Wooden pallet image segmentation based on Otsu and marker watershed

Abstract

Pallet detection is the key step of cargo handling for warehouse robots. In the visual detection, pallet needs to be segmented from the image background. In order to increase the recognition rate and reduce the influence of the surrounding environment and the pattern of the pallet, Otsu algorithm and marker watershed algorithm are combined to realize the image segmentation of the pallet contour. Based on the difference of colour information between the pallet and the background, Otsu algorithm is used to segment the pallet for the first time, and marker watershed is used to complete the target recognition. The experimental results show that the method can effectively solve the problem of over segmentation of the pallet object, and improve the recognition rate, which has some reference value for the design of the visual detection system of the warehouse robot wooden pallet.