Pub Date : 2013-12-01DOI: 10.1109/CISP.2013.6743946
Wencheng Yang, Jiankun Hu, Song Wang
Lack of the capability of cancellability in a biometric cryptosystem makes it impossible to be deployed broadly in real applications. This is because the compromise of a template in one application often means its loss in all other applications. To solve this problem, we equip a minutiae-based fuzzy vault with cancellability by applying a polar transformation to each Delaunay triangle group so as to enhance the security level of the biometric template. The unit of the non-invertible polar transformation is a triangle as a whole instead of a single minutiae point which makes the proposed method to be more stable and insensitive to non-linear distortion. Experimental results on public fingerprint databases FVC2002 DB1 and DB2 demonstrate the valid of the proposed method.
{"title":"A Delaunay triangle group based fuzzy vault with cancellability","authors":"Wencheng Yang, Jiankun Hu, Song Wang","doi":"10.1109/CISP.2013.6743946","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743946","url":null,"abstract":"Lack of the capability of cancellability in a biometric cryptosystem makes it impossible to be deployed broadly in real applications. This is because the compromise of a template in one application often means its loss in all other applications. To solve this problem, we equip a minutiae-based fuzzy vault with cancellability by applying a polar transformation to each Delaunay triangle group so as to enhance the security level of the biometric template. The unit of the non-invertible polar transformation is a triangle as a whole instead of a single minutiae point which makes the proposed method to be more stable and insensitive to non-linear distortion. Experimental results on public fingerprint databases FVC2002 DB1 and DB2 demonstrate the valid of the proposed method.","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":"124792379","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.2013.6745242
Hairong Ma, Xinwen Cheng, Xin Wang, Jinjin Yuan
Extracting road information rapidly and efficiently from high resolution images is one of the research hotspots and difficulties in remote sensing. This paper studied road extraction from high resolution remote sensing images based on threshold segmentation and mathematical morphology. Information like road seed points and orientations didn't need to be given manually in this algorithm, which to some extent improved automation of road extraction. The extraction process can be expressed as follows: Firstly, remote sensing images were segmented into binary images containing road information through threshold way. Then mathematical morphology operations are used to process binary image, extracting road regions according to road morphological characteristics. Finally, road centerline and contour were extracted by exploiting relevant mathematical morphology operations, which was proved by numerous experiments.
{"title":"Road information extraction from high resolution remote sensing images based on threshold segmentation and mathematical morphology","authors":"Hairong Ma, Xinwen Cheng, Xin Wang, Jinjin Yuan","doi":"10.1109/CISP.2013.6745242","DOIUrl":"https://doi.org/10.1109/CISP.2013.6745242","url":null,"abstract":"Extracting road information rapidly and efficiently from high resolution images is one of the research hotspots and difficulties in remote sensing. This paper studied road extraction from high resolution remote sensing images based on threshold segmentation and mathematical morphology. Information like road seed points and orientations didn't need to be given manually in this algorithm, which to some extent improved automation of road extraction. The extraction process can be expressed as follows: Firstly, remote sensing images were segmented into binary images containing road information through threshold way. Then mathematical morphology operations are used to process binary image, extracting road regions according to road morphological characteristics. Finally, road centerline and contour were extracted by exploiting relevant mathematical morphology operations, which was proved by numerous experiments.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"1 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":"129407031","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.2013.6743877
Cheolwoo Jo, Osho Gupta
This paper presents a basic research results to visualize the micro-movement of formant tracks for the purpose of using it to analyze speech signal on the research of speech analysis and synthesis etc. Current analysis method on formant is mainly focusing on analysis of static statistical characteristics. But dynamic information about changing vowels is hard to acquire using conventional statistical results such as mean, variance etc. In this experiment we propose a new visualization method to display the dynamic behavior of vowel changes according to the micro movement of formant on vowel space.
{"title":"On visualization of micro-movement of formant tracks on vowel space","authors":"Cheolwoo Jo, Osho Gupta","doi":"10.1109/CISP.2013.6743877","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743877","url":null,"abstract":"This paper presents a basic research results to visualize the micro-movement of formant tracks for the purpose of using it to analyze speech signal on the research of speech analysis and synthesis etc. Current analysis method on formant is mainly focusing on analysis of static statistical characteristics. But dynamic information about changing vowels is hard to acquire using conventional statistical results such as mean, variance etc. In this experiment we propose a new visualization method to display the dynamic behavior of vowel changes according to the micro movement of formant on vowel space.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"1 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":"127387066","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.2013.6743914
Chao Yan, Lin Zheng
The multiple-chirp-rate signal has been proved a good orthogonal characteristic in BFT (biorthogonal Fourier transform) domain. In this work, it is adopted to improve the modulation efficiency with BFT demodulator. However, the performance of BFT demodulator in real wireless environment was not evaluated yet. By our derivation, unfortunately, the orthogonality of multiple-chirp-rate chirp in BFT domain is proved to be damaged over multipath channel. Furthermore, the interference model is constructed, and the linear MMSE detection is proposed to recover the M-ary symbols. Numerical results indicate that the algorithm can simultaneously mitigate the inter-symbol interference (ISI) and the inter-rate interference of the BFT output in multipath environments.
{"title":"Multipath chirp signal detection based on biorthogonal fourier transform","authors":"Chao Yan, Lin Zheng","doi":"10.1109/CISP.2013.6743914","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743914","url":null,"abstract":"The multiple-chirp-rate signal has been proved a good orthogonal characteristic in BFT (biorthogonal Fourier transform) domain. In this work, it is adopted to improve the modulation efficiency with BFT demodulator. However, the performance of BFT demodulator in real wireless environment was not evaluated yet. By our derivation, unfortunately, the orthogonality of multiple-chirp-rate chirp in BFT domain is proved to be damaged over multipath channel. Furthermore, the interference model is constructed, and the linear MMSE detection is proposed to recover the M-ary symbols. Numerical results indicate that the algorithm can simultaneously mitigate the inter-symbol interference (ISI) and the inter-rate interference of the BFT output in multipath environments.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"36 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":"130518622","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.2013.6745205
Yuxi Luo, Y. Wan
The success of machine learning algorithms generally depends on data representation. So far there has been a great deal of literature on unsupervised feature learning and joint training of deep learning. There is little specific guidance, however, on combining hand-designed features or the operations on them with features which are learned from unsupervised learning. In this paper, using MNIST (“Modified National Institute of Standards and Technology”) handwritten digit database as an example, we propose a novel method for training sparse auto-encoders. In this method, we first get some small-scale features through training, then generate more features through operations such as rotation and translation. Finally, we use the whole dataset to fine-tune the network. This approach avoids optimizing cost function for all nodes in the traditional sparse auto-encoder training process, which is very time-consuming. Simulation results show that the proposed method can speed up the training process by over 50%, while keeping the recognition accuracy at the same level or even better. The present findings also contribute to the field's understanding of sparse representation that large-scale sparse features can be generated by small-scale sparse features.
机器学习算法的成功通常取决于数据表示。到目前为止，关于深度学习的无监督特征学习和联合训练已经有了大量的文献。然而，关于如何将手工设计的特征或对其进行的操作与从无监督学习中学习到的特征相结合，却很少有具体的指导。本文以MNIST (Modified National Institute of Standards and Technology)手写数字数据库为例，提出了一种训练稀疏自编码器的新方法。该方法首先通过训练得到一些小尺度的特征，然后通过旋转、平移等操作生成更多的特征。最后，我们使用整个数据集对网络进行微调。该方法避免了传统稀疏自编码器训练过程中对所有节点的代价函数进行优化的耗时问题。仿真结果表明，该方法可将训练速度提高50%以上，同时保持相同甚至更高的识别精度。本研究结果也有助于该领域对稀疏表示的理解，即大规模稀疏特征可以由小规模稀疏特征生成。
{"title":"A novel efficient method for training sparse auto-encoders","authors":"Yuxi Luo, Y. Wan","doi":"10.1109/CISP.2013.6745205","DOIUrl":"https://doi.org/10.1109/CISP.2013.6745205","url":null,"abstract":"The success of machine learning algorithms generally depends on data representation. So far there has been a great deal of literature on unsupervised feature learning and joint training of deep learning. There is little specific guidance, however, on combining hand-designed features or the operations on them with features which are learned from unsupervised learning. In this paper, using MNIST (“Modified National Institute of Standards and Technology”) handwritten digit database as an example, we propose a novel method for training sparse auto-encoders. In this method, we first get some small-scale features through training, then generate more features through operations such as rotation and translation. Finally, we use the whole dataset to fine-tune the network. This approach avoids optimizing cost function for all nodes in the traditional sparse auto-encoder training process, which is very time-consuming. Simulation results show that the proposed method can speed up the training process by over 50%, while keeping the recognition accuracy at the same level or even better. The present findings also contribute to the field's understanding of sparse representation that large-scale sparse features can be generated by small-scale sparse features.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"2007 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":"130573869","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.2013.6745293
Chunmeng Wang, Changhe Tu
In this paper, we present a new method for multi-exposure images fusion of dynamic scenes to remove ghost artefacts. First we unify the exposure level of each input image to that of the reference image and generate ghost detection binary images using difference method with an adaptive threshold strategy. Then we modify the binary images further with morphological operations. Finally, we redefine weight maps of exposure fusion according to the modified binary images to obtain the fused image. We can generate different deghosting result base on an optional reference image. Our method achieves high-quality deghosting effect even if the reference image is not well exposed. Besides, our method can process the misaligned exposures caused by camera shake effectively. Experimental results of various dynamic scenes prove the improvement over previous reference-based deghosting methods.
