Pub Date : 2019-04-01DOI: 10.1049/IET-SPR.2018.5123
Haifeng Li, Guoqi Liu, Jian Zou
In the underdetermined model Y ^ = A X + N , where X is a K-group sparse matrix (i.e. it has no more than K non-zero rows), the matrix A may be also perturbed. Theoretically, a more relaxed condition means that fewer measurements are required to ensure sparse recovery. In this study, a relaxed sufficient condition is proposed for greedy block coordinate descent (GBCD) under total perturbations based on the restricted isometry property in order to guarantee that the support of X is recovered. We also show that GBCD fails in a more general case when 1 / ( K + 1 ) ≤ δ K + 1 < 1 .
在待定模型Y ^ = A X + N中，其中X是K群稀疏矩阵(即不超过K个非零行)，矩阵A也可能被摄动。理论上，更宽松的条件意味着需要更少的测量来确保稀疏恢复。本文基于有限等距特性，提出了全扰动下贪心块坐标下降(GBCD)的松弛充分条件，以保证X的支撑得到恢复。当1 / (K + 1)≤δ K + 1 < 1时，GBCD在更一般的情况下失效。
{"title":"Sufficient condition for exact support recovery of sparse signals through greedy block coordinate descent","authors":"Haifeng Li, Guoqi Liu, Jian Zou","doi":"10.1049/IET-SPR.2018.5123","DOIUrl":"https://doi.org/10.1049/IET-SPR.2018.5123","url":null,"abstract":"In the underdetermined model Y ^ = A X + N , where X is a K-group sparse matrix (i.e. it has no more than K non-zero rows), the matrix A may be also perturbed. Theoretically, a more relaxed condition means that fewer measurements are required to ensure sparse recovery. In this study, a relaxed sufficient condition is proposed for greedy block coordinate descent (GBCD) under total perturbations based on the restricted isometry property in order to guarantee that the support of X is recovered. We also show that GBCD fails in a more general case when 1 / ( K + 1 ) ≤ δ K + 1 < 1 .","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120529333","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 : 2019-01-08DOI: 10.1049/IET-SPR.2018.5357
A. Ebrahimi, H. R. Abutalebi, M. Karimi
In this study, the authors address the problem of passive mixed near-field and far-field sources localisation using a uniform linear array (ULA) in which the signals received by the array may come from mixed sources. This study presents a new two stage cumulant-based multiple signal classification (MUSIC) algorithm for passive source localisation using fourth-order cumulants of a ULA data. The significant characteristic of the proposed algorithm is that it constructs a new special cumulant matrix to acquire more information of signals received by a ULA. Consequently, the proposed algorithm gives high direction of arrival (DOA) and range estimation accuracy, and alleviates the array aperture loss. Monte Carlo simulations are established to verify the effectiveness of the proposed method in increasing direction of arrival and range estimation accuracies.
{"title":"Generalised two stage cumulants-based MUSIC algorithm for passive mixed sources localisation","authors":"A. Ebrahimi, H. R. Abutalebi, M. Karimi","doi":"10.1049/IET-SPR.2018.5357","DOIUrl":"https://doi.org/10.1049/IET-SPR.2018.5357","url":null,"abstract":"In this study, the authors address the problem of passive mixed near-field and far-field sources localisation using a uniform linear array (ULA) in which the signals received by the array may come from mixed sources. This study presents a new two stage cumulant-based multiple signal classification (MUSIC) algorithm for passive source localisation using fourth-order cumulants of a ULA data. The significant characteristic of the proposed algorithm is that it constructs a new special cumulant matrix to acquire more information of signals received by a ULA. Consequently, the proposed algorithm gives high direction of arrival (DOA) and range estimation accuracy, and alleviates the array aperture loss. Monte Carlo simulations are established to verify the effectiveness of the proposed method in increasing direction of arrival and range estimation accuracies.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"118209172","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 : 2016-12-01DOI: 10.1049/iet-spr.2015.0564
Shu-wen Xu, P. Shui, Xue-Ying Yan
A novel non-coherent detection scheme for range-spread targets in the sea clutter is developed in this study. The detector utilises the bi-window non-linear shrinkage map (BNSM) to reduce sea clutter while preserving target echoes before energy integration (EI). The EI detector based on BNSM (EI-BNSM) is compared with the conventional non-coherent constant false alarm ratio detectors based on measured data, and the good performance of EI-BNSM detector is reported.
