Mura defect detection method for AMOLED display screen

The invention provides a Mura defect detection method for an AMOLED display screen. The method comprises the steps of firstly performing pre-segmentation on a Mura image of the AMOLED display screen by adopting a mean shift algorithm to obtain an initial profile required by a level set algorithm; and secondly performing segmentation on the Mura image of the AMOLED display screen by adopting the level set algorithm combining local and global information of the image. The method has the beneficial effects that the problem that a local image model is sensitive to the initial profile is solved; the proposed level set algorithm combines the local and global information of the image; and the problem that a global image model cannot process the image with grayscale unevenness is solved. 本发明提供了种AMOLED显示屏Mura缺陷检测方法，首先对AMOLED显示屏Mura图像采用均值漂移算法进行预分割，得到水平集算法所需的初始轮廓，然后再采用结合了图像的局部和全局信息的水平集算法对AMOLED显示屏Mura图像进行分割。本发明的有益效果是：解决了局部图像模型对初始轮廓敏感的问题，同时所提出的水平集算法结合了图像的局部和全局信息，解决了全局图像模型不能处理灰度不均图像的问题。