Low-illumination color image enhancement method and device

Abstract

The invention aims to overcome the defects in the prior art, provides the low-illumination color image enhancement method and device, which has a good inhibition effect on noise generated by imaging equipment, and effectively avoids noise pollution caused by a traditional enhancement algorithm and image information loss caused by a traditional denoising algorithm. According to the method, throughframe average method noise reduction and self-adaptive Gamma correction, an image sequence is sequentially converted from an RGB color space to a YUV color space, obtained YUV color space images are subjected to parallel processing, a non-global objective quality evaluation method is adopted to perform quality sorting on blind source separation noise reduction images, contrast-limited adaptive histogram equalization processing is performed on the optimal quality image, adaptive illumination equalization processing is performed on the fused image, and the illumination equalization processing image with U and V channel components of the frame average image are combined to obtain an enhanced image of the low-illumination color image.

Description

technical field [0001] The invention relates to a low-illuminance color image enhancement method, which belongs to the technical field of image enhancement. Background technique [0002] In today's society, 70-80% of the information that people obtain is obtained through the visual system, and the information contained in the image is more intuitive and rich. However, in low-light scenes, the information such as the color of the image is not intuitive and rich enough. Therefore, more and more Scholars began to study image restoration in low-light scenes. [0003] At present, it is mainly divided into two major research fields, one is the hardware field, and the other is the algorithm field. The field of hardware is mainly to obtain more and rich photosensitive information through the research and design of circuit devices, and at the same time overcome the noise of circuit devices; the field of algorithm is mainly to process the received information to restore scene informa...

AI Technical Summary

Problems solved by technology

At present, images of low-light scenes are mostly captured through devices such as image intensifiers and thermal imagers. However, these devices are not only expensive, but also cannot truly reflect the color information of the original scene.

[0004] Now there are various classic low-light enhancement algorithms, but they are generally enhanced for a single image, and the image enhancement also amplifies the noise component, and there are many disadvantages, such as color distortion, loss of details, etc.

Images