Pub Date : 2017-09-06DOI: 10.1109/ICFSP.2017.8097146
Liyun Gong, Miao Yu
This paper proposes a new method for tracking the whole trajectory of a ballistic missile from launch to impact on the ground. Multiple state models are applied for the ballistic missile movement descriptions during different phases, while the transition probabilities are modelled in a state-dependent way. A radar sensor is applied to obtain the missile range, azimuth angle and elevation angle measurements. Based on the state models and measurements, an interacting multiple model based particle filter method is applied for tracking. Simulation studies show that the proposed method outperforms the widely-applied extended Kalman filtering based interacting multiple model for tracking the ballistic missile.
{"title":"A new interacting multiple model particle filter based ballistic missile tracking method","authors":"Liyun Gong, Miao Yu","doi":"10.1109/ICFSP.2017.8097146","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097146","url":null,"abstract":"This paper proposes a new method for tracking the whole trajectory of a ballistic missile from launch to impact on the ground. Multiple state models are applied for the ballistic missile movement descriptions during different phases, while the transition probabilities are modelled in a state-dependent way. A radar sensor is applied to obtain the missile range, azimuth angle and elevation angle measurements. Based on the state models and measurements, an interacting multiple model based particle filter method is applied for tracking. Simulation studies show that the proposed method outperforms the widely-applied extended Kalman filtering based interacting multiple model for tracking the ballistic missile.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123544006","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 : 2017-09-06DOI: 10.1109/ICFSP.2017.8097157
Liyun Gong, Miao Yu, Timothy J. Gordon
This paper proposes a new background subtraction method by a moving camera for the object detection. Key points are firstly extracted and tracked. From the tracking results, spatial transformation relationships for the background scenes in consecutive frames are obtained while the current frame is warped to the previous image plane for the camera movement compensation. A codebook background model is constructed and updated in an online way by exploiting the full RGB color information, which is used to distinguish the foreground/background regions. Both qualitative and quantitative experimental results show that the proposed method outperforms its counterparts with a better performance.
{"title":"Online codebook modeling based background subtraction with a moving camera","authors":"Liyun Gong, Miao Yu, Timothy J. Gordon","doi":"10.1109/ICFSP.2017.8097157","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097157","url":null,"abstract":"This paper proposes a new background subtraction method by a moving camera for the object detection. Key points are firstly extracted and tracked. From the tracking results, spatial transformation relationships for the background scenes in consecutive frames are obtained while the current frame is warped to the previous image plane for the camera movement compensation. A codebook background model is constructed and updated in an online way by exploiting the full RGB color information, which is used to distinguish the foreground/background regions. Both qualitative and quantitative experimental results show that the proposed method outperforms its counterparts with a better performance.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121748874","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097155
A. Shirokanev, D. Kirsh, A. Kupriyanov
Each three-dimensional crystal structure consists of a set of unit cells which parameters comprehensively describe the location of atoms or atom groups in a crystal. However, the problem of ambiguity of unit cell choice significantly limits the application of existing methods of unit cell parameter identification and comparison. The article proposes a new lattice comparison method based on the unit cell nesting criterion to solve the problem of ambiguity. Results of computational experiments have showed that the developed method not only invariant to the unit cell choice, but also provides high stability to the distortion of a lattice structure.
{"title":"Development of crystal lattice comparison method invariant to bravais unit cell choice","authors":"A. Shirokanev, D. Kirsh, A. Kupriyanov","doi":"10.1109/ICFSP.2017.8097155","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097155","url":null,"abstract":"Each three-dimensional crystal structure consists of a set of unit cells which parameters comprehensively describe the location of atoms or atom groups in a crystal. However, the problem of ambiguity of unit cell choice significantly limits the application of existing methods of unit cell parameter identification and comparison. The article proposes a new lattice comparison method based on the unit cell nesting criterion to solve the problem of ambiguity. Results of computational experiments have showed that the developed method not only invariant to the unit cell choice, but also provides high stability to the distortion of a lattice structure.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126705641","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097058
Taotao Fu, Ge Yu, Lili Guo, Yan Wang, Ji Liang
There are many challenges in single-channel multi-person mixed speech separation, such as modeling the temporal continuity of the speech signals and improving the frame separation performance simultaneously. In this paper, a separation method based on Deep Clustering with local optimization by the improved Non-Negative Matrix Factorization (NMF) combined with Factorial Conditional Random Fields (FCRF) is proposed. Primarily, the separated voices are achieved by Deep Clustering model which are trained by the Bi-directional Long Short Term Memory (BLSTM) and clustered by the similar features. Then, separated voice are locally optimized by the improved NMF with K-means++ and FCRF iteratively. The results show the algorithm improves the separation performance, which satisfies both the local optimum of the speech signal on each frame and the continuity of the whole speech signal.
