Pub Date : 2002-08-26DOI: 10.1109/ICOSP.2002.1180085
Xiaoxiang Guo, M. Soumekh
Image registration is one of the crucial steps in detecting changes among the time series medical images. Due to variations in the imaging system over time, the impulse response of the imaging system, also known as its point spread function (PSF), exhibits a time-varying behavior. The registration is further complicated due to the subtle coordinate changes introduced by the patient. In this work, the registration problem is approached via a spatially varying multi-dimensional adaptive filtering method that relates one image in terms of an unknown linear combination of the other image and its spatially transformed versions. Using this model, we develop a scheme, which we refer to as signal subspace processing, to estimate a localized impulse response to calibrate relatively small regions. A criterion is designed to identify the localized PSFs that are not sensitive to the system noise or anatomical changes but accurately represent the spatially varying nature of the unknown miscalibration sources. Low order polynomials are used to sew the localized PSF together and construct a global spatially variant PSF. The anatomical changes between the time series images are achieved by calibrating the image with the global spatially variant PSF. Numerical experiments using MR images illustrate the effectiveness of the proposed algorithm.
{"title":"Signal subspace registration of time series medical imagery","authors":"Xiaoxiang Guo, M. Soumekh","doi":"10.1109/ICOSP.2002.1180085","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1180085","url":null,"abstract":"Image registration is one of the crucial steps in detecting changes among the time series medical images. Due to variations in the imaging system over time, the impulse response of the imaging system, also known as its point spread function (PSF), exhibits a time-varying behavior. The registration is further complicated due to the subtle coordinate changes introduced by the patient. In this work, the registration problem is approached via a spatially varying multi-dimensional adaptive filtering method that relates one image in terms of an unknown linear combination of the other image and its spatially transformed versions. Using this model, we develop a scheme, which we refer to as signal subspace processing, to estimate a localized impulse response to calibrate relatively small regions. A criterion is designed to identify the localized PSFs that are not sensitive to the system noise or anatomical changes but accurately represent the spatially varying nature of the unknown miscalibration sources. Low order polynomials are used to sew the localized PSF together and construct a global spatially variant PSF. The anatomical changes between the time series images are achieved by calibrating the image with the global spatially variant PSF. Numerical experiments using MR images illustrate the effectiveness of the proposed algorithm.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114636855","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1180095
D. Taskovski, S. Bogdanova, M. Bogdanov
In this paper we present extension of our low frequency watermarking scheme. We obtain a robustness improvement to most common image processing operations by embedding the watermark in the approximation image of die original image. in order to embed the watermark with minimum loss in image fidelity, die watermark strength is modulated according to the local image characteristics. We generate a visual mask based on the texture, edge and luminance masking effects of the human visual system. Experimental results show that the proposed technique is competitive with other watermarking techniques.
{"title":"Improved low frequency image adaptive watermarking scheme","authors":"D. Taskovski, S. Bogdanova, M. Bogdanov","doi":"10.1109/ICOSP.2002.1180095","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1180095","url":null,"abstract":"In this paper we present extension of our low frequency watermarking scheme. We obtain a robustness improvement to most common image processing operations by embedding the watermark in the approximation image of die original image. in order to embed the watermark with minimum loss in image fidelity, die watermark strength is modulated according to the local image characteristics. We generate a visual mask based on the texture, edge and luminance masking effects of the human visual system. Experimental results show that the proposed technique is competitive with other watermarking techniques.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123717376","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1180988
Wang Zhizhen, Wang Long, Yu Wensheng
A criterion based on LMI is established for robust stabilization of interval plants in this paper. Our main result is as follows: an arbitrary controller robustly stabilizes a family of interval plants if 16 specific vertex plants satisfy the corresponding LMI conditions.
{"title":"An LMI approach to robust stabilization of interval plants","authors":"Wang Zhizhen, Wang Long, Yu Wensheng","doi":"10.1109/ICOSP.2002.1180988","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1180988","url":null,"abstract":"A criterion based on LMI is established for robust stabilization of interval plants in this paper. Our main result is as follows: an arbitrary controller robustly stabilizes a family of interval plants if 16 specific vertex plants satisfy the corresponding LMI conditions.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124784281","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1179968
I.W.C. Lee, P. Dash
The paper presents a new approach for the classification of non-stationary signal patterns in an electric power network using a modified wavelet transform and neural network. The wavelet transform is phase corrected to yield a new transform known as the S-transform, which has an excellent time-frequency resolution characteristic. The phase correction absolutely references the phase of the wavelet transform to the zero time point, thus assuring that the amplitude peaks are regions of stationary phase. Once the features of a noisy time varying signal during steady state or transient conditions are extracted using the S-transform, they are passed through either a feedforward neural network or a probabilistic neural network for pattern classification. The average classification accuracy of the noisy signals due to disturbances in the power network is of the order 98%.
