Pub Date : 1900-01-01DOI: 10.1109/ICSIPA.2013.6708048
T. Adam, M. Salam, T. Gunawan
Traditional cepstral analysis methods are often used as part of feature extraction process in speech recognition. However the cepstral analysis method uses the Discrete Fourier Transform (DFT) in one of its computation process. The DFT uses fixed frame resolution to analyze frames of signal thus it will result in an analysis that would not accurately analyze localized events. This paper investigates the use of the Discrete Wavelet Transform (DWT) for calculating the cepstrum coefficients. Two wavelet types with different decomposition level are experimented to yield the cepstrum which is called the Wavelet Cepstral Coefficient (WCC). To test the WCC speech recognizing task of recognizing 26 English alphabets were conducted. Under same number of feature dimension the WCC outperformed the MFCC with about 20% in terms of recognition rate under both speaker dependent and speaker independent task.
{"title":"Wavelet based Cepstral Coefficients for neural network speech recognition","authors":"T. Adam, M. Salam, T. Gunawan","doi":"10.1109/ICSIPA.2013.6708048","DOIUrl":"https://doi.org/10.1109/ICSIPA.2013.6708048","url":null,"abstract":"Traditional cepstral analysis methods are often used as part of feature extraction process in speech recognition. However the cepstral analysis method uses the Discrete Fourier Transform (DFT) in one of its computation process. The DFT uses fixed frame resolution to analyze frames of signal thus it will result in an analysis that would not accurately analyze localized events. This paper investigates the use of the Discrete Wavelet Transform (DWT) for calculating the cepstrum coefficients. Two wavelet types with different decomposition level are experimented to yield the cepstrum which is called the Wavelet Cepstral Coefficient (WCC). To test the WCC speech recognizing task of recognizing 26 English alphabets were conducted. Under same number of feature dimension the WCC outperformed the MFCC with about 20% in terms of recognition rate under both speaker dependent and speaker independent task.","PeriodicalId":440373,"journal":{"name":"2013 IEEE International Conference on Signal and Image Processing Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115168266","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 : 1900-01-01DOI: 10.1109/ICSIPA.2013.6707992
A. B. Dehkordi, S. Abu-Bakar
Iris recognition is known to be as one of the most accurate biometric modalities. In iris image processing, the unwanted textures in the iris region such as those belong to pupil, eyelashes, eyelids, shadows and light reflections are defined as noises. These unwanted noises have strong gray values which cause wrong threshold value selection and thus, result in reducing the performance of the iris recognition system. In this paper, we proposed a multiple thresholding method for detection of eyelids, eyelash textures and light reflections and pupil pixels. The threshold values related to these noises are selected based on the information obtained from the histogram of the normalized iris image. The proposed method was applied to the CASIA V.3 iris image database, version three, from the institute of automation, Chinese academy of science and has 99.62% recognition rate with 0.04 false rejection rate (FRR).
{"title":"Noise reduction in iris recognition using multiple thresholding","authors":"A. B. Dehkordi, S. Abu-Bakar","doi":"10.1109/ICSIPA.2013.6707992","DOIUrl":"https://doi.org/10.1109/ICSIPA.2013.6707992","url":null,"abstract":"Iris recognition is known to be as one of the most accurate biometric modalities. In iris image processing, the unwanted textures in the iris region such as those belong to pupil, eyelashes, eyelids, shadows and light reflections are defined as noises. These unwanted noises have strong gray values which cause wrong threshold value selection and thus, result in reducing the performance of the iris recognition system. In this paper, we proposed a multiple thresholding method for detection of eyelids, eyelash textures and light reflections and pupil pixels. The threshold values related to these noises are selected based on the information obtained from the histogram of the normalized iris image. The proposed method was applied to the CASIA V.3 iris image database, version three, from the institute of automation, Chinese academy of science and has 99.62% recognition rate with 0.04 false rejection rate (FRR).","PeriodicalId":440373,"journal":{"name":"2013 IEEE International Conference on Signal and Image Processing Applications","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128679802","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 : 1900-01-01DOI: 10.1109/ICSIPA.2013.6707997
K. B. Low, U. U. Sheikh
Local Ternary Pattern (LTP) is usually applied for texture classification problems. In this work, we propose LTP for human gait characterization for the purpose of human identification. Our proposed method is based on the Gait Energy Image (GEI) whereby edge information over a complete gait cycle is extracted. However, GEI does not contain enough human body structure information for human recognition purpose. Therefore, LTP is used to extract texture information from all pixels in the human gait region which preserves more discriminative features of the subject. Gait cycle estimation is computed by using the aspect ratio of the subject's bounding box. After that, LTP features are averaged over a full gait cycle and a 2D joint histogram of the LTP is computed. At the end, K nearest-neighbor (k-NN) is used to obtain the final recognition results. The proposed method achieved higher accuracy compared to other methods when tested on the CMU MoBo human gait database. The proposed LTP method is easy to implement and also has the advantage of significantly lower computation time.
