Showing papers by "Ioannis Pitas published in 1998"

Robust audio watermarking in the time domain

Abstract: The audio watermarking method presented below offers copyright protection to an audio signal by modifying its temporal characteristics. The amount of modification embedded is limited by the necessity that the output signal must not be perceptually different from the original one. The watermarking method presented here does not require the original signal for watermark detection. The watermark key is simply a seed known only by the copyright owner. This seed creates the watermark signal to be embedded. Watermark embedding depends also on the audio signal amplitude in a way that minimizes the audibility of the watermark signal. The embedded watermark is robust to MPEG audio coding, filtering, resampling and requantization.

A novel method for automatic face segmentation, facial feature extraction and tracking

Abstract: The present paper describes a novel method for the segmentation of faces, extraction of facial features and tracking of the face contour and features over time. Robust segmentation of faces out of complex scenes is done based on color and shape information. Additionally, face candidates are verified by searching for facial features in the interior of the face. As interesting facial features we employ eyebrows, eyes, nostrils, mouth and chin. We consider incomplete feature constellations as well. If a face and its features are detected once reliably, we track the face contour and the features over time. Face contour tracking is done by using deformable models like snakes. Facial feature tracking is performed by block matching. The success of our approach was verified by evaluating 38 different color image sequences, containing features as beard, glasses and changing facial expressions.

Optical flow estimation and moving object segmentation based on median radial basis function network

Abstract: Various approaches have been proposed for simultaneous optical flow estimation and segmentation in image sequences. In this study, the moving scene is decomposed into different regions with respect to their motion, by means of a pattern recognition scheme. The inputs of the proposed scheme are the feature vectors representing still image and motion information. Each class corresponds to a moving object. The classifier employed is the median radial basis function (MRBF) neural network. An error criterion function derived from the probability estimation theory and expressed as a function of the moving scene model is used as the cost function. Each basis function is activated by a certain image region. Marginal median and median of the absolute deviations from the median (MAD) estimators are employed for estimating the basis function parameters. The image regions associated with the basis functions are merged by the output units in order to identify moving objects.

Digital watermarking: An overview

Abstract: In this paper we describe a general framework for image copyright protection through digital watermarking In particular we present the main features of an efficient watermarking scheme, discuss robustness issues and describe the three main stages of a watermarking algorithm namely watermark generation, embedding and detection

Facial feature extraction in frontal views using biometric analogies

Abstract: Face detection and facial feature extraction are considered to be key requirements in many applications, such as access control systems, model-based video coding, content-based video browsing and retrieval Thus, accurate face localization and facial feature extraction are most desirable A face detection and facial feature extraction in frontal views algorithm is described in this paper The algorithm is based on principles described in [1] but extends the work by considering: (a) the mirror-symmetry of the face in the vertical direction and (b) facial biometrie analogies depending on the size of the face estimated by the face localization method Further improvements have been added to the face localization method to enhance its performance The proposed algorithm has been applied to frontal views extracted from the European ACTS M2VTS database with very good results

A method for watermark casting on digital image

Abstract: Watermark casting on digital images is an important problem since it affects many aspects of the information market. We propose a method for casting digital watermarks on images, and we analyze its effectiveness. The satisfaction of some basic demands in this area is examined, and a method for producing digital watermarks is proposed. Moreover, issues like immunity to subsampling and image-dependent watermarks are examined, and simulation results are provided for the verification of the above-mentioned topics.

Chaotic watermarks for embedding in the spatial digital image domain

Abstract: Binary (bi-valued) digital signals, which are suitable watermarks for digital images are presented. They are generated by using chaotic dynamical systems that provide sufficient watermark complexity and controlled lowpass characteristics. Watermark detection is performed without resorting to the original image and its reliability is studied. Efficient robustness under lossy compression, lowpass filtering and other image processing can be achieved. Possible watermark detection after geometrical transformations is discussed.

Nonlinear processing and analysis of angular signals

Abstract: Physical quantities referring to angles, like vector direction, color hue, etc., exhibit an inherently periodic nature. Due to this periodicity, digital filters and edge operators proposed for data on the line cannot be applied on such data. We introduce filters for angular signals (circular mean, circular median, circular a-trimmed mean, circular modified trimmed mean). Particular emphasis is given to the circular median filter, for which some interesting properties are derived. We also use estimators of circular dispersion to introduce edge detectors for angular signals. Three variations for the extension of quasirange to circular data are proposed, and expressions for their output PDF are derived. These "circular" quasiranges have good and user-controlled properties as edge detectors in noisy angular signals. The performance of the proposed edge operators is evaluated on angular edges, using certain quantitative criteria. Finally, a series of experiments featuring one-dimensional (1-D) angular signals and hue images is used to illustrate the operation of the new filters and edge detectors.

