Low-light video real-time enhancement method based on bright channel

Abstract

The invention provides a low-light video real-time enhancement method based on a bright channel. The method includes the steps of determining the enhancement coefficient of each pixel point according to the calculated bright channel value, regulating the overall enhancement degree of an image / video according to the mean value of the image, calculating the mean value meanD of low-light video frames, setting the time threshold value T, calculating the bright channel value Jbright(x) of the low-light video frames, calculating the illumination factor gamma(x), and calculating the enhanced video frames BC(x). The blue correction enhancement effect is good. The colors of the enhanced video are true, the over enhancement phenomenon in a traditional method is effectively avoided. The window filtering operation process is avoided. The massive effect problem frequently seen in the prior art is fundamentally solved. The calculation speed is high. The problem that in the prior art, the operation speed is not high enough is solved, the 25 frame per second high-definition video is processed in real time, and the real-time processing requirement is met. The technical scheme is easy to achieve and is free of assistance of any complex peripheral device, the method can be achieved on a personal computer, and 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