In this paper, we present a decision level fused local Morphological Pattern Spectrum (PS) and Local Binary Pattern (LBP) approach for an efficient shape representation and classification. This method makes use of Earth Movers Distance (EMD) as the measure in feature matching and shape retrieval process. The proposed approach has three major phases: Feature Extraction, Construction of hybrid spectrum knowledge base and Classification. In the first phase, feature extraction of the shape is done using pattern spectrum and local binary pattern method. In the second phase, the histograms of both pattern spectrum and local binary pattern are fused and stored in the knowledge base. In the third phase, the comparison and matching of the features, which are represented in the form of histograms, is done using Earth Movers Distance (EMD) as metric. The top-n shapes are retrieved for each query shape. The accuracy is tested by means of standard Bulls eye score method. The experiments are conducted on publicly available shape datasets like Kimia-99, Kimia-216 and MPEG-7. The comparative study is also provided with the well known approaches to exhibit the retrieval accuracy of the proposed approach.
{"title":"Shape Representation and Classification through Pattern Spectrum and Local Binary Pattern -- A Decision Level Fusion Approach","authors":"B. H. Shekar, Bharathi Pilar","doi":"10.1109/ICSIP.2014.41","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.41","url":null,"abstract":"In this paper, we present a decision level fused local Morphological Pattern Spectrum (PS) and Local Binary Pattern (LBP) approach for an efficient shape representation and classification. This method makes use of Earth Movers Distance (EMD) as the measure in feature matching and shape retrieval process. The proposed approach has three major phases: Feature Extraction, Construction of hybrid spectrum knowledge base and Classification. In the first phase, feature extraction of the shape is done using pattern spectrum and local binary pattern method. In the second phase, the histograms of both pattern spectrum and local binary pattern are fused and stored in the knowledge base. In the third phase, the comparison and matching of the features, which are represented in the form of histograms, is done using Earth Movers Distance (EMD) as metric. The top-n shapes are retrieved for each query shape. The accuracy is tested by means of standard Bulls eye score method. The experiments are conducted on publicly available shape datasets like Kimia-99, Kimia-216 and MPEG-7. The comparative study is also provided with the well known approaches to exhibit the retrieval accuracy of the proposed approach.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125458188","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The information available in the World Wide Web is stored using many real Web databases (e.g. vehicle database). Accessing the information from these real Web databases has become increasingly important for the users to find the desired information. Web users search for the desired information by querying these Web databases, when the number of query results generated is large, it is very difficult for the Web user to select the most relevant information from the large result set generated. Users today, have become more and more demanding in terms of the quality of information that is provided to them while searching the Web databases. The most common solution to solve the problem involves ranking the query results returned by the Web databases. Earlier approaches have used query logs, user profiles and frequencies of database values. The problem in all of these techniques is that ranking is performed in a user and query independent manner. This paper, proposes an automated ranking of query results returned by Web databases by analyzing user, query and workload similarity. The effectiveness of this approach is discussed considering a vehicle Web database as an example.
{"title":"Similarity Based Ranking of Query Results from Real Web Databases","authors":"Harish Kumar, D. Chowdary","doi":"10.1109/ICSIP.2014.58","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.58","url":null,"abstract":"The information available in the World Wide Web is stored using many real Web databases (e.g. vehicle database). Accessing the information from these real Web databases has become increasingly important for the users to find the desired information. Web users search for the desired information by querying these Web databases, when the number of query results generated is large, it is very difficult for the Web user to select the most relevant information from the large result set generated. Users today, have become more and more demanding in terms of the quality of information that is provided to them while searching the Web databases. The most common solution to solve the problem involves ranking the query results returned by the Web databases. Earlier approaches have used query logs, user profiles and frequencies of database values. The problem in all of these techniques is that ranking is performed in a user and query independent manner. This paper, proposes an automated ranking of query results returned by Web databases by analyzing user, query and workload similarity. The effectiveness of this approach is discussed considering a vehicle Web database as an example.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116854961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
C. Sujatha, A. R. Chivate, R. Tabib, U. Mudenagudi
In this paper, we propose multilevel framework for summarization of surveillance videos using motion entropy by maintaining chronology of activities. We aim to reduce the size of the video to give a meaningful summary by retaining important activities without destroying the temporal relationship of the activities. Initially input video is divided into blocks and then blocks into segments in a multilevel framework with a non-uniform number of frames in a block at each level. We present a mechanism to select salient blocks which can contribute in the video summarization by propagating motion information from frame level to segment level and further to block level. We generate the summarized video at each level of the framework and is stopped at a level where optimization of condensed ratio between successive levels is minimal. The experimental results on surveillance videos shows better condensed ratios.
