This paper deals with the low-level joint processing of color and motion for robust face analysis within a feature-based approach. To gain robustness and contrast under unsupervised viewing conditions, a nonlinear color transform relevant for hue segmentation is derived from a logarithmic model. A hierarchical segmentation scheme is based on Markov random field modeling, that combines hue and motion detection within a spatiotemporal neighborhood. Relevant face regions are segmented without parameter tuning. The accuracy of the label fields enables not only face detection and tracking but also geometrical measurements on facial feature edges, such as lips or eyes. Results are shown both on typical test sequences and on various sequences acquired from micro- or mobile cameras. The efficiency of the method makes it suitable for real-time applications aiming at audiovisual communication in unsupervised environments.

The advances in communication networks and web technologies, in conjunction with the improved connectivity of test and measurement devices make it possible to implement e-learning applications that encompass the whole learning process. In the field of electrical engineering, automation or mechatronics, it means not only lectures, tutorials, demos and simulations, but also practical labwork for training with real-world devices that are controlled remotely. To make e-labs attractive, they should be easily implemented and accessed on the web by a client. This keypoint raises technical issues that are recalled in this paper. Nonetheless pedagogical issues are equally important. The benefit of a remote lab must be evaluated and compared to simulation labs or hands-on. Here, to foster student motivation, a game-like scenario embedded in a learning management system is proposed.

Abstract-This paper describes the real-time implementation of a simple and robust motion detection algorithm based on Markov random field (MRF) modeling. MRF-based algorithms often require a significant amount of computations. The intrinsic parallel property of MRF modeling has led most of implementations toward parallel machines and neural networks, but none of these approaches offers an efficient solution for real-world (i.e., industrial) applications. Here, an alternative implementation for the problem at hand is presented yielding a complete, efficient and autonomous real-time system for motion detection. This system is based on a hybrid architecture, associating pipeline modules with one asynchronous module to perform the whole process, from video acquisition to moving object masks visualization. A board prototype is presented and a processing rate of 15 images/s is achieved, showing the validity of the approach.

An algorithm for speaker's lip contour extraction is presented in this paper. A color video sequence of speaker's face is acquired, under natural lighting conditions and without any particular make-up. First, a logarithmic color transform is performed from RGB to HI (hue, intensity) color space. A bayesian approach segments the mouth area using Markov random field modelling. Motion is combined with red hue lip information into a spatiotemporal neighbourhood. Simultaneously, a Region Of Interest and relevant boundaries points are automatically extracted. Next, an active contour using spatially varying coefficients is initialised with the results of the preprocessing stage. Finally, an accurate lip shape with inner and outer borders is obtained with good quality results in this challenging situation.

No abstract

No abstract

This paper deals with an entropic approach as unsupervised thresholding technique for image processing, in order to extract a relevant binary information from noisy data. It is dedicated to situations where a signal of relatively high energy is localized in the image whereas the noise is spread over the entire frame. The method is based on the computation of the entropy power of the information source, as defined by Shannon. The threshold used for binarization is proportional to the entropic deviation of the observation source. The performance of the approach is illustrated by two classical image preprocessing tasks, namely motion detection and edge detection. The evaluation set contains both synthetic data and real-world image sequences.Résumé Cet article propose une approche entropique comme technique de seuillage automatique en traitement d'images, dans le but d'extraire une information binaire pertinenteà partir de données bruitées. La méthode est applicable dans le cas d'un signal de forteénergie localisé dans le plan image, tandis que le bruit est majoritaire et diffus. La technique est basée sur la notion de puissance entropique d'une source d'observations, telle que définie par Shannon. Le seuil de binarisation est propotionnelà l'écart-type entropique de la source. L'efficacité de la méthode est illustrée par deux applications classiques du traitement d'images: la détection de mouvement par les changements temporels et la détection de contours par les maxima du gradient spatial. L'évaluation des performances est réaliséeà la fois sur des données suynthétiques et sur des images réelles.

Using a DC pulsed plasma for the polymer surface treatment allows the attainment of macroscopic modifications of the surface such as an important increase of the wettability. In the same time microscopic variations of the surface structure are mainly linked to low depth chemical modifications even if very weak roughness changes appear.As a consequence this technique presents two major interests. The first one is an economical interest because of the low power consumption compared to other techniques like radiofrequency or microwaves plasmas. The second one is the very significant treatment (macroscopic) realized in soft conditions without degradation of the polymer.These results composed of macroscopic and microscopic studies on polystyrene surfaces may allow to establish a macroscopic interpretation of the interaction between polymer and the DC pulsed plasma.