Pub Date : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599313
Xiangji Chen, Guo-qiang Han, Zhan Li, Xiuxiu Liao
A novel super-resolution reconstruction algorithm of multi-resolution image sequence integrating the improved super-resolution reconstruction based on neighbor embedding with scale invariant feature transform (SIFT) is proposed in this paper. Firstly, SIFT key points in images are extracted. Then SIFT-feature-based image registration is used to map input high-resolution images to target low-resolution images. Secondly, the mapped images are used as training images and the neighbor embedding is adopted to reconstruct the high-resolution image. The proposed method performs well for problems caused by image deformation, change in viewpoints and change in illumination, which ruin the quality of image super-resolution. Experiments show that the proposed method performs better in terms of lower quantitative errors and better high-frequency information preservation.
{"title":"Image super-resolution via multi-resolution image sequence","authors":"Xiangji Chen, Guo-qiang Han, Zhan Li, Xiuxiu Liao","doi":"10.1109/ICWAPR.2013.6599313","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599313","url":null,"abstract":"A novel super-resolution reconstruction algorithm of multi-resolution image sequence integrating the improved super-resolution reconstruction based on neighbor embedding with scale invariant feature transform (SIFT) is proposed in this paper. Firstly, SIFT key points in images are extracted. Then SIFT-feature-based image registration is used to map input high-resolution images to target low-resolution images. Secondly, the mapped images are used as training images and the neighbor embedding is adopted to reconstruct the high-resolution image. The proposed method performs well for problems caused by image deformation, change in viewpoints and change in illumination, which ruin the quality of image super-resolution. Experiments show that the proposed method performs better in terms of lower quantitative errors and better high-frequency information preservation.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129907804","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599287
Chu Zhang, Wen-Sheng Chen
Traditional Fisher linear discriminant analysis (FLDA) method is a promising algorithm for face recognition. However, FLDA does not utilize the geometric distribution information of the training face data, which will degrade its performance. In order to enhance the discriminant power of FLDA, this paper proposes a novel Fisher criterion by using geometric distribution information of the training samples. The geometric distribution information based LDA (GLDA) algorithm is then developed for face recognition. The proposed GLDA approach has been evaluated with two publicly available face databases, namely ORL and FERET databases. Experimental results demonstrate the effectiveness of our GLDA approach.
{"title":"A novel fisher criterion based approach for face recognition","authors":"Chu Zhang, Wen-Sheng Chen","doi":"10.1109/ICWAPR.2013.6599287","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599287","url":null,"abstract":"Traditional Fisher linear discriminant analysis (FLDA) method is a promising algorithm for face recognition. However, FLDA does not utilize the geometric distribution information of the training face data, which will degrade its performance. In order to enhance the discriminant power of FLDA, this paper proposes a novel Fisher criterion by using geometric distribution information of the training samples. The geometric distribution information based LDA (GLDA) algorithm is then developed for face recognition. The proposed GLDA approach has been evaluated with two publicly available face databases, namely ORL and FERET databases. Experimental results demonstrate the effectiveness of our GLDA approach.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"194 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115349813","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599331
Yi Tang, Qi Wang
Dictionary learning and sparse representation are efficient methods for single-image super-resolution. We propose a new approach to learn a set of dictionaries and then choose the suitable one for a given test image patch of low resolution. Firstly, the training image patches are clustered into K groups with the information of the test image patches. Secondly, a best basis is learned to model each cluster using sparse prior. Finally, we employ this dictionary to estimate the high resolution patch for the given low resolution patch. This method reduces the complexity of dictionary learning greatly and also makes the representation of patches more compact compared to state-of-the-art methods, which learn a universal dictionary. Experimental results show the effectiveness of our method.
