Binary Image Watermarking in Ridgelet Domain
Binary images have only two distinct pixel color values so the capability of data hiding is very limited. To improve the robustness of watermarking algorithm, we proposed a novel ridgelet based watermarking for binary images. Ridgelet transform is efficient for representing images with line singularities. So, binary host image is partitioned into several non-overlapping blocks to make edges in each block similar to straight edges. Ridgelet transform is applied to each single block. To embed the watermark bits, directions with highest variance are selected in ridgelet coefficients matrix. To extract the watermark logo, detector response is computed for several sample watermarks and the maximum value is chosen as the extracted watermark. The proposed method has great robustness against different kinds of attacks.
Binary Image Watermarking Finite Ridgelet Transform Blind Extraction
Marzieh Amini Hamidreza Sadreazami
Department of Electrical and Computer Engineering Semnan University Semnan, Iran Department of Electrical and Computer Engineering Shahid Beheshti University Tehran, Iran
国际会议
2010 IEEE 10th International Conference on Signal Processing(第十届信号处理国际会议 ICSP 2010)
北京
英文
1813-1816
2010-08-24（万方平台首次上网日期，不代表论文的发表时间）