{"title":"Non-coherent detection of radar target in heavy-tailed sea clutter using bi-window non-linear shrinkage map","authors":"Shu-wen Xu, P. Shui, Xue-Ying Yan","doi":"10.1049/iet-spr.2015.0564","DOIUrl":"https://doi.org/10.1049/iet-spr.2015.0564","url":null,"abstract":"A novel non-coherent detection scheme for range-spread targets in the sea clutter is developed in this study. The detector utilises the bi-window non-linear shrinkage map (BNSM) to reduce sea clutter while preserving target echoes before energy integration (EI). The EI detector based on BNSM (EI-BNSM) is compared with the conventional non-coherent constant false alarm ratio detectors based on measured data, and the good performance of EI-BNSM detector is reported.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121376137","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 : 2016-10-31DOI: 10.1049/iet-spr.2016.0125
M. Jeeva, T. Nagarajan, Vijayalakshmi Parthasarathy
Conventional multiband speech enhancement involves splitting the spectrum into various frequency bins and performing speech enhancement in each band independently. However, owing to the pole-interaction problem in the spectral domain, estimation of clean speech from the formants, suppressed by the influence of the formants in the neighbouring bands, may result in poor quality. To reduce the influence of stronger formants over the neighbouring bands, in the current work, clean speech is estimated by filtering unprocessed speech in the temporal domain into various equivalent rectangular bandwidth based subbands followed by discrete cosine transform (DCT) based spectral speech enhancement in each band using spectral subtraction/minimum mean square error (MMSE). To further enhance speech, a spectral subtraction-based approach that incorporates band-specific weighting factor obtained using respective band signal-to-noise ratio (SNR), and an MMSE estimator that calculates apriori speech presence/absence probability based on local and global apriori SNR rather than a fixed/equiprobable value are proposed. The performance of the algorithms is evaluated using perceptual evaluation of speech quality and composite speech quality measure. It is observed that DCT-derived spectrum based temporal-domain multiband speech enhancement algorithm outperforms the existing techniques for car, babble, train, white, and factory noise in the 0-10 dB SNR levels.
{"title":"Discrete cosine transform-derived spectrum-based speech enhancement algorithm using temporal-domain multiband filtering","authors":"M. Jeeva, T. Nagarajan, Vijayalakshmi Parthasarathy","doi":"10.1049/iet-spr.2016.0125","DOIUrl":"https://doi.org/10.1049/iet-spr.2016.0125","url":null,"abstract":"Conventional multiband speech enhancement involves splitting the spectrum into various frequency bins and performing speech enhancement in each band independently. However, owing to the pole-interaction problem in the spectral domain, estimation of clean speech from the formants, suppressed by the influence of the formants in the neighbouring bands, may result in poor quality. To reduce the influence of stronger formants over the neighbouring bands, in the current work, clean speech is estimated by filtering unprocessed speech in the temporal domain into various equivalent rectangular bandwidth based subbands followed by discrete cosine transform (DCT) based spectral speech enhancement in each band using spectral subtraction/minimum mean square error (MMSE). To further enhance speech, a spectral subtraction-based approach that incorporates band-specific weighting factor obtained using respective band signal-to-noise ratio (SNR), and an MMSE estimator that calculates apriori speech presence/absence probability based on local and global apriori SNR rather than a fixed/equiprobable value are proposed. The performance of the algorithms is evaluated using perceptual evaluation of speech quality and composite speech quality measure. It is observed that DCT-derived spectrum based temporal-domain multiband speech enhancement algorithm outperforms the existing techniques for car, babble, train, white, and factory noise in the 0-10 dB SNR levels.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124745526","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 : 2016-10-01DOI: 10.1049/iet-spr.2015.0288