{"title":"Super-resolution via K-means sparse coding","authors":"Yi Tang, Qi Wang","doi":"10.1109/ICWAPR.2013.6599331","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599331","url":null,"abstract":"Dictionary learning and sparse representation are efficient methods for single-image super-resolution. We propose a new approach to learn a set of dictionaries and then choose the suitable one for a given test image patch of low resolution. Firstly, the training image patches are clustered into K groups with the information of the test image patches. Secondly, a best basis is learned to model each cluster using sparse prior. Finally, we employ this dictionary to estimate the high resolution patch for the given low resolution patch. This method reduces the complexity of dictionary learning greatly and also makes the representation of patches more compact compared to state-of-the-art methods, which learn a universal dictionary. Experimental results show the effectiveness of our method.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114112923","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599302
Gang Liu, Fanguang Li, Guang-Lei Wen, Shang-Kun Ning, Si-Guo Zheng
This paper proposes a method to identify and classify power quality disturbances (PQD) based on independent component analysis (ICA) and support vector machine (SVM). Firstly, PQD signals are decomposed into 10 layers by db4-wavelet with multi-resolution analysis. Energy Differences (ED) of every level between PQD signals and standard signals are extracted as eigenvectors. Then, Principal Component Analysis (PCA) is adopted to reduce the dimensions of eigenvectors and ICA is used to bleach eigenvectors, which forms new feature vectors. Finally, these new feature vectors are used for power quality disturbance classification using SVM. The results show this method meets the classification accuracy, has a strong resistance to noise, improves classification speed, and is suitable for the classification of PQD.
{"title":"Classification of power quality disturbances based on independent component analysis and support vector machine","authors":"Gang Liu, Fanguang Li, Guang-Lei Wen, Shang-Kun Ning, Si-Guo Zheng","doi":"10.1109/ICWAPR.2013.6599302","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599302","url":null,"abstract":"This paper proposes a method to identify and classify power quality disturbances (PQD) based on independent component analysis (ICA) and support vector machine (SVM). Firstly, PQD signals are decomposed into 10 layers by db4-wavelet with multi-resolution analysis. Energy Differences (ED) of every level between PQD signals and standard signals are extracted as eigenvectors. Then, Principal Component Analysis (PCA) is adopted to reduce the dimensions of eigenvectors and ICA is used to bleach eigenvectors, which forms new feature vectors. Finally, these new feature vectors are used for power quality disturbance classification using SVM. The results show this method meets the classification accuracy, has a strong resistance to noise, improves classification speed, and is suitable for the classification of PQD.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114805994","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599332
S. Lin, C. Wong, T. Ren, N. Kwok
This paper provides a performance evaluation on the Scale- invariant Feature Transform (SIFT) descriptors that utilise different sizes of image patches to represent the SIFT keypoints in images. Although SIFT has been widely employed in numerous applications such as object recognition and image registration, its performances against different image complexities and transformations are still unclear. Thus, an evaluation is commenced to examine SIFT descriptor's performance while its dimension (i.e., information volume) is varied. This paper is started by providing the general concept of SIFT descriptor, then the experimental setup and evaluation metrics are described for detailing the performance evaluation. The experimental results are shown by two evaluation metrics that are repeatability and recall-precision. Lastly, discussions and conclusions are included to emphasise the significances observed in the experimental results and highlight possible directions for future work.
{"title":"The impact of information volume on SIFT descriptor","authors":"S. Lin, C. Wong, T. Ren, N. Kwok","doi":"10.1109/ICWAPR.2013.6599332","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599332","url":null,"abstract":"This paper provides a performance evaluation on the Scale- invariant Feature Transform (SIFT) descriptors that utilise different sizes of image patches to represent the SIFT keypoints in images. Although SIFT has been widely employed in numerous applications such as object recognition and image registration, its performances against different image complexities and transformations are still unclear. Thus, an evaluation is commenced to examine SIFT descriptor's performance while its dimension (i.e., information volume) is varied. This paper is started by providing the general concept of SIFT descriptor, then the experimental setup and evaluation metrics are described for detailing the performance evaluation. The experimental results are shown by two evaluation metrics that are repeatability and recall-precision. Lastly, discussions and conclusions are included to emphasise the significances observed in the experimental results and highlight possible directions for future work.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129441455","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599335
Rong Sun, Y. Marye, Hua-An Zhao
The application of digital signal processing for detection and preservation of different species has been progressing rapidly. In this paper, one such better approach as applied to wild bird species is presented. Feature extraction is done by first performing wavelet transform on sampled bird sounds. After which frequency conversion and determination of mean value that determine the strength of the frequency ingredient is obtained; furthermore, the uniqueness of the modulation spectrum is used as an additional input for the detection mechanism of the birdcall's frequency. The obtained feature quantities then become input to the neural network to simplify classification of nocturnal wild bird species.
