A Framework Of Barcode Localization For Mobile Robots

Traditional automatic barcode localization method in mobile service robots for assisting shopping services has a conflict between wide field of view for barcode detection and high resolution barcode images for barcode localization. To resolve this conflict, we use two auto-focus zoom cameras in the framework: one static camera with a short focal length is used to capture the wide field image for barcode regions detection, and the other one is a pan-tilt-zoom camera to capture the sharp and high resolution barcode images for barcode localization. In addition, a novel barcode localization algorithm of field mode is also proposed. By using the framework, higher flexibility on the camera-to-object distance is beneficial for mobile robots in automatic barcode localization. The experimental results show that the proposed framework with the proposed barcode localization algorithm can accurately localize the barcodes on product packaging with complex imprinted patterns.