基于谱间相似性的高光谱图像稀疏超分辨率算法（英文）
|
摘要
为解决高光谱图像空间分辨率较低的问题,文中提出了一种基于谱间相似性的高光谱图像稀疏超分辨率算法。该算法在最大似然估计准则下,构建了基于混合高斯的稀疏超分辨率编码模型,针对不同的分解残差自适应分配权重,提高了重建图像的空间分辨率和算法对噪声的鲁棒性;该算法构建了基于谱间相似性的图像超分辨率模型,将高光谱图像中普遍存在的像元光谱相关性作为稀疏约束项,保证了图像重建时光谱信息的准确性。实验表明,与Bicubic、Yang、Pan算法相比,文中算法在主观视觉效果、客观评价指标等方面均具有一定优势,验证了算法的有效性。最后将算法各项参数对重建效果的影响进行了分析,为图像检测、分类等应用提供了有效前提。
Hyperspectral image sparse super-resolution algorithm based on spectral similarity was proposed to improve low spatial resolution of hyperspectral images. The super resolution algorithm, based on the criterion of maximum likelihood estimation and Gaussian mixture sparse representation, assigned various weights to different coding residuals to improve spatial resolution of reconstructed images and the robustness to noise. Based on spectral similarity, the super-resolution model which added sparsity constraints using pixel spectral similarity was proposed to ensure the accuracy of the spectrum images.The experiments have been run to prove that this model achieves a better result than Bicubic, Yang and Pan algorithms in both visual effect and objective measures. Additionally, various parameters in the reconstruction were analyzed in order to provide better image detection and classification.
Hyperspectral image sparse super-resolution algorithm based on spectral similarity was proposed to improve low spatial resolution of hyperspectral images. The super resolution algorithm, based on the criterion of maximum likelihood estimation and Gaussian mixture sparse representation, assigned various weights to different coding residuals to improve spatial resolution of reconstructed images and the robustness to noise. Based on spectral similarity, the super-resolution model which added sparsity constraints using pixel spectral similarity was proposed to ensure the accuracy of the spectrum images.The experiments have been run to prove that this model achieves a better result than Bicubic, Yang and Pan algorithms in both visual effect and objective measures. Additionally, various parameters in the reconstruction were analyzed in order to provide better image detection and classification.
引文
[1]Tang Yi,Wan Jianwei,Nian Yongjian.Distributed near lossless compression of hyperspectral images[J].Acta Optica Sinica,2015,35(3):0310001.
[2]Juefei-Xu F,Savvides M.Single face image superresolution via solo dictionary learning[C]//IEEE International Conference on Image Processing(ICIP),2015:2239-2243.
[3]Wang Yiqun,Yan Changxiang,Miao Chun′an.Choice of spectral-splitting modes in space-borne hyperspectral imager[J].Chinese Journal of Optics and Applied Optics,2009,2(4):304-308.(in Chinese)
[4]Xiao J,Pang G,Zhang Y,et al.Adaptive shock filter for image super-resolution and enhancement[J].Journal of Visual Communication and Image Representation,2016,40:168-177.
[5]Zhang K,Tao D,Gao X,et al.Learning multiple linear mappings for efficient single image super-resolution[J].IEEE Transactions on Image Processing,2015,24(3):846-861.
[6]Harris J L.Diffraction and resolving power[J].Journal of the Optical Society of America,1964,54(7):931-936.
[7]Goodman J W.Introduction to Fourier Optics[M].New York:Mc Graw-Hill,1968.
[8]Batz M,Eichenseer A,Selier J,et al.Hybrid superresolution combining example-based single-image and interpolation-based multi-image reconstruction approaches[C]//2015 IEEE International Conference on Image Processing(ICIP),2015:58-62.
[9]Zhou Jinghong,Zhou Cui,Zhu Jianjun,et al.A method of super-resolution reconstruction for remote sensing image based on non-subsampled contourlet transform[J].Acta Optica Sinica,2015,35(1):0110001.(in Chinese)
[10]Lian Qiusheng,Zhang Wei.Image super-resolution algorithms based on sparse representation of classified image patches[J].Acta Electronica Sinica,2012,40(5):920-925.(in Chinese)
[11]Zhao Y Q,Yang J X,Zhang Q Y,et al.Hyperspectral imagery super-resolution by sparse representation and spectral regularization[J].EURASIP Journal on Advances in Signal Processing,2011,2011(1):87.
[12]Pan Zongxu,Yu Jing,Xiao Chuangbai,et al.Spectral similarity-based super resolution for hyperspectral images[J].Acta Automatica Sinica,2014,40(12):2797-2807.(in Chinese)
[13]Fu Fazuo.Hyperspectral image super-resolution based on non-negative dictionary learning[D].Xi′an:Xidian University,2018.(in Chinese)
[14]Tropp J A,Wright S J.Computational methods for sparse solution of linear inverse problems[J].Proceedings of the IEEE,2010,98(6):948-958.
