A Hybrid Method for Underwater Image Preprocessing

Author(s): Hengxu Ma, Kun Wang

This paper provides a hybrid method for underwater image restoration. Firstly, for the color distortion caused by different light attenuation characteristics, the automatic color balance algorithm which based on dynamic threshold is used to correct the color of underwater images. Secondly, the image dehazing based on the dark primary color principle is adopted, which transmissivity and background light are modified according to the propagation characteristics of underwater light. Finally, aiming at the over exposure of underwater images after dehazing, the gray world light compensation algorithm is employed to improve the visual display of underwater images. Experiments have shown the effectiveness of the system designed in this paper.

Underwater Image, Color Distortion Correction, Dehazing Algorithm, Exposure Compensation

1. Zhang Yin. Underwater target image processing and shape recognition technology research [D], Heilongjiang, Harbin Engineering University， 2015
2. Yang Aiping, Zheng Jia, Wang Jian, He Yuqing. Underwater image restoration based on color distortion removal and dark channel prior [J]. Journal of electronics and information technology, 2015，37(11)：145-150