A Fragile Associative Watermarking on 2D Barcode for Data Authentication

Two-dimensional (2D) barcode has improved the information encoded capacity, and it also has enriched the applications of barcode technique. Recently, there are researches dealing with watermark technique on 2D barcode to prevent it from counterfeited or prepensely tampered. The existent methods still have to limit the size of embedded watermark in a relatively small portion. Furthermore, it also needs to utilize original watermark or other auxiliary verification mechanism to achieve the barcode verification. In this paper, we propose a method called associative watermarking which is conducted by the concept of Association Rules (ARs) and the idea of Vector Quantization (VQ). Our method is a kind of blind watermarking, and it also can free the size limitation of an embedded watermark. Performing associative watermarking to 2D barcode can reduce the embedded information amount, and using VQ indexing scheme can easily recall the embedded watermark for the purpose of barcode data authentication. The experiment demonstrates that our method can significantly save the information hiding capacity of 2D barcode and detects a counterfeited or prepensely tampered 2D barcode data correctly.