基于空譜聯(lián)合的多假設(shè)預(yù)測高光譜圖像壓縮感知重構(gòu)算法
doi: 10.11999/JEIT150480 cstr: 32379.14.JEIT150480
基金項(xiàng)目:
國家自然科學(xué)基金(61071171)和西北工業(yè)大學(xué)博士論文創(chuàng)新基金(CX201424)
Compressed Sensing Reconstruction of Hyperspectral Images Based on Spatial-spectral Multihypothesis Prediction
Funds:
The National Natural Science Foundation of China (61071171)
-
摘要: 為充分利用高光譜圖像的空間相關(guān)性和譜間相關(guān)性，該文提出一種基于空譜聯(lián)合的多假設(shè)預(yù)測壓縮感知重構(gòu)算法。將高光譜圖像分組為參考波段圖像和非參考波段圖像，參考波段圖像利用光滑Landweber投影算法重構(gòu)，對于非參考波段圖像，引入空譜聯(lián)合的多假設(shè)預(yù)測模型，提高重構(gòu)精度。非參考波段圖像中每個(gè)圖像塊的預(yù)測值不僅來自非參考波段圖像未經(jīng)預(yù)測的初始重構(gòu)值的相鄰圖像塊，而且來自參考波段重構(gòu)圖像相應(yīng)位置及其鄰近的圖像塊，利用預(yù)測值得到測量域中的殘差，然后對殘差進(jìn)行重構(gòu)并對預(yù)測值進(jìn)行修正，此殘差比原圖像更稀疏，且算法采用迭代方式提高重構(gòu)圖像的精度。借助Tikhonov正則化方法求解多假設(shè)預(yù)測的權(quán)重系數(shù)，并基于結(jié)構(gòu)相似性判斷是否改變多假設(shè)預(yù)測搜索窗口大小，最后利用交叉驗(yàn)證計(jì)算重構(gòu)算法終止迭代的判據(jù)參數(shù)。實(shí)驗(yàn)結(jié)果表明，所提算法優(yōu)于僅利用空間相關(guān)性或譜間相關(guān)性進(jìn)行預(yù)測和不預(yù)測的重構(gòu)算法，其重構(gòu)圖像的峰值信噪比提高2 dB以上。
-
關(guān)鍵詞:
- 高光譜圖像 /
- 壓縮感知 /
- 空譜聯(lián)合的多假設(shè)預(yù)測 /
- Tikhonov正則化 /
- 結(jié)構(gòu)相似性
Abstract: Compressed Sensing (CS) reconstruction of hyperspectral images driven by spatial-spectral multihypothesis prediction is proposed in order to take full advantage of spatial and spectral correlation of hyperspectral images. The hyperspectral images are grouped into reference band images and non-reference band images, and the reference band images are reconstructed by Smoothed Projected Landweber (SPL) algorithm. For the non-reference band images, the spatial-spectral multihypothesis prediction model is introduced to improve the reconstruction accuracy. Multihypothesis predictions drawn for an image block of non-reference band image are made not only from spatially surrounding image blocks within an initial non-predicted reconstruction of non-reference band image, but also from the corresponding position and neighboring image blocks within the reconstruction of reference band image. The resulting predictions are used to generate residuals in the projection domain, and the residuals are reconstructed to revise the prediction values. The residuals being typically more compressible than the original images and the iterative execution mode lead to improved reconstruction quality. Tikhonov regularization is utilized to solve the weight coefficients of multihypothesis prediction and structural similarity is used as a criterion to decide whether to change the search window size or not. Cross validation is presented to compute the criterion parameter of iteration termination. Experimental results demonstrate that the proposed algorithm outperforms alternative strategies only using spatial correlation or spectral correlation to predict or not employing prediction and the peak signal-to-noise ratio of its reconstructed images is increased by more than 2 dB. -
Heras D B, Argello F, and Quesada-Barriuso P. Exploring ELM-based spatialspectral classification of hyperspectral images[J]. International Journal of Remote Sensing, 2014, 35(2): 401-423. Zhao C, Li X, Ren J, et al.. Improved sparse representation using adaptive spatial support for effective target detection in hyperspectral imagery[J]. International Journal of Remote Sensing, 2013, 34(24): 8669-8684. Tan C, Samanta A, Jin X, et al.. Using hyperspectral vegetation indices to estimate the fraction of photosynthetically active radiation absorbed by corn canopies[J]. International Journal of Remote Sensing, 2013, 34(24): 8789-8802. Xie X, Li Y, Li R, et al.. Hyperspectral characteristics and growth monitoring of rice (Oryza sativa) under asymmetric warming[J]. International Journal of Remote Sensing, 2013, 34(23): 8449-8462. Donoho D. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306. Cands E, Romberg J, and Tao T. Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information[J]. IEEE Transactions on Information Theory, 2006, 52(2): 489-509. Provost J and Lesage F. The application of compressed sensing for photo-acoustic tomography[J]. IEEE Transactions on Medical Imaging, 2009, 28(4): 585-594. Riccardo M, Giorao Q, Michele R, et al.. A Bayesian analysis of compressive sensing data recovery in wireless sensor networks[C]. International Conference on Ultra Modern Telecommunications＆Workshops, St. Petersburg, 2009: 1-6. Shu X and Ahuja N. Imaging via three-dimensional compressive sampling[C]. Paper presented at the International Conference on Computer Vision (ICCV), Barcelona, ES, 2011: 439-446. 沈志博, 董春曦, 黃龍, 等. 基于壓縮感知的寬頻段二維DOA估計(jì)算法[J]. 