Underwater Image Enhancement Method via the Fusion of Retinex and Pulse Coupled Neural Network

To tackle the problems of color deviation, blur and low contrast in underwater images, this paper proposes an underwater image enhancement method based on the fusion of Retinex and pulse coupled neural network. First, this paper uses the multi-scale Retinex algorithm to dispose the original image and obtain the image with high saturation. Second, this paper uses the adaptive color correction algorithm to process the original image. And then this paper uses the pulse coupled neural network to further process the image, so as to obtain the image with clear texture and details. Finally, this paper uses the multi-scale fusion algorithm to obtain the final underwater clear image. The experimental results show that our method can improve the sharpness and contrast of images, and solve the problems of underwater images better than other algorithms.