Coevolutionary design of a watermark embedding scheme and an extraction algorithm for detecting replicated two-dimensional barcodes

Two-dimensional barcodes (2D codes) are currently used to authenticate airplane boarding passes and online payments. However, little attention has been paid to the malicious replication or fabrication of 2D codes. In a previous study, the authors have proposed a semi-fragile watermark to distinguish an authentic 2D code from a replicated one and have attempted to design the watermarking scheme automatically using optimization. The optimization of a watermark extraction algorithm and watermarking scheme for stable watermark extractions from 2D codes displayed on various display types is the actual desire. However, it is difficult to formulate the two tasks, as a unified optimization problem and solve it with an optimization algorithm because the two problems have different structures and properties. Consequently, this study proposes a sparsely synchronized heterogeneous coevolutionary method for the simultaneous optimization of a watermark embedding scheme and extracting algorithm to detect the replication of 2D codes displayed on mobile phone screens. Experimental results have shown that the proposed method can design a watermark with desirable semi-fragileness that works well on different types of mobile phone displays.