Osman Büyük
In this paper, we make use of hidden Markov model (HMM) state alignment information in i-vector/probabilistic linear discriminant analysis (PLDA) framework to improve the verification performance in a text-dependent single utterance (TDSU) task. In the TDSU task, speakers repeat a fixed utterance in both enrollment and authentication sessions. Despite Gaussian mixture models (GMMs) have been the dominant modeling technique for text-independent applications, an HMM based method might be better suited for the TDSU task since it captures the co-articulation information better. Recently, powerful channel compensation techniques such as joint factor analysis (JFA), i-vectors and PLDA have been proposed for GMM based text-independent speaker verification. In this study, we train a separate i-vector/PLDA model for each sentence HMM state in order to utilize the alignment information of the HMM states in a TDSU task. The proposed method is tested using a multi-channel speaker verification database. In the experiments, it is observed that HMM state based i-vector/PLDA (i-vector/PLDA-HMM) provides approximately 67% relative reduction in equal error rate (EER) when compared to the i-vector/PLDA. The proposed method also outperforms the baseline GMM and sentence HMM methods. It yields approximately 51% relative reduction in EER over the best performing sentence HMM method.
本文利用i-vector/probabilistic linear discriminant analysis (PLDA)框架中的隐马尔可夫模型(HMM)状态对齐信息来提高文本依赖单话语(TDSU)任务的验证性能。在TDSU任务中，说话者在注册会话和身份验证会话中重复固定的话语。尽管高斯混合模型(GMMs)一直是文本无关应用程序的主要建模技术，但基于HMM的方法可能更适合TDSU任务，因为它可以更好地捕获协同发音信息。近年来，人们提出了联合因子分析(JFA)、i-vectors和PLDA等强大的通道补偿技术，用于基于GMM的文本无关说话人验证。在本研究中，我们为每个句子HMM状态训练一个单独的i-vector/PLDA模型，以便在TDSU任务中利用HMM状态的对齐信息。利用多通道说话人验证数据库对该方法进行了测试。在实验中，我们观察到基于HMM状态的i-vector/PLDA (i-vector/PLDA-HMM)与i-vector/PLDA相比，在等错误率(EER)方面提供了大约67%的相对降低。该方法也优于基线GMM和句子HMM方法。与表现最好的句子HMM方法相比，它产生了大约51%的相对EER降低。
{"title":"Sentence-HMM state-based i-vector/PLDA modelling for improved performance in text dependent single utterance speaker verification","authors":"Osman Büyük","doi":"10.1049/iet-spr.2015.0288","DOIUrl":"https://doi.org/10.1049/iet-spr.2015.0288","url":null,"abstract":"In this paper, we make use of hidden Markov model (HMM) state alignment information in i-vector/probabilistic linear discriminant analysis (PLDA) framework to improve the verification performance in a text-dependent single utterance (TDSU) task. In the TDSU task, speakers repeat a fixed utterance in both enrollment and authentication sessions. Despite Gaussian mixture models (GMMs) have been the dominant modeling technique for text-independent applications, an HMM based method might be better suited for the TDSU task since it captures the co-articulation information better. Recently, powerful channel compensation techniques such as joint factor analysis (JFA), i-vectors and PLDA have been proposed for GMM based text-independent speaker verification. In this study, we train a separate i-vector/PLDA model for each sentence HMM state in order to utilize the alignment information of the HMM states in a TDSU task. The proposed method is tested using a multi-channel speaker verification database. In the experiments, it is observed that HMM state based i-vector/PLDA (i-vector/PLDA-HMM) provides approximately 67% relative reduction in equal error rate (EER) when compared to the i-vector/PLDA. The proposed method also outperforms the baseline GMM and sentence HMM methods. It yields approximately 51% relative reduction in EER over the best performing sentence HMM method.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121721993","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 : 2016-07-18DOI: 10.1049/iet-spr.2015.0508
Huhe Dai, Liyan Yin, Ye Li
