Underwater image enhancement method based on adaptive fractional order multi-scale entropy fusion

Abstract

The invention discloses an underwater image enhancement method based on adaptive fractional order multi-scale entropy fusion. The underwater image enhancement method comprises the steps: dividing an original underwater image into non-overlapping image blocks; establishing an order reference matrix of the fractional order of the gradient domain; constructing a multi-scale fractional order referencematrix; obtaining a fractional order-based enhanced output image consistent with the scale number; calculating the information entropy and contrast of the fused image; determining an output image with enhanced contrast in the image block; obtaining a global contrast enhanced image; converting the original underwater image from the RGB image into a grayscale image; solving an adaptive gain function of the gradient image; obtaining a final adaptive gain underwater enhanced image based on fractional order multilayer entropy fusion; and outputting an adaptive gain underwater enhanced image basedon fractional order multilayer entropy fusion. The adaptive fractional order differential image enhancement method has a good effect on image enhancement of a texture region, and can adapt to changesof image gradient and brightness, so that the image enhancement effect is better.

Description

technical field [0001] The invention belongs to the field of image information processing, and in particular relates to an underwater image enhancement method for adaptive fractional-order multi-scale entropy fusion. Background technique [0002] Underwater target detection images often have special conditions such as non-uniform brightness, low signal-to-noise ratio, and low contrast, which lead to serious degradation of underwater images. Image enhancement is an effective method to process the original image, improve the display effect of the image, and convert it into a more suitable image analysis and processing method. Some scholars have improved the integer-order differential image enhancement. By weighting it or combining other methods, the enhancement effect of the integer-order differential can be improved. However, these methods still cannot effectively improve the shortcomings of integer-order differentials that are sensitive to noise and are easy to suppress ima...

AI Technical Summary

Problems solved by technology

However, these methods still cannot effectively improve the shortcomings of integer-order differentials that are sensitive to noise and are easy to suppress image texture details. At the same time, the fixed-order fractional-order differential image enhancement operator has a good effect on image enhancement in texture regions, but cannot adapt to The change of image gradient and brightness is limited for image enhancement and noise suppression

Images