as sequence, holistic and pixel-level semantics. Particularly, pixel and holistic level semantics can be utilized to recognize the texts and define the text shapes, respectively. Precisely, in mask and detection, we segment and recognize both character and word instances. Furthermore, we implement text detection through the outcome of instance segmentation on 2-D feature-space. Also, to tackle and identify the text issues of smaller and blurry texts, we consider text recognition by attention-based of optical character recognition (OCR) model with the mask R-CNN at sequential level. The OCR module is used to estimate character sequence through feature maps of the word instances in sequence to sequence. Finally, we proposed a fine-grained learning technique that trains a more accurate and robust model by learning models from the annotated datasets at the word level. Our proposed approach is evaluated on popular benchmark dataset ICDAR 2013 and ICDAR 2015.
ROBUST TEXT DETECTION AND EXTRACTION IN NATURAL SCENE IMAGES USING CONDITIONA...ijiert bestjournal
In Natural Scene Image,Text detection is important tasks which are used for many content based image analysis. A maximally stable external region based method is us ed for scene detection .This MSER based method incl udes stages character candidate extraction,text candida te construction,text candidate elimination & text candidate classification. Main limitations of this method are how to detect highly blurred text in low resolutio n natural scene images. The current technology not focuses on any t ext extraction method. In proposed system a Conditi onal Random field (CRF) model is used to assign candidat e component as one of the two classes (text& Non Te xt) by Considering both unary component properties and bin ary contextual component relationship. For this pur pose we are using connected component analysis method. The proposed system also performs a text extraction usi ng OCR
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
A Study on Translucent Concrete Product and Its Properties by Using Optical F...IJMER
- Translucent concrete is a concrete based material with light-transferring properties,
obtained due to embedded light optical elements like Optical fibers used in concrete. Light is conducted
through the concrete from one end to the other. This results into a certain light pattern on the other
surface, depending on the fiber structure. Optical fibers transmit light so effectively that there is
virtually no loss of light conducted through the fibers. This paper deals with the modeling of such
translucent or transparent concrete blocks and panel and their usage and also the advantages it brings
in the field. The main purpose is to use sunlight as a light source to reduce the power consumption of
illumination and to use the optical fiber to sense the stress of structures and also use this concrete as an
architectural purpose of the building
Developing Cost Effective Automation for Cotton Seed DelintingIJMER
A low cost automation system for removal of lint from cottonseed is to be designed and
developed. The setup consists of stainless steel drum with stirrer in which cottonseeds having lint is mixed
with concentrated sulphuric acid. So lint will get burn. This lint free cottonseed treated with lime water to
neutralize acidic nature. After water washing this cottonseeds are used for agriculter purpose
Study & Testing Of Bio-Composite Material Based On Munja FibreIJMER
The incorporation of natural fibres such as munja fiber composites has gained
increasing applications both in many areas of Engineering and Technology. The aim of this study is to
evaluate mechanical properties such as flexural and tensile properties of reinforced epoxy composites.
This is mainly due to their applicable benefits as they are light weight and offer low cost compared to
synthetic fibre composites. Munja fibres recently have been a substitute material in many weight-critical
applications in areas such as aerospace, automotive and other high demanding industrial sectors. In
this study, natural munja fibre composites and munja/fibreglass hybrid composites were fabricated by a
combination of hand lay-up and cold-press methods. A new variety in munja fibre is the present work
the main aim of the work is to extract the neat fibre and is characterized for its flexural characteristics.
The composites are fabricated by reinforcing untreated and treated fibre and are tested for their
mechanical, properties strictly as per ASTM procedures.
