基于路径跟踪的水下焊接V形焊缝识别中强干扰的去除

介绍机器人水下焊接V形焊缝图像处理中基于模糊推理的阈值自动选取的算法,提出基于路径跟踪的强干扰的去除方法.制作9种不同形状的窗口模板,以列扫描方式求取这些窗口所覆盖的区域内最大的平均灰度值,同时获取取得最大灰度平均值时的垂直方向的位置,然后根据获取的最大平均灰度值以及相邻区域最大平均灰度值的改变量进行模糊推理,判断获得这些窗口所覆盖区域的阈值.最后跟踪取得最大灰度平均值时的所有位置,在最低位置以下以及最高位置以上检测经阈值处理后的白点,可以消除因水泡、较强弧光形成的干扰.

Removal of strong arc interference in detection of V-shape weld based on path tracking under water

The algorithm of threshold automatic selection based on fuzzy reasoning for image processing of arc-v-shape weld seam is introduced. The removal method of the strong arc interference is proposed based on path tracking. By designing nine kinds of window templates, and acquiring the maximum average gray value in the region where these templates can cover through column scanning mode to get the vertical positions where the average gray value are maximum, the threshold is obtained through fuzzy reasoning using the maximum gray value and its increment. After that, all of the positions are tracked to find the white spots below the lowest one and above the tiptop one, and then eliminate the disturbances formed by blebs or little strong arcs below the lowest position and upon the tiptop.