Atrial activity (AA) signal must first be extracted from atrial fibrillation electrocardiogram (AF ECG) before it is used to characterise AF. However, extracting AA signal is not an easy task, especially from single-lead ECG recording. The AA signals within QRS intervals extracted by the existing single-lead extraction methods are often heavily distorted due to the existence of large QRS residuals. In this study, the authors focus on reducing the QRS residuals in the extracted AA signals, and propose a novel signal extrapolation based method. AA signal is assumed to be band-limited, and a dedicated extrapolation formula is derived. Based on this extrapolation formula, the AA samples within QRS interval are reconstructed by using the ones in the adjacent SQ segments. The experiments with simulated AF ECGs showed that, after using the proposed method, the normalised mean square error of the AA signal extracted by average beat subtraction method decreased by 26–50%, 15–36%, 12–40 and 42–63% for simulated AF ECGs in lead I, II, V 1 and V 6, respectively. Experiments with real AF ECG also proved that the proposed method is able to greatly reduce the ventricular residuals of the extracted AA signal.
{"title":"QRS residual removal in atrial activity signals extracted from single lead: a new perspective based on signal extrapolation","authors":"Huhe Dai, Liyan Yin, Ye Li","doi":"10.1049/iet-spr.2015.0508","DOIUrl":"https://doi.org/10.1049/iet-spr.2015.0508","url":null,"abstract":"Atrial activity (AA) signal must first be extracted from atrial fibrillation electrocardiogram (AF ECG) before it is used to characterise AF. However, extracting AA signal is not an easy task, especially from single-lead ECG recording. The AA signals within QRS intervals extracted by the existing single-lead extraction methods are often heavily distorted due to the existence of large QRS residuals. In this study, the authors focus on reducing the QRS residuals in the extracted AA signals, and propose a novel signal extrapolation based method. AA signal is assumed to be band-limited, and a dedicated extrapolation formula is derived. Based on this extrapolation formula, the AA samples within QRS interval are reconstructed by using the ones in the adjacent SQ segments. The experiments with simulated AF ECGs showed that, after using the proposed method, the normalised mean square error of the AA signal extracted by average beat subtraction method decreased by 26–50%, 15–36%, 12–40 and 42–63% for simulated AF ECGs in lead I, II, V 1 and V 6, respectively. Experiments with real AF ECG also proved that the proposed method is able to greatly reduce the ventricular residuals of the extracted AA signal.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125491638","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 : 2016-07-18DOI: 10.1049/iet-spr.2016.0140
X. Mao, Yunlong Yang, Hong Hong, Weibo Deng
Filters with antenna arrays are widely used for interference suppression in the temporal domain, frequency domain, space domain and so on. Meanwhile, the interference mitigation performance of a single-domain-based filter depends on a noticeable difference between the target signal and interference in the corresponding domain. However, the interference cannot be efficiently suppressed by a single-domain-based filter when the difference in this individual domain is small. To solve this problem, a multi-domain collaborative oblique projection filter is proposed in this study. First, multi-domain spaces are theoretically derived to distinguish the target signal and the interference, and then a multi-domain oblique projection operator is provided to recover the original target and suppress the interference. Depending on the multi-domain spaces, the filter has excellent performance when the difference is insignificant in each individual domain, whereas the performances of the single-domain-based filter and cascade filter degrade significantly. Finally, a space–polarisation–frequency domain collaborative filter based on oblique projection is given as an illustration. Performance analysis and simulation results are provided to illustrate the superiority of the proposed filter for interference mitigation.
