X射线管道焊缝探伤中的视觉同步技术

研究了一种基于机器视觉的同步技术。实现同步的关键在于检测管内外的转角差，通过对X射线图像进行特征提取获取管内外转角差。在特征提取时，先对图像进行形态学滤波，然后用遗传算法（GA）进行图像分割，最后用Hough变换从图像中提取特征量。试验结果表明，该算法从X射线图像中准确检测出管内外的转角差，且检测速度超过了25fps的实时性要求。该算法再结合一定的控制规律控制伺服电机，即可实现管内外同步旋转。

Visual Synchronous Technology in X-ray Pipeline-weld Inspection

The paper proposes a kind of synchronous algorithms based on machine vision.What is the most important to realize synchronization is to detect the deviation angle.The method detects the deviation angle by extracting character from X-ray images.First,the method processes the image with a morphology filter,and then segments it with genetic algorithm(GA),finally extracts character with Hough transform.The experimental results show that,the algorithms detect the deviation angle precisely,and the detection velocity reaches 25 fps.Thus by controlling the servo-motor with the algorithms and a certain control rule,synchronization can be achieved.