{"title":"Gait recognition using Local Ternary Pattern (LTP)","authors":"K. B. Low, U. U. Sheikh","doi":"10.1109/ICSIPA.2013.6707997","DOIUrl":"https://doi.org/10.1109/ICSIPA.2013.6707997","url":null,"abstract":"Local Ternary Pattern (LTP) is usually applied for texture classification problems. In this work, we propose LTP for human gait characterization for the purpose of human identification. Our proposed method is based on the Gait Energy Image (GEI) whereby edge information over a complete gait cycle is extracted. However, GEI does not contain enough human body structure information for human recognition purpose. Therefore, LTP is used to extract texture information from all pixels in the human gait region which preserves more discriminative features of the subject. Gait cycle estimation is computed by using the aspect ratio of the subject's bounding box. After that, LTP features are averaged over a full gait cycle and a 2D joint histogram of the LTP is computed. At the end, K nearest-neighbor (k-NN) is used to obtain the final recognition results. The proposed method achieved higher accuracy compared to other methods when tested on the CMU MoBo human gait database. The proposed LTP method is easy to implement and also has the advantage of significantly lower computation time.","PeriodicalId":440373,"journal":{"name":"2013 IEEE International Conference on Signal and Image Processing Applications","volume":"48 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133457018","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 : 1900-01-01DOI: 10.1109/ICSIPA.2013.6708024
S. Nunoo, U. Chude-Okonkwo, R. Ngah
There has been extensive discussions on the performance of the LMS algorithm in nonstationary environment. In some instances its performance in such environments has been compared with that of other adaptive algorithms. However, most existing works treated in this respect focus on narrowband channels. In this paper, we investigate the performance of the LMS, NLMS and LMF algorithms in tracking time-varying ultra wideband (UWB) channels. Channel measurements were conducted for an indoor environment and the resultant channel impulse response is used in the analysis. The results show that the LMF and NLMS algorithms outperform the LMS algorithm in time-varying UWB channels with the NLMS providing the best performance.
{"title":"Performance of LMS, NLMS and LMF algorithms in tracking time-varying UWB channels","authors":"S. Nunoo, U. Chude-Okonkwo, R. Ngah","doi":"10.1109/ICSIPA.2013.6708024","DOIUrl":"https://doi.org/10.1109/ICSIPA.2013.6708024","url":null,"abstract":"There has been extensive discussions on the performance of the LMS algorithm in nonstationary environment. In some instances its performance in such environments has been compared with that of other adaptive algorithms. However, most existing works treated in this respect focus on narrowband channels. In this paper, we investigate the performance of the LMS, NLMS and LMF algorithms in tracking time-varying ultra wideband (UWB) channels. Channel measurements were conducted for an indoor environment and the resultant channel impulse response is used in the analysis. The results show that the LMF and NLMS algorithms outperform the LMS algorithm in time-varying UWB channels with the NLMS providing the best performance.","PeriodicalId":440373,"journal":{"name":"2013 IEEE International Conference on Signal and Image Processing Applications","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122043759","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 : 1900-01-01DOI: 10.1109/ICSIPA.2013.6708046
R. Ullah, N. Fisal, Hashim Safdar, W. Maqbool, Z. Khalid, A. Khan
Interference Management (IM) is one of the major challenges of next generation wireless communication. Fractional Frequency Reuse (FFR) has been acknowledged as an efficient IM technique, which offers significant capacity enhancement and improve cell edge coverage with low complexity. In literature, FFR has been analyzed mostly with cellular networks described by Hexagon Grid Model, which is neither tractable nor scalable to the dense deployment of next generation wireless networks. Moreover, the perfect geometry based grid model tends to overestimate the system performance and not able to reflect the reality. In this paper, we use the stochastic geometry approach, FFR is analyzed with cellular network modeled by homogeneous Poisson Point Process (PPP). A dynamic frequency allocation scheme is proposed which take into account the randomness of the cell coverage area describe by Voronoi tessellation. It is shown that the proposed scheme outperforms the traditional fixed frequency allocation schemes in terms of per user capacity and capacity density.
