Sylvie Lelandais

Matching segments in stereoscopic vision

We have shown that its possible to realize a stereoscopic sensor with poor cameras. We developed image processing that is robust and allows us to quickly obtain results for the matching algorithm. We computed an important number of features on each segment, and with these features, we built 16-component vector used in the classification step. After an exhaustive study, we decided to combine two methods, Bayesian and neural, to construct an efficient classifier. The tests for indoor images had better than 90% good matching. With segment couples, it is possible to compute the 3D coordinates of the objects. Therefore, the mobile robot is able to localize and move about in the environment.

Global interior robot localisation by a colour content image retrieval system

We propose a new global localisation approach to determine a coarse position of a mobile robot in structured indoor space using colour-based image retrieval techniques. We use an original method of colour quantisation based on the bakers transformation to extract a two-dimensional colour pallet combining as well space and vicinity-related information as colourimetric aspect of the original image. We conceive several retrieving approaches bringing to a specific similarity measure integrating the space organisation of colours in the pallet. The bakers transformation provides a quantisation of the image into a space where colours that are nearby in the original space are also nearby in the output space, thereby providing dimensionality reduction and invariance to minor changes in the image. Whereas the distance provides for partial invariance to translation, sight point small changes, and scale factor. In addition to this study, we developed a hierarchical search module based on the logic classification of images following rooms. This hierarchical module reduces the searching indoor space and ensures an improvement of our system performances. Results are then compared with those brought by colour histograms provided with several similarity measures. In this paper, we focus on colour-based features to describe indoor images. A finalised system must obviously integrate other type of signature like shape and texture.

Face localization by neural networks trained with Zernike moments and Eigenfaces feature vectors. A comparison

Face localization using neural network is presented in this communication. Neural network was trained with two different kinds of feature parameters vectors; Zernike moments and eigenfaces. In each case, coordinate vectors of pixels surrounding faces in images were used as target vectors on the supervised training procedure. Thus, trained neural network provides on its output layer a coordinates vector (rho,thetas) representing pixels surrounding the face contained in treated image. This way to proceed gives accurate faces contours which are well adapted to their shapes. Performances obtained for the two kinds of training feature parameters were recorded using a quantitative measurement criterion according to experiments carried out on the XM2VTS database.

A Pruning Approach Improving Face Identification Systems

We propose, in this paper, a new biometric identification approach which aims to improve recognition performances in identification systems. We aim to split the identity database into well separated partitions in order to simplify the identification task. In this paper we develop a face identification system and we use the reference algorithms of Eigenfaces and Fisherfaces in order to extract different features describing each identity. These features, which describe faces, are generally optimized to establish the required identity in a classical identification process. In this work, we develop a novel criterion to extract features used to partition the identity database. We develop database partitioning with clustering methods which split the gallery by bringing together identities which have similar features and separating dissimilar features in different bins. Pruning the most dissimilar bins from the query identity features allows us to improve the identification performances. We report results from the XM2VTS database.

Implementation of the watershed method in the HSI color space for the face extraction

The color segmentation is an important preprocessing step in the face detection methods. In this paper we present effective face color segmentation in the HSI (Hue, saturation and intensity) space. We propose to modify the merging algorithm of the catchment basins obtained by the watershed segmentation method by adding a criterion based on the relevance of the Hue component. The obtained experimental results are presented to demonstrate the effectiveness of our approach.

2D Face Recognition in the IV2 Evaluation Campaign

In this paper, the first evaluation campaign on 2D-face images using the multimodal IV2 database is presented. The five appearance-based algorithms in competition are evaluated on four experimental protocols, including experiments with challenging illumination and pose variabilities. The results confirm the advantages of the Linear Discriminant Analysis (LDA) and the importance of the training set for the Principal Component Analysis (PCA) based approaches. The experiments show the robustness of the Gabor based approach combined with LDA, in order to cope with challenging face recognition conditions. This evaluation shows the interest and the richness of the IV2 multimodal database.

Adaptive filtering and hypothesis testing: Application to cancerous cells detection

We propose a new method to detect cells in microscopic imagery, the problem under study being the analysis of cancerous cells experiencing metastasis i.e. cells susceptible to migration. This work would help medical researchers to study the evolution of a cancer. The peculiar nature of the images due to the acquisition protocol causes some difficulties. These are resolved through tailored preprocessing involving correction of uneven illumination and enhancement of cellular information. Detection and counting of cells are performed by our proposed filtering that provokes peaks in its convolution space wherever cells are present. We compare our counting results with those provided by human experts and with a Hough transform developed for similar purposes. The validity of the cell segmentation from the peaks is then established by a statistical test of the closeness of the segmented cell to a cell model.

A New Clustering Approach for Face Identification

We propose in this paper a search approach which aim to improve identification in biometric databases. We work with face images and we develop appearance-based Eigenfaces and Fisherfaces methods to generate holistic and discriminant features and attributes. These features, which describe faces, are often used to establish the required identity in a classical identification process. In this work we introduce a clustering process upstream the identification process which divides faces into partitions according to their features similarities. Indeed, we aim to split biometric databases into partitions in order to simplify the recognition task within these databases. This paper describes the proposed clustering approach, the Eigenfaces and Fisherfaces representation methods and preliminary clustering results on the XM2VTS data corpus.

Improvements of Object Grabbing Method by Using Color Images and Neural Networks Classification

The works presented in this paper allow the improvement of an object grabbing method developed for the ARPH project (Robotic Assistance for Disabled people). The basic tool of this project is a mobile robot with an embedded MANUS arm. The main aim of the method developed here is to make a manipulator arm equipped with a gripper and two kinds of sensors (camera and sonar) able to seizure and handle several objects in the human environment. Ultrasonic echo is recovered by a neural network classifier in order to recognize the object shape. At the same time, a little wireless HF camera takes a color image of the target, which allows to compute the coordinates of the object in the image then in the camera frame. The combination of these various information about the object with the camera geometrical model and the visual servoing principle allows to adopt an iterative strategy to approach the object under good conditions and to seize it at the end. Two kinds of object are used (spherical and cylindrical) with several colors and diameters

Poor man vote with M-ary classifiers: application to iris recognition

Achieving good performance in biometrics requires matching the capacity of the classifier or a set of classifiers to the size of the available training set. A classifier with too many adjustable parameters (large capacity) is likely to learn the training set without difficulty but be unable to generalize properly to new patterns. If the capacity is too small, the training set might not be learned without appreciable error. There is thus advantage to control the capacity through a variety of methods involving not only the structure of the classifiers themselves, but also the property of the input space. Ths paper proposes an original non parametric method to combine optimaly multiple classifier responses. Highly favorable results have been obtained using the above method.