Hybrid Engine (Stirling Engine + IC Engine + Electric Motor)IJMER
Hybrid engine is a combination of Stirling engine, IC engine and Electric motor. All these 3 are
connected together to a single shaft. The power source of the Stirling engine will be a Solar Panel. The aim of
this is to run the automobile using a Hybrid engine
Fabrication & Characterization of Bio Composite Materials Based On Sunnhemp F...IJMER
The present day technology demands eco-friendly developments. In this era the
composite material are playing a vital roal in different field of Engineering .The composite materials
are using as a principle materials. Nowaday the composite materials are utilizing as a important
component of engineering field .Where as the importance of the applications of composites is well
known, but thrust on the use of natural fibres in it for reinforcement has been given priority for some
times. But changing from synthetic fibres to natural fibres provides only half green-composites. A
partial green composite will be achieved if the matrix component is also eco-friendly. Keeping this in
view, a detailed literature surveyed has been carried out through various issues of the Journals
related to this field. The material systems used are sunnhemp fibres. Some epoxy and hardener has
been also added for stability and drying of the bio-composites. Various graphs and bar-charts are
super-imposed on each other for comparison among themselves and Graphs is plotted on MAT LAB
and ORIGIN 6.0 software. To determining tensile strengths, Various properties for different biocomposites
have been compared among themselves. Comparison of the behaviour of bio-composites of
this work has been also compare with other works. The bio-composites developed in this work are
likely to get applications in fall ceilings, partitions, bio-degradable packagings, automotive interiors,
sports things (e.g. rackets, nets, etc.), toys etc.
Geochemistry and Genesis of Kammatturu Iron Ores of Devagiri Formation, Sandu...IJMER
The Greenstone belts of Karnataka are enriched in BIFs in Dharwar craton, where Iron
formations are confined to the basin shelf, clearly separated from the deeper-water iron formation that
accumulated at the basin margin and flanking the marine basin. Geochemical data procured in terms of
major, trace and REE are plotted in various diagrams to interpret the genesis of BIFs. Al2O3, Fe2O3 (T),
TiO2, CaO, and SiO2 abundances and ratios show a wide variation. Ni, Co, Zr, Sc, V, Rb, Sr, U, Th,
ΣREE, La, Ce and Eu anomalies and their binary relationships indicate that wherever the terrigenous
component has increased, the concentration of elements of felsic such as Zr and Hf has gone up. Elevated
concentrations of Ni, Co and Sc are contributed by chlorite and other components characteristic of basic
volcanic debris. The data suggest that these formations were generated by chemical and clastic
sedimentary processes on a shallow shelf. During transgression, chemical precipitation took place at the
sediment-water interface, whereas at the time of regression. Iron ore formed with sedimentary structures
and textures in Kammatturu area, in a setting where the water column was oxygenated.
Experimental Investigation on Characteristic Study of the Carbon Steel C45 in...IJMER
In this paper, the mechanical characteristics of C45 medium carbon steel are investigated
under various working conditions. The main characteristic to be studied on this paper is impact toughness
of the material with different configurations and the experiment were carried out on charpy impact testing
equipment. This study reveals the ability of the material to absorb energy up to failure for various
specimen configurations under different heat treated conditions and the corresponding results were
compared with the analysis outcome
Non linear analysis of Robot Gun Support Structure using Equivalent Dynamic A...IJMER
Robot guns are being increasingly employed in automotive manufacturing to replace
risky jobs and also to increase productivity. Using a single robot for a single operation proves to be
expensive. Hence for cost optimization, multiple guns are mounted on a single robot and multiple
operations are performed. Robot Gun structure is an efficient way in which multiple welds can be done
simultaneously. However mounting several weld guns on a single structure induces a variety of
dynamic loads, especially during movement of the robot arm as it maneuvers to reach the weld
locations. The primary idea employed in this paper, is to model those dynamic loads as equivalent G
force loads in FEA. This approach will be on the conservative side, and will be saving time and
subsequently cost efficient. The approach of the paper is towards creating a standard operating
procedure when it comes to analysis of such structures, with emphasis on deploying various technical
aspects of FEA such as Non Linear Geometry, Multipoint Constraint Contact Algorithm, Multizone
meshing .
Static Analysis of Go-Kart Chassis by Analytical and Solid Works SimulationIJMER
This paper aims to do modelling, simulation and performing the static analysis of a go
kart chassis consisting of Circular beams. Modelling, simulations and analysis are performed using 3-D
modelling software i.e. Solid Works and ANSYS according to the rulebook provided by Indian Society of
New Era Engineers (ISNEE) for National Go Kart Championship (NGKC-14).The maximum deflection is
determined by performing static analysis. Computed results are then compared to analytical calculation,
where it is found that the location of maximum deflection agrees well with theoretical approximation but
varies on magnitude aspect.