{"title":"Multilevel Framework for Summarization of Surveillance Videos","authors":"C. Sujatha, A. R. Chivate, R. Tabib, U. Mudenagudi","doi":"10.1109/ICSIP.2014.48","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.48","url":null,"abstract":"In this paper, we propose multilevel framework for summarization of surveillance videos using motion entropy by maintaining chronology of activities. We aim to reduce the size of the video to give a meaningful summary by retaining important activities without destroying the temporal relationship of the activities. Initially input video is divided into blocks and then blocks into segments in a multilevel framework with a non-uniform number of frames in a block at each level. We present a mechanism to select salient blocks which can contribute in the video summarization by propagating motion information from frame level to segment level and further to block level. We generate the summarized video at each level of the framework and is stopped at a level where optimization of condensed ratio between successive levels is minimal. The experimental results on surveillance videos shows better condensed ratios.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124462604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Diabetic retinopathy (DR) is one of the leading causes of blindness in the world among patients suffering from diabetes. It is an ocular disease and progressive by nature. It is characterized by many pathologies, namely microaneurysms, hard exudates, soft exudates, hemorrhages, etc, among them presence of exudates is the prominent sign of non-proliferative DR. Both hard and soft exudates play a vital role in grading DR into different stages. In this paper, we present an efficient method to identify and classify the exudates as hard and soft exudates. The retinal image in CIELAB color space is pre-processed to eliminate noise. Next, blood vessels network is eliminated to facilitate detection and elimination of optic disc. Optic disc is eliminated using Hough transform technique. The candidate exudates are then detected using k-means clustering technique. Finally, the exudates are classified as hard and soft exudates based on their edge energy and threshold. The proposed method has yielded encouraging results.
{"title":"Detection and Classification of Exudates Using K-Means Clustering in Color Retinal Images","authors":"G. G. Rajput, Preethi N. Patil","doi":"10.1109/ICSIP.2014.25","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.25","url":null,"abstract":"Diabetic retinopathy (DR) is one of the leading causes of blindness in the world among patients suffering from diabetes. It is an ocular disease and progressive by nature. It is characterized by many pathologies, namely microaneurysms, hard exudates, soft exudates, hemorrhages, etc, among them presence of exudates is the prominent sign of non-proliferative DR. Both hard and soft exudates play a vital role in grading DR into different stages. In this paper, we present an efficient method to identify and classify the exudates as hard and soft exudates. The retinal image in CIELAB color space is pre-processed to eliminate noise. Next, blood vessels network is eliminated to facilitate detection and elimination of optic disc. Optic disc is eliminated using Hough transform technique. The candidate exudates are then detected using k-means clustering technique. Finally, the exudates are classified as hard and soft exudates based on their edge energy and threshold. The proposed method has yielded encouraging results.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131979521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Though annotations are integral part of symbols in drawings, due attention is yet to be given for their identification, interpretation, and storage. A reconstructed drawing from the vector format is thus deficient in the complete description of the original, and hence requires costly and time-consuming manual intervention. This paper presents a method for segmentation and identification of annotations associated with electrical symbols in a circuit diagram, which may be used with the vectorizer to make it complete. The proposed method first separates the text region around an intended circuit symbol, and then identifies the annotation part of the segmented text corresponding to that particular symbol. Morphological operations are used in identification phase. Finally, an efficient OCR is used to get the numerical values of the symbols along with their units. The performance of the algorithm is tested on a variety of images with ample variations in annotation. Some of the results are presented in this paper to demonstrate its efficiency.
