A Stereoscopic Warehouse Stocktaking Method Based on Machine Vision

Abstract

In order to solve the problem of high security risk, low efficiency and low accuracy of the manual stocktaking during stocktaking of stereoscopic warehouse, the image processing method of python+OpenCV visual library software was adopted. Firstly, the sampling camera is controlled by the stacker PLC to traverse all cargo positions and obtain physical images of the goods. Then, OpenCV platform is used to provide operators for image preprocessing, image segmentation and one-dimensional barcode recognition, so as to identify the barcode information and determine the material type. Finally, the inventory report is generated by comparing with WMS system. The research shows that the OpenCV-based method can quickly and accurately identify the one-dimensional barcode, and realize the automatic unmanned intelligent inventory, effectively reduce the workload of staff, improve the production efficiency, avoid the potential risk of manual inventory, and achieve the expected goal.