圖像化二維條碼之實體輸出及深度學習辨識研究

The development of smart phones has led to the rise of the use of mobile device. QR codes that can connect data on physical documents have also become the most commonly used two-dimensional barcodes. The QR code is composed of black and white modules, which detracts from its aesthetic appearance. Therefore, many scholars have conducted related research on beautifying QR code. Although there is a lot of research on beautification, it has not been widely adopted. The key reason is: when printed, due to size, dot gain, and other printing conditions, barcode information is easily distorted, yielding poor recognition results. We propose a systematic method of embedding information of aesthetics QR code and deep learning recognition error analysis method of physical printout of QR code. The experimental results show that the proposed method is compatible with current output equipment. After the information intensity is embedded, the output aesthetic QR code still has better visual quality. Through deep learning recognition, it improves the decoding rates of aesthetic QR codes. In addition to beautifying the QR code to change its appearance, the aesthetic QR code can also be stably decoded on physical output, and can be used in commercial value-added applications.