Lionel Robinault

A cognitive and video-based approach for multinational License Plate Recognition

License Plate Recognition (LPR) is mainly regarded as a solved problem. However, robust solutions able to face real-world scenarios still need to be proposed. Country-specific systems are mostly, designed, which can (artificially) reach high-level recognition rates. This option, however, strictly limits their applicability. In this paper, we propose an approach that can deal with various national plates. There are three main areas of novelty. First, the Optical Character Recognition (OCR) is managed by a hybrid strategy, combining statistical and structural algorithms. Secondly, an efficient probabilistic edit distance is proposed for providing an explicit video-based LPR. Last but not least, cognitive loops are introduced at critical stages of the algorithm. These feedback steps take advantage of the context modeling to increase the overall system performances, and overcome the inextricable parameter settings of the low-level processing. The system performances have been tested in more than 1200 static images with difficult illumination conditions and complex backgrounds, as well as in six different videos containing 525 moving vehicles. The evaluations prove our system to be very competitive among the non-country specific approaches.

A testing framework for background subtraction algorithms comparison in intrusion detection context

Identifying objects from a video stream is a fundamental and critical task in many computer-vision applications. A popular approach is the background subtraction, which consists in separating foreground (moving objects) from background. Many methodologies have been developed for automatic background segmentation but this fundamental task is still challenging. We focus here on a particular application of computer vision: intrusion detection in video surveillance. We propose in this paper a multi-level methodology for evaluating and comparing background subtraction algorithms. Three levels are studied: first, pixel level to evaluate the accuracy of the segmentation algorithm to attribute the right class to each pixel. Second, image level, measuring the rate of right decision on each frame (intrusion vs no intrusion) and finally sequence level, measuring the accordance with the time span where objects appear. Moreover, we also propose a new similarity measure, called D-Score, adapted to the context of intrusion detection.

ADAPTIVE BACKGROUND SUBTRACTION IN H.264/AVC BITSTREAMS BASED ON MACROBLOCK SIZES

In this article, we propose a novel approach to detect moving objects in H.264 compressed bitstreams. More precisely, we describe a multi-modal background subtraction technique that uses the size of macroblocks in order to label them as belonging to the background of the observed scene or not. Here, we integrate an adaptive Gaussian mixture-based scheme to model the background. We evaluate our contribution using the PETS video dataset and a realist synthetic video sequence rendered by a 3-D urban environment simulator. We compare two different background models, and we show that the Gaussian mixture-based is the best and outperforms other techniques that use macro bloc sizes.

A 3D Tracker for Ground-Moving Objects

Multi-object tracking is a major area of research because of its wide application scope. In this paper we describe a set of improvements, toward video surveillance context, to the multi-object tracker proposed by [1]. First, we generalize the tracking by removing the specialization made for pedestrians. Then, we integrate easily available scene knowledge in order to allow three-dimensional reasoning and better handle occlusions. Additionally, we improve the group creation and destruction mechanism by adding an association pass and an overlap similarity criterion. We evaluate the proposed method on several synthetic and real-world videos.

Self-Calibration and Control of a PTZ Camera Based on a Spherical Mirror

In video surveillance applications, PTZ cameras can focus and analyze in details specific zones of the scene. In a computer supervised intrusion detection, a single PTZ camera is unable to visualize the entire scene at once. This article proposes an original solution to this problem, by using an additional spherical mirror. Besides the equations needed to control the PTZ camera, this article presents also a self calibration processes of the camera with the mirror.

Panoramic mosaicing optimization

Motorized dome-type cameras, also called PTZ camera, allow the creation of panoramas. These panoramas represent the whole of the scene seen by the camera. In the case of a PTZ camera and with certain constraints, the scene seen by the camera can be considered as a sphere. The creation of a panorama consists in traversing a sphere in an exhaustive way. The acquired images are then projected on unspecified support which can be a cylinder, a cube or others. The projection of the rectangular images onto a sphere inevitably involves partial overlap between images. These overlaps lead to useless calculations. In order to limit the number of images we propose the calculation of an optimal trajectory for the camera according to intrinsic and extrinsic constraints.