Print-Resilient Binary Image Watermarking Based on Grayscale Conversion and DCT Transformation

Current binary image watermarking methods are mainly spatial domain methods, and most of them cannot resist printing and scanning. Based on this consideration, this paper introduces a new print-resilient watermarking method tailored for binary images. For data embedding, by dividing the cover binary image into disjoint blocks and counting the black pixels in each block, it is first converted into a form similar to a grayscale image. Then, by applying DCT to the transformed image, two DCT coefficients are selected and modified to embed the watermark. Next, after modification, the transformed image is converted back to the spatial domain to identify the number of pixels that need to be changed in each binary image block. Finally, for the cover binary image, edge detection is employed to select and modify its pixels (white to black, or black to white) without reducing the image quality, effectively realizing the watermark embedding. Experimental results show that the proposed method can well preserve the visual quality while resist the print-scan attack.