استنتاج بیزین تغییراتی در حذف نویز از تصاویر فراطیفی با استفاده از متغیرهای پنهان مبتنی بر خوشه‌بندی
محورهای موضوعی : مهندسی برق و کامپیوترطاهره بحرینی 1 , عباس ابراهیمی مقدم 2 , مرتضی خادمی 3 * , هادی صدوقی یزدی 4
1 - دانشگاه فردوسی مشهد
2 - دانشگاه فردوسی مشهد
3 - دانشگاه فردوسی مشهد
4 - دانشگاه فردوسی مشهد
کلید واژه: حذف نویز تصویرمتغیر پنهانتصاویر فراطیفیتجزیه ماتریس مرتبه پاییناستنتاج بیزین تغییراتینویز ترکیبی,
چکیده مقاله :
حذف نویز از تصاویر فراطیفی گامی اجتناب‌ناپذیر برای بهبود کیفیت این نوع تصاویر است و روش‌های بسیاری در این زمینه توسط محققان پیشنهاد شده است. اغلب این روش‌ها به شباهت‌های مکانی- طیفی به طور هم‌زمان توجه ندارند. زمانی که روش حذف نویز، داده را به صورت سراسری بدون در نظر گرفتن شباهت‌های مکانی- طیفی به کار می‌برد، معمولاً بر روی پیکسل‌های با سطح پایین نویز تأثیر نامطلوب می‌گذارد؛ آن هم زمانی که در داده فراطیفی، تعداد زیادی از پیکسل‌ها نویز کمی دارند و تعداد اندکی از پیکسل‌ها به وسیله سطح بالای نویز تخریب می‌شوند. در این مقاله، ابتدا شباهت‌های مکانی- طیفی موجود در تصاویر با تعریف متغیر پنهان مبتنی بر خوشه‌بندی استخراج می‌شود. در ادامه، یک روش تجزیه ماتریس رتبه‌پایین مبتنی بر این متغیرهای پنهان برای حذف نویز تصاویر فراطیفی و بهبود مقاومت در مقابل انواع نویز (در مقایسه با سایر روش‌ها) پیشنهاد می‌شود. کارایی روش پیشنهادی با شش روش جدید بر روی تصاویر واقعی آلوده به نویز به صورت بصری مقایسه شده و برای مقایسه کمی، همان آزمایش‌ها روی تصاویر بدون نویزی که با شش نوع نویز ترکیب شده و تصاویری نزدیک به داده واقعی ایجاد کرده‌اند مقایسه شده است. نتایج شبیه‌سازی نشان می‌دهد که با اعمال متغیر پنهان در چارچوب استنتاج بیزین تغییراتی، عملکرد روش حذف نویز بهبود می‌یابد و روش پیشنهادی عملکرد بهتری نسبت به سایر روش‌های مورد مقایسه دارد.
Removing noise from hyperspectral images is an inevitable step to improve the quality of these types of images. Many methods have been proposed by researchers in this field. Most of these methods do not address simultaneous spatial-spectral similarities. When the noise removal method applies data globally without regard to spatial-spectral similarities, it usually has a negative effect on low-level pixels; when in the spectral data, a large number of pixels have little noise and a small number of pixels are destroyed by the high level of noise. In this paper, we first extract spatial-spectral similarities in images by defining cluster-based latent variables. In the following, a low-rank matrix factorization method based on these latent variables is proposed to eliminate the noise of hyperspectral images and to improve the resistance to noise (as compared to other methods). The performance of the proposed method is compared visually with six new methods on real noise-contaminated images. For quantitative comparison, the same experiments are done on clean images combined with six types of simulated noise. The simulation results show that by applying latent variables in the Bayesian inference framework, the performance of the noise removal method is improved and the proposed method performs better than the other methods.
[1] Y. Chen, X. Cao, Q. Zhao, D. Meng, and Z. Xu, "Denoising hyperspectral image with non-i.i.d. noise structure," IEEE Trans. on Cybernetics, vol. 48, no. 3, pp. 1054-1066, Jul. 2017.