{"title":"An S-transform based neural pattern classifier for non-stationary signals","authors":"I.W.C. Lee, P. Dash","doi":"10.1109/ICOSP.2002.1179968","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1179968","url":null,"abstract":"The paper presents a new approach for the classification of non-stationary signal patterns in an electric power network using a modified wavelet transform and neural network. The wavelet transform is phase corrected to yield a new transform known as the S-transform, which has an excellent time-frequency resolution characteristic. The phase correction absolutely references the phase of the wavelet transform to the zero time point, thus assuring that the amplitude peaks are regions of stationary phase. Once the features of a noisy time varying signal during steady state or transient conditions are extracted using the S-transform, they are passed through either a feedforward neural network or a probabilistic neural network for pattern classification. The average classification accuracy of the noisy signals due to disturbances in the power network is of the order 98%.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124860073","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1180135
H. Luo, H. Arndt
Digital signal processing has found applications in almost every industry and the hearing aid industry is no exception. Building practical DSP platforms for hearing aids has been difficult due to the physical limitations of small size, low power consumption and low operating voltage, and although progress was initially very slow, real advances have been made during the last decade. Now DSP based hearing aids make use of many well-known digital signal processing technologies developed for communication, speech processing, radar and sonar systems. More sophisticated digital platforms are going to be created and more advanced digital signal processing technologies will be developed and implemented on this new digital hardware. This paper addresses DSP hardware, algorithm, and their application in hearing aids.
{"title":"Digital signal processing technology and applications in hearing aids","authors":"H. Luo, H. Arndt","doi":"10.1109/ICOSP.2002.1180135","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1180135","url":null,"abstract":"Digital signal processing has found applications in almost every industry and the hearing aid industry is no exception. Building practical DSP platforms for hearing aids has been difficult due to the physical limitations of small size, low power consumption and low operating voltage, and although progress was initially very slow, real advances have been made during the last decade. Now DSP based hearing aids make use of many well-known digital signal processing technologies developed for communication, speech processing, radar and sonar systems. More sophisticated digital platforms are going to be created and more advanced digital signal processing technologies will be developed and implemented on this new digital hardware. This paper addresses DSP hardware, algorithm, and their application in hearing aids.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123412144","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1181072
Guo Li, Lina Ning, Guo Yan, Zhou Jiong-pan
The constant modulus algorithm (CMA) enjoys widespread popularity as methods for blind beamforming and equalization of communication signals. CMA is straightforward to implement, robust, and computationally of modest complexity. Despite its effectiveness and apparent simplicity, adaptive implementation of the CMA comes along with several complicating factors that have never really been solved. In particular, convergence can be unpredictable and slow depending on the stepsize. In this paper, Convergence behaviors of the constant modulus algorithm based on "1-2" cost function (CMA/sub 1-2/) and "2-1" cost function (CMA/sub 2-1/) are investigated. We found that certain signals could be quickly removed from the output data choosing special stepsize if at least two signals were of different power. Simulation examples confirm the results.
{"title":"Convergence behavior of the constant modulus algorithm controlled by special stepsize","authors":"Guo Li, Lina Ning, Guo Yan, Zhou Jiong-pan","doi":"10.1109/ICOSP.2002.1181072","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1181072","url":null,"abstract":"The constant modulus algorithm (CMA) enjoys widespread popularity as methods for blind beamforming and equalization of communication signals. CMA is straightforward to implement, robust, and computationally of modest complexity. Despite its effectiveness and apparent simplicity, adaptive implementation of the CMA comes along with several complicating factors that have never really been solved. In particular, convergence can be unpredictable and slow depending on the stepsize. In this paper, Convergence behaviors of the constant modulus algorithm based on \"1-2\" cost function (CMA/sub 1-2/) and \"2-1\" cost function (CMA/sub 2-1/) are investigated. We found that certain signals could be quickly removed from the output data choosing special stepsize if at least two signals were of different power. Simulation examples confirm the results.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121739786","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1180019
Jinwei Wen, S. Luo
The HME network divides a task into small tasks by the principle of divide and conquer to improve the performance of a single network. This approach often brings simple, elegant and efficient algorithms. By studying the dual manifold architecture for mixtures of neural networks and analyzing the probability of knowledge-increasable model based on information geometry, the paper proposes a new method to achieve the multi-HME model that has knowledge-increasable and structure-extendible ability. The method helps to provide an explanation of the transformation mechanism of the human recognition system and understand the theory of the global architecture of the neural network.