在单通道多人混合语音分离中，如何同时建立语音信号的时间连续性模型和提高帧分离性能是一个亟待解决的问题。本文提出了一种基于改进的非负矩阵分解(NMF)与阶乘条件随机场(FCRF)相结合的局部优化深度聚类分离方法。首先，通过双向长短期记忆(bidirectional Long - Short Term Memory, BLSTM)训练的深度聚类模型，根据相似特征聚类，实现语音分离。然后，利用改进的NMF结合k -means++和FCRF对分离后的语音进行局部优化。结果表明，该算法提高了分离性能，既满足了每帧语音信号的局部最优，又满足了整个语音信号的连续性。
{"title":"Single-channel speech separation based on deep clustering with local optimization","authors":"Taotao Fu, Ge Yu, Lili Guo, Yan Wang, Ji Liang","doi":"10.1109/ICFSP.2017.8097058","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097058","url":null,"abstract":"There are many challenges in single-channel multi-person mixed speech separation, such as modeling the temporal continuity of the speech signals and improving the frame separation performance simultaneously. In this paper, a separation method based on Deep Clustering with local optimization by the improved Non-Negative Matrix Factorization (NMF) combined with Factorial Conditional Random Fields (FCRF) is proposed. Primarily, the separated voices are achieved by Deep Clustering model which are trained by the Bi-directional Long Short Term Memory (BLSTM) and clustered by the similar features. Then, separated voice are locally optimized by the improved NMF with K-means++ and FCRF iteratively. The results show the algorithm improves the separation performance, which satisfies both the local optimum of the speech signal on each frame and the continuity of the whole speech signal.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129714281","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097059
Van Long Do, T. B. Nguyen, Vu Kien Dao, C. H. Nguyen
In this paper, we propose a novel pre-processing method for the direction-of-arrival (DOA) estimation of multiple sources in multipath environments. In such environments, radio signals impinging on an antenna array can be either uncorrelated, partially correlated or fully correlated (or coherent signals). The proposed pre-processing technique consists of receiving the radio signals at different time instants while moving the uniform circular array (UCA) in predefined paths. The signal covariance matrix is rendered full rank by averaging the sample covariance matrix estimated from the radio signals impinging on the moving UCA. As a result, conventional super-resolution algorithms such as MUSIC can be applied. The simulation results have shown the superiority of the proposed method in comparison with classical pre-processing techniques.
{"title":"Direction finding in multipath environments using moving uniform circular arrays","authors":"Van Long Do, T. B. Nguyen, Vu Kien Dao, C. H. Nguyen","doi":"10.1109/ICFSP.2017.8097059","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097059","url":null,"abstract":"In this paper, we propose a novel pre-processing method for the direction-of-arrival (DOA) estimation of multiple sources in multipath environments. In such environments, radio signals impinging on an antenna array can be either uncorrelated, partially correlated or fully correlated (or coherent signals). The proposed pre-processing technique consists of receiving the radio signals at different time instants while moving the uniform circular array (UCA) in predefined paths. The signal covariance matrix is rendered full rank by averaging the sample covariance matrix estimated from the radio signals impinging on the moving UCA. As a result, conventional super-resolution algorithms such as MUSIC can be applied. The simulation results have shown the superiority of the proposed method in comparison with classical pre-processing techniques.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127351085","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097057
M. Marczyk, J. Polańska, A. Polański
In recent years mass spectrometry became the leading measurement technique in proteomics, giving the opportunity to construct many methods for detection of signal peaks, that are the most important elements of each spectrum. An efficient approach for detecting peaks is partitioning of mass spectrum into fragments and modeling each fragment separately using Gaussian mixture decomposition. The partitioning may be obtained using unique algorithm or any existing peak detection method. In this work two commonly used peak detection algorithms were examined, namely Cromwell and Mass Spec Wavelet. Additionally, a built-in algorithm was proposed. To show that Gaussian mixture modeling of mass spectrum can improve the peak detection performance obtained by using existing solutions, many synthetic spectra with different number of true peaks and real mass spectrometry data were analyzed. In synthetic data mixture modeling of mass spectra gave higher sensitivity and lower false discovery rate of peak detection than existing peak detection algorithms. In real data the coefficient of variation of estimated peak amplitude among biological replicates was reduced.