{"title":"Wavelet transform digital sound processing to identify wild bird species","authors":"Rong Sun, Y. Marye, Hua-An Zhao","doi":"10.1109/ICWAPR.2013.6599335","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599335","url":null,"abstract":"The application of digital signal processing for detection and preservation of different species has been progressing rapidly. In this paper, one such better approach as applied to wild bird species is presented. Feature extraction is done by first performing wavelet transform on sampled bird sounds. After which frequency conversion and determination of mean value that determine the strength of the frequency ingredient is obtained; furthermore, the uniqueness of the modulation spectrum is used as an additional input for the detection mechanism of the birdcall's frequency. The obtained feature quantities then become input to the neural network to simplify classification of nocturnal wild bird species.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126105931","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599330
Mao-Hsu Yen, Ha-Yong Shin, C.-C. Lai
Feature detection is widely used in image processing, image recognition and machine vision. Points, lines and regions are usually understood as features. In addition, point features are such as object corners because of their variances are not to be impacted by geometry property, and they are simple to recognize for every man. Hence, more and more individual corner detections are proposed. However, rounded corners are seldom discussed in image recognition, and we find that helical compression spring is a great object of study. In this paper we propose rounded corner detection for detecting outside diameter of helical compression spring. The method uses slope comparison to search sites of rounded corner on the helical compression spring image. Through experiments and statistics for computing the outside diameter of spring, this method can steadily detect the rounded corners between 0 degree and 45 degrees, and the deviation of diameter is less than 0.3 percent. Furthermore, it does not have complicated operations in steps, so it can provide stable and accurate results swiftly.
{"title":"Spring gauge system by using R-radius corner detection","authors":"Mao-Hsu Yen, Ha-Yong Shin, C.-C. Lai","doi":"10.1109/ICWAPR.2013.6599330","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599330","url":null,"abstract":"Feature detection is widely used in image processing, image recognition and machine vision. Points, lines and regions are usually understood as features. In addition, point features are such as object corners because of their variances are not to be impacted by geometry property, and they are simple to recognize for every man. Hence, more and more individual corner detections are proposed. However, rounded corners are seldom discussed in image recognition, and we find that helical compression spring is a great object of study. In this paper we propose rounded corner detection for detecting outside diameter of helical compression spring. The method uses slope comparison to search sites of rounded corner on the helical compression spring image. Through experiments and statistics for computing the outside diameter of spring, this method can steadily detect the rounded corners between 0 degree and 45 degrees, and the deviation of diameter is less than 0.3 percent. Furthermore, it does not have complicated operations in steps, so it can provide stable and accurate results swiftly.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127583330","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599300
Zhong Zhang, Shotaro Hosokawa, H. Toda, T. Imamura, T. Miyake
A variable filter band discrete wavelet transform (VFB-DWT) is a kind of discrete wavelet transform (DWT), which has variable band filter, and it can extract desired signal. However, its calculation speed becomes slower than that of the base DWT. In this study, in order to improve the calculation speed of the VFB-DWT, a complex wavelet, RI-Spline wavelet is used to base DWT by Lifting Scheme, and the VFB-DWT is achieved by added the band-pass and band-reject filters on the CDWT. As a result of numerical experimentation, it was shown that signal extraction can be performed correctly.
