18 results on '"PETRETO, Andrea"'
Search Results
2. Meteorix: High performance computer vision application for Meteor detection from a CubeSat
- Author
-
Millet, Maxime, Rambaux, Nicolas, Cassagne, Adrien, Bouyer, Manuel, Petreto, Andrea, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Institut de Mécanique Céleste et de Calcul des Ephémérides (IMCCE), Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université de Lille-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[SDU.ASTR.EP]Sciences of the Universe [physics]/Astrophysics [astro-ph]/Earth and Planetary Astrophysics [astro-ph.EP] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,[SPI.AUTO]Engineering Sciences [physics]/Automatic - Abstract
International audience
- Published
- 2022
3. Meteorix - A new processing chain for real-time detection and tracking of meteors from space
- Author
-
Millet, Maxime, Rambaux, Nicolas, Petreto, Andrea, Lemaitre, Florian, Lacassagne, Lionel, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, and Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-AR] Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-SE] Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-RB] Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-DS] Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[SPI.AUTO]Engineering Sciences [physics]/Automatic ,[INFO.INFO-DM] Computer Science [cs]/Discrete Mathematics [cs.DM] ,[SPI.AUTO] Engineering Sciences [physics]/Automatic ,[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-TI] Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-AO] Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing - Abstract
International audience; this article describes a processing chain for meteor detection from space. Unlike ground detection using stationary cameras with a still background, detection aboard a nano-satellite needs to deal with a camera in motion and a moving background. The main parts of this chain are an estimation of the apparent movement and the computation of angular statistics. The first results show a detection probability close to 96% on the whole set of Chiba videos from the PERC (Planetary Exploration Research Center) Meteor Project.
- Published
- 2022
4. Meteorix camera tests for space-based meteor observations
- Author
-
Rambaux, Nicolas, Vaubaillon, Jérémie, Derelle, Sophie, Jacquart, Marc, Millet, Maxime, Lacassagne, Lionel, Petreto, Andrea, Simoneau, Pierre, Baillié, K, Desmars, J, Galayko, Dimitri, Chotin, Roselyne, Institut de Mécanique Céleste et de Calcul des Ephémérides (IMCCE), Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université de Lille-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), DOTA, ONERA, Université Paris Saclay [Palaiseau], ONERA-Université Paris-Saclay, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Circuits Intégrés Numériques et Analogiques (CIAN)
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,[SPI.AUTO]Engineering Sciences [physics]/Automatic - Abstract
International audience; Meteorix is a Universitary CubeSat dedicated to be a demonstrator for the detection and characterisation of meteors and space debris. Its payload is an onboard camera and detection chain. Usually, cameras on CubeSat are used for daylight observations and this proceeding present some tests realized with a sensitive CMOS camera, which is also used in the martian rovers for imaging purposes. In-lab tests were conducted to measure the spectral response. First on-sky images were performed during the 2020 Geminids meteor shower from the Paris area. Follow-up tests were performed from Observatoire de Haute-Provence during the 2021 η-Aquariids meteor shower. Capabilities and needed modifications for meteor detection were identified.
- Published
- 2021
5. METEORIX: a CubeSat mission for detection and characterization of meteors and space debris
- Author
-
Millet, Maxime, Rambaux, Nicolas, Petreto, Andrea, Vaubaillon, Jeremie, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Institut de Mécanique Céleste et de Calcul des Ephémérides (IMCCE), Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université de Lille-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Lerity - Alcen
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,ComputingMilieux_MISCELLANEOUS ,[SPI.AUTO]Engineering Sciences [physics]/Automatic - Abstract
International audience
- Published
- 2021
6. Meteorix: a new processing chain for detection and tracking of meteors from space
- Author
-
Millet, Maxime, Rambaux, Nicolas, PETRETO, Andrea, lemaitre, florian, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Institut de Mécanique Céleste et de Calcul des Ephémérides (IMCCE), Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire de Paris, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université de Lille-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,ComputingMilieux_MISCELLANEOUS ,[SPI.AUTO]Engineering Sciences [physics]/Automatic - Abstract
International audience
- Published
- 2021
7. Meteorix: A cubesat mission dedicated to the detection of meteors and space debris
- Author
-
Rambaux, Nicolas, Vaubaillon, Jérémie, Derelle, Sophie, Jacquart, Marc, Millet, Maxime, Lacassagne, Lionel, PETRETO, Andrea, Simoneau, Pierre, Baillié, K, Desmars, J, Galayko, Dimitri, Chotin, Roselyne, Institut de Mécanique Céleste et de Calcul des Ephémérides (IMCCE), Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université de Lille-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), DOTA, ONERA, Université Paris Saclay [Palaiseau], ONERA-Université Paris-Saclay, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Circuits Intégrés Numériques et Analogiques (CIAN)