[2] A. Green, M. Berman, P. Switzer, and M. D. Craig, "A transformation for ordering multispectral data in terms of image quality with implications for noise removal," IEEE Trans. Geosci. Remote Sens., vol. 26, no. 1, pp. 65-74, Jan. 1988.
[3] K. Dabov, A. Foi, V. Katkovnik, and K. Egiazarian, "Image denoising by sparse 3-D transform-domain collaborative filtering," IEEE Trans. Image Process., vol. 16, no. 8, pp. 2080-2095, Aug. 2007.
[4] M. Elad and M. Aharon, "Image denoising via sparse and redundant representations over learned dictionaries," IEEE Trans. Image Process., vol. 15, no. 12, pp. 3736-3745, Dec. 2006.
[5] A. Buades, B. Coll, and J. M. Morel, "A non-local algorithm for image denoising," in Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. CVPR'05, vol. 2, pp. 60-65, San Diego, CA, USA, 20-26 Jun. 2005.
[6] L. Liu, L. Chen, C. L. P. Chen, Y. Y. Tang, and C. M. Pun, "Weigthed joint sparse representation for removing mixed noise in image," IEEE Trans. Cybern., vol. 47, no. 3, pp. 600-611, Mar. 2017.
[7] L. Shao, R. Yan, X. Li, and Y. Liu, "From heuristic optimization to dictionary learning: a review and comprehensive comparison of image denoising algorithms," IEEE Trans. Cybern., vol. 44, no. 7, pp. 1001-1013, Jul. 2014.
[8] R. Yan, L. Shao, and Y. Liu, "Nonlocal hierarchical dictionary learning using wavelets for image denoising," IEEE Trans. Image Process., vol. 22, no. 12, pp. 4689-4698, Dec. 2013.
[9] Q. Wang, L. Zhang, Q. Tong, and F. Zhang, "Hyperspectral imagery denoising based on oblique subspace projection," IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 7, no. 6, pp. 2468-2480, Jun. 2014.
[10] C. I. Chang and Q. Du, "Interference and noise-adjusted principal components analysis," IEEE Trans. Geosci. Remote Sens., vol. 37, no. 5, pp. 2387-2396, Sept. 1999.
[11] G. Chen and S. Qian, "Denoising of hyperspectral imagery using principal component analysis and wavelet shrinkage," IEEE Trans. Geosci. Remote Sens., vol. 49, no. 3, pp. 973-980, Mar. 2011.
[12] F. Bollenbeck, A. Backhaus, and U. Seiffert, "A multivariate wavelet PCA denoising-filter for hyperspectral images," in Proc. 3rd Workshop Hyperspectral Image Signal Process. Evol. Remote Sens. WHISPERS'11, 4 pp., Lisbon, Portugal, 6-9 Jun. 2011.
[13] H. Zhang, W. He, L. Zhang, H. Shen, and Q. Yuan, "Hyperspectral image restoration using low-rank matrix recovery," IEEE Trans. Geosci. Remote Sens., vol. 52, no. 8, pp. 4729-4743, Aug. 2014.
[14] G. Chen and S. E. Qian, "Simultaneous dimensionality reduction and denoising of hyperspectral imagery using bivariate wavelet shrinking and principal component analysis," Can. J. Remote Sens., vol. 34, no. 5, pp. 447-454, 2008.
[15] Q. Yuan, L. Zhang, and H. Shen, "Hyperspectral image denoising employing a spectral-spatial adaptive total variation model," IEEE Trans. Geosci. Remote Sens., vol. 50, no. 10, pp. 3660-3677, Oct. 2012.
[16] S. L. Chen, X. Y. Hu, and S. L. Peng, "Hyperspectral imagery denoising using a spatial-spectral domain mixing prior," Comput. Sci. Technol., vol. 27, no. 4, pp. 851-861, Jul. 2012.
