A Grain-Controllable Low-Illumination Image Enhancement Method

Abstract

The invention provides a method for enhancing a granularity-controllable low-illuminance image. The method includes the steps that firstly, an light radiation field of an original low-illuminance image is obtained, then through estimation of the brightness of the original low-illuminance image, an adaptive brightness repair factor omega and basic light intensity A are prepared, and by the utilization of the standard deviation of the light radiation field, two adaptive Gaussian filters are prepared; finally, noise removal and fuzzification are conducted on the light radiation field through the two Gaussian filters and a recovered image is obtained through the adaptive brightness repair factor omega and the basic light intensity A. Due to the fact that an airglow scattering model reflecting the visible light imaging low is adopted, the image recovery effect can meet the human vision characteristic; the Gaussian filters are designed adaptively and the brightness repair factor and the basic light intensity are calculated adaptively, the adaption to scene changes is achieved; the method is suitable for color images, or gray level images or optical images or other spectrum images and has universality.

Description

technical field [0001] The invention relates to a method for enhancing low-illuminance images with controllable granularity, which can be used for single images or continuous videos, belongs to the field of image processing, and relates to scalable computational complexity, adaptive Gaussian filters based on regional consistency, and self-adaptive Image processing methods such as brightness restoration coefficients. Background technique [0002] At present, in order to enhance the safety factor of the people, video surveillance systems are installed in various important places in the society, such as houses, banks, roads, etc., so that once a crime or other emergency occurs, the video surveillance screen can be used. Monitor specific scenarios after the fact or in real time. However, most lawbreakers generally choose night or places with low brightness to carry out criminal activities. At this time, the monitoring system cannot capture a clearer picture because the ambient ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the shortcomings of the previous low-illuminance image enhancement algorithm, such as high complexity, general visual effect, and inability to adjust the brightness according to the scene adaptively, and propose a low-illuminance image enhancement method with controllable granularity, using atmospheric light scattering The model can restore high-quality pictures by adjusting the parameters of the Gaussian filter and the brightness repair factor according to different scenes and brightness. It has a general enhancement effect on videos and images in low-light environments, and has high robustness. After enhancement The visual effect of the image is better

Images