Secure spread spectrum watermarking for multimedia

Abstract

This paper presents a secure (tamper-resistant) algorithm for watermarking images, and a methodology for digital watermarking that may be generalized to audio, video, and multimedia data. We advocate that a watermark should be constructed as an independent and identically distributed (i.i.d.) Gaussian random vector that is imperceptibly inserted in a spread-spectrum-like fashion into the perceptually most significant spectral components of the data. We argue that insertion of a watermark under this regime makes the watermark robust to signal processing operations (such as lossy compression, filtering, digital-analog and analog-digital conversion, requantization, etc.), and common geometric transformations (such as cropping, scaling, translation, and rotation) provided that the original image is available and that it can be successfully registered against the transformed watermarked image. In these cases, the watermark detector unambiguously identifies the owner. Further, the use of Gaussian noise, ensures strong resilience to multiple-document, or collusional, attacks. Experimental results are provided to support these claims, along with an exposition of pending open problems.

Print signatures for document authentication

“…The transferred image is then fixed permanently to the paper by fusing the toner using pressure and heat (5). The last step cleans off all excess toner and electrostatic charge from the photoreceptor to make it ready for the next cycle (6).…”

“…Such approaches include bar codes [19,20] and information hiding (notably digital watermarking) [6,22]. These methods add some machine readable information onto the document to serve as a digital signature.…”

Rotation, scale and translation invariant spread spectrum digital image watermarking

“…In many cases, image contrast may be distorted, for example by a scanner, in which case the e ects of change of contrast must be compensated for in some way. Cox et al [5,6] describe a method known as dynamic histogram warping [7] to carry this out. …”

“…Objective criteria for measuring the degree to which an image component is signi cant in watermarking have gradually evolved from being based purely on energy content [16,5,6] to statistical [20] and psychovisual [27,10] criteria. Digital watermarking is also fundamentally a problem in digital communications [16,25,5,6]. In parallel with the increasing sophistication in modelling and exploiting the properties of the human visual system, there has been a corresponding development in communication techniques.…”

“…Information has been embedded using the Discrete Cosine Transform (DCT) [16,34,5,6] Discrete Fourier Transform magnitude and phase [15], Wavelets [16], Linear Predictive Coding [13] and Fractals [9,22]. The key to making watermarks robust has been the recognition that in order for a watermark to be robust it must be embedded in the perceptually signi cant components of the image [16,5,6]. The term \perceptually signi cant" is somewhat subjective but it suggests that a good watermark is one which takes account of the behaviour of human visual system.…”