[17] P. Zhong and R. Wang, "Multiple-spectral-band CRFs for denoising junk bands of hyperspectral imagery," IEEE Trans. Geosci. Remote Sens., vol. 51, no. 4, pp. 2260-2275, Apr. 2013.
[18] D. Letexier and S. Bourennane, "Noise removal from hyperspectral images by multidimensional filtering," IEEE Trans. Geosci. Remote Sens., vol. 46, no. 7, pp. 2061-2069, Jul. 2008.
[19] X. Liu, S. Bourennane, and C. Fossati, "Non white noise reduction in hyperspectral images," IEEE Geosci. Remote Sens. Lett., vol. 9, no. 3, pp. 368-372, May. 2012.
[20] X. Liu, S. Bourennane, and C. Fossati, "Denoising of hyperspectral images using the PARAFAC model and statistical performance analysis," IEEE Trans. Geosci. Remote Sens., vol. 50, no. 10, pp. 3717-3724, Oct. 2012.
[21] A. Karami, M. Yazdi, and A. Z. Asli, "Noise reduction of hyperspectral images using kernel non-negative Tucker decomposition," IEEE J. Sel. Topics Signal Process., vol. 5, no. 3, pp. 487-493, Jun. 2011.
[22] Y. Peng, et al., "Decomposable nonlocal tensor dictionary learning for multispectral image denoising," in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 2949-2956, Columbus, OH, USA, 23-28 Jun. 2014.
[23] Q. Xie, et al., "Multispectral images denoising by intrinsic tensor sparsity regularization," in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 1692-1700, Seattle, WA, USA, 27-30 Jun. 2016.
[24] K. Dabov, A. Foi, and K. Egiazarian, "Video denoising by sparse 3D transform-domain collaborative filtering," in Proc. European Signal Processing Conf., pp. 145-149, Poznan, Poland, 3-7 Sept. 2007.
[25] M. Maggioni, V. Katkovnik, K. Egiazarian, and A. Foi, "Nonlocal transform-domain filter for volumetric data denoising and reconstruction," IEEE Trans. Image Process., vol. 22, no. 1, pp. 119-133, Jan. 2013.
[26] H. Othman and S. E. Qian, "Noise reduction of hyperspectral imagery using hybrid spatial-spectral derivative-domain wavelet shrinkage," IEEE Trans. Geosci. Remote Sens., vol. 44, no. 2, pp. 397-408, Feb. 2006.
[27] G. Chen and W. P. Zhu, "Signal denoising using neighbouring dualtree complex wavelet coefficients," IET Signal Process., vol. 6, no. 2, pp. 143-147, Apr. 2012.
[28] J. Martn-Herrero, "Anisotropic diffusion in the hypercube," IEEE Trans. Geosci. Remote Sens., vol. 45, no. 5, pp. 1386-1398, May 2007.
[29] H. Zhang, "Hyperspectral image denoising with cubic total variation model," in Proc. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci., vol. I-7, pp. 95-98, Melbourne, Australia, 25 Aug.-1 Sept. 2012.
[30] Q. Yuan, L. Zhang, and H. Shen, "Hyperspectral image denoising with a spatial-spectral view fusion strategy," IEEE Trans. Geosci. Remote Sens., vol. 52, no. 5, pp. 2314-2325, May 2014.
[31] T. Lin and S. Bourennane, "Hyperspectral image processing by jointly filtering wavelet component tensor," IEEE Trans. Geosci. Remote Sens., vol. 51, no. 6, pp. 3529-3541, Jun. 2013.
[32] B. Rasti, J. R. Sveinsson, M. O. Ulfarsson, and J. A. Benediktsson, "Hyperspectral image denoising using first order spectral roughness penalty in wavelet domain," IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 7, no. 6, pp. 2458-2467, Jun. 2014.
[33] G. Ely, S. Aeron, and E. L. Miller, Exploiting Structural Complexity for Robust and Rapid Hyperspectral Imaging, arXiv preprint arXiv:1305.2170, 2013.
