Abstract. We present in this paper a system for passengers counting in buses based on stereovision. The objective of this work is to provide a precise counting system well adapted to buses environment. The processing chain corresponding to this counting system involves several blocks dedicated to the detection, segmentation, tracking and counting. From original stereoscopic images, the system operates primarily on the information contained in disparity maps previously calculated with a novel algorithm. We show that one can obtain a counting accuracy of 99% on a large data set including specific scenarios played in laboratory and on some video sequences shot in a bus during exploitation period.

Global Navigation Satellite Systems (GNSS) are widely spread (with Global Positioning System -GPS) in intelligent transport systems and offer a low cost, continuous and global solution for positioning. Unfortunately, urban users are often the most demanding of accurate localization but receive a degraded service because of signal propagation conditions. Several mitigation solutions can be developed. We propose, within CAPLOC project (2010)(2011)(2012)(2013) to deal with inaccuracy by associating image processing techniques and signal propagation knowledge. In this paper, we focus on the contribution of image processing in more accurate position estimation. Thus, we use a laboratory vehicle, which is equipped with a fisheye camera and two GNSS receivers. The camera is located on the roof and oriented upwards to capture images of the sky. The GNSS receivers are used to obtain raw data, the position of the vehicle and the reference trajectory. The proposed approach consists in determining where satellites are located in the fisheye image, and then excluding those located in non-sky regions when calculating the position. For that, the strategy is based on an image simplification step coupled with a pixels classification. The image-based exclusion procedure is compared with the classical one based on the application of a threshold on carrier-to-noise (CN0) ratio to separate LOS and NLOS signals. Accuracy improvement is satisfying with the CN0-based method and show an improvement from 13m to 4,5m. Image-based detection shows mixed improvements but promising: good in a static area and too harsh in another configuration of the scenario.

The problem described in this paper consists in re-identifying moving people in different sites which are completely covered with non-overlapping cameras. Our proposed framework relies on the spectral classification of the appearance-based signatures extracted from the detected person in each sequence. We first propose a new feature called "color-position" histogram combined with several illumination invariant methods in order to characterize the silhouettes in static images. Then, we develop an algorithm based on spectral analysis and Support Vector Machines (SVM) for the re-identification of people. The performance of our system is evaluated on real data sets collected on INRETS premises. The experimental results show that our approach provides promising results for security applications.

Abstract. In this paper, a new learning method is proposed to build Support Vector Machines (SVM) Binary Decision Function (BDF) of reduced complexity, efficient generalization and using an adapted hybrid color space. The aim is to build a fast and efficient SVM classifier of pixels. The Vector Quantization (VQ) is used in our learning method to simplify the training set. This simplification step maps pixels of the training set to representative prototypes. A criterion is defined to evaluate the Decision Function Quality (DFQ) which blends recognition rate and complexity of a BDF. A model selection based on the selection of the simplification level, of a hybrid color space and of SVM hyperparameters is performed to optimize this DFQ. Search space for selecting the best model being huge. Our learning method uses Tabu Search (TS) metaheuritics to find a good sub-optimal model on tractable times. Experimental results show the efficiency of the method.