A Video Watermarking Framework Resistant to Super Strong Cropping Attacks Based on NMF with Sparseness Constraints on Parts of the Basis Matrix

Firstly, the Non-negative Matrix Factorization with Sparseness Constraints on Parts of the Basis Matrix　(NMFSCPBM) method is proposed in this paper. Secondly, the encrypted watermark is embedded into the big coefficients of the basis matrix by NMFSCPBM. At the same time, the watermark embedding strength is adaptively adjusted by the video motion characteristics extracted by NMFSCPBM. Finally, when detecting the watermark, as long as the residual video contains the numbers of least remaining sub-blocks, the complete basis matrix can be completely recovered, and then the complete watermark can be extracted. The experimental results show that the performance of resisting the strong cropping attacks of this paper is improved greatly compared with other similar methods.