{"title":"Improving peak detection by Gaussian mixture modeling of mass spectral signal","authors":"M. Marczyk, J. Polańska, A. Polański","doi":"10.1109/ICFSP.2017.8097057","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097057","url":null,"abstract":"In recent years mass spectrometry became the leading measurement technique in proteomics, giving the opportunity to construct many methods for detection of signal peaks, that are the most important elements of each spectrum. An efficient approach for detecting peaks is partitioning of mass spectrum into fragments and modeling each fragment separately using Gaussian mixture decomposition. The partitioning may be obtained using unique algorithm or any existing peak detection method. In this work two commonly used peak detection algorithms were examined, namely Cromwell and Mass Spec Wavelet. Additionally, a built-in algorithm was proposed. To show that Gaussian mixture modeling of mass spectrum can improve the peak detection performance obtained by using existing solutions, many synthetic spectra with different number of true peaks and real mass spectrometry data were analyzed. In synthetic data mixture modeling of mass spectra gave higher sensitivity and lower false discovery rate of peak detection than existing peak detection algorithms. In real data the coefficient of variation of estimated peak amplitude among biological replicates was reduced.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"205 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133145467","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097153
Shaojia Ge, Jianchun Lu, Hong Gu, Zeshi Yuan, W. Su
Inspired by recent successful deep learning methods, this paper presents a new approach for polarimetric synthetic aperture radar (PolSAR) image classification. It combines both advantages of pixel-based and object-based methods. An improved simple linear iterative clustering (SLIC) superpixel segmentation algorithm is used to obtain spatial information in the PolSAR image. Then, a Deep Belief Network (DBN) is introduced to make full use of the limited training data sets, which is trained in an unsupervised manner to extract high-level features from the unlabeled pixels. The DBN's preliminary classification results are finally refined according to the spatial information contained in superpixels. Experimental results over real PolSAR data show that the proposed approach is more efficient with less training data and higher classification accuracy compared with the conventional manners.
{"title":"Polarimetrie SAR image classification based on deep belief network and superpixel segmentation","authors":"Shaojia Ge, Jianchun Lu, Hong Gu, Zeshi Yuan, W. Su","doi":"10.1109/ICFSP.2017.8097153","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097153","url":null,"abstract":"Inspired by recent successful deep learning methods, this paper presents a new approach for polarimetric synthetic aperture radar (PolSAR) image classification. It combines both advantages of pixel-based and object-based methods. An improved simple linear iterative clustering (SLIC) superpixel segmentation algorithm is used to obtain spatial information in the PolSAR image. Then, a Deep Belief Network (DBN) is introduced to make full use of the limited training data sets, which is trained in an unsupervised manner to extract high-level features from the unlabeled pixels. The DBN's preliminary classification results are finally refined according to the spatial information contained in superpixels. Experimental results over real PolSAR data show that the proposed approach is more efficient with less training data and higher classification accuracy compared with the conventional manners.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125397138","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097060
Jihen Zeremdini, M. B. Messaoud, A. Bouzid
This paper describes three methods for multiple fundamental frequencies estimation based on the multi-scale product analysis. The three methods use the autocorrelation of the multi-scale product analysis for the target pitch estimation. For the intrusion pitch, each one has its techniques. The first one uses the classic comb filtering. The second method employs the rectangular comb filter followed by the dynamic programming and the third one uses multiple-comb filters. These methods are evaluated on the Cooke database by calculating the gross error and the root means square error and compared to each other.
{"title":"Multiple fundamental frequencies estimation approaches based on multi-scale product analysis","authors":"Jihen Zeremdini, M. B. Messaoud, A. Bouzid","doi":"10.1109/ICFSP.2017.8097060","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097060","url":null,"abstract":"This paper describes three methods for multiple fundamental frequencies estimation based on the multi-scale product analysis. The three methods use the autocorrelation of the multi-scale product analysis for the target pitch estimation. For the intrusion pitch, each one has its techniques. The first one uses the classic comb filtering. The second method employs the rectangular comb filter followed by the dynamic programming and the third one uses multiple-comb filters. These methods are evaluated on the Cooke database by calculating the gross error and the root means square error and compared to each other.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131366163","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097147
Zhumabek Zhantayev, A. Bibossinov, A. Fremd, B. Iskakov, D. Talgarbayeva, A. Kikkarina, A. Yelisseyeva
As a rule, geodynamic monitoring uses complex approach and monitoring commences from the start of field exploration throughout the whole period of field exploitation. For this purpose specialised geodynamic polygons used for periodic or permanent monitoring and further study of tectonic, technogenic, physical and chemical and other processes that lead to hazardous deformations in earth core. Mineral fields are referred to a number of priority facilities. In connection with the aforementioned, there was carried out radar survey data processing by differential interferometry in the territory of the Karaganda coal basin. The maps of vertical displacements based on SAR data of interferometry were constructed and analyzed, as well as a comprehensive interpretation and comparison of displacement results with underworked territories and adjacent infrastructure aimed at increasing the detection efficiency of catastrophic subsidence of the earth's surface under the influence of technogenic factors and seismogenerating zones.
