Rectification of images of QR codes posted on cylinders by conic segmentation

Powered by the pervasive penetration of inexpensive mobile devices and relevant APP's into our daily life, QR codes have found a wide range of applications in recent years. Therefore, the processing of the images of QR codes (also referred to as QR images) is an important task. In this paper, we try to deal with the scenario wherein the QR code is printed or pasted (altogether also referred to as "posted") on the surface of a cylinder. In the scenario considered in this work, the posting need not be aligned to the axis of the cylinder. When a picture is taken on such a QR code, the picture can be seriously distorted, and thus suffers a high probability of decoding failure. In this paper, a scheme called rectification by conic segmentation (CS) is proposed to tackle this kind of problems. Experiments show that the proposed scheme is effective, in the sense that many failed trials at QR decoding become successful after the rectification.