Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036448
Arvind K Sharma, D. Mittal
Hash Functions have a distinct paramount significance in the sub domain of Networking like Network Security, Computer Security and Internet Security as compare to Symmetric and Public Key Encryption-Decryption Techniques. Major issues primarily which resolved by any hash algorithm are to managing the Integrity of Plaint-text Message(s) which are to be transmitting between communicating parties and to prove the Authenticity of Resources (Users/Machines), with digital signatures as well. Hash function also utilized for computing random secrect key of fixed length which further feeds to Symmetric and Public Key Cryptosystems in particular Key Management. Different level of security provided by different algorithms depending on how difficult is to break them. The most well-known hash algorithms are MD4, MD5, SHA, JH, Skein, Grøstl, Blake, Hamsi, Fugue, Crush, Whirlpool, Tav etc. In this paper we are discussing importance of hash functions, hash functions widely used in networking their application, literature and most importantly various Attacks applicable on hash functions and compression functions utilized by hash functions.
{"title":"Cryptography & Network Security Hash Function Applications, Attacks and Advances: A Review","authors":"Arvind K Sharma, D. Mittal","doi":"10.1109/ICISC44355.2019.9036448","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036448","url":null,"abstract":"Hash Functions have a distinct paramount significance in the sub domain of Networking like Network Security, Computer Security and Internet Security as compare to Symmetric and Public Key Encryption-Decryption Techniques. Major issues primarily which resolved by any hash algorithm are to managing the Integrity of Plaint-text Message(s) which are to be transmitting between communicating parties and to prove the Authenticity of Resources (Users/Machines), with digital signatures as well. Hash function also utilized for computing random secrect key of fixed length which further feeds to Symmetric and Public Key Cryptosystems in particular Key Management. Different level of security provided by different algorithms depending on how difficult is to break them. The most well-known hash algorithms are MD4, MD5, SHA, JH, Skein, Grøstl, Blake, Hamsi, Fugue, Crush, Whirlpool, Tav etc. In this paper we are discussing importance of hash functions, hash functions widely used in networking their application, literature and most importantly various Attacks applicable on hash functions and compression functions utilized by hash functions.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114801042","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036421
Shivangi Dhariwal, Hemant Makwana
the cloud computing offers the efficient computation and the storage for providing the high performance computing. In this context the different job management schemes and scheduling techniques are implemented for improving the performance of cloud servers. Because the cloud servers are much costly it is required to utilize the maximum resources for gaining the profit from the cloud infrastructure. In order to perform this task two major cases are considered for providing the solution. First the server waiting for job and all the resources are free, secondly the job achieving their waiting time and rent the third party server temporarily. In this context for demonstrating the issues and challenges first the job allocation system is explained using the suitable algorithm steps. In further for improving the performance of the server three additional algorithms are implemented which computes the maximum possible profit on the basis of available server speed and size. The implementation of the proposed system is provided using the cloudsim simulation tool and with the help of JAVA technology. In addition of that the performance of the system is also identified in terms of net profit, optimal profit and the maximum profit. The proposed technique is found acceptable for the real world scenarios also.
