基于机器视觉的瓷砖素坯表面缺陷无损检测算法研究

Nondestructive detection algorithm research of the surface defects of ceramic tile billet based on machine vision

针对瓷砖生产中对瓷砖素坯表面缺陷检测的高速度和高准确率的要求,本文提出了基于机器视觉的表面缺陷无损检测算法。首先对采集到的瓷砖素坯图像采用双边滤波器进行图像预处理,降低噪声,提高图像质量,然后利用Canny边缘算子提取图像边缘,在图像边缘的基础上,采用最佳阈值分割算法,实现图像分割,利用圆形度对缺陷特征进行描述,并实现缺陷判别,最后,通过实验验证表面缺陷检测的准确度和稳定性。实验结果表明,该方法可以实现瓷砖素坯表面缺陷的无损检测,准确率达到95%以上,能应用于瓷砖表面缺陷质量检测的生产实践中。

Aiming at the high speed and high accuracy requirements of the surface defects of ceramic tile billet,the paper presents nondestructive detection algorithms of surface defects based on machine vision. Firstly,in order to reduce noise and improve image quality,the image of ceramic tile billet is processed by bilateral filter. Then,image edge is extracted by Canny edge operator and image segmentation is realized by optimal threshold segmentation algorithm based on image edge.Next,defect characteristics are described and identified by degree of circularity. Finally,the accuracy and stability of the surface defect detection is verified through the experiment. The experimental results show that,nondestructive detection of the surface defects of ceramic tile billet is realized by the method and the detection accuracy in the experiments is over 95%.It can be applied to production of ceramic tiles on quality detection of the surface defects of the ceramic tiles.