Author(s)

Jiabao Liu, Yang Fu, Bin Wang

Corresponding Author

Jiabao Liu

Abstract

The automatic identification technology of goods in intelligent three-dimensional warehouse has the problems of high cost of RFID electronic tag and low recognition rate of goods image recognition technology. This paper presents an automatic cargo image recognition method based on the combination of feature matching and template matching. Firstly, the minimum container rectangle (MBR) in the image is extracted as the template for image matching. Secondly, the scale-invariant feature transform (SIFT) algorithm is used for rough matching of cargo images. Finally, the deformable multiple similarity measure (DDIS) method was used to match accurately. Experimental results show that the proposed method has high recognition accuracy and relatively fast recognition speed, and has certain practical value.

Keywords

Automatic image recognition; SIFT feature matching; DDIS template matching; Minimal Rectangle Extraction