{"title":"Maximize the Cloud Profit to Improved QoS in Cloud Computing: Design and Analysis","authors":"Shivangi Dhariwal, Hemant Makwana","doi":"10.1109/ICISC44355.2019.9036421","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036421","url":null,"abstract":"the cloud computing offers the efficient computation and the storage for providing the high performance computing. In this context the different job management schemes and scheduling techniques are implemented for improving the performance of cloud servers. Because the cloud servers are much costly it is required to utilize the maximum resources for gaining the profit from the cloud infrastructure. In order to perform this task two major cases are considered for providing the solution. First the server waiting for job and all the resources are free, secondly the job achieving their waiting time and rent the third party server temporarily. In this context for demonstrating the issues and challenges first the job allocation system is explained using the suitable algorithm steps. In further for improving the performance of the server three additional algorithms are implemented which computes the maximum possible profit on the basis of available server speed and size. The implementation of the proposed system is provided using the cloudsim simulation tool and with the help of JAVA technology. In addition of that the performance of the system is also identified in terms of net profit, optimal profit and the maximum profit. The proposed technique is found acceptable for the real world scenarios also.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"195 Pt B 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116353040","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036416
Rakhi Gupta, Nashrah Gowalkar, S.D Joshi, S. Patil
Twitter sentiment analysis is an application of sentiment analysis on data from Twitter (tweets) for sentiment extraction. It automatically identifies whether a tweet expresses positive, negative or neutral opinion about and individual or an entity. The aim of this paper is to give detail explanation about the process of Crime mitigation by analyzing sentiment over the data generated by a social platform twitter using machine learning. This study focuses on a comparison done between, two classifier models, for data streamed from twitter. The paper then uses the comparative study for crime mitigation. Classification of sentiment can be done at various level. In our paper, the focus is on Feature level classification. [2]Risk analysis and profiling is then done on the predicted classes.
{"title":"Comparative study of classifier models for Crime Mitigation and Risk Modelling","authors":"Rakhi Gupta, Nashrah Gowalkar, S.D Joshi, S. Patil","doi":"10.1109/ICISC44355.2019.9036416","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036416","url":null,"abstract":"Twitter sentiment analysis is an application of sentiment analysis on data from Twitter (tweets) for sentiment extraction. It automatically identifies whether a tweet expresses positive, negative or neutral opinion about and individual or an entity. The aim of this paper is to give detail explanation about the process of Crime mitigation by analyzing sentiment over the data generated by a social platform twitter using machine learning. This study focuses on a comparison done between, two classifier models, for data streamed from twitter. The paper then uses the comparative study for crime mitigation. Classification of sentiment can be done at various level. In our paper, the focus is on Feature level classification. [2]Risk analysis and profiling is then done on the predicted classes.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121984374","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036426
Sunil kumar Bidhan, Laxmi Ahuja, S. Khatri, S. Som
A large quantity of statistics is generated because of the explosive boom inside the wide variety of devices connected to the Internet of Things. However, such data large amount of data is not useful without the power of analytics. Different IOT, big data, and analytics have enable individuals to get significant understanding into vast measure of information produced by IoT gadgets. Nonetheless, these arrangements are still in their underlying state, and the area comes up short on a far reaching overview. The contribution of this paper is to purpose a five layer architecture for big IOT data. The functionality of five different layers of big IOT data architecture is explained. The opportunities for big IoT data architecture are discussed. Further more challenges face by the architecture are also explained.
{"title":"Anatomy of Big Iot Data analytics","authors":"Sunil kumar Bidhan, Laxmi Ahuja, S. Khatri, S. Som","doi":"10.1109/ICISC44355.2019.9036426","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036426","url":null,"abstract":"A large quantity of statistics is generated because of the explosive boom inside the wide variety of devices connected to the Internet of Things. However, such data large amount of data is not useful without the power of analytics. Different IOT, big data, and analytics have enable individuals to get significant understanding into vast measure of information produced by IoT gadgets. Nonetheless, these arrangements are still in their underlying state, and the area comes up short on a far reaching overview. The contribution of this paper is to purpose a five layer architecture for big IOT data. The functionality of five different layers of big IOT data architecture is explained. The opportunities for big IoT data architecture are discussed. Further more challenges face by the architecture are also explained.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129663935","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036378
Navjot Kour, Smvdu Katra, N. Gondhi
Content Based Image retrieval(CBIR) has a important place in research field. This paper deals with the realization of different approaches used in image retrieval based on content. In CBIR, a query image is searched from larger database by selecting features from image and then exact match is retrieved using efficient algorithms. The various CBIR approaches are reviewed such as relevance feedback, SVM, graphs, wavelet transform, Gabor filter, Semantic templates, GMM, Fuzzy logic, Object ontology, Machine intelligence and Histogram. In this paper, the various graph theory algorithms used in CBIR have been discussed and compared viz. NN Graphs, Collocation Tree, Graphlets, Efficient Manifold Ranking, A* algorithm and segmentation algorithm.
