Pub Date : 2013-12-01DOI: 10.1109/CISP.2013.6743990
E. Zhang, Kaihui Lv, Yongchao Li, J. Duan
The captured images in the frog climatic condition are often blurred and will degrade the performance of the video surveillance system. In order to meet the real-time requirements of video image defogging, this paper proposes a fast video image defogging algorithm based on the dark channel prior. Firstly, atmospheric light value is estimated by more reasonable way. Secondly, the tolerance mechanism is included for adjusting the transmission rate map in the bright areas in order to eliminate color distortion phenomenon in the bright areas of the image. Finally, by reducing the transmission rate map resolution reasonably, the calculating time is shorten but the processing effects is not affected at the same time. The experiment results show that the proposed method can get a good defogging effect.
{"title":"A fast video image defogging algorithm based on dark channel prior","authors":"E. Zhang, Kaihui Lv, Yongchao Li, J. Duan","doi":"10.1109/CISP.2013.6743990","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743990","url":null,"abstract":"The captured images in the frog climatic condition are often blurred and will degrade the performance of the video surveillance system. In order to meet the real-time requirements of video image defogging, this paper proposes a fast video image defogging algorithm based on the dark channel prior. Firstly, atmospheric light value is estimated by more reasonable way. Secondly, the tolerance mechanism is included for adjusting the transmission rate map in the bright areas in order to eliminate color distortion phenomenon in the bright areas of the image. Finally, by reducing the transmission rate map resolution reasonably, the calculating time is shorten but the processing effects is not affected at the same time. The experiment results show that the proposed method can get a good defogging effect.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114570493","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 : 2013-12-01DOI: 10.1109/CISP.2013.6745289
Shiyuan Su, Fuxiang Wang
A novel image fusion rule named “variance-choosemax” based on Structure Similarity Index is proposed in this paper. Firstly, the sparse representation of source image patches are acquired through bases training algorithm K-SVD and pursuit algorithm Orthogonal Matching Pursuit. Then, we group image patches into relevant patches and independent patches according to the Structure Similarity Index of each patch pair. Finally, we fuse the corresponding sparse coefficients of relevant patches and independent patches with “coefficient-choose-max” rule and a new fusion rule named “variance-choose-max” respectively. According to the experiments, our proposed method gains a good performance in visual quality of fused image and also in objective metric.
{"title":"A novel image fusion rule based on Structure Similarity indices","authors":"Shiyuan Su, Fuxiang Wang","doi":"10.1109/CISP.2013.6745289","DOIUrl":"https://doi.org/10.1109/CISP.2013.6745289","url":null,"abstract":"A novel image fusion rule named “variance-choosemax” based on Structure Similarity Index is proposed in this paper. Firstly, the sparse representation of source image patches are acquired through bases training algorithm K-SVD and pursuit algorithm Orthogonal Matching Pursuit. Then, we group image patches into relevant patches and independent patches according to the Structure Similarity Index of each patch pair. Finally, we fuse the corresponding sparse coefficients of relevant patches and independent patches with “coefficient-choose-max” rule and a new fusion rule named “variance-choose-max” respectively. According to the experiments, our proposed method gains a good performance in visual quality of fused image and also in objective metric.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"3 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114019539","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 : 2013-12-01DOI: 10.1109/CISP.2013.6744060
Baojuan Luo, Weilan Wang, Yanjun Jia, W. Gao
Before inpainting damaged Thangka image using digital technology, it's necessary to segment the damaged regions. A segmentation algorithm is proposed to segment the spotted regions of damaged Thangka mage, which combines grayscale morphology with maximum entropy threshold method. First of all, the mathematical morphology is used to act on RGB channels respectively in order to segment the tiny spots. Afterwards, maximum entropy threshold method is applied to segment the large spots. Finally, the above results are merged. The final segmentation result is achieved. Experimental results demonstrate the effectiveness of the proposed algorithm.
