Optimization of Low Illumination Image Enhancement Algorithm Based on Dark Channel Prior

摘要

The dark channel prior algorithm has achieved a certain effect in the enhancement of low-illuminance images, but this method has a large calculation amount and high memory requirements, and the dark channel calculation is not accurate enough. In this paper, based on the dark channel prior theory, by performing edge detection on the brightness channel of the image to generate a mask, the edge area of the object in the image and the large area flat area are processed separately. Using this method, it is possible to restore the basic method that is basically consistent with the original method or even more accurate. As a result, it can greatly reduce consumption and increase the calculation speed.