Thierry Pun

Rotation, scale and translation invariant spread spectrum digital image watermarking

Abstract A digital watermark is an invisible mark embedded in a digital image which may be used for a number of different purposes including image captioning and copyright protection. This paper describes how a combination of spread spectrum encoding of the embedded message and transform-based invariants can be used for digital image watermarking. In particular, it is described how a Fourier–Mellin-based approach can be used to construct watermarks which are designed to be unaffected by any combination of rotation and scale transformations. In addition, a novel method of CDMA spread spectrum encoding is introduced which allows one to embed watermark messages of arbitrary length and which need only a secret key for decoding. The paper also describes the usefulness of Reed Solomon error-correcting codes in this scheme.

Robust template matching for affine resistant image watermarks

Digital watermarks have been proposed as a method for discouraging illicit copying and distribution of copyrighted material. This paper describes a method for the secure and robust copyright protection of digital images. We present an approach for embedding a digital watermark into an image using the Fourier transform. To this watermark is added a template in the Fourier transform domain to render the method robust against general linear transformations. We detail a new algorithm based on polar maps for the accurate and efficient recovery of the template in an image which has undergone a general affine transformation. We also present results which demonstrate the robustness of the method against some common image processing operations such as compression, rotation, scaling, and aspect ratio changes.

Performance evaluation in content-based image retrieval: overview and proposals

Abstract Evaluation of retrieval performance is a crucial problem in content-based image retrieval (CBIR). Many different methods for measuring the performance of a system have been created and used by researchers. This article discusses the advantages and shortcomings of the performance measures currently used. Problems such as defining a common image database for performance comparisons and a means of getting relevance judgments (or ground truth) for queries are explained. The relationship between CBIR and information retrieval (IR) is made clear, since IR researchers have decades of experience with the evaluation problem. Many of their solutions can be used for CBIR, despite the differences between the fields. Several methods used in text retrieval are explained. Proposals for performance measures and means of developing a standard test suite for CBIR, similar to that used in IR at the annual Text REtrieval Conference (TREC), are presented.

A new method for grey-level picture thresholding using the entropy of the histogram

This paper describes an automatic threshold selection method for picture segmentation, using the entropy of the grey level histogram. It is shown that, by an a priori maximation of an entropy determined a posteriori, a picture can successfully be thresholded into a two-level image. Several experimental results are presented to show the validity of the method. An extension to multithresholding and to multidimensional histogram processing is also discussed.

Rotation, scale and translation invariant digital image watermarking

A digital watermark is an invisible mark embedded in a digital image which may be used for copyright protection. This paper proposes that Fourier-Mellin transform-based invariants can be used for digital image watermarking. The embedded marks may be designed to be unaffected by any combination of rotation, scale and translation transformations. The original image is not required for extracting the embedded mark.

Entropic thresholding, a new approach

This paper describes an automatic threshold selection method for picture segmentation. The basic concept is the definition of an anisotropy coefficient, which is related to the asymmetry of the grey-level histogram. Its use permits the derivation of the entropic thresholds, which has been successfully applied to images having various kinds of histograms. Several experimental results are presented. An extension to multithresholding is also suggested.

A Stochastic Approach to Content Adaptive Digital Image Watermarking

This paper presents a new stochastic approach which can be applied with different watermark techniques. The approach is based on the computation of a Noise Visibility Function (NVF) that characterizes the local image properties, identifying textured and edge regions where the mark should be more strongly embedded. We present precise formulas for the NVF which enable a fast computation during the watermark encoding and decoding process. In order to determine the optimal NVF, we first consider the watermark as noise. Using a classical MAP image denoising approach, we show how to estimate the ”noise”. This leads to a general formulation for a texture masking function, that allows us to determine the optimal watermark locations and strength for the watermark embedding stage. We examine two such NVFs, based on either a non-stationary Gaussian model of the image, or a stationary Generalized Gaussian model. We show that the problem of the watermark estimation is equivalent to image denoising and derive content adaptive criteria. Results show that watermark visibility is noticeably decreased, while at the same time enhancing the energy of the watermark.

A Multimodal Database for Affect Recognition and Implicit Tagging

MAHNOB-HCI is a multimodal database recorded in response to affective stimuli with the goal of emotion recognition and implicit tagging research. A multimodal setup was arranged for synchronized recording of face videos, audio signals, eye gaze data, and peripheral/central nervous system physiological signals. Twenty-seven participants from both genders and different cultural backgrounds participated in two experiments. In the first experiment, they watched 20 emotional videos and self-reported their felt emotions using arousal, valence, dominance, and predictability as well as emotional keywords. In the second experiment, short videos and images were shown once without any tag and then with correct or incorrect tags. Agreement or disagreement with the displayed tags was assessed by the participants. The recorded videos and bodily responses were segmented and stored in a database. The database is made available to the academic community via a web-based system. The collected data were analyzed and single modality and modality fusion results for both emotion recognition and implicit tagging experiments are reported. These results show the potential uses of the recorded modalities and the significance of the emotion elicitation protocol.

Emotion assessment: arousal evaluation using EEG's and peripheral physiological signals

The arousal dimension of human emotions is assessed from two different physiological sources: peripheral signals and electroencephalographic (EEG) signals from the brain. A complete acquisition protocol is presented to build a physiological emotional database for real participants. Arousal assessment is then formulated as a classification problem, with classes corresponding to 2 or 3 degrees of arousal. The performance of 2 classifiers has been evaluated, on peripheral signals, on EEGs, and on both. Results confirm the possibility of using EEGs to assess the arousal component of emotion, and the interest of multimodal fusion between EEGs and peripheral physiological signals.

Attack modelling: towards a second generation watermarking benchmark

Abstract Digital image watermarking techniques for copyright protection have become increasingly robust. The best algorithms perform well against the now standard benchmark tests included in the Stirmark package. However the stirmark tests are limited since in general they do not properly model the watermarking process and consequently are limited in their potential to removing the best watermarks. Here we propose a stochastic formulation of watermarking attacks using an estimation-based concept. The proposed attacks consist of two main stages: (a) watermark or cover data estimation; (b) modification of stego data aiming at disrupting the watermark detection and of resolving copyrights, taking into account the statistics of the embedded watermark and exploiting features of the human visual system. In the second part of the paper we propose a “second generation benchmark”. We follow the model of the Stirmark benchmark and propose the 6 following categories of tests: denoising attacks and wavelet compression, watermark copy attack, synchronization removal, denoising/compression followed by perceptual remodulation, denoising and random bending. Our results indicate that even though some algorithms perform well against the Stirmark benchmark, almost all algorithms perform poorly against our benchmark. This indicates that much work remains to be done before claims about “robust” watermarks can be made. We also propose a new method of evaluating image quality based on the Watson metric which overcomes the limitations of the PSNR.