Open AccessA threshold segmentation method for non-uniform illumination image based on brightness equalization
Author(s) -
Guo Hui,
Peng Chen,
Shouren Huang,
Cong Hu,
Lijun Zhang
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/569/5/052046
Subject(s) - artificial intelligence , brightness , computer vision , histogram equalization , image segmentation , gradation , computer science , segmentation , region growing , inflection point , pattern recognition (psychology) , mathematics , image (mathematics) , histogram , scale space segmentation , optics , physics , geometry
In the process of image acquisition, non-uniform illumination images are common due to poor lighting, surface reflection, or a combination of these two factors. In order to improve the quality of image segmentation, an image threshold segmentation method is proposed for non-uniform illumination images. First, the brightness of different regions in the image is compensated to make the brightness background of the whole image consistent. Second, the gradation histogram of the processed image is obtained and simulated into Gaussian distribution curve, then the inflection points of the curve are calculated. Finally, the two inflection points are set as thresholds to segment the image after brightness equalization. This method eliminates the influence of non-uniform illumination to a certain extent and gets a better segmentation effect. Experiments were carried out on images containing several different types of non-uniform illumination. The results demonstrate that the proposed method outperforms the compared enhancement algorithms in threshold segmentation.