In récent year various vehicle introduced in market but due to limitation in
carbon émission and BS Séries limitd speed availability vehicle in the market and causing of
environnent pollution over few year There is need to decrease dependancy on fuel vehicle.
bicycle is to be modified for optional in the future To implement new technique using change in
pedal assembly and variable speed gearbox such as planetary gear optimise speed of vehicle
with variable speed ratio.To increase the efficiency of bicycle for confortable drive and to
reduce torque appli éd on bicycle. we introduced epicyclic gear box in which transmission done
throgh Chain Drive (i.e. Sprocket )to rear wheel with help of Epicyclical gear Box to give
number of différent Speed during driving.To reduce torque requirent in the cycle with change in
the pedal mechanism
Integration of Struts & Spring & Hibernate for Enterprise ApplicationsIJMER
The proposal of this paper is to present Spring Framework which is widely used in
developing enterprise applications. Considering the current state where applications are developed using
the EJB model, Spring Framework assert that ordinary java beans(POJO) can be utilize with minimal
modifications. This modular framework can be used to develop the application faster and can reduce
complexity. This paper will highlight the design overview of Spring Framework along with its features that
have made the framework useful. The integration of multiple frameworks for an E-commerce system has
also been addressed in this paper. This paper also proposes structure for a website based on integration of
Spring, Hibernate and Struts Framework.
Microcontroller Based Automatic Sprinkler Irrigation SystemIJMER
Microcontroller based Automatic Sprinkler System is a new concept of using
intelligence power of embedded technology in the sprinkler irrigation work. Designed system replaces
the conventional manual work involved in sprinkler irrigation to automatic process. Using this system a
farmer is protected against adverse inhuman weather conditions, tedious work of changing over of
sprinkler water pipe lines & risk of accident due to high pressure in the water pipe line. Overall
sprinkler irrigation work is transformed in to a comfortableautomatic work. This system provides
flexibility & accuracy in respect of time set for the operation of a sprinkler water pipe lines. In present
work the author has designed and developed an automatic sprinkler irrigation system which is
controlled and monitored by a microcontroller interfaced with solenoid valves.
On some locally closed sets and spaces in Ideal Topological SpacesIJMER
In this paper we introduce and characterize some new generalized locally closed sets
known as
δ
ˆ
s-locally closed sets and spaces are known as
δ
ˆ
s-normal space and
δ
ˆ
s-connected space and
discussed some of their properties
Intrusion Detection and Forensics based on decision tree and Association rule...IJMER
This paper present an approach based on the combination of, two techniques using
decision tree and Association rule mining for Probe attack detection. This approach proves to be
better than the traditional approach of generating rules for fuzzy expert system by clustering methods.
Association rule mining for selecting the best attributes together and decision tree for identifying the
best parameters together to create the rules for fuzzy expert system. After that rules for fuzzy expert
system are generated using association rule mining and decision trees. Decision trees is generated for
dataset and to find the basic parameters for creating the membership functions of fuzzy inference
system. Membership functions are generated for the probe attack. Based on these rules we have
created the fuzzy inference system that is used as an input to neuro-fuzzy system. Fuzzy inference
system is loaded to neuro-fuzzy toolbox as an input and the final ANFIS structure is generated for
outcome of neuro-fuzzy approach. The experiments and evaluations of the proposed method were
done with NSL-KDD intrusion detection dataset. As the experimental results, the proposed approach
based on the combination of, two techniques using decision tree and Association rule mining
efficiently detected probe attacks. Experimental results shows better results for detecting intrusions as
compared to others existing methods
Natural Language Ambiguity and its Effect on Machine LearningIJMER
"Natural language processing" here refers to the use and ability of systems to process
sentences in a natural language such as English, rather than in a specialized artificial computer
language such as C++. The systems of real interest here are digital computers of the type we think of as
personal computers and mainframes. Of course humans can process natural languages, but for us the
question is whether digital computers can or ever will process natural languages. We have tried to
explore in depth and break down the types of ambiguities persistent throughout the natural languages
and provide an answer to the question “How it affects the machine translation process and thereby
machine learning as whole?” .
