Underwater image enhancement method based on brightness compensation residual network

Abstract

The invention discloses an underwater image enhancement method based on a brightness compensation residual network. The method comprises the following steps: acquiring a data set; preprocessing data: cutting the pictures into the same size, converting the RGB type pictures into the HSV type pictures, extracting V channels and independently storing the V channels; model training: putting the RGB type training set into an RGB multi-scale residual sub-network for training to obtain an enhanced network 1 and an output result; putting the V-type training set into a V-channel brightness compensation sub-network for training to obtain an enhanced network 2 and an output result; the two output results are sent to an enhanced reconstruction sub-network to obtain a final output result; and model testing: respectively putting the test set into the trained enhancement network 1 and enhancement network 2, and sending the output results into the enhancement reconstruction sub-network together to obtain an enhanced underwater picture, the method can solve the problems of color distortion, blurring and low contrast of the underwater image, and further improves the detail information of the image through brightness compensation.

Description

technical field [0001] The invention relates to an underwater image enhancement method based on a brightness compensation residual network, and belongs to the technical field of underwater image enhancement. Background technique [0002] In recent years, high-quality underwater images are of great significance for surveying marine resources, protecting marine life, and defending marine security. However, due to the complexity of underwater imaging, the captured images have serious quality problems, which are not conducive to information acquisition and in-depth research. Therefore, underwater images need to be enhanced and repaired to further extract more useful information. [0003] Problems such as blurring, low contrast, and color distortion of underwater images are due to the special nature of light propagation underwater. Due to the different attenuation of light of different wavelengths when propagating underwater, the problem of color distortion in underwater images...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide an underwater image enhancement method based on brightness compensation residual network, which can solve the problems of underwater image color distortion, blur, low contrast, etc., and at the same time through brightness compensation Further improve the detail information of the image

Images