Enhancement MSRCR algorithm of color fog image based on the adaptive scale

Yin Gao, Lijun Yun, Junsheng Shi, Feiyan Chen, Liansha Lei · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2014

To deal with the image hue and saturation distortion problems in the traditional Retinex algorithm with color restoration coefficient, enhancement MSRCR algorithm of color fog image based on the adaptive scale is proposed. In the RGB color space, the color restoration coefficient is confirmed firstly. Then according to the each channel pixel values, the local weight correction function is introduced and the Gaussian kernel of the required scale is calculated. Doing local correction for the reflection component estimation and obtaining the multi-scale image by weighting. At last, the obtained image is used to contrast stretching and global Gamma correction in order to enhance the image. Through the subjective observation and objective evaluation, the algorithm is better than the traditional MSRCR algorithm in the overall and details.

Read the paper · More papers on PaperTik