基于Blob分析的手机屏幕缺陷检测方法

Mobile Phone Screen Defect Detection Method Based on Blob Analysis

针对手机屏幕缺陷检测的高效率、高精度的要求,提出一种基于Blob分析的手机屏幕缺陷检测方法。首先使用基于PatMax算法的模板定位技术对图像进行定位,其次使用基于仿射变换的图像校正技术实现模板图像与测试图像的逐像素对齐,最后采用Blob分析算法获取缺陷特征的位置、面积、方向等信息。采用文中所提出的方法、差影法对200幅手机屏幕图像进行缺陷检测,实验结果显示,所提方法的准确率为98%,误检率为1.5%,漏检率为0.5%,每张图像的平均检测时间为52.19 ms。相较于差影法,所提出的方法具有更高的准确率和检测速度。

This paper proposes a defect detection method for mobile phone screen based on Blob analysis to meet the high efficiency and high precision requirements of mobile phone screen defects. Firstly, the template localization technology based on PatMax algorithm is used to locate the image, and the image-correction technique based on affine transformation is used to realize the pixel-by-pixel alignment of the template image and the test image. Finally, the Blob analysis algorithm is used to obtain the location, area, direction and other information of defect features.The defects of 200 mobile phone screen images were detected by the proposed method and the difference image method. The experimental results show that the accuracy of the proposed method is 98%, the false detection rate is 1.5%, and the missed detection rate is 0.5%. The average detection time of each image is 52.19 ms. Compared with the difference image method, the proposed method has higher accuracy and detection speed.