Low-illumination image enhancement method

Abstract

A low-illumination image enhancement method comprises the following steps: (1) brightness detection: calculating a brightness value of a detected target image, and determining whether the brightness is too low; (2) image inversion: when the to-be-detected image has the problem of too low brightness, performing image inversion operation on the to-be-detected image; and if not, directly outputting the image. (3) image defogging: after the image with relatively low brightness is inverted, image features and histogram distribution are similar to those of a foggy image, so that defogging processing is carried out on the inverted image; and (4) image output: carrying out reverse normalization and inversion on the defogged image, and then outputting the defogged image. According to the invention, system design is carried out by using a dark channel prior theory, so that the system can effectively improve the quality of an acquired image in a low-illumination scene, after low-illumination scene image enhancement processing, the definition of an image shot in a low-illumination environment is remarkably improved, and the color and texture information of the image is also greatly improved.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a low-illuminance image enhancement method. Background technique [0002] In outdoor scenes, especially in areas with harsh conditions such as mountains and river valleys, it is very common to use drones for inspections. However, in environments with low illumination at night, such as moonlight environments and starlight environments, it is difficult to ensure that drones The quality of the collected image to be detected has a significant increase in the false detection rate and missed detection rate in the target detection task. The signal-to-noise ratio of the image to be detected collected in the above scene is reduced, the color and texture information are seriously lost, and the image recognition is poor, which greatly reduces the application value of the image. This type of image is called a low-light image. In low-light conditions, the imaging brightness of low-l...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is: to overcome the deficiencies in the prior art, a low-illuminance image enhancement method is provided, which includes the following steps: (1) brightness detection: calculate the brightness value of the detection target image, and determine whether its brightness is too low ;(2) Image inversion: when there is a problem of low brightness in the image to be detected, perform an image inversion operation on it; if it does not exist, directly output the image

Images