基于小波变换与分形结合的图像压缩算法
|
摘要
针对分形图像压缩过程中匹配编码效率和保证重构图像质量的冲突问题,在定义一种图像子块的新特征—相似比的基础上,提出一种基于小波变换与分形编码相结合的图像压缩算法。该算法首先利用小波变换对图像进行处理,由于经过小波变换后的原图像自相似性被破坏,在引入分形特征时,对于低频区域图像信息不再进行分形压缩,直接保存处理;在高频区域则利用提出的相似比特征,定义每个range块和domain块的相似比,建立它与匹配均方根误差间的关系不等式,可把寻找range块的最佳匹配domain块的全局搜索转化为局部搜索。仿真实验结果表明,与同类特征算法相比,该算法不仅缩短了图像编解码的时间,还提高了重构图像的质量。
Aiming at the problem of the matching coding efficiency and the reconstructed image quality in fractal image compression,on the basis of the similarity ratio,a newfeature of image sub-block defined,we propose a image compression algorithm based on combination of wavelet transform and fractal coding. The algorithm first processes the image by wavelet transform. As the self-similarity of original image after wavelet transform is destroyed,when introduction of fractal features,the fractal compression is no longer carried out for the image information of the low-frequency region,and it is saved directly. In the high-frequency region,using the similarity ratio features proposed,the similarity ratio between each range block and the domain block is defined,and the relation inequality between it and the root mean square error( RMSE) is established. The global search of the best matching domain block to find the range block can be transformed into local search. Simulation shows that the proposed algorithm not only shortens the time of image coding and decoding but also improves the quality of the reconstructed image compared with other similar algorithms.
Aiming at the problem of the matching coding efficiency and the reconstructed image quality in fractal image compression,on the basis of the similarity ratio,a newfeature of image sub-block defined,we propose a image compression algorithm based on combination of wavelet transform and fractal coding. The algorithm first processes the image by wavelet transform. As the self-similarity of original image after wavelet transform is destroyed,when introduction of fractal features,the fractal compression is no longer carried out for the image information of the low-frequency region,and it is saved directly. In the high-frequency region,using the similarity ratio features proposed,the similarity ratio between each range block and the domain block is defined,and the relation inequality between it and the root mean square error( RMSE) is established. The global search of the best matching domain block to find the range block can be transformed into local search. Simulation shows that the proposed algorithm not only shortens the time of image coding and decoding but also improves the quality of the reconstructed image compared with other similar algorithms.
引文
[1]法尔科内(英).分形几何:数学基础及其应用[M].第2版.北京:人民邮电出版社,2007.
[2]BARNSLEY MF,HURD L P. Fractal image compression[M]//Fractal image compression. Natick,MA USA:A. K.Peters,Ltd.,2013.
[3]CHONG S T,MAN W.Adaptive approximate nearest neighbor search for fractal image compression[J]. IEEE Transactions on Image Processing,2002,11(6):605-615.
[4]DU Songlin,YAN Yaping,MA Yide.Quantum-accelerated fractal image compression:an interdisciplinary approach[J].IEEE Signal Processing Letters,2015,22(4):499-503.
[5]JACQUIN A E.Fractal image coding:a review[J].Proceedings of the IEEE,1993,81(10):1451-1465.
[6]ZHANG Lin,ZHANG Lei,MOU Xuanqin,et al. FSIM:a feature similarity index for image quality assessment[J].IEEE Transactions on Image Processing,2011,20(8):2378-2386.
[7]李高平,刘莉.图像子块特征匹配的快速分形编码算法[J].计算机工程与应用,2017,53(1):195-200.
[8]ZHANG Aihua,SHENG Fei,SUN Xuemin.A fast fractal encoding algorithm based on sub-block subtraction[C]//Ninth international conference on natural computation. Shenyang,China:IEEE,2013:1204-1208.
[9]袁宗文,鲁业频,杨汉生.半叉迹特征的快速分形图像编码[J].计算机工程与应用,2016,52(3):197-201.
[10]BIS. Improved method for predicting the peak signal-tonoise ratio quality of decoded images in fractal image coding[J].Journal of Electronic Imaging,2017,26(1):013024.
[11]俞玉莲.一种改进的分形图像压缩算法[J].信息技术,2015,39(6):55-57.
[12]ZHU Shiping,ZONG Xianzi.Fractal lossy hyperspectral image coding algorithm based on prediction[J]. IEEE Access,2017,5:21250-21257.
[13]吴国新,丁春艳,徐小力,等.基于分形与小波相结合的东巴经典古籍图像压缩方法研究[J].北京信息科技大学学报:自然科学版,2017,32(1):9-12.
[14]马俐,赵红东,Hafiz Shehzad Ahmed,等.提升小波变换与分形结合的图像压缩算法[J].电视技术,2017,41(2):11-15.
[15]张爱华,何雨虹,张璟.基于小波与分形理论的图像压缩编码算法[J].计算机技术与发展,2017,27(6):46-50.
[16]康佳星,李尧,唐国鑫.基于小波变换与分形理论的图像边缘检测[J].科技展望,2017,27(23):217.
[2]BARNSLEY MF,HURD L P. Fractal image compression[M]//Fractal image compression. Natick,MA USA:A. K.Peters,Ltd.,2013.
[3]CHONG S T,MAN W.Adaptive approximate nearest neighbor search for fractal image compression[J]. IEEE Transactions on Image Processing,2002,11(6):605-615.
[4]DU Songlin,YAN Yaping,MA Yide.Quantum-accelerated fractal image compression:an interdisciplinary approach[J].IEEE Signal Processing Letters,2015,22(4):499-503.
[5]JACQUIN A E.Fractal image coding:a review[J].Proceedings of the IEEE,1993,81(10):1451-1465.
[6]ZHANG Lin,ZHANG Lei,MOU Xuanqin,et al. FSIM:a feature similarity index for image quality assessment[J].IEEE Transactions on Image Processing,2011,20(8):2378-2386.
[7]李高平,刘莉.图像子块特征匹配的快速分形编码算法[J].计算机工程与应用,2017,53(1):195-200.
[8]ZHANG Aihua,SHENG Fei,SUN Xuemin.A fast fractal encoding algorithm based on sub-block subtraction[C]//Ninth international conference on natural computation. Shenyang,China:IEEE,2013:1204-1208.
[9]袁宗文,鲁业频,杨汉生.半叉迹特征的快速分形图像编码[J].计算机工程与应用,2016,52(3):197-201.
[10]BIS. Improved method for predicting the peak signal-tonoise ratio quality of decoded images in fractal image coding[J].Journal of Electronic Imaging,2017,26(1):013024.
[11]俞玉莲.一种改进的分形图像压缩算法[J].信息技术,2015,39(6):55-57.
[12]ZHU Shiping,ZONG Xianzi.Fractal lossy hyperspectral image coding algorithm based on prediction[J]. IEEE Access,2017,5:21250-21257.
[13]吴国新,丁春艳,徐小力,等.基于分形与小波相结合的东巴经典古籍图像压缩方法研究[J].北京信息科技大学学报:自然科学版,2017,32(1):9-12.
[14]马俐,赵红东,Hafiz Shehzad Ahmed,等.提升小波变换与分形结合的图像压缩算法[J].电视技术,2017,41(2):11-15.
[15]张爱华,何雨虹,张璟.基于小波与分形理论的图像压缩编码算法[J].计算机技术与发展,2017,27(6):46-50.
[16]康佳星,李尧,唐国鑫.基于小波变换与分形理论的图像边缘检测[J].科技展望,2017,27(23):217.