Single Underwater Image Enhancement Based On UCIQE
Jun Ao, Mingming Zhang, Chun-bo MA · Journal of Physics Conference Series · 2019
Abstract In order to deal with the problems of low contrast and color distortion of underwater images, a single underwater image enhancement method based on UCIQE is designed in this paper. It estimates background light intensity by a quad-tree search algorithm and divides the image into many blocks based on the assumption that the transmittance of local blocks is equal. The objective function of each block is composed of UCIQE and the information loss function, and the transmittance is obtained at maximum value of the objective function. The global optimization of the overall transmittance image is carried out by using the guided filter. Experimental results demonstrate that this method can effectively solve the problem of color deviation and blur of underwater degraded images.