基于机器视觉的光纤倒像器缺陷检测系统

Defect Detection System for Fiber Image Inverter Based on Machine Vision

设计了基于机器视觉的光纤倒像器缺陷检测系统,通过对成像系统的机理研究并结合光纤倒像器缺陷特点,选用远心工业镜头配合千万像素的工业相机,实现了光纤倒像器表面缺陷的一次性成像,并保证了足够的分辨率。在检测算法上利用形态学运算中的膨胀和腐蚀运算得到其边缘轮廓,通过求取边缘像素的加权平均灰度值作为原图像的分割阈值。仿真结果表明,该算法能够准确、有效地实现缺陷分割。同时,该系统与人工检测相比效率提高近2倍,速度可达2 s/个,重复精度大于98.8%,整体系统运行稳定,能够满足市场需求。

To overcome the problem of low efficiency detection, large error and high cost for the apparent defect inspection technology for fiber image inverter, in this paper, a defect detection system is designed based on machine vision for fiber optical inverter and by the mechanism research of the imaging system and the combination of optical fiber image inverter defect characteristics. The system uses telecentric industrial lens with millions of pixels of industrial camera, realizes one-time imaging for the surface defects of optical fiber image inverter, and ensures the sufficient resolution. The detection algorithm uses morphological operations to obtain the edge contour, the image threshold segmentation is calculated by using weighted average to the original edge pixel gray value. The algorithm is accurate and effective to realize the defect segmentation. Experimental results show than the system gets efficiency nearly twice compared with manual detection. The speed of the system is 2 s/a, and the accuracy of the system is more than 98.8%. The whole system is stable and can meet the demand of the market.