Research on Defect Detection of the Outer Side of Bottle Cap Based on High Angle and Multi-View Vision System

In the defect detection of the outer side of the bottle cap, it is difficult to achieve uniform illumination of the circular arc surface to be detected. And there is a texture variation of non-skid bars on the surface, which can easily cause wrong segmentation of defects. To solve these problems, this paper proposes a high-angle and multi-view inspection system (HAMV), in which the chief ray of the light source is incident uniformly along the arc direction to circular arc surface by means of high-angle annular illumination. Also, the complete imaging of the outer side is accomplished in four imaging views. Further, a defect detection algorithm based on the background reconstruction of line structure element (BRLSE) is established. It reconstructs the background texture using line structure element and reduces the interference of non-skid bars texture on defect detection. At the same time, the adaptive segmentation of different sub-regions is completed by combining the contour coordinates of the bottle cap to be inspected. Finally, different kinds of defects are extracted in different sub-regions. In this paper, the HAMV is applied to the actual inspection scenario, and the inspection speed is better than 400 pcs/min and the inspection accuracy is better than 95%, which can meet the actual production line capacity as well as the inspection requirements.