Dirt detection method based on concentric circle segmentation positioning

The invention relates to the technical field of camera detection, and particularly discloses a dirt detection method based on concentric circle segmentation positioning, which comprises the following steps: acquiring multi-frame image data; carrying out gray scale, gray level and edge cutting processing on the image; carrying out concentric circle segmentation processing and region division on the cut image; calculating a region deviation ratio and gradient processing according to the divided regions; then, carrying out dirt detection through a multi-region double-card air standard; and if the threshold value is exceeded, judging that dirt exists, otherwise, judging that no dirt exists. By adopting the technical scheme of the invention, the dirt of the camera module can be effectively detected, the accuracy of dirt detection is improved, and different users can conveniently and autonomously adjust dirt card control standards. 本发明涉及摄像头检测技术领域，具体公开了一种基于同心圆分割定位的脏污检测方法，包括获取多帧图像数据；对图像进行灰阶、灰度及边缘剪切处理；对剪切后图像进行同心圆分割处理、