基于ε-SVR的嵌入式小波图像编码算法
|
摘要
针对基于支持向量机的小波图像编码算法难以实现嵌入式特性问题,在小波域构建一种回归树结构作为回归基本数据集合,同时利用子带内和子带间小波系数的相关性,提出一种线性动态阈值选取方法,以利于逐次逼近并保证回归数据的均衡性,并基于选定的阈值动态选取ε误差参数对小波系数进行多次回归,保证了重要系数被优先编码,使压缩算法具有嵌入式特性,对获得的支持向量及其权重进行自适应算术编码。实验结果表明,在压缩比相近的情况下,重构图像的PSNR(Peak Signal to Noise Ratio)比同类算法提高1~3 d B。
The problem that support vector machine based wavelet image coding is difficult to achieve embedded characteristics was studied. Firstly,a regression tree structure was constructed in wavelet domain to act as the basic regression data set,which can use the inner-subband and inter-subband correlation simultaneously.Secondly,a linear dynamic threshold selecting method was put forward to facilitate the successive approximation and to make the regression data to be harmonious. Thirdly,based on the threshold selected,the SVM(Support Vector Machine) error parameters ε was dynamically determined in order to achieve multiple regression to the wavelet coefficients; the significant coefficients could be encoded prior and the impression algorithm was endowed with embedded characteristics. Finally,the adaptive arithmetic coding method was used to encode the support vectors and their weights. The experimental results show that,compared with the current similar algorithms,the PSNR(Peak Signal to Noise Ratio) of the reconstructed image is improved by 1 ~ 3 d B.
The problem that support vector machine based wavelet image coding is difficult to achieve embedded characteristics was studied. Firstly,a regression tree structure was constructed in wavelet domain to act as the basic regression data set,which can use the inner-subband and inter-subband correlation simultaneously.Secondly,a linear dynamic threshold selecting method was put forward to facilitate the successive approximation and to make the regression data to be harmonious. Thirdly,based on the threshold selected,the SVM(Support Vector Machine) error parameters ε was dynamically determined in order to achieve multiple regression to the wavelet coefficients; the significant coefficients could be encoded prior and the impression algorithm was endowed with embedded characteristics. Finally,the adaptive arithmetic coding method was used to encode the support vectors and their weights. The experimental results show that,compared with the current similar algorithms,the PSNR(Peak Signal to Noise Ratio) of the reconstructed image is improved by 1 ~ 3 d B.
引文
[1]CHAUDHARI R E,DHOK S B.Wavelet Transformed Based Fast Fractal Image Compression[C]∥Proceedings of International Conference on Circuits,Systems,Communication and Information Technology Applications.Piscataway,NJ,USA:IEEE,2014:65-69.
[2]HUSSAIN ABIR JAAFAR,AL-JUMEILY DHIYA,RADI NAEEM,et al.Hybrid Neural Network Predictive-Wavelet Image Compression System[J].Neurocomputing,2015,151(P3):975-984.
[3]田润澜,肖卫华,齐兴龙.几种图像变换算法性能比较[J].吉林大学学报:信息科学版,2010,28(5):439-444.TIAN Runlan,XIAO Weihua,QI Xinglong.Comparion of Servral Image Transform[J].Journal of Jilin University:Information Science Edition,2010,28(5):439-444.
[4]ARVIND T,PREM K K.Relevance Vector Machine with Adaptive Wavelet Kernels for Efficient Image Coding[J].Neurocomputing,2010,73(7):1417-1424.
[5]AHMED R.Wavelet-Based Image Compression Using Support Vector Machine Learning and Encoding Techniques[C]∥Proceedings of the Eighth International Conference on Computer Graphics and Imaging.Honolulu,Hawaii,USA:ACTA,2005:162-166.
[6]CHANG C C,LIAO C T.An Image Coding Scheme Using SMVQ and Support Vector Machines[J].Neurocomputing,2006,69(16/18):2327-2335.
[7]JIAO R,LI Y,WANG Q,et al.SVM Regression and Its Application to Image Compression[C]∥Proceedings of the International Conference on Intelligent Computing.Heidelberg:Springer-Verlag,2005:747-756.
[8]KARAMI A,BEHESHTI S,YAZDI M.Hyperspectral Image Compression Using 3D Discrete Cosine Transform and Support Vector Machine Learning[C]∥Proceedings of Information Science,Signal Processing and Their Applications.Piscataway,NJ,USA:IEEE,2012:809-812.
