The Application of Single Shot MultiBox Detector in multiple-object QR codes recognition

QR code recognition is one of the key and core technologies in the Internet of Things industry. In actual QR code recognition scenarios, there are often multiple objects to be recognized. However, existing QR code recognition technologies are limited to single object recognition. Even if multiple QR codes can be scanned, they cannot be recognized simultaneously. Based on research into the Single Shot MultiBox Detector (SSD), a multi-target detection convolutional neural network, we propose a method for real-time positioning and identification of multiple QR codes. First, The SSD network was trained to detect multiple QR codes, allowing each individual QR code to be identified for further analysis. Second, the images of QR codes were improved by integrating image enhancement techniques across various environments. Third, Zbar-based QR code recognition technology is utilized for recognizing single QR codes which were detected by SSD. The proposed approach enables simultaneous localization and recognition of QR codes, even during camera movement. The efficiency of QR code information reading and the accuracy of identification will be significantly improved.