{"title":"Multi-domain collaborative filter for interference suppressing","authors":"X. Mao, Yunlong Yang, Hong Hong, Weibo Deng","doi":"10.1049/iet-spr.2016.0140","DOIUrl":"https://doi.org/10.1049/iet-spr.2016.0140","url":null,"abstract":"Filters with antenna arrays are widely used for interference suppression in the temporal domain, frequency domain, space domain and so on. Meanwhile, the interference mitigation performance of a single-domain-based filter depends on a noticeable difference between the target signal and interference in the corresponding domain. However, the interference cannot be efficiently suppressed by a single-domain-based filter when the difference in this individual domain is small. To solve this problem, a multi-domain collaborative oblique projection filter is proposed in this study. First, multi-domain spaces are theoretically derived to distinguish the target signal and the interference, and then a multi-domain oblique projection operator is provided to recover the original target and suppress the interference. Depending on the multi-domain spaces, the filter has excellent performance when the difference is insignificant in each individual domain, whereas the performances of the single-domain-based filter and cascade filter degrade significantly. Finally, a space–polarisation–frequency domain collaborative filter based on oblique projection is given as an illustration. Performance analysis and simulation results are provided to illustrate the superiority of the proposed filter for interference mitigation.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128068687","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 : 2016-07-18DOI: 10.1049/iet-spr.2016.0298
S. Mahata, S. Saha, R. Kar, D. Mandal
This study presents a novel approach to design wideband infinite impulse response fractional order digital integrators (FODIs) for the half and one-fourth order integrators based on a parameter independent metaheuristic algorithm called colliding bodies optimisation (CBO). The performance of CBO-based FODIs have been compared with the designs based on three well-known benchmark evolutionary optimisation algorithms namely, real coded genetic algorithm (RGA), particle swarm optimisation (PSO), and differential evolution (DE) in terms of robustness, consistency, parameter sensitivity, convergence speed, and computational time. Simulations results confirm that the proposed CBO-based designed FODIs achieve consistently superior magnitude responses in a computationally efficient manner as compared with the designs based on RGA, PSO, and DE. The proposed CBO-based FODIs also significantly outperform all state-of-the-art designs reported in literature in terms of two different magnitude response error metrics. This research work underlines the potential of CBO as an efficient optimisation tool for the design of accurate digital approximations to the fractional order integrators.
{"title":"Optimal design of wideband infinite impulse response fractional order digital integrators using colliding bodies optimisation algorithm","authors":"S. Mahata, S. Saha, R. Kar, D. Mandal","doi":"10.1049/iet-spr.2016.0298","DOIUrl":"https://doi.org/10.1049/iet-spr.2016.0298","url":null,"abstract":"This study presents a novel approach to design wideband infinite impulse response fractional order digital integrators (FODIs) for the half and one-fourth order integrators based on a parameter independent metaheuristic algorithm called colliding bodies optimisation (CBO). The performance of CBO-based FODIs have been compared with the designs based on three well-known benchmark evolutionary optimisation algorithms namely, real coded genetic algorithm (RGA), particle swarm optimisation (PSO), and differential evolution (DE) in terms of robustness, consistency, parameter sensitivity, convergence speed, and computational time. Simulations results confirm that the proposed CBO-based designed FODIs achieve consistently superior magnitude responses in a computationally efficient manner as compared with the designs based on RGA, PSO, and DE. The proposed CBO-based FODIs also significantly outperform all state-of-the-art designs reported in literature in terms of two different magnitude response error metrics. This research work underlines the potential of CBO as an efficient optimisation tool for the design of accurate digital approximations to the fractional order integrators.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121231853","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 : 2016-07-15DOI: 10.1049/iet-spr.2016.0067
Yan Zhang, Hua Peng
One sample per person face recognition (OSPP) is a challenging problem in face recognition community. Lack of samples leads to performance deterioration. Extended sparse representation-based classifier (ESRC) demonstrates excellent performance on OSPP. However, because there are intra-class variant atoms in the dictionary of ESRC, the number of atoms in the dictionary is always large and it will spend a long time during recognition. In this study, the authors propose a new OSPP face recognition algorithm via sparse representation (OSPP-SR). A compressed dictionary and a new identification strategy are provided in OSPP-SR. It is proved theoretically and experimentally that OSPP-SR reaches better or similar performance but spends less time than ESRC. Experiments are conducted on three different databases (extended Yale Face database B, AR database and FERET database) to show the validity of OSPP-SR. Images under clean and noise conditions are also tested to evaluate the robustness of OSPP-SR.