{"title":"A segmentation method for spotted-partten damaged Thangka image combining grayscale morphology with maximum entropy threshold","authors":"Baojuan Luo, Weilan Wang, Yanjun Jia, W. Gao","doi":"10.1109/CISP.2013.6744060","DOIUrl":"https://doi.org/10.1109/CISP.2013.6744060","url":null,"abstract":"Before inpainting damaged Thangka image using digital technology, it's necessary to segment the damaged regions. A segmentation algorithm is proposed to segment the spotted regions of damaged Thangka mage, which combines grayscale morphology with maximum entropy threshold method. First of all, the mathematical morphology is used to act on RGB channels respectively in order to segment the tiny spots. Afterwards, maximum entropy threshold method is applied to segment the large spots. Finally, the above results are merged. The final segmentation result is achieved. Experimental results demonstrate the effectiveness of the proposed algorithm.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"206 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114596855","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 : 2013-12-01DOI: 10.1109/CISP.2013.6744003
H. Wen, Jie Wen
In this paper, a kind of method for image denoising is proposed which is based on Pulse Coupled Neural Network (PCNN) and conjugate gradient method. Firstly, we compress and add noise on a high resolution image. Then, we remove noise based on PCNN within 5 neighborhoods and 3 neighborhoods. Finally, we use adaptive median filter to modify the image pixels and use the matrix conjugate gradient method to rebuild the image. The computer simulation experiment results show that the method is perfect for images with noise of salt and pepper and has a good ability in keeping the details of images.
{"title":"Image denoising and restoration using Pulse Coupled Neural Networks","authors":"H. Wen, Jie Wen","doi":"10.1109/CISP.2013.6744003","DOIUrl":"https://doi.org/10.1109/CISP.2013.6744003","url":null,"abstract":"In this paper, a kind of method for image denoising is proposed which is based on Pulse Coupled Neural Network (PCNN) and conjugate gradient method. Firstly, we compress and add noise on a high resolution image. Then, we remove noise based on PCNN within 5 neighborhoods and 3 neighborhoods. Finally, we use adaptive median filter to modify the image pixels and use the matrix conjugate gradient method to rebuild the image. The computer simulation experiment results show that the method is perfect for images with noise of salt and pepper and has a good ability in keeping the details of images.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"308 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114415927","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}
This paper provides a detailed description of a novel multivariant optimization algorithm (MOA) for multi-modal optimization with the main idea to share search information by organizing all search atoms into a special designed structure. Its multiple and variant group property make MOA capable on multi-modal optimization problems. The capability of the MOA method in locating and maintaining multi optima in one execution is discussed in details in this paper and two experiments are carried out to validate its feasibility in multi-modal optimization problems. The experimental results are also compared with those obtained by the species-based PSO, the adaptive sequential niche PSO and the memetic PSO. The experiment results show that MOA has high success rate and convergence speed in multi-modal optimization problems.
{"title":"A novel multivariant optimization algorithm for multimodal optimization","authors":"Changxing Gou, Xinling Shi, Baolei Li, Tiansong Li, Lan-juan Liu, Qinhu Zhang, Yajie Liu","doi":"10.1109/CISP.2013.6743936","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743936","url":null,"abstract":"This paper provides a detailed description of a novel multivariant optimization algorithm (MOA) for multi-modal optimization with the main idea to share search information by organizing all search atoms into a special designed structure. Its multiple and variant group property make MOA capable on multi-modal optimization problems. The capability of the MOA method in locating and maintaining multi optima in one execution is discussed in details in this paper and two experiments are carried out to validate its feasibility in multi-modal optimization problems. The experimental results are also compared with those obtained by the species-based PSO, the adaptive sequential niche PSO and the memetic PSO. The experiment results show that MOA has high success rate and convergence speed in multi-modal optimization problems.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122257928","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 : 2013-12-01DOI: 10.1109/CISP.2013.6743905
Xiaoke Qi, Yu Li, Haining Huang
Equalizer is widely applied in communication systems to eliminate Inter-Symbol Interference mainly caused by multipath over wireless channels. Various algorithms are developed for coefficients update of the equalizer when tracking the channel. However, advantages and drawbacks coexist for single updating algorithm. In this paper, instead of single algorithm applied in the whole frame, two algorithms, recursive least square (RLS) and least mean square (LMS), are intelligently combined in our algorithm. For each iteration, one of two algorithms is chosen by comparing the windowed estimated error autocorrelation with a pre-selected threshold. Since the combined algorithm reaches to convergence using RLS algorithm, the convergence rate is fast and the length of training sequence can be decreased as a result of the effective rate increase. Extended simulations show that our proposed combination algorithm has better mean square error (MSE) and bit error rate (BER) performance compared with single LMS algorithm and lower complexity compared with RLS algorithm. Moreover, the proposed algorithm can track time-varying channel with small performance degradation and dramatic complexity reduction.