Today in era of software industry there is no perfect software framework available for
analysis and software development. Currently there are enormous number of software development
process exists which can be implemented to stabilize the process of developing a software system. But no
perfect system is recognized till yet which can help software developers for opting of best software
development process. This paper present the framework of skillful system combined with Likert scale. With
the help of Likert scale we define a rule based model and delegate some mass score to every process and
develop one tool name as MuxSet which will help the software developers to select an appropriate
development process that may enhance the probability of system success.
Material Parameter and Effect of Thermal Load on Functionally Graded CylindersIJMER
The present study investigates the creep in a thick-walled composite cylinders made
up of aluminum/aluminum alloy matrix and reinforced with silicon carbide particles. The distribution
of SiCp is assumed to be either uniform or decreasing linearly from the inner to the outer radius of
the cylinder. The creep behavior of the cylinder has been described by threshold stress based creep
law with a stress exponent of 5. The composite cylinders are subjected to internal pressure which is
applied gradually and steady state condition of stress is assumed. The creep parameters required to
be used in creep law, are extracted by conducting regression analysis on the available experimental
results. The mathematical models have been developed to describe steady state creep in the composite
cylinder by using von-Mises criterion. Regression analysis is used to obtain the creep parameters
required in the study. The basic equilibrium equation of the cylinder and other constitutive equations
have been solved to obtain creep stresses in the cylinder. The effect of varying particle size, particle
content and temperature on the stresses in the composite cylinder has been analyzed. The study
revealed that the stress distributions in the cylinder do not vary significantly for various combinations
of particle size, particle content and operating temperature except for slight variation observed for
varying particle content. Functionally Graded Materials (FGMs) emerged and led to the development
of superior heat resistant materials.
Energy Audit is the systematic process for finding out the energy conservation
opportunities in industrial processes. The project carried out studies on various energy conservation
measures application in areas like lighting, motors, compressors, transformer, ventilation system etc.
In this investigation, studied the technical aspects of the various measures along with its cost benefit
analysis.
Investigation found that major areas of energy conservation are-
1. Energy efficient lighting schemes.
2. Use of electronic ballast instead of copper ballast.
3. Use of wind ventilators for ventilation.
4. Use of VFD for compressor.
5. Transparent roofing sheets to reduce energy consumption.
So Energy Audit is the only perfect & analyzed way of meeting the Industrial Energy Conservation.
An Implementation of I2C Slave Interface using Verilog HDLIJMER
The focus of this paper is on implementation of Inter Integrated Circuit (I2C) protocol
following slave module for no data loss. In this paper, the principle and the operation of I2C bus protocol
will be introduced. It follows the I2C specification to provide device addressing, read/write operation and
an acknowledgement. The programmable nature of device provide users with the flexibility of configuring
the I2C slave device to any legal slave address to avoid the slave address collision on an I2C bus with
multiple slave devices. This paper demonstrates how I2C Master controller transmits and receives data to
and from the Slave with proper synchronization.
The module is designed in Verilog and simulated in ModelSim. The design is also synthesized in Xilinx
XST 14.1. This module acts as a slave for the microprocessor which can be customized for no data loss.
Discrete Model of Two Predators competing for One PreyIJMER
This paper investigates the dynamical behavior of a discrete model of one prey two
predator systems. The equilibrium points and their stability are analyzed. Time series plots are obtained
for different sets of parameter values. Also bifurcation diagrams are plotted to show dynamical behavior
of the system in selected range of growth parameter
Discrete Model of Two Predators competing for One Prey
Cc31331335
1. International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.3, Issue.1, Jan-Feb. 2013 pp-331-335 ISSN: 2249-6645
Detecting and Localizing Color Text in Natural Scene Images
Using Region Based & Connected Component Method
Mohanabharathi.R, 1 Surender.K, 2 Selvi.C3
1, 2, 3
Asst. Professor /Department of Computer Science and Engineering Selvam College of Technology, Namakkal.
