Digital watermarking by geometric warping

Abstract

A new method is presented f o r watermarking digital images and video streams. This technique embeds information in a n image by changing the geometric features of the image. The method is essentially different from existing techniques in which a pseudonoise pattern is added t o the image. T h e new method has the advantage that the detection is computationally faster than existing methods. Furthermore, it is easier to detect the watermark an images that have been rotated, scaled, or distorted by a geometric transformation. The watermark is formed by a predefined dense pixel pattern, such as a collection of lines. Salient points in a n image are warped into the vicinity of the line pattern in such a way that the changes t o the image are imperceptible.

Region-based image watermarking

“…Thus, the detector expresses the difference of two sample means. The mean value and variance of the detector output are (9) In the case that the watermark is embedded on the entire embedding region, the detector output is assumed to follow a normal distribution. If the correct watermark is embedded on the image, then the mean value is and the variance is , where is the variance of the initial image and is the variance of the watermark, as is adapted for the certain region.…”

“…An approach that is somehow related to our method was proposed in [9], where the core of the technique was to find image points that could be warped according to their distance to specific line segments that form the watermark. However, the cost of producing a theoretically infinite set of different line patterns in the detection stage, together with the questionable prominence of the selected points to be warped, render this technique unsuitable.…”

“…However, the cost of producing a theoretically infinite set of different line patterns in the detection stage, together with the questionable prominence of the selected points to be warped, render this technique unsuitable. The main weak aspect of the technique [9] is that the chosen pattern is completely random and does not take any spatial properties into consideration. Another method that tries to face the problem of geometric attacks is presented in [10], where a template consisting of a fixed number of points is inserted in the DFT domain of the image.…”

Robust digital watermarking in videos based on geometric transformations

“…In 1998, Maes and van Overveld already proposed to modify geometric features instead of color components [2]. However, they applied local changes by geometric warping, which has several disadvantages: First of all, the modifications are much more likely to be perceived by viewers unless the amount of the modification is relatively small.…”

Digital watermarking: algorithms and applications

“…It has been established that for image data, the phase information is perceptually more important than the magnitude data. Other novel approaches for watermarking image data include fractal-based approaches [42], [43] and geometric feature based watermarking [44]. In [44], salient points in an image are found and warped according to a dense line pattern representing the watermark and generated randomly.…”