{"title":"A HME neural network knowledge-increasable model","authors":"Jinwei Wen, S. Luo","doi":"10.1109/ICOSP.2002.1180019","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1180019","url":null,"abstract":"The HME network divides a task into small tasks by the principle of divide and conquer to improve the performance of a single network. This approach often brings simple, elegant and efficient algorithms. By studying the dual manifold architecture for mixtures of neural networks and analyzing the probability of knowledge-increasable model based on information geometry, the paper proposes a new method to achieve the multi-HME model that has knowledge-increasable and structure-extendible ability. The method helps to provide an explanation of the transformation mechanism of the human recognition system and understand the theory of the global architecture of the neural network.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121419543","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1179967
Xu Hongbo, Tian-yun Yan, Su Jianzhong, Tian Jinwen, Liu Jian
A terrain aided navigation using a range image from imaging laser radar is presented. A real-time terrain elevation map is generated from the range image. Because of undulating terrain, the recovered terrain elevation map is non-uniform. It needs to be resampled. Ridge lines are extracted as basic 3D terrain features. The Hausdorff distance between planar sets of points is known to be a good method to compare binary images. We present an algorithm to match terrain using the modified Hausdorff distance as the measure of the difference between two images. The experimental results show that the proposed method is effective for terrain matching.
{"title":"Terrain matching based on imaging laser radar","authors":"Xu Hongbo, Tian-yun Yan, Su Jianzhong, Tian Jinwen, Liu Jian","doi":"10.1109/ICOSP.2002.1179967","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1179967","url":null,"abstract":"A terrain aided navigation using a range image from imaging laser radar is presented. A real-time terrain elevation map is generated from the range image. Because of undulating terrain, the recovered terrain elevation map is non-uniform. It needs to be resampled. Ridge lines are extracted as basic 3D terrain features. The Hausdorff distance between planar sets of points is known to be a good method to compare binary images. We present an algorithm to match terrain using the modified Hausdorff distance as the measure of the difference between two images. The experimental results show that the proposed method is effective for terrain matching.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123841116","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1180005
Chunhong Jiang, Chen Zhe
This paper presents a novel multisensor fusion method using a Hopfield neural network in the INS/SMGS (inertial navigation system/scene matching guidance system) integrated systems. The state estimation of INS/SMGS systems has multirate and unequal interval characteristics due to the stochastic results of SMGS, so the classical state estimator such as Kalman filter is not competent. The method presented in this paper obtains the optimal fusion state estimation by minimizing the energy function of the Hopfield neural network, and this method is named the hop-filter. Simulation results show that the hop-filter performs much better than the Kalman filter in many factors such as fast convergence, unbias and high precision. Also as the parallel computational mode is easily carried out in hardware of the Hopfield neural network, this fusion method can improve the navigation/guidance accuracy, real time ability and practicability of the INS/SMGS.
{"title":"Multisensor fusion using Hopfield neural network in INS/SMGS integrated system","authors":"Chunhong Jiang, Chen Zhe","doi":"10.1109/ICOSP.2002.1180005","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1180005","url":null,"abstract":"This paper presents a novel multisensor fusion method using a Hopfield neural network in the INS/SMGS (inertial navigation system/scene matching guidance system) integrated systems. The state estimation of INS/SMGS systems has multirate and unequal interval characteristics due to the stochastic results of SMGS, so the classical state estimator such as Kalman filter is not competent. The method presented in this paper obtains the optimal fusion state estimation by minimizing the energy function of the Hopfield neural network, and this method is named the hop-filter. Simulation results show that the hop-filter performs much better than the Kalman filter in many factors such as fast convergence, unbias and high precision. Also as the parallel computational mode is easily carried out in hardware of the Hopfield neural network, this fusion method can improve the navigation/guidance accuracy, real time ability and practicability of the INS/SMGS.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125260944","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 : 2002-08-26DOI: 10.1109/ICOSP.2002.1180028
Xiaoqiu Wang, Jianming Lu, Hua Lin, Nuo Zhang, H. Sekiya, T. Yahagi
In this paper, we propose a novel receiver structure by combining adaptive RNN (recurrent neural network) equalizer with a SOM (self-organizing map) detector under serious ISI and nonlinear distortion in QAM system. The performance of the proposed scheme is shown to be quite effective in channel equalization under nonlinear distortion.
{"title":"Combining RNN equalizer with SOM detector","authors":"Xiaoqiu Wang, Jianming Lu, Hua Lin, Nuo Zhang, H. Sekiya, T. Yahagi","doi":"10.1109/ICOSP.2002.1180028","DOIUrl":"https://doi.org/10.1109/ICOSP.2002.1180028","url":null,"abstract":"In this paper, we propose a novel receiver structure by combining adaptive RNN (recurrent neural network) equalizer with a SOM (self-organizing map) detector under serious ISI and nonlinear distortion in QAM system. The performance of the proposed scheme is shown to be quite effective in channel equalization under nonlinear distortion.","PeriodicalId":159807,"journal":{"name":"6th International Conference on Signal Processing, 2002.","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126459457","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：