Abstract: Large amounts of information are embedded in natural scenes which are often required to be automatically
recognized and processed. This requires automatic detection, segmentation and recognition of visual text entities in natural
scene images. In this paper, we present a hybrid approach to detect color texts in natural scene images. The approaches
used in this project are region based and connected component based approach. A text region detector is designed to
estimate the probabilities of text position and scale, which helps to segment candidate text components with an efficient local
binarization algorithm. To combine unary component properties and binary contextual component relationships, a
conditional random field (CRF) model with supervised parameter learning is proposed. Finally, text components are
grouped into text lines/words with a learning-based energy minimization method. In our proposed system, a selective metric-
based clustering is used to extract textual information in real-world images, thus enabling the processing of character
segmentation into individual components to increase final recognition rates. This project is evaluated on natural scene
image dataset.
Keywords: Conditional random field (CRF); connected component analysis (CCA); text detection; text localization.
I. INTRODUCTION
Image processing is a physical process used to convert an image signal into a physical image. Fig.1 shows, Image
acquisition is the first process in image processing that is used to acquire digital image. Image enhancement is the simplest
and most appealing areas of digital image processing. The idea behind enhancement techniques is to bring out detail that is
obscured, or simply to highlight certain features of interest in an image. Recognition is the process that assigns a label to an
object based on its descriptors. This is the act of determining the properties of represented region for processing the images.
Information Extraction (IE) is a type of information retrieval whose goal is to automatically extract structured information
from unstructured and/or semi-structured machine-readable documents. In most of the cases, this activity concerns
processing human language texts by means of Natural Language Processing (NLP). Recent activities in multimedia
document processing like automatic annotation and concept extraction out of images/audio/video could be seen as
information extraction. [2] [3]
Fig.1.Text information extraction
Existing system presented a hybrid approach to robustly detect and localize texts in natural scene images by taking
advantages of both region-based and CC-based methods. This system consists of three stages are the pre-processing stage,
the connected component analysis stage and text grouping. At the pre-processing stage, a text region detector is designed to
detect text regions in each layer of the image pyramid and project the text confidence and scale information back to the
original image, scale-adaptive local binarization is then applied to generate candidate text components. At the connected
component analysis stage, [4][5][7] a CRF model combining unary component properties and binary contextual component
relationships is used to filter out non-text components. At the last stage, neighboring text components are linked with a
learning-based minimum spanning tree (MST) algorithm and between-line/word edges are cut off with an energy
minimization model to group text components into text lines or words. And also describes the binary contextual component
relationships, in addition to the unary component properties, are integrated in a CRF model, whose parameters are jointly
www.ijmer.com 331 | Page
2. International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.3, Issue.1, Jan-Feb. 2013 pp-331-335 ISSN: 2249-6645
optimized by supervised learning. But this approach fails on some hard-to-segment texts. Although the existing methods
have reported promising localization performance, there still remain several problems to solve. For region-based methods,
the speed is relatively slow and the performance is sensitive to text alignment orientation. On the other hand, CC-based
methods cannot segment text components accurately without prior knowledge of text position and scale. Here, designing fast
and reliable connected component analyzer is difficult since there are many non-text components which are easily confused
with texts when analyzed individually.
This paper is organised as follows: Section II briefly reviews the related work. Section III describes the
preprocessing of image. Section IV explains the connected component analysis using CRF model. Section V describes the
text line/word grouping method. Clustering based method for text extraction and character segmentation is discussed in
section VI. Experimental results and conclusion are presented in section VII.
II. RELATED WORKS
Most region-based methods are based on observations that text regions have distinct characteristics from non-text
regions such as the distribution of gradient strength and texture properties. Generally, a region-based method consists of two
stages: 1) text detection to estimate text existing confidence in local image regions by classification, and 2) text localization
to cluster local text regions into text blocks, and text verification to remove non-text regions for further processing.
An earlier method proposed by Wu et al. [44] uses a set of Gaussian derivative filters to extract texture features
from local image regions. With the corresponding filter responses, all image pixels are assigned to one of three classes
(“text”, “non text” and “complex background”), then c-means clustering and morphological operators are used to group text
pixels into text regions.
Li et al. [16] proposed an algorithm for detecting texts in video by using first- and second-order moments of
wavelet decomposition responses as local region features classified by a neural network classifier. Text regions are then
merged at each pyramid layer and further projected back to the original image map.
