Abstract-The Scale Invariant Feature Transform (SIFT) algorithm is widely used in computer vision to match features between images or to localize and recognize objets. However, mostly because of speckle noise, it does not perform well on synthetic aperture radar (SAR) images. We present here an improvement of this algorithm for SAR images, named SAR-SIFT. A new gradient computation, yielding an orientation and a magnitude robust to speckle noise, is first introduced. It is then used to adapt several steps of the SIFT algorithm to SAR images. We study the improvement brought by this new algorithm, compared to existing approaches. We present an application of SAR-SIFT for the registration of SAR images in different configurations, especially with different incidence angles.

The scale invariant feature transform (SIFT) algorithm, commonly used in computer vision, does not perform well on synthetic aperture radar (SAR) images, in particular because of the strong intensity and the multiplicative nature of the noise. We present an improvement of this algorithm for SAR images. First, a robust yet simple way to compute gradient on radar images is introduced. This step is first used to develop a new keypoints extraction algorithm, based on the Harris criterion. Second, we rely on this gradient definition to adapt the computation of both the main orientation and the geometric descriptor to SAR image specificities. We validate this new algorithm with different experiments and present an application of our new SAR-SIFT algorithm.

In disaster situations, remote sensing images are very useful to quickly assess damages. However, the choice of available images for the studied area is frequently limited. It is often needed to compare images acquired by different sensors and with different acquisition conditions. We propose a new feature-based approach to detect changes between a pair of either optical or radar images. This approach is based on the SIFT algorithm and an a contrario approach. It can deal with multi-resolutions, multi-sensors and multi-incidence angles situations, and it offers promising results.

L’objet du projet présenté, support de réflexion et d’expérimentation, est la prise en compte de la gestion des eaux pluviales par infiltration dans le cadre de la rénovation urbaine d’un quartier en politique de la ville où, jusqu’à présent, toutes les eaux collectées étaient renvoyées aux réseaux. L’étude de cas est celle de la rénovation des espaces publics sur une surface de 7 ha du quartier de l’Arlequin de la Villeneuve à Grenoble en 2017 et 2018, pour un montant d’investissement de 6,50 M€ HT. En partenariat avec les services de la métropole de Grenoble, les espaces réaménagés sont progressivement déconnectés du réseau, qui, en conséquence, est pour partie abandonné. Le travail porte autant sur les expérimentations des systèmes de gestion hydrauliques en renouvellement urbain que sur le développement conjoint d’un projet urbain et paysager, sur l’évolution des palettes végétales, sur le confort des espaces publics et la réduction des îlots de chaleur, sur la sécurité et sur la gestion de la voiture dans le quartier.

Camera-based Deep Learning algorithms are increasingly needed for perception in Automated Driving systems. However, constraints from the automotive industry challenge the deployment of CNNs by imposing embedded systems with limited computational resources. In this paper, we propose an approach to embed a multitask CNN network under such conditions on a commercial prototype platform, i.e. a low power System on Chip (SoC) processing four surround-view fisheye cameras at 10 FPS. The first focus is on designing an efficient and compact multi-task network architecture. Secondly, a pruning method is applied to compress the CNN, helping to reduce the runtime and memory usage by a factor of 2 without lowering the performances significantly. Finally, several embedded optimization techniques such as mixed-quantization format usage and efficient data transfers between different memory areas are proposed to ensure real-time execution and avoid bandwidth bottlenecks. The approach is evaluated on the hardware platform, considering embedded detection performances, runtime and memory bandwidth. Unlike most works from the literature that focus on classification task, we aim here to study the effect of pruning and quantization on a compact multi-task network with object detection, semantic segmentation and soiling detection tasks.