Online Detection of Beer Bottle Based on Optical Flow Field Segmentation and SUSAN Algorithm

To realize fast defect detection on moving object, a fusion algorithm based on segmentation of optical flow field and Susan´s operator a method was put forward under the condition of camera being fixed. The method first finishes the area segmentation depending upon the image segmentation algorithm of optical flow field, then takes advantage of edge detection precision of SUSAN operator to get the detailed edge map of object area. The experiments based on the fusion algorithm have shown encouraging results of fast detecting the surface defect of cylindrical moving object on product line.