Noise Variance Estimation Method Based on Regression Analysis and Principal Component Analysis
WU Jiang① YOU Fei① JIANG Ping①②
①(College of Information Engineering, Yulin University, Yulin 719000, China) ②(School of Computer Science and Technology, Xidian University, Xi’an 710071, China)
Abstract：Accurate and reliable blind noise estimation is an important research topic of digital image processing. The main challenge is how to extract pure noise information for estimating. In recent years, many algorithms use principal component analysis technology to exclude the interference of image textures information, and estimate noise level by using the minimal eigenvalue. So that, the image textures have smallest effect on the minimal eigenvalue, thus this kind of methods performs well for high frequency image (image with abundant textures). The minimal eigenvalue is actually smaller than the true noise variance because of limited image blocks, and the bias is the bigger if the number of image patches is the smaller. If the noise level is estimated as the smallest eigenvalue, the final result will be underestimated. It is found that the relation between the ratio of estimated result to real noise variance and the number of image blocks is power function by using regression analysis, thus the true noise level can be computed by using the minimal eigenvalue and the power function. The experiment results show that the proposed algorithm works well over a large range of visual content and noise conditions, and can process multiply Gaussian noise too.
HANIF M and SEGHOUANE A K. Non-local noise estimation for adaptive image denoising[C]. IEEE Digital Image Computing: Techniques and Applications, Adelaide, AUS, 2015: 1-5. doi: 10.1109/DICTA.2015.7371290.
JIANG P and ZHANG J Z. No-reference image quality assessment based on local maximum gradient[J]. Journal of Electronics & Information Technology, 2015, 37(11): 2587-2593. doi: 10.11999/JEIT141447.
[3]
KUMAR K S, SREENIVASULU G, and RAJAN S V. Block based SVD approach for additive white Gaussian noise level estimation in satellite images[C]. 3rd International Conference on Computing for Sustainable Global Development, New Delhi, IND, 2016: 1464-1468.
[4]
DONOHO D. De-noising by soft-thresholding[J]. IEEE Transaction on Information Theory, 1995, 41(3): 613-627. doi: 10.1109/18.382009.
[5]
LIU W and LIN W. Additive white Gaussian noise level estimation in SVD domain for images[J]. IEEE Transactions on Image Processing, 2013, 22(3): 872-883. doi: 10.1109/TIP. 2012.2219544.
[6]
GHAZI M M and ERDOGAN H. Image noise level estimation based on higher-order statistics[J]. Multimedia Tools & Applications, 2016, 76(2): 1-19. doi: 10.1007/s11042- 015-3169-1.
[7]
IMMERKÆR J. Fast noise variance estimation[J]. Computer Vision and Image Understanding, 1996, 64(2): 300-302. doi: 10.1006/cviu.1996.0060.
[8]
RAKHSHANFAR M and AMER A. Estimation of Gaussian, Poissonian-Gaussian, and processed visual noise and its level function[J]. IEEE Transactions on Image Processing, 2016, 25(9): 4172-4185. doi: 10.1109/TIP.2016.2588320.
[9]
MIKOLAJCZAK G and PEKSINSKI J. Estimation of the variance of noise in digital images using a median filter[C]. 39th International Conference on Telecommunications and Signal Processing, Vienna, AUT, 2016: 489-492. doi: 10.1109 /TSP.2016.7760927.
[10]
BOUHRARA M, BONNY J M, ASHINSKY B G, et al. Noise estimation and reduction in magnetic resonance imaging using a new multispectral nonlocal maximum-likelihood filter[J]. IEEE Transactions on Medical Imaging, 2017, 36(1): 181-193. doi: 10.1109/TMI.2016.2601243.
[11]
DONG L, ZHOU J T, and TANG Y Y. Noise level estimation for natural images based on scale-invariant kurtosis and piecewise stationarity[J]. IEEE Transactions on Image Processing, 2017, 26(2): 1017-1030. doi: 10.1109/TIP.2016. 2639447.
[12]
AMER A and DUBOIS E. Fast and reliable structure- oriented video noise estimation[J]. IEEE Transactions on Circuits System and Video Technology, 2005, 15(1): 113-118. doi: 10.1109/TCSVT.2004.837017.
[13]
JIANG P and ZHANG J Z. Fast and reliable noise estimation algorithm based on statistical hypothesis test[C]. 2012 IEEE Vision Communication and Image Processing, San Diego, USA, 2012: 27-30. doi: 10.1109/VCIP.2012.6410754.
[14]
JIANG P and ZHANG J Z. A noise estimation method based on principal components analysis using intensity- homogeneity and sharpness[J]. Information-An International Interdisciplinary Journal, 2012, 15(2): 805-813.
[15]
JIANG P and ZHANG J Z. Fast and reliable noise level estimation based on local statistic[J]. Pattern Recognition Letters, 2016, 78: 8-13. doi: 10.1016/j.patrec.2016.03.026.
[16]
LIU X, TANAKA M, and OKUTOMI M. Noise level estimation using weak textured patches of a single noisy image[C]. 19th IEEE International Conference on Image Processing, Orlando, USA, 2012: 665-668. doi: 10.1109/ICIP. 2012.6466947.
[17]
PYATYKH S, HESSER J, and ZHENG L. Image noise level estimation by principal component analysis[J]. IEEE Transactions on Image Processing, 2013, 22(3): 687-699. doi: 10.1109/TIP.2012.2221728.
[18]
CHEN G, ZHU F, and HENG P A. An efficient statistical method for image noise level estimation[C]. 15th International Conference on Computer Vision, Santiago, Chile, 2015: 477-485. doi: 10.1109/ICCV.2015.62.
[19]
PONOMARENKO N, JIN L, IEREMEIEV O, et al. Image database TID2013: Peculiarities, results and perspectives[J]. Signal Processing: Image Communication, 2015, 30: 57-77. doi: 10.1016/j.image.2014.10.009.
[20]
SHEIKH H R, WANG L C Z, and BOVIK A C. LIVE image quality assessment database release 2. [OL]. http://live.ece.utexas.edu/research/quality, 2006.
[21]
LARSON E C and CHANDLER D M. Categorical image quality (CSIQ) database [EB/OL], http://vision.okstate. edu/csiq, 2009.
[22]
ARBELAEZ P, MAIRE M, FOWLKES C, et al. Contour detection and hierarchical image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(5): 898-916. doi: 10.1109/TPAMI.2010.161.