Digital restoration of painting cracks

Abstract: In this paper we develop a method for the restoration of cracks on a painting. First, we detect the local minima (they can be either cracks or painting brush strokes), by using a morphological high-pass operator, called top-hat transformation. The crack filling procedure must be applied only on the cracks and not on these dark brush strokes, which are also detected. In order to separate these brush strokes from cracks, we use the Hue and Saturation information in the HSV or HSI color space. The separation is obtained by classification through the implementation of the MRBF neural network. Alternatively, a semi-automatic method is described for this separation. The primitive geometric shape-matching property of the morphological opening can be used to separate brush strokes, which have a specific shape. Finally, we propose two crack filling methods, one which is based on order statistics and another one using anisotropic diffusion. The results on painting crack restoration were very good.

Image watermarking using block site selection and DCT domain constraints.

Abstract: In this paper we propose an image watermarking algorithm based on constraints in the Discrete Cosine Transform (DCT) domain. An image watermarking algorithm has two stages: signature casting (embedding) and signature detection. In the first stage it embeds an identifying label in the image. This is recognized in the second stage. The proposed algorithm has two processing steps. In the first step certain pixel blocks are selected using a set of parameters while in the second step a DCT coefficient constraint is embedded in the selected blocks. Two different constraint rules are suggested for the parametric modification of the DCT frequency coefficients. The first one embeds a linear constraint among certain selected DCT coefficients and the second defines circular detection regions according to the given parameters. The watermarks cast by the proposed algorithm are resistant to JPEG compression and filtering.

Facial Feature Extraction and Determination of Pose

Abstract: The present paper describes a set of methods for the extraction of facial features as well as for the determination of the gaze direction. The ultimate goal of the approach followed is to define a sufficient set of feature distances so that a unique description of the structure of a face is produced. Eyebrows, eyes, nostrils, mouth, cheeks and chin are considered as interesting features. The candidates for eyes, nostrils and mouth are determined by searching for minima and maxima in the x- and y- projections of the greylevel relief. The candidates for cheek borders and chin are determined by performing an adaptive Hough transform on a relevant subimage defined according to the position of an ellipse containing the main face region of the image. A technique based on dynamic programming is applied that exploits this ellipse in order to acquire a more accurate model of the face. The candidates for eyebrows are determined by adapting a proper greylevel mask to an area defined by the eye position. Finally, the orientation of face is determined using the symmetric properties of certain facial features. The algorithms presented were tested on the M2VTS multimodal face database.

Analog implementation of fast min/max filtering

Abstract: An analog implementation of running min/max filters based on current-mode techniques is presented. Switched-current delay cells and current/voltage two inputs min/max selectors are used either for current or voltage inputs respectively. The voltage two input min/max circuit is designed using current conveyors and a modified structure of this is used to implement the running min/max filter for window size n=8. Simulation results demonstrate the feasibility of the proposed implementation, which can be extended to a higher window size.

Frontal face authentication using variants of dynamic link matching based on mathematical morphology

Abstract: Two variants of dynamic link matching based on mathematical morphology are developed and tested for frontal face authentication, namely, the morphological dynamic link architecture and the morphological signal decomposition-dynamic link architecture. Local coefficients which weigh the contribution of each node in elastic graph matching according to its discriminatory power are derived. The performance of the proposed algorithms is evaluated in terms of their receiver operating characteristic and the equal error rate (EER) achieved in the M2VTS database. The comparison with other frontal face authentication algorithms developed within M2VTS project indicates that morphological dynamic link architecture with discriminatory power coefficients is ranked as the best algorithm in terms of the EER.