[9]LI Y,WANG Y,XIAO R,et al.Curvelet Based Image Compression via Core Vector Machine[J].Optik-International Journal for Light and Electron Optics,2013,124(21):4859-4866.
[10]HE Qian.Application and Simulation Analysis of SVM and WSVM Hybrid Algorithm in Image Gray Compression[J].Journal of Convergence Information Technology,2013,8(7):178-185.
[11]唐国维.嵌入式小波图像编码算法及应用研究[D].哈尔滨:哈尔滨工程大学计算机科学与技术学院,2010.TANG Guowei.Research on Embedded Wavelet Image Coding Algorithm and Application[D].Harbin:School of Computer Science and Technology,Harbin Engineering University,2010.
[12]邓乃扬,田英杰.数据挖掘中的新方法:支持向量机[M].上海:科学技术出版社,2004:228-245.DENG Naiyang,TIAN Yingjie.Data Mining New Method:SVM[M].Shanghai:Scientific&Technical Press,2004:228-245.
[13]倪虹霞,杨信昌,陈贺新.基于小波自适应阈值的图像去噪方法[J].吉林大学学报:信息科学版,2005,23(4):445-448.NI Hongxia,YANG Xinchang,CHEN Hexin.Denoising Method Based on Adaptive Wavelet Thresholding[J].Journal of Jilin University:Information Science Edition,2005,23(4):445-448.
[14]李元诚,焦润海,李波.一种基于支持向量机的小波图像压缩方法[J].北京航空航天大学学报,2006,32(5):598-602.LI Yuancheng,JIAO Runhai,LI Bo.Wavelet Image Compression Based on Support Vector Machines[J].Journal of Beijing University of Aeronautics and Astronautics,2006,32(5):598-602.
[15]易荣庆,李文辉,孟宇.基于嵌入式零树小波编码的视频编码解码系统[J].吉林大学学报:理学版,2004,42(3):393-396.YI Rongqing,LI Wenhui,MENG Yu.Code and Decode System of Video Based on Embedded Zerotree Wavelet Encoder[J].Journal of Jilin University:Science Edition,2004,42(3):393-396.
[16]张立保,王珂.一种基于整数小波变换的图像编码算法[J].软件学报,2003,14(8):1433-1438.ZHANG Libao,WANG Ke.An Image Compression Algorithm Based on Integer Wavelet Transform[J].Journal of Software,2003,14(8):1433-1438.
[17]赵楠楠,孙红星,徐新和.基于小波变换和SVM的图像压缩仿真研究[J].系统仿真学报,2006,18(11):3034-3037.ZHAO Nannan,SUN Hongxing,XU Xinhe.Approach of Image Compression Based on Wavelet Transform and SVM[J].Journal of System Simulation,2006,18(11):3034-3037.
[18]CHEN J M,LI L,NIE L Y.Wavelet Image Compression by Using Hybrid Kernel SVM[C]∥Proceedings of the Seventh International Conference on Machine Learning and Cybernetics.Piscataway,NJ,USA:IEEE,2008:3056-3060.
[19]SHE Q,SU H,DONG L.Support Vector Machine with Adaptive Parameters in Image Coding[J].International Journal of Innovative Computing,Information and Control,2008,4(2):359-367.
[20]ARVIND T,PREM K K.WSVM with Morlet Wavelet Kernel for Image Compression[C]∥Proceedings of International Conference on System of Systems Engineering.Piscataway,NJ,USA:IEEE,2007:1-5.
[2]HUSSAIN ABIR JAAFAR,AL-JUMEILY DHIYA,RADI NAEEM,et al.Hybrid Neural Network Predictive-Wavelet Image Compression System[J].Neurocomputing,2015,151(P3):975-984.
[3]田润澜,肖卫华,齐兴龙.几种图像变换算法性能比较[J].吉林大学学报:信息科学版,2010,28(5):439-444.TIAN Runlan,XIAO Weihua,QI Xinglong.Comparion of Servral Image Transform[J].Journal of Jilin University:Information Science Edition,2010,28(5):439-444.
[4]ARVIND T,PREM K K.Relevance Vector Machine with Adaptive Wavelet Kernels for Efficient Image Coding[J].Neurocomputing,2010,73(7):1417-1424.