{"title":"Assessment on various Approaches for Content Based Image Retrieval","authors":"Navjot Kour, Smvdu Katra, N. Gondhi","doi":"10.1109/ICISC44355.2019.9036378","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036378","url":null,"abstract":"Content Based Image retrieval(CBIR) has a important place in research field. This paper deals with the realization of different approaches used in image retrieval based on content. In CBIR, a query image is searched from larger database by selecting features from image and then exact match is retrieved using efficient algorithms. The various CBIR approaches are reviewed such as relevance feedback, SVM, graphs, wavelet transform, Gabor filter, Semantic templates, GMM, Fuzzy logic, Object ontology, Machine intelligence and Histogram. In this paper, the various graph theory algorithms used in CBIR have been discussed and compared viz. NN Graphs, Collocation Tree, Graphlets, Efficient Manifold Ranking, A* algorithm and segmentation algorithm.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"02 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127192498","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036468
Manjhari T., J. S. Kumar, Shabareeshwaran T.
In wireless sensor network (WSN) every node is capable of sense and monitor the parameters in various fields such as industry, agriculture, forest etc., and data forward to sink node in wireless. In WSN network lifetime and energy is key issue because all nodes operated in limited battery power and cannot be replaced easily. Energy efficient clustering technique can be used in order to give better network existence and low energy utilization. In this work, the Secondary Cluster Head (SCH) selected in the obtainable clustering system and performance was analyzed. A comparison among Low-Energy Adaptive Clustering Hierarchy (LEACH), Fuzzy C-Means and K-Means were done based on the amount of dead and alive nodes. The proposed method significantly extends the network lifetime and fuzzy C-means protocol performed well in terms of first and last dead nodes.
{"title":"Performance Analysis of Secondary Cluster Head in Wireless Sensor Network","authors":"Manjhari T., J. S. Kumar, Shabareeshwaran T.","doi":"10.1109/ICISC44355.2019.9036468","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036468","url":null,"abstract":"In wireless sensor network (WSN) every node is capable of sense and monitor the parameters in various fields such as industry, agriculture, forest etc., and data forward to sink node in wireless. In WSN network lifetime and energy is key issue because all nodes operated in limited battery power and cannot be replaced easily. Energy efficient clustering technique can be used in order to give better network existence and low energy utilization. In this work, the Secondary Cluster Head (SCH) selected in the obtainable clustering system and performance was analyzed. A comparison among Low-Energy Adaptive Clustering Hierarchy (LEACH), Fuzzy C-Means and K-Means were done based on the amount of dead and alive nodes. The proposed method significantly extends the network lifetime and fuzzy C-means protocol performed well in terms of first and last dead nodes.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"430 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132201364","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036461
Arya Rajiv Chaloli, Karthik Bhat, Vishwas N S
Conventional approaches to solving the Gait-Generation problem in humanoids are usually static in nature and are modeled based on a particular environment. These approaches can work well in a constrained context but fail to adapt to different environments. Hence turning our view and drawing inspiration from the various processes in nature, observing its principles and understanding the underlying concepts behind them, greatly improves our perspective in tackling this problem. There has been significant research in particular areas of such an approach yet the integration of these principles to form a wholesome approach has not been looked at with great detail. The approach taken intends to solve this problem and generate gait on a humanoid robot (the INDRA platform) by implementing a biologically inspired control approach called Central Pattern Generator (CPG) with neural oscillators. The parameters of the neural oscillator are tuned using Genetic algorithms and a policy is created to help the robot adapt to new environments of which there is no previous knowledge with the help of Reinforcement Learning. Therefore, this paper attempts to bring out a solution that combines multiple biologically derived approaches to generate a robust and stable gait.
