A Robust SLIC Segmentation Based Zero Watermarking Scheme Using DWT And Partial Pivoting LU Decomposition

Abstract

This article presents a novel simple linear iterative clustering (SLIC) superpixels and partial pivoting lower triangular upper triangular (PPLU) factorization based zero-watermarking (ZWM) scheme is proposed. The novelty of this article is that the use of SLIC superpixels and PPLU decomposition for ZWM. The original image is subjected to SLIC segmentation and non-overlapping high entropy blocks are selected. On the selected blocks discrete wavelet transform (DWT) is applied and those blocks are undergone PPLU factorization to get three matrices $L, U$ , and $P$ . The product matrix $L\times U$ is used to construct zero-watermark. This work is validated with BOWS, and SIPI datasets, and shown robust against the attacks. The test results demonstrate that this work has better performance than the recent ZWM schemes.