Optimal-order image enhancement method based on fractional differential image enhancement algorithm

Abstract

The invention discloses an optimal-order image enhancement method based on a fractional differential image enhancement algorithm, and the method comprises: calculating the gradient, information entropy, image brightness, human eye feeling brightness, and human eye contrast sensitivity functions of an image, so as to judge the characteristics of the image; performing normalization calculation on the gradient, the information entropy, the image brightness, the human eye feeling brightness and the human eye contrast sensitivity function of the image to obtain a value S; numerical values S representing image texture information, brightness and human eye visual perception are put into the logarithm compression curve to be compressed, and a corresponding optimal fractional order is obtained; Andusing the optimal fractional order in the fractional order differential algorithm of the image to realize self-adaptive image enhancement. Normalized local statistical information capable of representing image characteristics is constructed, a function relation between the orders and the normalized local statistical information is established, the normalized local information is used as an independent variable, a fractional order differential algorithm of the optimal order is obtained according to a logarithm function, and image enhancement is carried out.

Description

technical field [0001] The present invention specifically relates to the technical field of image processing, in particular to an optimal order image enhancement method based on a fractional order differential image enhancement algorithm. Background technique [0002] Image enhancement is the processing of images to improve image quality for better analysis and processing. Traditional differential image enhancement methods are based on integer-order differential enhancement algorithms, such as Sobel, Roberts, Prewitt algorithms and Laplacian algorithms based on second-order integer differentials. These algorithms can extract the image edge information very well, but they will greatly attenuate the middle and low frequency information of the image, and the contour information of the smooth area of ​​the image will be lost. In the existing technology, many scholars have made improvements to the integer-order differential algorithm, but no essential breakthrough has been made....

AI Technical Summary

Problems solved by technology

In recent years, many scholars have constructed a fractional differential image enhancement algorithm based on the mathematical theory of fractional differential combined with image processing methods, and verified the advantages of the application of fractional differential in image enhancement, but there is a problem that has become a technical bottleneck: in the algorithm In the application of the present invention, how to select the best number of sub-orders to achieve image enhancement has become a technical problem. In the existing applications, the number of sub-orders is artificially set, and the optimal sub-order cannot be determined adaptively through quantitative analysis. number for best enhancement

Images