Variants of dynamic link architecture based on mathematical morphology for frontal face authentication

Abstract: Two novel variants of dynamic link architecture that are based on mathematical morphology and incorporate coefficients which weigh the contribution of each node in elastic graph matching according to its discriminatory power are developed. They are the so called Morphological Dynamic Link Architecture and the Morphological Signal Decomposition-Dynamic Lint Architecture. The proposed variants are tested for face authentication in a cooperative scenario where the candidates claim an identity to be checked. Their performance is evaluated in terms of their receiver operating characteristic and the equal error rate achieved in M2VTS database. An equal error rate in the range 3.7-6.8% is reported.

Morphological techniques in the iterative closest point algorithm

Abstract: This paper describes a method for the accurate and computationally efficient registration of 3-D shapes. The method is based on the iterative closest point (ICP) algorithm and improves it by dramatically decreasing the computational cost of the algorithm's most inefficient step, namely the implementation of the closest point operator. The decrease is achieved with the help of a 3-D volume containing the points to be registered. Prior to the implementation of the ICP algorithm, the Voronoi diagram of the "model" points is constructed in the volume, by means of the morphological Voronoi tesselation method with respect to the Euclidean distance metric. The use of the tesselated volume renders the calculation of the closest point operator extremely fast and speeds up the ICP algorithm tremendously.

Adaptive order statistic filters for the removal of noise from corrupted images

Abstract: Adaptive order statistic filters for noise smoothing in digital images are presented. Two classes of adaptive filters are studied, namely, the least mean squares (LMS)-based adaptive order statistic filters and the signal-adaptive filters. The filter structures in the first class require a noise-free image to be used as a reference image, whereas those in the second class do not require a reference image. Two filter structures from the former class are examined: the adaptive locationinvariant L-filter and the adaptive Ll -filter. A novel signal-adaptive filter, namely, the morphological signal-adaptive median (MSAM) filter is proposed in the second class. It employs an anisotropic window adaptation procedure based on mathematical morphology operations. The noisesmoothing capabilities and the computational complexity of the LMSbased adaptive order statistic filters studied serves as a baseline in the assessment of the properties of the proposed MSAM filter. Quantitative criteria (e.g., the SNR, the peak SNR, the mean absolute error, and the mean squared error) as well as qualitative criteria (e.g., the perceived visual quality of the processed images) are employed to assess the performance of the filters in various corruption cases by different noise models.

Speaker dependent video indexing based on audio-visual interaction

Abstract: A content-based video indexing method is presented that aims at temporally indexing a video sequence according to the actual speaker. This is achieved by the integration of audio and visual information. Audio analysis leads to the extraction of a speaker identity label versus time diagram. Visual analysis includes scene cut detection, face shot determination, mouth region extraction and tracking and finally talking face shot determination. Results from both sources are combined to improve speaker dependent video indexing. Such a task enables flexible video retrieval or browsing in cases where queries according to speaker identities are imposed. Speaker recognition errors are reduced to 2%.

Morphological signal adaptive median filter for still image and image sequence filtering

Abstract: A novel extension of the classical signal-adaptive median filter (SAM) is proposed in this paper, namely the morphological signal-adaptive median filter (MSAM). Three modifications are introduced in the SAM filter aiming at: (1) enhancing the SAM impulse detection mechanism so that it also detects randomly-valued impulses, (2) employing an anisotropic window adaptation based on binary morphological erosions/dilations with predefined structuring sets and (3) extending its design for its application to image sequences as well. Its performance has been tested for noise suppression in both still images and image sequences. Its robustness against a wide variety of noise distributions as well as its superiority to the classical SAM filter are proved judging from both objective (SNR, MAE) and subjective (perceived image quality) criteria.

Old painting digital color restoration

Abstract: Many old paintings suffer from the effects of certain physicochemical phenomena, that can seriously degrade their overall visual appearance.Digital image processing techniques can be utilized for the purpose of restoring the original appearance of a painting, with minimal physical interaction with the painting surface. In this paper, a number of methods are presented which can yield satisfactory results. Indeed, simulation results indicate that acceptable restoration performance may be attained, despite the small size of painting surface data utilized.

Face verification based on morphological shape decomposition

Abstract: Morphological shape decomposition is used to model a facial image region as a sum of components and to extract a feature vector at the nodes of a sparse grid overlaid over the facial area in dynamic link matching. The feature vector is comprised of the greylevel values at this node in the reconstructed images at several decomposition levels. This feature vector is subsequently employed in Dynamic Link Architecture to verify the identity of each person from a training set. The experimental results indicate that the proposed combination of morphological shape decomposition and dynamic link matching practically offers the same verification capability to the standard dynamic link matching with Gabor wavelets.

