Research on an Improved Segmentation Recognition Algorithm of Overlapping .

The accurate identification of overlapping in a factory environment is one of the challenges faced by automated picking. In order to better segment the complex adhesion between , this paper proposes a segmentation recognition algorithm for overlapping . This algorithm calculates the global gradient threshold and divides the image according to the image edge gradient feature to obtain the binary image.