A real-time enhancement method for low-light video based on bright channel

Abstract

A real-time enhancement method for low-illuminance video based on the bright channel. The method determines the enhancement coefficient of each pixel according to the calculated bright channel value, and adjusts the overall enhancement degree of the image / video according to the average value of the image. Calculate the mean meanD of the low-illumination video frame; set the time threshold T; calculate the bright channel value Jbright(x) of the low-illumination video frame; calculate the illumination factor γ(x); calculate the enhanced video frame BC(x); over-blue correction The enhancement effect is good. The color of the enhanced video is true, which effectively avoids the "over-enhancement" phenomenon in the traditional method, avoids the window filtering operation process, and fundamentally avoids the block effect problem commonly found in existing inventions; the calculation speed is fast. It effectively solves the problem of insufficient computing speed in the existing invention, processes high-definition video at 25 frames per second in real time, and meets real-time processing requirements; the technical solution is simple to implement. The technical scheme of the present invention can be realized on a personal computer without the assistance of any complicated peripheral equipment, and the cost is low.

Description

technical field [0001] The invention belongs to the field of digital image / video signal processing, and relates to a method for real-time enhancement of low-illuminance video based on bright channels. Background technique [0002] The images / videos captured under low-light conditions have problems such as dark colors and insufficient color contrast, which lead to obvious degradation of the collected images / videos, and in severe cases, the video surveillance system will not work properly. Therefore, it is particularly important to study a real-time low-light video enhancement method to improve the subjective perception of low-quality video. [0003] The research on low-light image / video enhancement methods has received widespread attention. Traditional low-light image / video enhancement methods mainly use methods such as grayscale curve mapping and histogram equalization to process images / videos. Although such methods can effectively improve the contrast of images / videos, th...

AI Technical Summary

Problems solved by technology

This method runs relatively fast, and can process SD video frames (720×576) in real time with 4 CPUs, but still cannot process HD video (1280×720) in real time

[0006] In summary, although the existing technology can basically achieve the effect of low-light enhancement, it cannot meet the requirements of real-time processing.

With the continuous upgrading of electronic acquisition equipment, the images / videos collected by the monitoring system are gradually developing from the original standard definition format (720×576) to high definition format (1280×720), which will lead to a significant increase in the amount of video data

Such a huge amount of data poses a serious challenge to the real-time performance of the enhancement method, and the existing methods cannot meet the real-time processing requirements of high-definition video

Images