Abstract
Aiming at the color distortion and low visibility in the images with low light conditions, we proposed a low light image enhancement algorithm based on Retinex theory.Firstly, the light component is obtained by multi-scale Gaussian fusion and then are refined. Then, two input images are obtained by the brightness enhancement function and contrast enhancement function, respectively, and the white shark optimization algorithm is applied to perform updating the weights of the Gaussian-Laplace pyramids. Again, the two input images derived from the brightness function and contrast enhancement function are fused by Gaussian-Laplace pyramids. Finally, the final enhanced image are obtained from the multiplication between the adjusted lighting component and the reflection component based on Retinex theory.The experimental results show that the proposed method in this paper has better enhancement effects on images captured under uneven lighting and low lighting conditions.