{"title":"Voronoi cell geometry based dynamic Fractional Frequency Reuse for OFDMA cellular networks","authors":"R. Ullah, N. Fisal, Hashim Safdar, W. Maqbool, Z. Khalid, A. Khan","doi":"10.1109/ICSIPA.2013.6708046","DOIUrl":"https://doi.org/10.1109/ICSIPA.2013.6708046","url":null,"abstract":"Interference Management (IM) is one of the major challenges of next generation wireless communication. Fractional Frequency Reuse (FFR) has been acknowledged as an efficient IM technique, which offers significant capacity enhancement and improve cell edge coverage with low complexity. In literature, FFR has been analyzed mostly with cellular networks described by Hexagon Grid Model, which is neither tractable nor scalable to the dense deployment of next generation wireless networks. Moreover, the perfect geometry based grid model tends to overestimate the system performance and not able to reflect the reality. In this paper, we use the stochastic geometry approach, FFR is analyzed with cellular network modeled by homogeneous Poisson Point Process (PPP). A dynamic frequency allocation scheme is proposed which take into account the randomness of the cell coverage area describe by Voronoi tessellation. It is shown that the proposed scheme outperforms the traditional fixed frequency allocation schemes in terms of per user capacity and capacity density.","PeriodicalId":440373,"journal":{"name":"2013 IEEE International Conference on Signal and Image Processing Applications","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117171171","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 : 1900-01-01DOI: 10.1109/ICSIPA.2013.6708026
U. Chude-Okonkwo, S. Nunoo, R. Ngah, Chollette C. Chude-Olisah, T. A. Rahman, Anthony A. Okafor
In some applications, multivariable stochastic processes that are composed of sequentially arranged independent weakly-stationary processes, may arise. Such multivariable process can be categorized as a class of non-summable processes with very complex probability density function. In this paper, we present the formal definition of such non-summable process, and provide a method of parameterizing and defining the statistical trend associated with the process. The illustration of a typical example of a multivariable non-summable process and how a bounded statistical trend can be obtained for the process is presented. The typical example is obtained from the simulation of a time-varying wideband wireless channel.
{"title":"Characterization and parameterization of a class of multivariable non-summable stochastic processes with bounded stochastic trends","authors":"U. Chude-Okonkwo, S. Nunoo, R. Ngah, Chollette C. Chude-Olisah, T. A. Rahman, Anthony A. Okafor","doi":"10.1109/ICSIPA.2013.6708026","DOIUrl":"https://doi.org/10.1109/ICSIPA.2013.6708026","url":null,"abstract":"In some applications, multivariable stochastic processes that are composed of sequentially arranged independent weakly-stationary processes, may arise. Such multivariable process can be categorized as a class of non-summable processes with very complex probability density function. In this paper, we present the formal definition of such non-summable process, and provide a method of parameterizing and defining the statistical trend associated with the process. The illustration of a typical example of a multivariable non-summable process and how a bounded statistical trend can be obtained for the process is presented. The typical example is obtained from the simulation of a time-varying wideband wireless channel.","PeriodicalId":440373,"journal":{"name":"2013 IEEE International Conference on Signal and Image Processing Applications","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132526332","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：