{"title":"Identification of Annotations for Circuit Symbols in Electrical Diagrams of Document Images","authors":"Paramita De, Sekhar Mandal, Partha Bhowmick","doi":"10.1109/ICSIP.2014.53","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.53","url":null,"abstract":"Though annotations are integral part of symbols in drawings, due attention is yet to be given for their identification, interpretation, and storage. A reconstructed drawing from the vector format is thus deficient in the complete description of the original, and hence requires costly and time-consuming manual intervention. This paper presents a method for segmentation and identification of annotations associated with electrical symbols in a circuit diagram, which may be used with the vectorizer to make it complete. The proposed method first separates the text region around an intended circuit symbol, and then identifies the annotation part of the segmented text corresponding to that particular symbol. Morphological operations are used in identification phase. Finally, an efficient OCR is used to get the numerical values of the symbols along with their units. The performance of the algorithm is tested on a variety of images with ample variations in annotation. Some of the results are presented in this paper to demonstrate its efficiency.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125178693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Medical imaging and computer aided diagnosis traditionally focus on organ or disease based applications. CAD is used by radiologists as a second opinion in detecting tumors, accessing the extent of diseases and making diagnostic decision. Automatic segmentation of tumor from CT images is difficult due to size, shape, position and presence of other objects with the same intensity present in the image. Therefore, first segment the liver so that tumor can then be segmented accurately from it. In this paper, an approach for segmentation of liver and tumor from CT Images is mainly used for computer aided diagnosis of liver is proposed. The method uses region growing with optimized threshold algorithm. The liver is segmented using region growing method that starts from a seed point automatically detected and efficiently close around the vessels and tumors.
{"title":"Extracting the Liver and Tumor from Abdominal CT Images","authors":"M. Jayanthi, B. Kanmani","doi":"10.1109/ICSIP.2014.24","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.24","url":null,"abstract":"Medical imaging and computer aided diagnosis traditionally focus on organ or disease based applications. CAD is used by radiologists as a second opinion in detecting tumors, accessing the extent of diseases and making diagnostic decision. Automatic segmentation of tumor from CT images is difficult due to size, shape, position and presence of other objects with the same intensity present in the image. Therefore, first segment the liver so that tumor can then be segmented accurately from it. In this paper, an approach for segmentation of liver and tumor from CT Images is mainly used for computer aided diagnosis of liver is proposed. The method uses region growing with optimized threshold algorithm. The liver is segmented using region growing method that starts from a seed point automatically detected and efficiently close around the vessels and tumors.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121187650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As the exchange of data over the open networks and Internet is rapidly growing, security of the data becomes a major concern. One possible solution to this problem is to encrypt the data. The data can be text, image, audio, video etc.. In today's world most of the multimedia applications involve images. Earlier image encryption techniques like AES,DES,RSA etc. exhibit low levels of security and also weak anti attack ability. This problem was overcome by using chaos based cryptography. The chaotic systems are very sensitive to initial conditions and control parameters which make them suitable for image encryption. Many works have been done in the field of chaos based image encryption. In this survey paper an attempt has been made to review the aspects and approaches of the design used for image encryption.