{"title":"An exposure fusion approach without ghost for dynamic scenes","authors":"Chunmeng Wang, Changhe Tu","doi":"10.1109/CISP.2013.6745293","DOIUrl":"https://doi.org/10.1109/CISP.2013.6745293","url":null,"abstract":"In this paper, we present a new method for multi-exposure images fusion of dynamic scenes to remove ghost artefacts. First we unify the exposure level of each input image to that of the reference image and generate ghost detection binary images using difference method with an adaptive threshold strategy. Then we modify the binary images further with morphological operations. Finally, we redefine weight maps of exposure fusion according to the modified binary images to obtain the fused image. We can generate different deghosting result base on an optional reference image. Our method achieves high-quality deghosting effect even if the reference image is not well exposed. Besides, our method can process the misaligned exposures caused by camera shake effectively. Experimental results of various dynamic scenes prove the improvement over previous reference-based deghosting methods.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"45 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":"130595567","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}
Bridge cable is an important part of the cable-stayed bridge, and reliable methods are necessary to ensure its safety. This paper proposes an algorithm which realizes real-time monitoring the health of bridge cable by the b-value analysis and acoustic emission technique. The algorithm can eliminate the impulse noise and the continuous noise. Discrete wavelet transform decomposes the AE signal, and then wavelet coefficients are processed. By comparing the b-value with the predefined threshold, the method can monitor the health of bridge cable. And this method can realize the real-time monitoring.
{"title":"Real-time monitoring of the bridge cable health by acoustic emission and b-value analysis","authors":"Jian Wang, Guangming Li, Ruijuan Jiang, Yiyan Chen","doi":"10.1109/CISP.2013.6743881","DOIUrl":"https://doi.org/10.1109/CISP.2013.6743881","url":null,"abstract":"Bridge cable is an important part of the cable-stayed bridge, and reliable methods are necessary to ensure its safety. This paper proposes an algorithm which realizes real-time monitoring the health of bridge cable by the b-value analysis and acoustic emission technique. The algorithm can eliminate the impulse noise and the continuous noise. Discrete wavelet transform decomposes the AE signal, and then wavelet coefficients are processed. By comparing the b-value with the predefined threshold, the method can monitor the health of bridge cable. And this method can realize the real-time monitoring.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"26 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":"130619461","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.2013.6745285
T. Rabie
This work builds on a novel framework for image hiding that exploits spatial domain color properties of natural images combined with spectral properties of the Fourier magnitude and phase of these images. The main idea behind this steganography paradigm is that as long as the Fourier phase of an image is maintained intact, the overall appearance of an image remains specious if the Fourier magnitude of the image is slightly modified. In this paper we present improvements in data hiding capacity without compromising the high fidelity, security, and robustness to tampering that the original paradigm developed. Experimental results are presented in this work to demonstrate the effectiveness of these improvements.
