表面弱边缘瑕疵检测算法及应用

Surface Weak Edge Flaw Detection Algorithm and Its Application

工业环境中,产品表面质量是产品质量的重要组成部分,也是商品商业价值的重要保障。产品表面瑕疵由于受到光源、光照方式等因素的影响,容易形成灰度变化相对缓慢的过渡边缘即弱边缘。在工业生产中,弱边缘瑕疵由于其灰度缓慢变化,所以相对强边缘瑕疵较难被检测出。针对这一问题,提出了一种有效检测弱边缘瑕疵的方法。该方法利用Scharr算子能突出弱边缘的优势,结合形态学开闭滤波,从而达到强化弱边缘的效果。根据形成的弱边缘增强图像,利用最大熵阈值对图像进行分割得到瑕疵。对光缆生产流水线上采集的图像利用该方法进行测试分析。实验结果表明,使用改进的弱边缘瑕疵检测算法能更好地检测出弱边缘,从而更加有效地识别表面瑕疵,提高了产品质量。

In the industrial environment,the surface quality of products has become an important part of product quality as well as an important guarantee for the commercial value of commodities.Influenced by factors such as light source,light mode and so on,product surface defects can be prone to form weak edge that the changes in their gray scale are relatively slow.In industrial production,compared with strong edge blemishes,weak edge blemishes are more difficult to be detected due to their gradual change in gray level.Therefore,we propose a method to detect the weak edge defects effectively.In order to strengthen the weak edge,the advantage that Scharr operator can highlight the weak edge is utilized,combining with the morphological opening and closing filtering,and then the maximum entropy threshold is used to segment the image to get the flaw.So far,the images collected on the production line of optical fiber cable are tested and analyzed.Experiment shows that the improved weak edge detection algorithm can better detect weak edge flaws,which identifies surface defects more effectively and accordingly improves product quality.