[34] B. Rasti, J. R. Sveinsson, and M. O. Ulfarsson, "Wavelet-based sparse reduced-rank regression for hyperspectral image restoration," IEEE Trans. Geosci. Remote Sens., vol. 52, no. 10, pp. 6688-6698, Oct. 2014.
[35] T. Hu, H. Zhang, H. Shen, and L. Zhang, "Robust registration by rank minimization for multiangle hyper/multispectral remotely sensed imagery," IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 7, no. 6, pp. 2443-2457, Jun. 2014.
[36] Q. Li, H. Li, Z. Lu, Q. Lu, and W. Li, "Denoising of hyperspectral images employing two-phase matrix decomposition," IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 7, no. 9, pp. 3742-3754, Sept. 2014.
[37] W. He, H. Zhang, L. Zhang, and H. Shen, "A noise-adjusted iterative randomized singular value decomposition method for hyperspectral image denoising," in Proc. IEEE Int. Geosci. Remote Sens. Symp. (IGARSS), pp. 1536-1539, Quebec City, Canada, 13-18 Jul. 2014.
[38] W. He, H. Zhang, L. Zhang, and H. Shen, "Total-variation-regularized low-rank matrix factorization for hyperspectral image restoration," IEEE Trans. Geosci. Remote Sens., vol. 54, no. 1, pp. 178-188, Jan. 2016.
[39] Q. Zhao, D. Meng, Z. Xu, W. Zuo, and L. Zhang, "Robust principal component analysis with complex noise," in Proc. 31st Int. Conf. Mach. Learn., pp. 55-63 Beijing, China, 21-26 Jun. 2014.
[40] Y. Zheng, G. Liu, S. Sugimoto, S. Yan, and M. Okutomi, "Practical low-rank matrix approximation under robust L1-norm," in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 1410-1417, Providence, RI, 16-21 Jun. 2012.
[41] D. Meng, Z. Xu, L. Zhang, and J. Zhao, "A cyclic weighted median method for L1 low-rank matrix factorization with missing entries," in Proc. 27th Assoc. Adv. Artif. Intell. Conf. on Artificial Intelligence, pp., 704-710, Bellevue, Washington, USA, 14-18 Jun. 2013.
[42] M. Colom, et al., "BBD: a new Bayesian bi-clustering denoising algorithm for IASI-NG hyperspectral images," in Proc. IEEE 8th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS'16, 5 pp., Los Angeles, CA, USA, 21-24 Aug. 2016.
[43] A. Skrondal, Generalized Latent Variable Modeling: Multilevel, Longitudinal, and Structural Equation Models, 2nd Edition (Monographs on Statistics and Applied Probability), 2004 Chapman & Hall/CRC.
[44] C. Bishop, Pattern Recognition and Machine Learning, Springer, Published April 6th 2011.
[45] H. Songa, G. Wanga, and K. Zhang, "Hyperspectral image denoising via low-rank matrix recovery," Remote Sensing Letters, Taylor and Francis, vol. 5, no. 10, pp. 872-881, Oct. 2014.
[46] Y. Wang, et al., "Hyperspectral image restoration via total variation regularized low-rank tensor decomposition," IEEE J. of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 11, no. 4, pp. 1227-1243, Apr. 2018.
[47] J. Liu, P. Musialski, P. Wonka, and J. Ye, "Tensor completion for estimating missing values in visual data," IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 1, pp. 208-220, Jan. 2013.
[48] Q. Xie, Q. Zhao, D. Meng, and Z. Xu, "Kronecker-basis-representation based tensor sparsity and its applications to tensor recovery," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 40, no. 8, pp. 1888-1902, Aug. 2018.
[49] C. Zou and Y. Xia, "Poissonian hyperspectral image superresolution using alternating direction optimization," IEEE J. of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 9, no. 9, pp. 4464-4479, Sept. 2016.
[50] F. Fana, et al., "Hyperspectral image denoising with superpixel segmentation and low-rank representation," Information Sciences, vol. 397-398, pp. 48-68, Aug. 2017.