{"title":"Image Encryption Using Chaotic Maps: A Survey","authors":"Priya R Sankpal, P A Vijaya","doi":"10.1109/ICSIP.2014.80","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.80","url":null,"abstract":"As the exchange of data over the open networks and Internet is rapidly growing, security of the data becomes a major concern. One possible solution to this problem is to encrypt the data. The data can be text, image, audio, video etc.. In today's world most of the multimedia applications involve images. Earlier image encryption techniques like AES,DES,RSA etc. exhibit low levels of security and also weak anti attack ability. This problem was overcome by using chaos based cryptography. The chaotic systems are very sensitive to initial conditions and control parameters which make them suitable for image encryption. Many works have been done in the field of chaos based image encryption. In this survey paper an attempt has been made to review the aspects and approaches of the design used for image encryption.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127946177","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Intrauterine growth restriction (IUGR) is the failure of the fetus to achieve his/her intrinsic growth potential. IUGR results in significant perinatal and long-term complications, including the development of insulin resistance/metabolic syndrome in adulthood [5]. Accurate and effective monitoring of fetal growth is one of the key component of prenatal care [3]. Ultrasound evaluation is considered the cornerstone of diagnosis and surveillance of the growth-restricted fetus [2]. Ultrasound measurements play a significant role in obstetrics as an accurate means for the estimation of the fetal age. Several parameters are used as aging parameters, the most important of which are the biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC) and femur length (FL). Serial measurement of these parameters over time is used to determine the fetal condition. Hence, consistency and reproducibility of measurements is an important issue. Consequently the automatic segmentation of anatomical structures in ultrasound imagery is a real challenge due to acoustic interferences (speckle noise) and artifacts which are inherent in these images. In this paper, a novel method is proposed for developing a Computer Aided Diagnosis (CAD) system for diagnosis and classification of IUGR foetuses. Diagnosis is performed by segmenting and extracting the required foetus features from an ultrasound image, using the Re-initialization free level set with Reaction Diffusion (RD) technique. An artificial neural network (ANN) classifier is developed, the features extracted are provided to the designed ANN model. The ANN then classifies normal and abnormal fetuses based on features provided.
{"title":"Features Based IUGR Diagnosis Using Variational Level Set Method and Classification Using Artificial Neural Networks","authors":"Akhilraj V. Gadagkar, K. S. Shreedhara","doi":"10.1109/ICSIP.2014.54","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.54","url":null,"abstract":"Intrauterine growth restriction (IUGR) is the failure of the fetus to achieve his/her intrinsic growth potential. IUGR results in significant perinatal and long-term complications, including the development of insulin resistance/metabolic syndrome in adulthood [5]. Accurate and effective monitoring of fetal growth is one of the key component of prenatal care [3]. Ultrasound evaluation is considered the cornerstone of diagnosis and surveillance of the growth-restricted fetus [2]. Ultrasound measurements play a significant role in obstetrics as an accurate means for the estimation of the fetal age. Several parameters are used as aging parameters, the most important of which are the biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC) and femur length (FL). Serial measurement of these parameters over time is used to determine the fetal condition. Hence, consistency and reproducibility of measurements is an important issue. Consequently the automatic segmentation of anatomical structures in ultrasound imagery is a real challenge due to acoustic interferences (speckle noise) and artifacts which are inherent in these images. In this paper, a novel method is proposed for developing a Computer Aided Diagnosis (CAD) system for diagnosis and classification of IUGR foetuses. Diagnosis is performed by segmenting and extracting the required foetus features from an ultrasound image, using the Re-initialization free level set with Reaction Diffusion (RD) technique. An artificial neural network (ANN) classifier is developed, the features extracted are provided to the designed ANN model. The ANN then classifies normal and abnormal fetuses based on features provided.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114777255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The present research work explores application of perimeter method to determine fractal dimension for both glaucoma and non glaucoma eye. Fractal dimension is a real value treated as over simplified diagnostics parameter to determine earlier detection of glaucoma. A perimeter method is a well known method adopted to determine the fractal dimension of natural objects such as leaves and some biological organs. But no attempt has been made to consider importance of selecting an initial point and rotation of object to determine the fractal dimension. Hence this work explains procedure to determine fractal dimension considering rotation of object and selecting the initial point. The results obtained signify that rotation of object and selecting the initial point definitely influences fractal dimension for glaucoma eye, as found in this present case study. The fractal dimension found to be lesser for glaucoma eye because of attaining regular shape of optic cup and it is accounted by losses of optic nerve and blood vessels in different stages. The fractal dimension could be treated as an alternative method for manual deduction of glaucoma through cup-disc diameter ratio by ophthalmologist. The author concludes that as soon as eye attacked by glaucoma disease, fractal dimension starts to decrease leading to lesser values. The fractal dimension values found to be in the range of 1.4242 to 1.5986 for glaucoma and for non glaucoma 1.8643 to 2.3027. Hence fractal dimension as soon as it reaches 1.5986 (1.6) signifies earlier detection of glaucoma. The rotation of object also considered in the interval of 600 (00 to 3000) to determine the fractal dimension. One more finding is that transforming from irregular shape of optic cup to regular shape from non-glaucoma to glaucoma could be visualized in different stages of images, under section four.
