Performance analysis of a 2-D-multipulse amplitude modulation scheme for data hiding and watermarking of still images

Abstract

In this paper a watermarking scheme for copyright protection of still images is modeled and analyzed. In this scheme a signal following a key-dependent two-dimensional multipulse modulation is added to the image for ownership enforcement purposes. The main contribution of this paper is the introduction of an analytical point of view to the estimation of performance measurements. Two topics are covered in the analysis: the ownership verification process, also called watermark detection test, and the data-hiding process. In the first case, bounds and approximations to the receiver operating characteristic are derived. These results can be used to determine the threshold associated to a required probability of false alarm and the corresponding probability of detection. The data-hiding process is modeled as a communications system and approximations for the bit error rate are derived. Finally, analytical expressions are contrasted with experimental results.

Reversible Watermark Using an Accurate Predictor and Sorter Based on Payload Balancing

“…For instance, if the histogram bins (T n2 , T n1 ) = (-4, -3) and (T p1 , T p2 ) = (3,4) are selected during the encoding process, then pixel values corresponding to bin -2, -1, 0, 1, and 2 do not have to change their values and only bins outside -4 and 4 are shifted, creating distortion after watermarking.…”

“…In the case of a still image, one of the simplest watermarking methods is to replace the least significant bit plane of pixel values with a payload. Techniques including spread spectrum [1], quantization [2], and amplitude modulation [3] are designed to achieve various purposes such as robustness and efficiency, but they are all lossy. This lossy property of watermarking is problematic to some experts, such as graphic designers and physicians, who rely upon high image quality in their professions.…”

Copy protection for DVD video

“…Our estimates of the required false positive rate are about one in 10 11 or 10 12 distinct frames. Recent models for predicting the false positive rate can be found in [10] and [11].…”

Introduction

“…The F M frequency band however, has been chosen as the embedding region as well as to provide such additional resistance to JPEG lossy compression techniques, while avoiding significance of being perceptible to the human eye due to the sensitivity of the human eye for the meaning of detecting the existence of noises within the F L region [12]. To o c co om mm me en nc ce e, we first need to go through the initial method that is the encoding and break the image into 8 x 8 blocks then apply the DCT function and computing all the image blocks separately and converting them into frequency components based on the frequency domain and having the whole details and components of the cover image and the watermark converted from the spatial domain into the frequency domain.…”