Recently, Weinman et al. [14] use a CRF model for patch-based text detection. This method justifies the benefit of
adding contextual information to traditional local region-based text detection methods. Their experimental results show that
this method can deal with texts of variable scales and alignment orientations. To speed up text detection, Chen and Yuille [5]
proposed a fast text detector using a cascade AdaBoost classifier, whose weak learners are selected from a feature pool
containing gray-level, gradient and edge features. Detected text regions are then merged into text blocks, from which text
components are segmented by local binarization. Their results on the ICDAR 2005 competition dataset [15] show that this
method performs competitively and is more than 10 times faster than the other methods.
Unlike region-based methods, CC-based methods are based on observations that texts can be seen as a set of
connected components, each of which has distinct geometric features, and neighboring components have close spatial and
geometric relationships. These methods normally consist of three stages: 1) CC extraction to segment candidate text
components from images; 2) CC analysis to filter out non-text components using heuristic rules or classifiers; and 3) post-
processing to group text components into text blocks (e.g., words and lines).
The method of Liu et al. [17] extracts candidate CCs based on edge contour features and removes non-text
components by wavelet feature analysis. Within each text component region, a GMM is used for binarization by fitting the
gray-level distributions of the foreground and background pixel clusters. Zhang et al. [19] presented a Markov random field
(MRF) method for exploring the neighboring information of components. The candidate text components are initially
segmented with a mean-shift process. After building up a component adjacency graph, a MRF model integrating a first-order
component term and a higher-order contextual term is used for labeling components as “text” or “non-text”. For multilingual
text localization, Liu et al. proposed a method [18] which employs a GMM to fit third-order neighboring information of
components using a specific training criterion: maximum minimum similarity (MMS). Their experiments show good
performance on their multilingual image datasets.
III. PRE-PROCESSING
In this module, preprocessing stage of the overall process is discussed. At the preprocessing stage, a text region
detector is designed to detect text regions in each layer of the image pyramid and project the text confidence and scale
information back to the original image, scale-adaptive local binarization is then applied to generate candidate text
components. To extract and utilize local text region information, a text region detector is designed by integrating a widely
used feature descriptor: Histogram of oriented gradients (HOG) and waldboost classifier to estimate the text confidence and
the corresponding scale, based on which candidate text components can be segmented and analyzed accurately. Initially, the
original color image is converted into a gray level image. To measure the text confidence for each image patch in a window,
no matter it is accepted or rejected. [2] [3] [9] [10] The posterior probability of a label yi, yi ε{„text‟, „non-text‟}conditioned
on its detection state , si, si ε{„accepted‟, „rejected‟} at the stage t, can be estimated based on the Bayes formula as defined as
www.ijmer.com 332 | Page
3. International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.3, Issue.1, Jan-Feb. 2013 pp-331-335 ISSN: 2249-6645
Where all the stage likelihoods Pt (si/yi)are calculated on a validation dataset during training. The text scale map is
used in local binarization for adaptively segmenting candidate CCs and the confidence map is used later in CCA for
component classification. The formula to binaries each pixel x is
,
Here and are mean and standard deviation with radius r. Figure 3 shows the example of
preprocessing stage. They calculate the radius from the text scale map which is more stable under noisy conditions. After
local binarization assume that within each local region, gray-level values of foreground pixels are higher or lower than the
average intensity.
Fig.3.Example of preprocessing stage
IV. CONNECTED COMPONENT ANALYSIS
This module presents the connected component analysis (CCA) stage using a CRF model combining unary
component properties and binary contextual component relationships is used to filter out non-text components. Conditional
random field (CRF) [4] [7] is proposed model to assign candidate components as one of the two classes (“text” and “non-
text”) by considering both unary component properties and binary contextual component relationships. CRF is a probabilistic
graphical model which has been widely used in many areas such as natural language processing. Next considering that
neighboring text components normally have similar width or height, build up a component neighborhood graph by defining a
component linkage rule. And also use the CRF model to explore contextual component relationships as well as unary
component properties. During the test process, to alleviate the computation overhead of graph inference, some apparent non-
text components are first removed by using thresholds on unary component features. The thresholds are set to safely accept
almost all text components in the training set.
