复杂光照环境下手工装配件视觉检测方法研究

手工装配流水线上的工件摆放位置随机,且光照复杂多变,现有的视觉检测方法易受影响.针对此工况下的插座螺钉装配检测需求,设计了一套复杂光照环境下手工装配件视觉检测方法.首先在HSV颜色空间利用颜色进行图像分割,找到工件外轮廓,通过几何位置换算找到螺钉孔所在的大致区域;然后在此区域中对自适应阈值二值化图像进行霍夫变换圆检测,...

Research on vision inspection method for manual assembly under complex illumination

Manual assemblies on a production line are placed randomly,and the illumination is complex and changeable.The existing visual detection methods are easily affected by this.A vision inspection method for manual assembly under complex illumination was designed to meet the testing requirements of socket screw assembly under this condition.Firstly,the image was segmented by color in HSV color space to find the outer contour of the workpiece,and the general area where the screw hole was located was found through geometric position conversion.Then,in this region,the Hough transform circle of the adaptive threshold binarization image was detected to obtain the precise position of the screw hole.Finally,the threshold value was set based on the average gray value of the whole image,and the gray value of the screw hole position was used for judgement.By that,the detection result was obtained.The experimental results showed that the proposed method had strong robustness and adaptability,and it had a good reference value in engineering.