{"title":"Nature Inspired Approaches combined with Dynamic Learning to generate stable gait in Humanoids","authors":"Arya Rajiv Chaloli, Karthik Bhat, Vishwas N S","doi":"10.1109/ICISC44355.2019.9036461","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036461","url":null,"abstract":"Conventional approaches to solving the Gait-Generation problem in humanoids are usually static in nature and are modeled based on a particular environment. These approaches can work well in a constrained context but fail to adapt to different environments. Hence turning our view and drawing inspiration from the various processes in nature, observing its principles and understanding the underlying concepts behind them, greatly improves our perspective in tackling this problem. There has been significant research in particular areas of such an approach yet the integration of these principles to form a wholesome approach has not been looked at with great detail. The approach taken intends to solve this problem and generate gait on a humanoid robot (the INDRA platform) by implementing a biologically inspired control approach called Central Pattern Generator (CPG) with neural oscillators. The parameters of the neural oscillator are tuned using Genetic algorithms and a policy is created to help the robot adapt to new environments of which there is no previous knowledge with the help of Reinforcement Learning. Therefore, this paper attempts to bring out a solution that combines multiple biologically derived approaches to generate a robust and stable gait.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125384694","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036427
B. A. Reddy, D. Sowjanya
In classical boundary layer (First-order) SMC strategy, the tracking performance and robustness affected when this controller is designed to reduce chattering phenomenon. To improve this tracking performance higher order SMC strategies are preferred. In this paper, higher order smc algorithms are applied to regulate the armature control dc motor. These higher order sliding modes reduces the chattering without affecting tracking accuracy and robustness and also insensitive to disturbance. This higher-order SMC controller smoothens the control signal. This paper examine two different control algorithms related to second-order SMC, these are twisting and super twisting. The effectiveness of these control algorithms in regulating the speed and angular position of the DC motor is verified in simulation in MATLAB environment. The speed characteristics of DC motor under different initial conditions are shown in simulation using twisting and super twisting higher order SMC strategy.
{"title":"Regulation of A DC Motor Using Second-Order Sliding Mode Strategies","authors":"B. A. Reddy, D. Sowjanya","doi":"10.1109/ICISC44355.2019.9036427","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036427","url":null,"abstract":"In classical boundary layer (First-order) SMC strategy, the tracking performance and robustness affected when this controller is designed to reduce chattering phenomenon. To improve this tracking performance higher order SMC strategies are preferred. In this paper, higher order smc algorithms are applied to regulate the armature control dc motor. These higher order sliding modes reduces the chattering without affecting tracking accuracy and robustness and also insensitive to disturbance. This higher-order SMC controller smoothens the control signal. This paper examine two different control algorithms related to second-order SMC, these are twisting and super twisting. The effectiveness of these control algorithms in regulating the speed and angular position of the DC motor is verified in simulation in MATLAB environment. The speed characteristics of DC motor under different initial conditions are shown in simulation using twisting and super twisting higher order SMC strategy.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126536256","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036417
R. P, N. Mahakalkar, Tripty Singh
The early detection of Diabetic Retinopathy is necessary to prevent blindness. Retinal imaging is a common clinical procedure used to record the visualisation of the retina. The main difficulty in capturing the ocular fundus is image quality which is affected by medial opacities. Micro Aneurysms (MA) are the earliest clinical sign of Diabetic Retinopathy. The appearance and structure of blood vessels in retinal images are main problem in diagnosis of eye diseases. The detection of blood vessels is difficult in the automatic processing of retinal images. In proposed method, for segmentation of blood vessels, the contrast-limited adaptive histogram equalisation (CLAHE) is used along with canny edge technique for exact edge location. The method is tested on Digital Retinal Images for Vessel Extraction (DRIVE) database. Comparative results of all basic Edge detection techniques with respect to the quality metrics parameter such as energy, entropy, mean etc is analysed. It is noted that proposed method performs good in extracting the vascular pattern than traditional edge detection techniques.