{"title":"Fractal Dimension with Object Rotation: A Case Study with Glaucoma Eye","authors":"Dharmanna Lamani, T. Manjunath, Ramegowda","doi":"10.1109/ICSIP.2014.22","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.22","url":null,"abstract":"The present research work explores application of perimeter method to determine fractal dimension for both glaucoma and non glaucoma eye. Fractal dimension is a real value treated as over simplified diagnostics parameter to determine earlier detection of glaucoma. A perimeter method is a well known method adopted to determine the fractal dimension of natural objects such as leaves and some biological organs. But no attempt has been made to consider importance of selecting an initial point and rotation of object to determine the fractal dimension. Hence this work explains procedure to determine fractal dimension considering rotation of object and selecting the initial point. The results obtained signify that rotation of object and selecting the initial point definitely influences fractal dimension for glaucoma eye, as found in this present case study. The fractal dimension found to be lesser for glaucoma eye because of attaining regular shape of optic cup and it is accounted by losses of optic nerve and blood vessels in different stages. The fractal dimension could be treated as an alternative method for manual deduction of glaucoma through cup-disc diameter ratio by ophthalmologist. The author concludes that as soon as eye attacked by glaucoma disease, fractal dimension starts to decrease leading to lesser values. The fractal dimension values found to be in the range of 1.4242 to 1.5986 for glaucoma and for non glaucoma 1.8643 to 2.3027. Hence fractal dimension as soon as it reaches 1.5986 (1.6) signifies earlier detection of glaucoma. The rotation of object also considered in the interval of 600 (00 to 3000) to determine the fractal dimension. One more finding is that transforming from irregular shape of optic cup to regular shape from non-glaucoma to glaucoma could be visualized in different stages of images, under section four.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130154842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Since decades, real time hardware implementation of Text-To-Speech system has been drawing attention of the research community due to its various real time applications. These include reading aids for the blind, talking aid for the vocally handicapped and training aids and other commercial applications. All these applications demand the real time embedded platform to meet the real time specifications such as speed, power, space requirements etc. In this context the embedded processor ARM (Advanced RISC Machine), has been chosen as hardware platform to implement Text-To-Speech conversion. This conversion needs algorithms to perform various operations like parts of speech tagging, phrase marking, word to phoneme conversion and clustergen synthesis. These algorithms are coded and developed in C using eclipse IDE and finally implemented on commercially available ARM9 microcontroller (AT91SAM9263EJ-S). Experiments have been performed on ARM microcontroller using test cases. It has been observed that the performance of the ARM based implementation is very close to x86 implementation.
{"title":"ARM Based Implementation of Text-to-Speech (TTS) for Real Time Embedded System","authors":"A. Rawoof, Kulesh, K. C. Ray","doi":"10.1109/ICSIP.2014.36","DOIUrl":"https://doi.org/10.1109/ICSIP.2014.36","url":null,"abstract":"Since decades, real time hardware implementation of Text-To-Speech system has been drawing attention of the research community due to its various real time applications. These include reading aids for the blind, talking aid for the vocally handicapped and training aids and other commercial applications. All these applications demand the real time embedded platform to meet the real time specifications such as speed, power, space requirements etc. In this context the embedded processor ARM (Advanced RISC Machine), has been chosen as hardware platform to implement Text-To-Speech conversion. This conversion needs algorithms to perform various operations like parts of speech tagging, phrase marking, word to phoneme conversion and clustergen synthesis. These algorithms are coded and developed in C using eclipse IDE and finally implemented on commercially available ARM9 microcontroller (AT91SAM9263EJ-S). Experiments have been performed on ARM microcontroller using test cases. It has been observed that the performance of the ARM based implementation is very close to x86 implementation.","PeriodicalId":111591,"journal":{"name":"2014 Fifth International Conference on Signal and Image Processing","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117076437","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：