V. TEXT GROUPING METHOD
To group text components into text regions are lines and words, a learning-based method by clustering nearing
components into a tree with a minimum spanning tree (MST) algorithm and cutting off between-line (word) edges with an
energy minimization model is designed. Cluster text components into a tree with MST based on a learned distance metric,
which is defined between two components as a linear combination of some features. With the initial component tree built
with the MST algorithm, between-line/word edges need to be cut to partition the tree into subtrees, each of which
corresponds to a text unit. Finally, text words corresponding to partitioned subtrees can be extracted and the ones containing
too small components are removed as noises. With the initial component tree built with the MST algorithm, between-
line/word edges need to be cut to partition the tree into subtrees, each of which corresponds to a text unit (line or word).
A .Text Line Partition
A method to formulate the edge cutting in the tree is proposed as a learning-based energy minimization problem. In
the component tree, each edge is assigned one of two labels: “linked” and “cut”, and each subtree corresponding to a text line
are separated by cutting the “cut” edges. The objective of the proposed method is to find the optimal edge labels such that the
total energy of the separated subtrees is minimal. The total text line energy is defined as
www.ijmer.com 333 | Page
4. International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.3, Issue.1, Jan-Feb. 2013 pp-331-335 ISSN: 2249-6645
Where N is the number of subtrees (text lines), Fi is the feature vector of a text line, and Wline is the vector of
combining weights.
B. Text Word Partition
For comparing our system with previous methods which reported word localization results, further partition text
lines into words using a similar process as line partition. The major difference lies in the word-level features, which are
defined as: 1) word number; 2) component centroid distances of cut edges; 3) component bounding box distances of cut
edges; 4) bounding box distances between words separated by cut edges; 5) the ratio between the component centroid
distance of the cut edge and the average component centroid distance of the edges within separated words; and 6) bounding
box distance ratio between the cut edge and edges within separated words.
VI. SELECTIVE METRIC-BASED CLUSTERING USING LOG–GABOR FILTERS
This module discuss about the selective metric based clustering using log-Gabor filter. Hence, our selective metric-
based clustering is integrated into a dynamic method suitable for text extraction and character segmentation. This method
uses several metrics to merge similar color together for an efficient text-driven segmentation in the RGB color space.
However, color information by itself is not sufficient to solve all natural scene issues; hence complement it with intensity
and spatial information obtained using Log–Gabor filters, thus enabling the processing of character segmentation into
individual components to increase final recognition rates. Our selective metric-based clustering uses mainly color
information for text extraction and our system fails for natural scene images having embossed characters. In this case,
foreground and background have the same color presenting partial shadows around characters due to the relief but not
enough to separate textual foreground from background in a discriminative way as displayed.
a)
b)
Fig.3. Comparison between adaboost and waldboost classifier a) Execution Time b) Detection Rate
Gray-level information with the simultaneous use of a priori information on characters could be a solution to handle
these cases. Next we propose a new text validation measure M to find the most textual foreground cluster over the two
remaining clusters. Based on properties of connected components of each cluster, spatial information is already added at this
point to find the main textual cluster. The proposed validation measure, M, is based on the largest regularity of connected
components of text compared to those of noise and background. And also we use Log–Gabor filters that present globally
high responses to characters. Hence, in order to choose efficiently which clustering distance is better to handle text text
extraction, we perform an average of pixel values inside each mask. The mask which has the highest average is chosen as the
final segmentation.
www.ijmer.com 334 | Page
5. International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.3, Issue.1, Jan-Feb. 2013 pp-331-335 ISSN: 2249-6645
VII. RESULT AND CONCLUSION
From the results its incurred that waldboost classifer has better execution time and detection rate of text, when
compared with previously used adaboost classifier in preprocessing stage .Hence this classifier can be used for text
recognition to be integrated with text localization for complete text information extraction.
REFERENCES
[1] D. T. Chen, J.-M. Odobez, and H. Bourlard, “Text detection and recognition in images and video frames,” Pattern Recogn., vol. 37,
no. 5, pp. 595–608, 2004.
[2] X. L. Chen, J. Yang, J. Zhang, and A.Waibel, “Automatic detection and recognition of signs from natural scenes,” IEEE Trans.
Image Process., vol. 13, no. 1, pp. 87–99, Jan. 2004.
[3] X. R. Chen and A. L. Yuille, “Detecting and reading text in natural scenes,” in Proc. IEEE Conf. Computer Vision and Pattern
Recognition (CVPR’04), Washington, DC, 2004, pp. 366–373.
