OLED Display Panel Oriented Defect Detection Algorithm And System Development

OLED (Organic light-emitting display) has been thought to be the next generation display equipment for its brightness in color, large viewing angle, frivolity, and energy saving. Due to the hundreds of complex processes in the OLED manufacturing, defects on the display panel are unavoidable. Recognization and detection of OLED defects is important for quality improvement which can be used for defect statistics, repairing, and failure disqualified products. This paper was focused on the study of OLED defect detection algorithms and system development. The main research contents are described as follow:1) Investigate current research state and future development og the OLED technology, understand OLED manufacturing processes and possible defects caused.2) Based on the study of image processing theory, this thesis proposed a corner-points extracting method which are extracted from skeleton image and used as control points for the differential operation. This method can be used to solve the problems of precision image alignment.3) Due to the difference between the standard template and the actual image, dynamic adjusted template was used in the image alignment which is based on the initial standard template to overcome the influences of uncertain environment.4) Based on the Otsu algorithm, this thesis modified the algorithm and improved the speed of image segmentation. Through the recursive threshold determination, the algorithm can calculate the threshold based on the previous value and thus it can speed up the calculation.5) Based on the proposed algorithms about OLED panel defects detection, this thesis developed a OLED-oriented defect detection system through VC++6.0 programming. The system is validated through several OLED image examples. And the testing results show that the developed system can detect most of the defects on the panel.