Method for excavating optimum image based on information entropy and logarithm contrast weight sum

Abstract

The invention relates to a method for digging an optimal image based on information entropy and logarithm contrast weighted sum, which comprises the following steps: (1) detecting the gray scale spectrum distribution of original images, determining the existence of a descending gray scale spectrum distribution, and then calculating the gray scale spectrum continuous bandwidth BW of the original images; (2) establishing an image quality assessment function IQAF; (3) digging a bottom layer image, and calculating the value of the assessment function IQAF by using the information entropy InEn corresponding to a dug image and an average logarithm contrast lgC; and (4) using a digging parameter corresponding to the maximum value of the image quality assessment function IQAF to dig an image which is the image with the optimal quality according to the bottom layer image digging formula in the step (3). The method can quickly dig out the image with the optimal quality objectively from the images with the descending gray scale spectrum distribution of the original images, and the images shot under a condition with quite weak illumination can obtain clear optimal images with objective sceneries through the method.

Description

technical field [0001] The invention belongs to the technical field of computer digital image processing. Background technique [0002] The underlying image mining technology converts information that is not available to human vision into visually usable information, which can be used to mine image information captured under harsh conditions, such as car hit-and-run, and information hidden in monitoring images of financial institutions. However, in the mining process, it is impossible to judge what mining parameters to use to mine the best image quality, so in practice, repeated mining has to be done to make the mined image information reach the best quality of human vision, which is time-consuming and laborious, and the mining results are very high. The so-called best quality image depends on the subjective feelings of different individuals, not the best quality image in the objective sense. At present, how to directly obtain the best quality images in mining has become an...

AI Technical Summary

Problems solved by technology

However, in the mining process, it is impossible to judge what mining parameters to use to mine the best image quality, so in practice, repeated mining has to be done to make the mined image information reach the best quality of human vision, which is time-consuming and laborious, and the mining results are very high. The so-called best quality image depends on the subjective feelings of different individuals, not the best quality image in the objective sense

At present, at home and abroad, how to directly obtain the best quality images in mining has become an unsolved problem for human beings.

Images