An image watermarking approach based on Set Partitioning in Hierarchical Trees(SPIHT)algorithm

Abstract

In recent times, digital watermarking has become one of the advanced technologies to protect images from illicit manipulations. A non-blind robust Discrete Wavelet Transform domain watermarking approach for images is introduced in this paper. Based on the parent-child relationship between wavelet coefficients, the Set Partitioning in Hierarchical Trees (SPIHT) algorithm is applied on the transformed host image to find out the significant coefficients for embedding binary patterned watermark image. Higher level of security is achieved by scrambling the watermark using Arnold transform. The selected significant wavelet coefficients are adapted using Noise Visibility Function (NVF) to achieve enhanced imperceptibility. The scheme is tested against various image processing operations including JPEG compression, noise addition, filtering, cropping, histogram equalization and contrast adjustment. The experimental results showed that the proposed algorithm is highly robust. Keywords-Arnold transform, digital image watermarking, discrete wavelet transform (DWT), noise visibility function (NVF), set partitioning in hierarchical trees (SPIRT).

DWT-based blind and robust watermarking using SPIHT algorithm with applications in tele-medicine

“…The comparative results are given only for the standard images like Pirate or Lena. Table 10 shows that the proposed method achieves almost similar result or better in some cases even though the scheme proposed in [37] is a non-blind scheme and our proposed scheme is a blind scheme. Comparison of results according to correlation coefficient (C) with the blind scheme proposed in [38] is listed in table 11.…”

“…The proposed algorithm is compared to a non-blind scheme presented in [37] and a blind scheme presented in [38]. The comparison results are summarized in tables 10 and 11.…”

Anti-Print and Scan Halftone Watermarking Algorithm Based on Edge Detection

“…In order to improve the invisibility of watermark, edge points which have complex texture will be selected as the embedding location of watermark. So a noise visibility function [5] was introduced. The noise visibility function (NVF) can reflect the shelter ability of texture of the local image, and it can measure each pixel's sensitivity to noise .The NVF range from zero to one.…”