{"title":"A combined recursive least square and least mean square equalization scheme based on windowed error autocorrelation estimation","authors":"Xiaoke Qi, Yu Li, Haining Huang","doi":"10.1109/CISP.2013.6743905","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743905","url":null,"abstract":"Equalizer is widely applied in communication systems to eliminate Inter-Symbol Interference mainly caused by multipath over wireless channels. Various algorithms are developed for coefficients update of the equalizer when tracking the channel. However, advantages and drawbacks coexist for single updating algorithm. In this paper, instead of single algorithm applied in the whole frame, two algorithms, recursive least square (RLS) and least mean square (LMS), are intelligently combined in our algorithm. For each iteration, one of two algorithms is chosen by comparing the windowed estimated error autocorrelation with a pre-selected threshold. Since the combined algorithm reaches to convergence using RLS algorithm, the convergence rate is fast and the length of training sequence can be decreased as a result of the effective rate increase. Extended simulations show that our proposed combination algorithm has better mean square error (MSE) and bit error rate (BER) performance compared with single LMS algorithm and lower complexity compared with RLS algorithm. Moreover, the proposed algorithm can track time-varying channel with small performance degradation and dramatic complexity reduction.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122467055","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 : 2013-12-01DOI: 10.1109/CISP.2013.6743991
J. Zhu, Jianguo Wen, Yafeng Zhang
Speckle noise usually occurs in synthetic aperture radar (SAR) images , and SAR data is processed coherently. Speckle filters commonly are adaptive filters using local statistics such as mean and standard deviation, such as the Lee and its enhanced filters and median filter. They adapt the filter coefficients based on data within a fixed moving window, and this brings in contradiction between the quality of speckle noise suppression and the capability of preserving image details. The Lee filter decreases speckle noise well in homogeneous regions, and the enhanced filter performs well both in the homogeneous and heterogeneous areas. But it does not effectively maintain image edges and details, while depressing SAR image noise. The median filter does well in decreasing impulse noise. In this paper, we propose an improved algorithm fusing the enhanced Lee filter and median filter based on spatial filtering of SAR image speckle. The experiment proves it has a good performance in preserving edges and details while filtering images.
{"title":"A new algorithm for SAR image despeckling using an enhanced Lee filter and median filter","authors":"J. Zhu, Jianguo Wen, Yafeng Zhang","doi":"10.1109/CISP.2013.6743991","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743991","url":null,"abstract":"Speckle noise usually occurs in synthetic aperture radar (SAR) images , and SAR data is processed coherently. Speckle filters commonly are adaptive filters using local statistics such as mean and standard deviation, such as the Lee and its enhanced filters and median filter. They adapt the filter coefficients based on data within a fixed moving window, and this brings in contradiction between the quality of speckle noise suppression and the capability of preserving image details. The Lee filter decreases speckle noise well in homogeneous regions, and the enhanced filter performs well both in the homogeneous and heterogeneous areas. But it does not effectively maintain image edges and details, while depressing SAR image noise. The median filter does well in decreasing impulse noise. In this paper, we propose an improved algorithm fusing the enhanced Lee filter and median filter based on spatial filtering of SAR image speckle. The experiment proves it has a good performance in preserving edges and details while filtering images.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"481 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122746677","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 : 2013-12-01DOI: 10.1109/CISP.2013.6743875
Yihong Wang
The optimized speaker model is trained by many time iterative algorithm based on expectation maximization (Abbr. EM). In the process, the choice of speaker model initial value has great influence on the final recognition effect. The most common algorithms which are used to choose the initial value are K-means algorithm and LBG algorithm at present, but the two algorithms belong to a sort of local clustering arithmetic, therefore, it is difficult for them to provide the optimal initial value. For this reason, the ant colony algorithm combined with genetic arithmetic is proposed in the paper. The comparative experiment between this algorithm and K-means algorithm has been done, and the experimental results have been obtained to verify that this algorithm can bring better recognition rate than K-means algorithm.