[4] N. Dalal and B. Triggs, “Histograms of oriented gradients for human detection,” in Proc. IEEE Conf. Computer Vision and Pattern
Recognition (CVPR’05), San Diego, CA, 2005, pp. 886–893.
[5] S. L. Feng, R. Manmatha, and A. McCallum, “Exploring the use of conditional random field models and HMMs for historical
handwritten document recognition,” in Proc. Washington, DC, 2006, pp. 30–37.
[6] J. Gllavata, R. Ewerth, and B. Freisleben, “Text detection in images based on unsupervised classification of high-frequency wavelet
coefficients,” in Proc. 17th Int. Conf. Pattern Recognition (ICPR’04), Cambridge, U.K., 2004, pp. 425–428.
[7] J. M. Hammersley and P. Clifford, Markov Field on Finite Graphs and Lattices, 1971, unpublished. [10] X.-B. Jin, C.-L. Liu, and X.
Hou, “Regularized margin-based conditional log-likelihood loss for prototype learning,” Pattern Recogn., vol. 43, no. 7, pp. 2428–
2438, 2010.
[8] Y. Jin and S. Geman, “Context and hierarchy in a probabilistic image model,” in Proc. IEEE Conf. Computer Vision and Pattern
Recognition (CVPR’06), New York, NY, 2006, pp. 2145–2152.
[9] B.-H. Juang and S. Katagiri, “Discriminative learning for minimum error classification,” IEEE Trans. Signal Process., vol. 40, pp.
3043–3054, 1992.
[10] K. Jung, K. I. Kim, and A. K. Jain, “Text information extraction in images and video: A survey,” Pattern Recogn., vol. 37, no. 5, pp.
977–997, 2004.
[11] K. I. Kim, K. Jung, and J. H. Kim, “Texture-based approach for text detection in images using support vector machines and
continuously adaptive mean shift algorithm,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 25, no. 12, pp. 1631–1639, 2003.
[12] A Hybrid Approach to Detect and Localize Texts in Natural Scene Images Yi-Feng Pan, Xinwen Hou, and Cheng-Lin Liu, Senior
Member
[13] V.Wu, R. Manmatha, and E. M. Riseman, “Finding text in images,” in Proc. 2nd ACM Int. Conf. Digital Libraries (DL‟97), New
York, NY, 1997, pp. 3–12.
[14] J. Weinman, A. Hanson, and A. McCallum, “Sign detection in natural images with conditional random fields,” in Proc. 14th IEEE
Workshop on Machine Learning for Signal Processing (MLSP‟04), São Luís, Brazil, 2004, pp. 549–558.
[15] S. M. Lucas, “ICDAR 2005 text locating competition results,” in Proc.8th Int. Conf. Document Analysis and Recognition
(ICDAR‟05), Seoul, South Korea, 2005, pp. 80–84.
[16] H. P. Li, D. Doermann, and O. Kia, “Automatic text detection and tracking in digital video,” IEEE Trans. Image Process., vol. 9,
pp.147–156, Jan. 2000.
[17] Y. X. Liu, S. Goto, and T. Ikenaga, “A contour-based robust algorithm for text detection in color images,” IEICE Trans. Inf. Syst.,
vol. E89-D, no. 3, pp. 1221–1230, 2006.
[18] X. B. Liu, H. Fu, and Y. D. Jia, “Gaussian mixture modeling and learning of neighboring characters for multilingual text extraction
in images,” Pattern Recogn., vol. 41, no. 2, pp. 484–493, 2008.
[19] D.-Q. Zhang and S.-F. Chang, “Learning to detect scene text using a higher-order MRF with belief propagation,” in Proc. IEEE
Conf. ComputerVision and Pattern Recognition Workshop s (CVPRW‟04),Washington, DC, 2004, pp. 101–108.
[20] Y.-F. Pan, X. W. Hou, and C.-L. Liu, “A robust system to detect and localize texts in natural scene images,” in Proc. 8th IAPR
Workshop onDocument Analysis Syetems (DAS‟08), Nara, Japan, 2008, pp. 35–42.
www.ijmer.com 335 | Page