Chinese Optics Letters, 2007, 5 (1): 28, Published Online: Jan. 11, 2007
A color correction algorithm for noisy multi-view images Download: 656次
图像处理 颜色校正 噪声多视点图像 K-L变换(Karhunen-Loeve变换) 噪声方差 110.0110 Imaging systems 100.6890 Three-dimensional image processing 330.1690 Color
Abstract
A novel color correction algorithm for noisy multi-view images is presented. The key idea is to use the improved Karhunen-Loeve (K-L) transform to obtain correction matrix that can eliminate noise effect to the fullest extent. Noise variance estimation is first performed in the algorithm. In the end, wavelet transform is applied to denoise the corrected image. Experimental results show that, compared with traditional correction method, a well-performed correction result is achieved using the proposed method, and the visual effect of the denoised corrected image is almost consistent with ideal corrected image.
Feng Shao, Gangyi Jiang, Mei Yu, Ken Chen. A color correction algorithm for noisy multi-view images[J]. Chinese Optics Letters, 2007, 5(1): 28.