An adaptive method to correct the non-uniform illumination of images
Xuefei Xiong, Yang Shang · AOPC 2020: Optical Sensing and Imaging Technology · 2020
It is known there are some problems such as the image suffers from excessive brightness or darkness under non-uniform illumination, loss of detail in the enhanced image, halo effect and so on. In this paper, an adaptive correction algorithm is proposed for these problems. Firstly, based on the illumination-reflection model, the illumination component was extracted by weighted guided filtering integrated with gradient information. Afterwards, the index of 2-d gamma curve is adjusted with the illumination component to realize the adaptive illumination correction. Finally, the output image pixels are produced with their adjacent pixels to enhance the local contrast of the image. The experimental results show three aspects. One is the improved weighted guided filtering can distinguish the edges adaptively and avoid the halo caused by excessive smoothness near the edges. Two is the clarity of enhanced image was improved by over 86.5% on average. At the same time, due to the reduction of color distortion, the ability to enhance image detail was improved by 18.5% on average.