Fast computation of a class of running filters

Abstract: This paper focuses on the computation of a class of running filters defined as the n-ary extension of an associative, commutative, and idempotent binary operation T on an ordered sequence of operands. The well-known max/min filters are the prominent representatives of the class. For any arbitrary window filter of size n, the existence of a fast algorithm of complexity O(log/sub 2/ n) T operations is proven. A remarkable feature of the proof is its ability to generate a particular solution for every n. In addition to the theoretical results, practical implementation aspects regarding the flexibility of pipeline processors for fast computation of the one-dimensional (1-D) and two-dimensional (2-D) running filters are investigated.

Object segmentation in 3-D images based on alpha-trimmed mean radial basis function network

Abstract: This paper presents a new approach for 3-D object segmentation. Objects from a stack of images are represented as overlapping ellipsoids. Graylevel statistics and shape features are simultaneously employed for object modeling in an unsupervised approach. The extension of the Hough Transform in the 3-D space is used for finding the ellipsoid centers. Each ellipsoid is modeled by a Radial Basis Function (RBF) and the entire structure is represented by means of an RBF network. The proposed algorithm is applied for blood vessel segmentation from tooth pulp in a stack of microscopy images.

Nonlinear location and scale estimators of fuzzy numbers

Abstract: In this correspondence, the extension principle is used in order to fuzzify location and scale estimators when used on fuzzy numbers. First, fuzzy nonlinear means are defined as extensions of the corresponding crisp means. Fuzzy L location and scale estimators, which are based on fuzzy-order statistics, are defined as extensions of the crisp L location and scale estimators. The most widely used scale estimator, which is known as the sample standard deviation, is also extended to fuzzy numbers through the extension principle. Equivalent relations that can be used to calculate the fuzzy estimators by using crisp arithmetic are also given for each one of the proposed fuzzy estimators.

Object segmentation and modeling in volumetric images

Abstract: We propose a pattern classification based approach for simultaneous 3-D object modeling and segmentation in image volumes. The 3-D objects are described as a set of overlapping ellipsoids. The segmentation relies on the geometrical model and graylevel statistics. The extension of the Hough Transform algorithm in the 3-D space by employing the spherical coordinate system is used for ellipsoidal center estimation. The characteristic parameters of the ellipsoids and of the graylevel statistics are embedded in a Radial Basis Function (RBF) network and they are found by means of unsupervised training. We propose a new robust training algorithm for RBF networks based on α-Trimmed Mean statistics. The proposed algorithm is applied for tooth pulpal blood vessel segmentation in a stack of microscopy images.

Robust and adaptive techniques in self-organizing neural networks

Abstract: Robust and adaptive training algorithms aiming at enhancing the capabilities of self-organizing and Radial Basis Function (RBF) neural networks are reviewed in this paper. The following robust variants of Learning Vector Quantizer (LVQ) are described: the order statistics LVQ, the L 2 LVQ and the split-merge LVQ. Successful application of the marginal median LVQ that belongs to the class of order statistics LVQs in the self-organized selection of the centers in RBF neural networks is reported. Moreover, the use of the median absolute deviation in the estimation of the covariance matrix of the observations assigned to each hidden unit in RBF neural networks is proposed. Applications that prove the superiority of the proposed variants of LVQ and RBF neural networks in noisy color image segmentation, color-based image recognition, segmentation of ultrasonic images, motion-field smoothing and moving object segmentation are outlined.

Analog implementation of an order statistics filter

Abstract: An analog implementation of an order statistics filter based on current-mode techniques is presented The circuit is designed using switched-current delay lines and current maximum extractors These filters could be easily incorporated in smart sensors as well as in smart cameras SPICE simulation results demonstrate the feasibility of simple analog filters using current-mode techniques

Fuzzy vector median definition based on a fuzzy vector distance

Abstract: In this paper, the Fuzzy Vector Median is proposed, defined as an extension of Vector Median. It is based on a novel distance definition of multidimensional fuzzy numbers (fuzzy vectors), which satisfy the property of angle decomposition. The proposed distance of two fuzzy vectors depends on the classical distance of the fuzzy set centers and on the fuzziness that every fuzzy set holds. As a result the Fuzzy Vector Median of a set of fuzzy vectors is affected by the presence of fuzziness.