{"title":"One sample per person face recognition via sparse representation","authors":"Yan Zhang, Hua Peng","doi":"10.1049/iet-spr.2016.0067","DOIUrl":"https://doi.org/10.1049/iet-spr.2016.0067","url":null,"abstract":"One sample per person face recognition (OSPP) is a challenging problem in face recognition community. Lack of samples leads to performance deterioration. Extended sparse representation-based classifier (ESRC) demonstrates excellent performance on OSPP. However, because there are intra-class variant atoms in the dictionary of ESRC, the number of atoms in the dictionary is always large and it will spend a long time during recognition. In this study, the authors propose a new OSPP face recognition algorithm via sparse representation (OSPP-SR). A compressed dictionary and a new identification strategy are provided in OSPP-SR. It is proved theoretically and experimentally that OSPP-SR reaches better or similar performance but spends less time than ESRC. Experiments are conducted on three different databases (extended Yale Face database B, AR database and FERET database) to show the validity of OSPP-SR. Images under clean and noise conditions are also tested to evaluate the robustness of OSPP-SR.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127018843","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 : 2016-07-13DOI: 10.1049/iet-spr.2015.0252
Yang-yang Dong, Chun-xi Dong, Ying-Tong Zhu, Guoqing Zhao, Song-yang Liu
Non-uniform L-shaped array consisting of two nested arrays and its computationally efficient two-dimensional direction-of-arrival (DOA) estimation method are developed in this study. The basic idea of the proposed method is to utilise the property of nested arrays and the conjugate symmetry property of the signal auto-correlation function for different time lags to construct a conjugate augmented spatial–temporal cross-correlation matrix (CAST-CCM) and form joint diagonalisation structure from the signal subspace corresponding to the CAST-CCM. Hence, the DOAs are estimated and paired automatically via signal subspace joint diagonalisation technique. The proposed method can handle underdetermined DOA estimation with automatic matching and deal with the angle ambiguity problem when multiple sources have the same azimuth or elevation angles. Meanwhile, the proposed method is computationally efficient without multidimensional search. The effectiveness of the proposed method is verified through computer simulations.
{"title":"Two-dimensional DOA estimation for L-shaped array with nested subarrays without pair matching","authors":"Yang-yang Dong, Chun-xi Dong, Ying-Tong Zhu, Guoqing Zhao, Song-yang Liu","doi":"10.1049/iet-spr.2015.0252","DOIUrl":"https://doi.org/10.1049/iet-spr.2015.0252","url":null,"abstract":"Non-uniform L-shaped array consisting of two nested arrays and its computationally efficient two-dimensional direction-of-arrival (DOA) estimation method are developed in this study. The basic idea of the proposed method is to utilise the property of nested arrays and the conjugate symmetry property of the signal auto-correlation function for different time lags to construct a conjugate augmented spatial–temporal cross-correlation matrix (CAST-CCM) and form joint diagonalisation structure from the signal subspace corresponding to the CAST-CCM. Hence, the DOAs are estimated and paired automatically via signal subspace joint diagonalisation technique. The proposed method can handle underdetermined DOA estimation with automatic matching and deal with the angle ambiguity problem when multiple sources have the same azimuth or elevation angles. Meanwhile, the proposed method is computationally efficient without multidimensional search. The effectiveness of the proposed method is verified through computer simulations.","PeriodicalId":272888,"journal":{"name":"IET Signal Process.","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133102625","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：