Watermarked Compressive Sensing Measurements Reconstructed by the Greedy Algorithms

Abstract

This paper proposes a CS based watermarking algorithm. In this algorithm, the watermarking image embeds into the compressive measurement vectors by the encoder. The measurement vectors are sparse in suitable basis. The resulting watermarked measurements recover by using both the OMP and the OMP-PKS reconstruction algorithms. Then the decoder uses to extract the watermarking image. In an experimental study, results obtained by the OMP and the OMP-PKS algorithms are compared. The results show that the OMP-PKS algorithm achieves performance superior to that of the OMP reconstruction algorithm. Index Terms—Watermarking image, CS, OMP and OMP-PKS algorithms.