Image Watermarking Using All Phase Discrete Cosine Biorthogonal Transform in Selected Pixel Blocks

Abstract

In the past decade, a transform called all phase discrete cosine biorthogonal transform (APDCBT) appeared in the field of digital image processing. It is mainly used to solve the negativity found in discrete cosine transform (DCT), especially in a low bit rate. In this paper, the APDCBT is employed for watermark insertion based on selected regions of an image. The insertion is depending on the homogeneity of each part of the image. To determine the extent of homogeneity, there are two criteria: Mean and standard deviation are applied for the intensity of the image. Medium frequency bands of APDCBT in image pixel blocks are used to hold the embedded watermark. Then, the transform is inversed to obtain the resultant watermarked image. The robustness of APDCBT against watermark removal attacks is tested and the experimental results showed the superiority of the APDCBT over traditional DCT in the watermark embedding system. More specifically, when the LSB reset attack applied. For instance, when the contrast adjustment attack applied, the average of normalized cross-correlation (NCC) values of extracted watermark images with the original watermark, it was 0.992 for the proposed method. This is a promised result, if it is compared with the NCC of the DCT method, which was 0.423. The proposed method can be used for copyright protection purpose.

An Image Watermarking Technique Proposed Based on Discrete Cosine Transformation and Pseudo-Random Generator

“…According to (Khalifa and Saeed, 2020) in [13], MSE stands for the square of the variations in pixel values of the two images' respective pixels. Equation 6.1 gives the MSE of a NxM size image.…”

“…The discrete cosine transformation (DCT) is considered to have high energy concentration properties in the field of discrete cosine transformation (DCT) research. As a consequence, it's been applied to a variety of tasks, including image compression, watermarking, and pattern recognition [13].The signal is processed using discrete cosine transformation (DCT). The signal is converted from the spatial to the frequency domain.…”

A Comprehensive Approach to Image Protection in Digital Environments

“…In addition, the unauthorized manipulation and alteration of images is another growing problem in the digital environment. Image editing tools and software are widely available, making it easy to modify visual content fraudulently [2]. This can result in the spread of false information, misleading images, or even impersonation.…”