{"title":"High-capacity steganography","authors":"T. Rabie","doi":"10.1109/CISP.2013.6745285","DOIUrl":"https://doi.org/10.1109/CISP.2013.6745285","url":null,"abstract":"This work builds on a novel framework for image hiding that exploits spatial domain color properties of natural images combined with spectral properties of the Fourier magnitude and phase of these images. The main idea behind this steganography paradigm is that as long as the Fourier phase of an image is maintained intact, the overall appearance of an image remains specious if the Fourier magnitude of the image is slightly modified. In this paper we present improvements in data hiding capacity without compromising the high fidelity, security, and robustness to tampering that the original paradigm developed. Experimental results are presented in this work to demonstrate the effectiveness of these improvements.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"2 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":"130686801","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.2013.6745218
Hailin Zou, LongQuan Zheng, Chanjuan Liu
In order to settle the limitation of stochastic resonance to detect big parameters signal, frequency modulation stochastic resonance is proposed to detect the parameters of high frequency signal. The principle of frequency modulation stochastic resonance is to modulate the signal's frequency firstly and then detect the signal's frequency by the use of stochastic resonance. The paper analyzes the theory of stochastic resonance detection with multi-frequency signal and raises the process of the experiment. The experimental results show that stochastic resonance can achieve the frequency of the modulated high-frequency signal accurately and quickly, and Signal to Noise Ratio(SNR) can be improved effectively.
{"title":"Detecting parameters of high frequency signals with frequency modulation stochastic resonance","authors":"Hailin Zou, LongQuan Zheng, Chanjuan Liu","doi":"10.1109/CISP.2013.6745218","DOIUrl":"https://doi.org/10.1109/CISP.2013.6745218","url":null,"abstract":"In order to settle the limitation of stochastic resonance to detect big parameters signal, frequency modulation stochastic resonance is proposed to detect the parameters of high frequency signal. The principle of frequency modulation stochastic resonance is to modulate the signal's frequency firstly and then detect the signal's frequency by the use of stochastic resonance. The paper analyzes the theory of stochastic resonance detection with multi-frequency signal and raises the process of the experiment. The experimental results show that stochastic resonance can achieve the frequency of the modulated high-frequency signal accurately and quickly, and Signal to Noise Ratio(SNR) can be improved effectively.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"55 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":"130996224","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.2013.6744057
Yongzheng Geng, Jing Chen, Li Wang
Image segmentation is the first step in computer vision project. But it is also the key procedure from image processing to image analysis. With the continuous development of computer hardware and technology advances, more attention has been paid to the color image. Based on previous image segmentation technology, this paper proposes a novel color image segmentation method. The method improves the JSEG(Joint Systems Engineering Group) algorithm, and it uses the results of JSEG as the input of Ncut(Normalized Cuts). Experiments show that this proposed algorithm can obtain a good segmentation results. The two main contributions of this paper are: 1) it is effective to overcome the problems of the large computation, 2) the final image segmentation boundary is more accurate.
图像分割是计算机视觉研究的第一步。但它也是从图像处理到图像分析的关键步骤。随着计算机硬件的不断发展和技术的不断进步，彩色图像越来越受到人们的重视。在现有图像分割技术的基础上，提出了一种新的彩色图像分割方法。该方法改进了JSEG(Joint Systems Engineering Group)算法，并将JSEG算法的结果作为Ncut(Normalized Cuts)算法的输入。实验表明，该算法能获得较好的分割效果。本文的两个主要贡献是:1)有效地克服了计算量大的问题;2)最终的图像分割边界更加准确。
{"title":"A novel color image segmentation algorithm based on JSEG and Normalized Cuts","authors":"Yongzheng Geng, Jing Chen, Li Wang","doi":"10.1109/CISP.2013.6744057","DOIUrl":"https://doi.org/10.1109/CISP.2013.6744057","url":null,"abstract":"Image segmentation is the first step in computer vision project. But it is also the key procedure from image processing to image analysis. With the continuous development of computer hardware and technology advances, more attention has been paid to the color image. Based on previous image segmentation technology, this paper proposes a novel color image segmentation method. The method improves the JSEG(Joint Systems Engineering Group) algorithm, and it uses the results of JSEG as the input of Ncut(Normalized Cuts). Experiments show that this proposed algorithm can obtain a good segmentation results. The two main contributions of this paper are: 1) it is effective to overcome the problems of the large computation, 2) the final image segmentation boundary is more accurate.","PeriodicalId":442320,"journal":{"name":"2013 6th International Congress on Image and Signal Processing (CISP)","volume":"6 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":"127943725","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：