Title :
Sure-optimal two-dimensional Savitzky-Golay filters for image denoising
Author :
Menon, Sreeram V. ; Seelamantula, Chandra Sekhar
Author_Institution :
Dept. of Electr. Eng., Indian Inst. of Sci., Bangalore, India
Abstract :
Savitzky-Golay (SG) filters are linear, shift-invariant lowpass filters employed for data smoothing. In their pathbreaking paper published in Analytical Chemistry, Savitzky and Golay mathematically established that polynomial regression of data over local intervals and evaluation of their values at the center of the approximation window is equivalent to convolution with a finite impulse response filter. In this paper, we expound SURE (Stein´s unbiased risk estimate) based adaptive SG filters for image denoising. Our goal is to optimally choose SG filter parameters, namely, order and window length, the optimality defined in terms of the mean squared error (MSE). In practical scenarios, only a single realization of the noisy image is available and the ground truth is inaccessible. Hence, we propose SURE, which is an unbiased estimate of MSE, to solve the parameter selection problem. It is observed that bandwidth of the minimum MSE (MMSE)-optimum SG filter is small at relatively slowly varying portions of the underlying image, and vice versa at abrupt transitions, thereby enabling us to trade off bias and variance to obtain near-optimal performance. The denoising results obtained exhibit considerable peak signal-to-noise-ratio (PSNR) improvement. At low SNRs, the filter performance is further enhanced by using a regularized cost function.
Keywords :
FIR filters; image denoising; low-pass filters; mean square error methods; regression analysis; SG filter parameters; SURE based adaptive SG filters; approximation window; data smoothing; finite impulse response filter; image denoising; local intervals; mean squared error; minimum MSE optimum SG filter; parameter selection problem; peak signal-to-noise-ratio; polynomial regression; regularized cost function; shift-invariant lowpass filters; sure-optimal two-dimensional Savitzky-Golay filters; Approximation methods; Bandwidth; Noise measurement; Noise reduction; PSNR; Polynomials; Finite impulse response filters; Minimum mean-squared error; Polynomial regression; Stein´s unbiased risk estimator (SURE);
Conference_Titel :
Image Processing (ICIP), 2013 20th IEEE International Conference on
Conference_Location :
Melbourne, VIC
DOI :
10.1109/ICIP.2013.6738095