Abstract
The OLED defects generated in the manufacturing process restrict the development of OLED industry, machine vision based automatic OLED-inspection equipment can rapidly detect these defects and help to improve the OLED manufacturing process. The OLED images have the features of repeating texture background, uneven overall brightness of the image and the defects without obvious edge. In addition, an uncertain change in light and position of the inspection system increases the difficulty of the detection. Therefore, we propose the method to detect defects which take advance of the human eye characteristics of the Gabor filter and the unsupervised and fast segmentation features of the Fuzzy C-Means FCM algorithm. Through the combined 2-step segmentation, most OLED defects can be detected. The experimental tests are performed to validate the effectiveness of the proposed method. The result of the experiment shows that this method works well which can meet the requirements of robustness, automation of the fast and reliable of an online inspection system.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