電子與信息學(xué)報(bào), 2014, 36(12): 2935-2941. Shen Zhi-bo, Dong Chun-xi, Huang Long, et al.. Broadband 2-D DOA estimation based on compressed sensing[J]. Journal of Electronics Information Technology, 2014, 36(12): 2935-2941. 王軍, 閆鋒剛, 馬文潔, 等. 基于Laplace先驗(yàn)的Bayes壓縮感知波達(dá)方向估計(jì)[J]. 電子與信息學(xué)報(bào), 2015, 37(4): 817-823. Wang Jun, Yan Feng-gang, Ma Wen-jie, et al.. Direction-of-arrival estimation using Laplace prior based on bayes compressive sensing[J]. Journal of Electronics Information Technology, 2015, 37(4): 817-823. Yin J, Sun J, and Jia X. Sparse analysis based on generalized Gaussian model for spectrum recovery with compressed sensing theory[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(6): 2752-2759. Cands E and Tao T. Decoding by linear programming[J]. IEEE Transactions on Information Theory, 2005, 51(12): 4203-4215. Figueiredo M, Nowak R D, and Wright S J. Gradient projection for sparse reconstruction: application to compressed sensing and other inverse problems[J]. IEEE Journal of Selected Topics in Signal Processing, 2007, 1(4): 586-598. Baron S, Sarvoham R, and Baraniuk R G. Bayesian compressive sensing via belief propagation[J]. IEEE Transactions on Signal Processing, 2010, 58(1): 269?280. Mallat S G and Zhang Z. Matching pursuits with time- frequency dictionaries[J]. IEEE Transactions on Signal Processing, 1993, 41(12): 3397-3415. Fowler J E. Compressive-projection principal component analysis[J]. IEEE Transactions on Image Processing, 2009, 18(10): 2230-2242. Gan L. Block Compressed sensing of natural images[C]. 15th International Conference on Digital Signal Processing, Cardiff, Wales, UK, 2007: 403-406. Ji S, Dunson D, and Carin L. Multitask compressive sensing[J]. IEEE Transaction on Signal Processing, 2009, 57(1): 92-106. Claudia V C, Henry A, and Gonzalo R A. Fast lapped block reconstructions in compressive spectral imaging[J]. Applied Optics, 2013, 52(10): D32-45. Henry A, Hoover R, Wu Y, et al.. Higher-order computational model for coded aperture spectral imaging[J]. Applied Optics, 2013, 52(10): D12-21. 劉海英, 吳成柯, 呂沛, 等. 基于譜間預(yù)測和聯(lián)合優(yōu)化的高光譜壓縮感知圖像重構(gòu)[J]. 電子與信息學(xué)報(bào), 2011, 33(9): 2248-2252. Liu Hai-ying, Wu Cheng-ke, L Pei, et al.. Compressed hyperspectral image sensing reconstruction based on interband prediction and joint optimization[J]. Journal of Electronics Information Technology, 2011, 33(9): 2248-2252. 劉海英, 李云松, 吳成柯, 等. 一種高重構(gòu)質(zhì)量低復(fù)雜度的高光譜圖像壓縮感知[J]. 西安電子科技大學(xué)學(xué)報(bào)(自然科學(xué)版), 2011, 38(3): 37-41. Liu Hai-ying, Li Yun-song, Wu Cheng-ke, et al.. Compressed hyperspectral image sensing based on interband prediction[J]. Journal of Xidian University (Natural Science), 2011, 38(3): 37-41. Mun S and Fowler J E. Residual reconstruction for blocked-based compressed sensing of video[C]. Paper presented at the Data Compression Conference (DCC), Snowbird, UT, 2011: 183-192. 李然, 干宗良, 崔子冠, 等. 聯(lián)合時(shí)空特征的視頻分塊壓縮感知重構(gòu)[J]. 電子與信息學(xué)報(bào), 2014, 36(2): 285-292. Li Ran, Gan Zong-liang, Cui Zi-guan, et al.. Block compressed sensing reconstruction of video combined with temporal-spatial characteristics[J]. Journal of Electronics Information Technology, 2014, 36(2): 285-292. 賈應(yīng)彪, 馮燕, 袁曉玲, 等. 高光譜圖像分塊壓縮感知采樣及譜間預(yù)測重構(gòu)[J]. 應(yīng)用科學(xué)學(xué)報(bào), 2014, 32(3): 281-286. Jia Ying-biao, Feng Yan, Yuan Xiao-ling, et al.. Block compressed sensing sampling and reconstruction using spectral prediction for hyperspectral images[J]. Journal of Applied Science, 2014, 32(3): 281-286. Chen C, Tramel E W, and Fowler J E. Compressed-sensing recovery of images and video using multihypothesis predictions[C]. Proceedings of the 45th Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, 2011: 1193-1198. Chen C, Li W, Tramel E W, et al.. Reconstruction of hyperspectral imagery from random projections using multihypothesis prediction[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(1): 365-374. Fowler J E, Mun S, and Tramel E W. Block-based compressed sensing of images and video[J]. Foundations and Trends in Signal Processing, 2012, 4(4): 297-416. Mun S and Fowler J E. Block compressed sensing of images using directional transforms[C]. Paper presented at the International Conference on Image Processing (ICIP), Cairo, EG, 2009: 3021-3024. Johnson W B and Lindenstrauss J. Extensions of Lipschitz mappings into a Hilbert space[J]. Contemporary Mathematics, 1984, 26(1): 189-206. Tikhonov A N and Arsenin V Y. Solutions of Ill-posed Problems[M]. Washington, DC, USA: V. H. Winston Sons, 1977: 45-87. -
計(jì)量
- 文章訪問數(shù): 1247
- HTML全文瀏覽量: 104
- PDF下載量: 537
- 被引次數(shù): 0
下載:
