Image enhancement method for underwater target detection with adaptive multi-scale dark channel prior

Abstract

The invention discloses an adaptive multi-scale dark channel prior underwater target detection image enhancement method. The steps include: obtaining the original color image of the underwater target; calculating the illumination vector of the original image; calculating the dark channel image of the original image; Calculate the transmission images corresponding to windows of different scales of the dark channel image; apply the guided filtering method to calculate the guided filter output images corresponding to the transmission images of the original color image in windows of different scales; minimize the error between the guided filter output image and the transmission map, and initially select several The most appropriate minimum filter scale window; weighted average of the filtered guided filter output image; apply the dark channel prior theory to calculate the restored enhanced image of the original image; perform the restored image J from the aspects of mean, variance, contrast and information entropy Quantitative evaluation. The guided filter function used in the present invention can achieve balanced processing between the texture and smoothness of the image, so that the visual quality of the processed image is improved and the texture information is rich.

Description

technical field [0001] The invention belongs to the field of image information processing, and in particular relates to an adaptive multi-scale dark channel prior image enhancement method for underwater target detection. Background technique [0002] Underwater target detection images have special situations such as non-uniform brightness, low signal-to-noise ratio, and low contrast. Commonly used underwater target detection image enhancement algorithms are mainly divided into two categories: modifying the illumination of underwater images and suppressing image contrast to preserve image edges. , but it will inevitably reduce the visual quality of the detection image. Based on the light scattering physical model, the prior condition is used for image reverse restoration, the most classic Dr. He Kaiming's dark channel prior dehazing method. The dark channel prior method selects a scale parameter to calculate the minimum value of each pixel component in the corresponding scal...

AI Technical Summary

Problems solved by technology

Due to the influence of the optical characteristics of water and various particles, plankton and water flow in water, it is difficult to directly translate and transfer the research results of dark channel prior theory to enhance underwater detection images: 1. Because only one Scale parameters, for detection images with large bright areas, the processing result error is large; 2. The image enhancement process is an open-loop system, which does not have the possibility of adaptively adjusting the scale parameters according to the enhancement effect evaluation index parameters

Images