Research On Image Enhancement Algorithm Optimization Based On Histogram Equalization

Image enhancement is an important part of digital image processing,the purpose of which is to highlight the region of interest by improving the contrast of the image,while retaining the original image details as much as possible to improve the information identification of the image.As one of them,histogram equalization has been widely used because of its simplicity,ease of implementation and effectiveness.However,it also has some defects: the excessive stretch of gray level leads to the over-enhancement phenomenon and artifact,so that the image enhancement effect is not natural;The merging of gray levels leads to the loss of detail information and the blurring of edge information;The directional transfer of the average brightness of the enhanced image cannot meet the enhancement requirements of consumer electronics products.In view of the above defects,this study combined with different application scenarios on the basis of the existing research optimization,the specific work is as follows:(1)Aiming at the poor performance of histogram equalization algorithms in low-illumination grayscale image enhancement,a histogram equalization image enhancement algorithm based on luminance correction and limited contrast was proposed,including luminance correction and contrast correction of low-illumination image.This method optimizes the calculation method of the traditional histogram distribution range,and improves the problem of insufficient brightness correction.According to the histogram distribution characteristics of low-illumination images,a proportional segmentation method based on the distribution range is proposed,which is beneficial to improve the image brightness and have a reasonable contrast ratio.Experimental results show that the proposed algorithm is superior to other contrast algorithms in enhancing image brightness and preserving detail information.The enhancement results are natural,and good enhancement effects can be achieved for different types and degrees of low-illumination images.(2)In view of the problem that the existing contrast control methods cannot suppress the over-enhancement and gray level merging,a histogram reconstruction model is proposed to further control the contrast and detail information of the image flexibly.On this basis,a histogram equalization image enhancement algorithm based on histogram reconstruction and brightness preservation is proposed,which provides a relatively balanced treatment for the contradiction between brightness preservation and contrast enhancement,that is,the average brightness of the original image is taken as a distribution point of the remap interval.The experimental results show that this method is superior to other methods in subjective and objective evaluation for gray images with different brightness,namely dark,medium and bright.It can reasonably improve the contrast of the image while keeping the original image brightness as much as possible,and enhance the image with rich details.(3)In order to solve the problem of color distortion caused by over enhancement of histogram equalization algorithm in underwear image enhancement process,an underwater image enhancement algorithm based on exposure value and reconstructed histogram is proposed.The histogram segmentation method based on exposure value is optimized,and the mapping interval redistribution method based on the pixel proportion of sub-histogram is improved.Experimental results show that the optimized histogram segmentation and mapping interval reallocation method can improve the color information and clarity of enhanced images.The objective evaluation index results show that,compared with other contrast methods,the enhanced results generated by this method are more similar to the reference images in the underwater image dataset,and the enhanced results are natural and colorful.