[5]AHMED R.Wavelet-Based Image Compression Using Support Vector Machine Learning and Encoding Techniques[C]∥Proceedings of the Eighth International Conference on Computer Graphics and Imaging.Honolulu,Hawaii,USA:ACTA,2005:162-166.
[6]CHANG C C,LIAO C T.An Image Coding Scheme Using SMVQ and Support Vector Machines[J].Neurocomputing,2006,69(16/18):2327-2335.
[7]JIAO R,LI Y,WANG Q,et al.SVM Regression and Its Application to Image Compression[C]∥Proceedings of the International Conference on Intelligent Computing.Heidelberg:Springer-Verlag,2005:747-756.
[8]KARAMI A,BEHESHTI S,YAZDI M.Hyperspectral Image Compression Using 3D Discrete Cosine Transform and Support Vector Machine Learning[C]∥Proceedings of Information Science,Signal Processing and Their Applications.Piscataway,NJ,USA:IEEE,2012:809-812.
[9]LI Y,WANG Y,XIAO R,et al.Curvelet Based Image Compression via Core Vector Machine[J].Optik-International Journal for Light and Electron Optics,2013,124(21):4859-4866.
[10]HE Qian.Application and Simulation Analysis of SVM and WSVM Hybrid Algorithm in Image Gray Compression[J].Journal of Convergence Information Technology,2013,8(7):178-185.
[11]唐国维.嵌入式小波图像编码算法及应用研究[D].哈尔滨:哈尔滨工程大学计算机科学与技术学院,2010.TANG Guowei.Research on Embedded Wavelet Image Coding Algorithm and Application[D].Harbin:School of Computer Science and Technology,Harbin Engineering University,2010.
[12]邓乃扬,田英杰.数据挖掘中的新方法:支持向量机[M].上海:科学技术出版社,2004:228-245.DENG Naiyang,TIAN Yingjie.Data Mining New Method:SVM[M].Shanghai:Scientific&Technical Press,2004:228-245.
[13]倪虹霞,杨信昌,陈贺新.基于小波自适应阈值的图像去噪方法[J].吉林大学学报:信息科学版,2005,23(4):445-448.NI Hongxia,YANG Xinchang,CHEN Hexin.Denoising Method Based on Adaptive Wavelet Thresholding[J].Journal of Jilin University:Information Science Edition,2005,23(4):445-448.
[14]李元诚,焦润海,李波.一种基于支持向量机的小波图像压缩方法[J].北京航空航天大学学报,2006,32(5):598-602.LI Yuancheng,JIAO Runhai,LI Bo.Wavelet Image Compression Based on Support Vector Machines[J].Journal of Beijing University of Aeronautics and Astronautics,2006,32(5):598-602.
[15]易荣庆,李文辉,孟宇.基于嵌入式零树小波编码的视频编码解码系统[J].吉林大学学报:理学版,2004,42(3):393-396.YI Rongqing,LI Wenhui,MENG Yu.Code and Decode System of Video Based on Embedded Zerotree Wavelet Encoder[J].Journal of Jilin University:Science Edition,2004,42(3):393-396.
[16]张立保,王珂.一种基于整数小波变换的图像编码算法[J].软件学报,2003,14(8):1433-1438.ZHANG Libao,WANG Ke.An Image Compression Algorithm Based on Integer Wavelet Transform[J].Journal of Software,2003,14(8):1433-1438.
[17]赵楠楠,孙红星,徐新和.基于小波变换和SVM的图像压缩仿真研究[J].系统仿真学报,2006,18(11):3034-3037.ZHAO Nannan,SUN Hongxing,XU Xinhe.Approach of Image Compression Based on Wavelet Transform and SVM[J].Journal of System Simulation,2006,18(11):3034-3037.
[18]CHEN J M,LI L,NIE L Y.Wavelet Image Compression by Using Hybrid Kernel SVM[C]∥Proceedings of the Seventh International Conference on Machine Learning and Cybernetics.Piscataway,NJ,USA:IEEE,2008:3056-3060.
[19]SHE Q,SU H,DONG L.Support Vector Machine with Adaptive Parameters in Image Coding[J].International Journal of Innovative Computing,Information and Control,2008,4(2):359-367.
[20]ARVIND T,PREM K K.WSVM with Morlet Wavelet Kernel for Image Compression[C]∥Proceedings of International Conference on System of Systems Engineering.Piscataway,NJ,USA:IEEE,2007:1-5.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |