An Optical Detection System For AMOLED Screen Brightness Uniformity

Active-matrix organic light emitting diode(AMOLED)is the most popular display technology at present.AMOLED display has the characteristics of ultra-light,ultra-thin,flexible,true color,low power consumption,fast response,ultra-wide viewing angle and ultra-high contrast,and is widely used in smart phones,tablet computers and other mobile terminals.With the rapid growth of demand for AMOLED panels by various terminal devices,AMOLED panels are mass-produced on a large scale.However,due to the complicated manufacturing process of AMOLED panels,some AMOLED panels are inevitable.In the AMOLED panel,the Mura defect is the most complicated defect in the AMOLED panel due to its uneven local brightness,low contrast and blurred outline,and is also a defect most in need of detection.At present,all major AMOLED panel manufacturers in the world are making unremitting efforts to improve panel quality.The most effective and direct way to ensure the quality of AMOLED panels is to strictly control product quality and strictly test AMOLED panels.The AMOLED panel defects detection of this link,many panel manufacturers are still using artificial methods with the naked eye for defect detection.Due to different experience and subjective feeling of different inspectors,the detection results of Mura defects on the same panel may be different.In addition,the method of artificial visual inspection is affected by the subjective factors such as emotional changes and fatigue of the inspectors,which can not guarantee the reliability and consistency of the test results.In order to solve these shortcomings,the subject has designed an optical system for the brightness uniformity of the AMOLED screen,and used the machine vision method to detect AMOLED panel defects efficiently and accurately.Aiming at the expected requirement,a scheme of Mura defect detection is proposed.The characteristics of Mura defects of AMOLED panel and the requirement of industrial production are investigated deeply.Aiming at the requirement of imaging quality of Mura defect detection on AMOLED panel,the requirement of Mura defect accurately and fast segmentation,and the requirement of quantitative evaluation,the optical inspection system of AMOLED panel brightness uniformity is built.The algorithm of image distortion correction is designed.By using the result of camera calibration,the problem of image distortion caused by the error of image acquisition system is solved.A method of combining Gabor wavelet filter and low-pass filter is proposed,which solves the problem that the collected images show repetitive background texture due to R,G,B subpixels regularly arranged on the AMOLED screen.A Mura defect detection method based on mean shift and level set algorithm is proposed,which solves the problem that the local image model is sensitive to the initial contour of level set.Because the proposed level set algorithm combines the local and global Information,which solves the problem that the global image model can not effectively segment images of intensity inhomogeneity.Finally,according to the quantitative method of Mura defects of SEMU standard,by using the characteristic parameters of the Mura defect,the accuracy and reliability of the Mura defect level were evaluated,which provided the support for the improvement of AMOLED panel technology.