基于动态阈值和分层检测的图像缺陷识别算法的研究与应用

针对高精度印刷图像的质量检测中的检测速度和检测精度问题,本文提出了一种基于动态灰度阈值和分层检测的高精度印刷品质量检测的实时的图像缺陷识别算法.算法考虑到人的视觉感官特性,首先将所用的灰度阈值对采集到的印刷品图像进行缺陷的初步识别和标定,然后再按照高精度图像检测要求采用分层检测的方法进行搜索,解决了以往图像检测缺陷识别中的检测效果和速度不理想的问题.经过在高精度印刷图像的质量自动化检测系统中的应用,算法的速度和效果得到验证.

Research and Application of Image Defect Recognization Algorithm Based on Dynamic Threshold and Storied Inspect

Aimed at the problem of inspecting velocity and precision in the quality inspecting of high precision printing image, the paper presented a real_time image defect recognization algorithm, which is based on dynamic threshold and storied inspect. Considering person's vision, the algorithm first sets the dynamic threshold to recognize and locate defects primarily in printing image and then adopts storied inspect method according to he request of high precision image inspecting. The algorithm resolves the unfavorable effect and velocity problem in former image defect recognization. The algorithm has been used in the high precision automatic quality inspect system of printing image, and its velocity and effect are proved preferable.