[15]Boyd S,Van den berghe L.Convex Optimization[M].Cambridge:Cambridge University Press,2009:291-293.
[16]Yang M,Zhang L,Yang J,et al.Robust sparse coding for face recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2011:625-632.
[17]Xu Guoming,Xue Mogen,Yuan Guangling.Image superresolution reconstruction method via mixture gaussian sparse coding[J].Opto-Electronic Engineering,2013,40(3):94-101.(in Chinese)
[18]Yang J C,Wright J,Huang T,et al.Image super-resolution via sparse representation[J].IEEE Transactions on Image Processing,2010,19(11):2861-2873.
[19]Saad M A,Bovik A C,Charrier C.Model-based blind image quality assessment:a natural scene statistics approach in the DCT domain[J].IEEE Transactions on Image Processing,2012,21(8):3339-9952.
[20]Akgun T,Altunbasak Y,Mersereau R M.Super-resolution reconstruction of hyperspectral images[J].IEEE Transactions on Image Processing,2005,14(11):1860-1875.
[2]Juefei-Xu F,Savvides M.Single face image superresolution via solo dictionary learning[C]//IEEE International Conference on Image Processing(ICIP),2015:2239-2243.
[3]Wang Yiqun,Yan Changxiang,Miao Chun′an.Choice of spectral-splitting modes in space-borne hyperspectral imager[J].Chinese Journal of Optics and Applied Optics,2009,2(4):304-308.(in Chinese)
[4]Xiao J,Pang G,Zhang Y,et al.Adaptive shock filter for image super-resolution and enhancement[J].Journal of Visual Communication and Image Representation,2016,40:168-177.
[5]Zhang K,Tao D,Gao X,et al.Learning multiple linear mappings for efficient single image super-resolution[J].IEEE Transactions on Image Processing,2015,24(3):846-861.
[6]Harris J L.Diffraction and resolving power[J].Journal of the Optical Society of America,1964,54(7):931-936.
[7]Goodman J W.Introduction to Fourier Optics[M].New York:Mc Graw-Hill,1968.
[8]Batz M,Eichenseer A,Selier J,et al.Hybrid superresolution combining example-based single-image and interpolation-based multi-image reconstruction approaches[C]//2015 IEEE International Conference on Image Processing(ICIP),2015:58-62.
[9]Zhou Jinghong,Zhou Cui,Zhu Jianjun,et al.A method of super-resolution reconstruction for remote sensing image based on non-subsampled contourlet transform[J].Acta Optica Sinica,2015,35(1):0110001.(in Chinese)
[10]Lian Qiusheng,Zhang Wei.Image super-resolution algorithms based on sparse representation of classified image patches[J].Acta Electronica Sinica,2012,40(5):920-925.(in Chinese)
[11]Zhao Y Q,Yang J X,Zhang Q Y,et al.Hyperspectral imagery super-resolution by sparse representation and spectral regularization[J].EURASIP Journal on Advances in Signal Processing,2011,2011(1):87.
[12]Pan Zongxu,Yu Jing,Xiao Chuangbai,et al.Spectral similarity-based super resolution for hyperspectral images[J].Acta Automatica Sinica,2014,40(12):2797-2807.(in Chinese)
[13]Fu Fazuo.Hyperspectral image super-resolution based on non-negative dictionary learning[D].Xi′an:Xidian University,2018.(in Chinese)
[14]Tropp J A,Wright S J.Computational methods for sparse solution of linear inverse problems[J].Proceedings of the IEEE,2010,98(6):948-958.
[15]Boyd S,Van den berghe L.Convex Optimization[M].Cambridge:Cambridge University Press,2009:291-293.
[16]Yang M,Zhang L,Yang J,et al.Robust sparse coding for face recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2011:625-632.
[17]Xu Guoming,Xue Mogen,Yuan Guangling.Image superresolution reconstruction method via mixture gaussian sparse coding[J].Opto-Electronic Engineering,2013,40(3):94-101.(in Chinese)
[18]Yang J C,Wright J,Huang T,et al.Image super-resolution via sparse representation[J].IEEE Transactions on Image Processing,2010,19(11):2861-2873.
[19]Saad M A,Bovik A C,Charrier C.Model-based blind image quality assessment:a natural scene statistics approach in the DCT domain[J].IEEE Transactions on Image Processing,2012,21(8):3339-9952.
[20]Akgun T,Altunbasak Y,Mersereau R M.Super-resolution reconstruction of hyperspectral images[J].IEEE Transactions on Image Processing,2005,14(11):1860-1875.