{"title":"Novel Approach For Detection Of Early Diabetic Retinopathy","authors":"R. P, N. Mahakalkar, Tripty Singh","doi":"10.1109/ICISC44355.2019.9036417","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036417","url":null,"abstract":"The early detection of Diabetic Retinopathy is necessary to prevent blindness. Retinal imaging is a common clinical procedure used to record the visualisation of the retina. The main difficulty in capturing the ocular fundus is image quality which is affected by medial opacities. Micro Aneurysms (MA) are the earliest clinical sign of Diabetic Retinopathy. The appearance and structure of blood vessels in retinal images are main problem in diagnosis of eye diseases. The detection of blood vessels is difficult in the automatic processing of retinal images. In proposed method, for segmentation of blood vessels, the contrast-limited adaptive histogram equalisation (CLAHE) is used along with canny edge technique for exact edge location. The method is tested on Digital Retinal Images for Vessel Extraction (DRIVE) database. Comparative results of all basic Edge detection techniques with respect to the quality metrics parameter such as energy, entropy, mean etc is analysed. It is noted that proposed method performs good in extracting the vascular pattern than traditional edge detection techniques.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124945798","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}
Pub Date : 2019-01-01DOI: 10.1109/ICISC44355.2019.9036362
Girish Gidaye, J. Nirmal, Kadria Ezzine, M. Frikha
The presence of various vocal pathologies seriously affects the quality of the speech. These pathologies can treat better if they are diagnosed in primary stage. In this work, for early detection, we conceived non-intrusive automatic vocal fold pathologies recognition system. The sustained vowel /ah:/ with normal intonation for both healthy and pathologic subjects are extracted from PdA corpus. Glottal Inverse Filtering (GIF) is used to estimate glottal pulseform from frame of voiced speech signal. Various time and frequency domain descriptors are extracted from glottal pulseform and used for detection of voice disorder. For inverse filtering, Iterative Adaptive Inverse Filtering (IAIF) algorithm with Discrete All-Pole (DAP) model for vocal tract is used. The extracted descriptors are fed to classifier to separate healthy and pathologic subjects. The artificial neural network (ANN), support vector machine (SVM) and k-nearest neighbour (kNN) were used for classification. We have used box and density plots to investigate the discrimination ability of extracted glottal descriptors. To observe the discrimination ability of descriptors quantitatively, analysis of variance (ANOVA) and information gain feature scoring method is used. The time domain descriptors were found very rich in discrimination compared to frequency domain. The best classification rate achieved were 99.85%, 99.90% and 99.95% with kNN, SVM and ANN respectively.
{"title":"Effective Detection of Voice Dysfunction Using Glottic Flow Descriptors","authors":"Girish Gidaye, J. Nirmal, Kadria Ezzine, M. Frikha","doi":"10.1109/ICISC44355.2019.9036362","DOIUrl":"https://doi.org/10.1109/ICISC44355.2019.9036362","url":null,"abstract":"The presence of various vocal pathologies seriously affects the quality of the speech. These pathologies can treat better if they are diagnosed in primary stage. In this work, for early detection, we conceived non-intrusive automatic vocal fold pathologies recognition system. The sustained vowel /ah:/ with normal intonation for both healthy and pathologic subjects are extracted from PdA corpus. Glottal Inverse Filtering (GIF) is used to estimate glottal pulseform from frame of voiced speech signal. Various time and frequency domain descriptors are extracted from glottal pulseform and used for detection of voice disorder. For inverse filtering, Iterative Adaptive Inverse Filtering (IAIF) algorithm with Discrete All-Pole (DAP) model for vocal tract is used. The extracted descriptors are fed to classifier to separate healthy and pathologic subjects. The artificial neural network (ANN), support vector machine (SVM) and k-nearest neighbour (kNN) were used for classification. We have used box and density plots to investigate the discrimination ability of extracted glottal descriptors. To observe the discrimination ability of descriptors quantitatively, analysis of variance (ANOVA) and information gain feature scoring method is used. The time domain descriptors were found very rich in discrimination compared to frequency domain. The best classification rate achieved were 99.85%, 99.90% and 99.95% with kNN, SVM and ANN respectively.","PeriodicalId":419157,"journal":{"name":"2019 Third International Conference on Inventive Systems and Control (ICISC)","volume":"59 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114036980","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：