{"title":"Initialization in speaker model training based on expectation maximization","authors":"Yihong Wang","doi":"10.1109/CISP.2013.6743875","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743875","url":null,"abstract":"The optimized speaker model is trained by many time iterative algorithm based on expectation maximization (Abbr. EM). In the process, the choice of speaker model initial value has great influence on the final recognition effect. The most common algorithms which are used to choose the initial value are K-means algorithm and LBG algorithm at present, but the two algorithms belong to a sort of local clustering arithmetic, therefore, it is difficult for them to provide the optimal initial value. For this reason, the ant colony algorithm combined with genetic arithmetic is proposed in the paper. The comparative experiment between this algorithm and K-means algorithm has been done, and the experimental results have been obtained to verify that this algorithm can bring better recognition rate than K-means algorithm.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127203330","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 : 2013-12-01DOI: 10.1109/CISP.2013.6743897
Hongjuan Sun, Qinglei Zhu
In this paper, problems concerning blind speech signal separation in wireless sensor networks (WSNs) are discussed. The observations captured by sensors are assumed to be linear instantaneous mixtures of the speech sources in the sensing field. First, a framework is designed for the collection, transmission and separation of mixtures of speech signal in WSNs, and the corresponding algorithm appropriate for blind speech signal separation in wireless sensor networks is selected. We also propose the corresponding sensor selection scheme in WSNs on purpose of energy conservation while ensuring the performance of blind speech signal separation. Experimental results show that the mixtures of speech signal can be effectively separated in the assumed WSN scenario with the traditional blind signal separation (BSS) algorithm we select and the schemes for energy efficiency we propose.
{"title":"Blind speech signal separation in wireless sensor networks","authors":"Hongjuan Sun, Qinglei Zhu","doi":"10.1109/CISP.2013.6743897","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743897","url":null,"abstract":"In this paper, problems concerning blind speech signal separation in wireless sensor networks (WSNs) are discussed. The observations captured by sensors are assumed to be linear instantaneous mixtures of the speech sources in the sensing field. First, a framework is designed for the collection, transmission and separation of mixtures of speech signal in WSNs, and the corresponding algorithm appropriate for blind speech signal separation in wireless sensor networks is selected. We also propose the corresponding sensor selection scheme in WSNs on purpose of energy conservation while ensuring the performance of blind speech signal separation. Experimental results show that the mixtures of speech signal can be effectively separated in the assumed WSN scenario with the traditional blind signal separation (BSS) algorithm we select and the schemes for energy efficiency we propose.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126490196","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 : 2013-12-01DOI: 10.1109/CISP.2013.6745240
Jin Li, Shoudong Han, Yong Zhao
In this paper, an interactive image segmentation method is proposed base on the kernel density feature estimation. Compared with the traditional RGB value, it could be more accurate to model the color feature of pixel using corresponding kernel density estimation. To obtain the regional color feature, the mean of kernel densities of all pixels in this region is applied, and Bhattacharyya distance is used to measure the differences between two kernel densities. Consequently, an energy function is constructed according to the main idea of Chan-Vese Model, and it is optimized using the graph cuts technique. Experimental results demonstrate the advantages of our proposed method in terms of robustness and accuracy, especially for objects with thin elongated or concave parts.
{"title":"Kernel density feature based improved Chan-Vese Model for image segmentation","authors":"Jin Li, Shoudong Han, Yong Zhao","doi":"10.1109/CISP.2013.6745240","DOIUrl":"https://doi.org/10.1109/CISP.2013.6745240","url":null,"abstract":"In this paper, an interactive image segmentation method is proposed base on the kernel density feature estimation. Compared with the traditional RGB value, it could be more accurate to model the color feature of pixel using corresponding kernel density estimation. To obtain the regional color feature, the mean of kernel densities of all pixels in this region is applied, and Bhattacharyya distance is used to measure the differences between two kernel densities. Consequently, an energy function is constructed according to the main idea of Chan-Vese Model, and it is optimized using the graph cuts technique. Experimental results demonstrate the advantages of our proposed method in terms of robustness and accuracy, especially for objects with thin elongated or concave parts.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122940590","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：