{"title":"Application of the Lifting Scheme to variable filter band discrete wavelet transform","authors":"Zhong Zhang, Shotaro Hosokawa, H. Toda, T. Imamura, T. Miyake","doi":"10.1109/ICWAPR.2013.6599300","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599300","url":null,"abstract":"A variable filter band discrete wavelet transform (VFB-DWT) is a kind of discrete wavelet transform (DWT), which has variable band filter, and it can extract desired signal. However, its calculation speed becomes slower than that of the base DWT. In this study, in order to improve the calculation speed of the VFB-DWT, a complex wavelet, RI-Spline wavelet is used to base DWT by Lifting Scheme, and the VFB-DWT is achieved by added the band-pass and band-reject filters on the CDWT. As a result of numerical experimentation, it was shown that signal extraction can be performed correctly.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115507606","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599328
M. Zeng, Hao Yang, Qing-Hao Meng, H. Jia
Researches on variation of the wind field in a ventilated room can help to understand the mechanisms of odor/gas dispersal and provide useful clues for the optimization of odor/gas source localization algorithms. In order to investigate laws of changes of wind fields in the indoor environment, a tool of computation fluid dynamics (CFD), i.e. the Reynolds-based standard k-ε turbulence computation model, is applied to numerically simulate air flows in different height levels in the ventilated room. The variations of the wind direction in different height levels are systematically analyzed by means of velocity vector figures and histograms of the velocity direction, respectively. Simulation results show that wind directions in different height levels do not change dramatically in the height below the air inlet. This indicates that traditional two-dimensional plume models can work well in three-dimensional environment below a certain height and provides an important clue to fix the anemometer on a odor source localization robot.
{"title":"Research on variation of the wind direction in different height levels of a ventilated room","authors":"M. Zeng, Hao Yang, Qing-Hao Meng, H. Jia","doi":"10.1109/ICWAPR.2013.6599328","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599328","url":null,"abstract":"Researches on variation of the wind field in a ventilated room can help to understand the mechanisms of odor/gas dispersal and provide useful clues for the optimization of odor/gas source localization algorithms. In order to investigate laws of changes of wind fields in the indoor environment, a tool of computation fluid dynamics (CFD), i.e. the Reynolds-based standard k-ε turbulence computation model, is applied to numerically simulate air flows in different height levels in the ventilated room. The variations of the wind direction in different height levels are systematically analyzed by means of velocity vector figures and histograms of the velocity direction, respectively. Simulation results show that wind directions in different height levels do not change dramatically in the height below the air inlet. This indicates that traditional two-dimensional plume models can work well in three-dimensional environment below a certain height and provides an important clue to fix the anemometer on a odor source localization robot.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"140 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116275183","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 : 2013-07-14DOI: 10.1109/ICWAPR.2013.6599295
A. Morimoto, K. Ikebe, Yoshito Ishida, Yuji Oshima, Motoi Tatsumi, Hitoshi Tsuji
N-tree discrete wavelet transform, which is an extended version of the dual-tree complex discrete wavelet transform, is proposed. Application of N-tree discrete wavelet transform to digital watermarking is considered. Some experimental results demonstrate the validity of the proposed method.
{"title":"An application of N-tree discrete wavelet transform to digital watermarking","authors":"A. Morimoto, K. Ikebe, Yoshito Ishida, Yuji Oshima, Motoi Tatsumi, Hitoshi Tsuji","doi":"10.1109/ICWAPR.2013.6599295","DOIUrl":"https://doi.org/10.1109/ICWAPR.2013.6599295","url":null,"abstract":"N-tree discrete wavelet transform, which is an extended version of the dual-tree complex discrete wavelet transform, is proposed. Application of N-tree discrete wavelet transform to digital watermarking is considered. Some experimental results demonstrate the validity of the proposed method.","PeriodicalId":236156,"journal":{"name":"2013 International Conference on Wavelet Analysis and Pattern Recognition","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130879899","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：