A multi-scale relative gradient histogram equalization method and device

Abstract

The invention discloses a multi-scale relative gradient histogram equalization method and device. The method includes: step S1, calculating the relative gradient histogram of an image according to the definition of the relative gradient histogram; step S2, based on the relative gradient histogram, Calculate the multi-scale relative gradient histogram of the image; step S3, calculate the contrast increment of each gray level based on the multi-scale relative gradient histogram, and calculate the contrast increment of each gray level corresponding to the traditional histogram; step S4, based on step S3 The obtained multi-scale relative gradient histogram and the contrast increment corresponding to the traditional histogram, adopt the optimization algorithm to obtain the modified gray-level contrast increment; step S5, compare the modified multi-scale relative gradient histogram obtained in step S4 The gray-level contrast increments are accumulated to obtain the corresponding gray-level mapping function; in step S6, the gray-levels of all pixels of the original image are replaced by their mapped gray-levels to obtain an enhanced image.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a method and device for equalizing a multi-scale relative gradient histogram. Background technique [0002] Image enhancement is an important preprocessing step in image processing, which can effectively improve image quality and improve image visual effects. Among them, histogram equalization is one of the most widely used image enhancement algorithms. It makes the histogram of the image more evenly distributed in the whole gray level range through gray level mapping transformation. Histogram equalization can well expand the dynamic range of the gray level of the image and improve the contrast of the image. But it has the following problems: [0003] 1. The image is over-enhanced. The mapping function of histogram equalization is the accumulation of the probability density distribution of the gray level of the image. If there is a relatively steep peak in th...

AI Technical Summary

Problems solved by technology

The mapping function of histogram equalization is the accumulation of the probability density distribution of the gray level of the image. If there is a relatively steep peak in the histogram of the original image, the image is prone to over-enhancement.

[0004] 2. The relative gradient strength of the image cannot be improved

The histogram equalization algorithm is simple and efficient, and makes the gray level distribution of the enhanced image more uniform, but it cannot improve the relative gradient strength of the image, and research shows that the human visual system compares the relative brightness change (relative gradient) of the image. Sensitive, increasing the relative gradient strength of the image can effectively improve its subjective visual effect

Images