{"title":"SAR interferometry, as a method of area-based geodynamic control on mineral deposits and adjacent urbanized areas","authors":"Zhumabek Zhantayev, A. Bibossinov, A. Fremd, B. Iskakov, D. Talgarbayeva, A. Kikkarina, A. Yelisseyeva","doi":"10.1109/ICFSP.2017.8097147","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097147","url":null,"abstract":"As a rule, geodynamic monitoring uses complex approach and monitoring commences from the start of field exploration throughout the whole period of field exploitation. For this purpose specialised geodynamic polygons used for periodic or permanent monitoring and further study of tectonic, technogenic, physical and chemical and other processes that lead to hazardous deformations in earth core. Mineral fields are referred to a number of priority facilities. In connection with the aforementioned, there was carried out radar survey data processing by differential interferometry in the territory of the Karaganda coal basin. The maps of vertical displacements based on SAR data of interferometry were constructed and analyzed, as well as a comprehensive interpretation and comparison of displacement results with underworked territories and adjacent infrastructure aimed at increasing the detection efficiency of catastrophic subsidence of the earth's surface under the influence of technogenic factors and seismogenerating zones.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"5105 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133836725","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 : 2017-09-01DOI: 10.1109/ICFSP.2017.8097055
Belhedi Wiem, M. B. Messaoud, A. Bouzid
In real-life environment, the speech of interest is often correlated with different kinds of perturbation. Perturbation can be caused by speaking or non-speaking noise, or even by reverberation. This could make the speech signal auditable but not intelligible. In this case, speech cannot be exploited by other automated applications such as voice-command or speech/speaker identification and identification. Extracting a meaningful signal of good quality is a bigger challenge in monaural case. In this paper, we propose an extensible full joint system that deals with real-environment perturbations that include speaking and non-speaking background as well as reverberation. After introducing the input signal, a decision is taken on the process to opt for. The main chain blocks of the proposed system are speech denoising, speech separation and speech de-reverberation. The system operates in single channel case in a fully unsupervised manner. Furthermore, it requires minimal information about the reference signal. As the system is targeting real-time enhancement, results evaluation is conduct in terms of non-intrusive metrics in addition to intrusive metrics. The evaluation results prove the effectiveness of the proposed system in cancelling difficult noise types, in extracting desired speaker from speaking background, and in enhancing reverberated speech.
{"title":"Joint system for speech separation from speaking and non-speaking background, and de-reverberation: Application on real-world recordings","authors":"Belhedi Wiem, M. B. Messaoud, A. Bouzid","doi":"10.1109/ICFSP.2017.8097055","DOIUrl":"https://doi.org/10.1109/ICFSP.2017.8097055","url":null,"abstract":"In real-life environment, the speech of interest is often correlated with different kinds of perturbation. Perturbation can be caused by speaking or non-speaking noise, or even by reverberation. This could make the speech signal auditable but not intelligible. In this case, speech cannot be exploited by other automated applications such as voice-command or speech/speaker identification and identification. Extracting a meaningful signal of good quality is a bigger challenge in monaural case. In this paper, we propose an extensible full joint system that deals with real-environment perturbations that include speaking and non-speaking background as well as reverberation. After introducing the input signal, a decision is taken on the process to opt for. The main chain blocks of the proposed system are speech denoising, speech separation and speech de-reverberation. The system operates in single channel case in a fully unsupervised manner. Furthermore, it requires minimal information about the reference signal. As the system is targeting real-time enhancement, results evaluation is conduct in terms of non-intrusive metrics in addition to intrusive metrics. The evaluation results prove the effectiveness of the proposed system in cancelling difficult noise types, in extracting desired speaker from speaking background, and in enhancing reverberated speech.","PeriodicalId":382413,"journal":{"name":"2017 3rd International Conference on Frontiers of Signal Processing (ICFSP)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132632541","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：