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,ComputingMilieux_MISCELLANEOUS ,[SPI.AUTO]Engineering Sciences [physics]/Automatic - Abstract
International audience
- Published
- 2021
8. Détection temps réel de météores à bord d'un nanosatellite, application au projet Meteorix
- Author
-
Millet, Maxime, Rambaux, Nicolas, Petreto, Andrea, lemaitre, florian, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Lhéritier - Alcen, Institut de Mécanique Céleste et de Calcul des Ephémérides (IMCCE), Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Université de Lille-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Centre National de la Recherche Scientifique [CNRS]
- Subjects
flot optique ,étiquetage et analyse en composantes connexes ,Détection et suivi de météores ,nanosatellite ,système embarqué ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO]Computer Science [cs] ,statistiques circulaires - Abstract
National audience; Dans cet article, nous présentons une nouvelle chaîne de traitements pour la détection et le suivi de météores depuis un nanosatellite. Les contraintes d'embarquabilité sont fortes : temps de traitement et consommation électrique.Nous avons donc choisi un ensemble d'algorithmes avec un rapport le plus équilibré possible entre complexité et précision.Les spécificités architecturales ont également orienté nos choix algorithmiques en vue de futures optimisations.La chaîne de traitement présentée ici repose principalement sur une estimation de vitesse par flot optique et des calculs de statistiques circulaires. Si les autres méthodes de l'état de l'art sont efficaces pour la détection de météores depuis le sol, nous montrons que celle présentée ici l'est pour une détection depuis l'espace.
- Published
- 2021
9. Implementations Impact on Iterative Image Processing for Embedded GPU
- Author
-
Romera, Thomas, Petreto, Andrea, Lemaitre, Florian, Bouyer, Manuel, Meunier, Quentin, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Lacassagne, Lionel
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-AR] Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing ,Image Processing ,[INFO.INFO-SE] Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,GPU ,[INFO.INFO-DS] Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,02 engineering and technology ,Energy Consumption ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[SPI.AUTO]Engineering Sciences [physics]/Automatic ,[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,TV-L1 ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,0202 electrical engineering, electronic engineering, information engineering ,[INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,Embedded System ,[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing ,Optical flow ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-RB] Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,020202 computer hardware & architecture ,[INFO.INFO-DM] Computer Science [cs]/Discrete Mathematics [cs.DM] ,[SPI.AUTO] Engineering Sciences [physics]/Automatic ,[INFO.INFO-TI] Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,020201 artificial intelligence & image processing ,[INFO.INFO-AO] Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing - Abstract
International audience; The emergence of low-power embedded Graphical Processing Units (GPUs) with high computation capabilities has enabled the integration of image processing chains in a wide variety of embedded systems. Various optimisation techniques are however needed in order to get the most out of an embedded GPU. This paper explores several optimisation methods for iterative stencil-like image processing algorithms on embedded NVIDIA GPUs using the Compute Unified Device Architecture (CUDA) API. We chose to focus our architectural optimisations on the TV-L1 algorithm, an optical flow estimation method based on total variation (TV) regularisation and the L1 norm. It is widely used as a model for more complex optical flow estimations and is used in many recent video processing applications. In this work we evaluate the impact of architecture-oriented optimisations on both execution time and energy consumption on several Nvidia Jetson GPU embedded boards. Results show a speedup up to 3× compared to State-of-the-Art versions as well as a 2.6× decrease in energy consumption.
- Published
- 2021
10. Débruitage temps réel embarqué pour vidéos fortement bruitées
- Author
-
Petreto, Andrea, Romera, Thomas, Lemaitre, Florian, Masliah, Ian, Gaillard, Boris, Bouyer, Manuel, Meunier, Quentin, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and PETRETO, Andrea
- Subjects
[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-ES]Computer Science [cs]/Embedded Systems ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,ComputingMilieux_MISCELLANEOUS ,[INFO.INFO-ES] Computer Science [cs]/Embedded Systems - Abstract
International audience
- Published
- 2019
11. Implementations Impact on Iterative Image Processing for Embedded GPU
- Author
-
Romera, Thomas, primary, Petreto, Andrea, additional, Lemaitre, Florian, additional, Bouyer, Manuel, additional, Meunier, Quentin, additional, and Lacassagne, Lionel, additional
- Published
- 2021
- Full Text
- View/download PDF
12. Real-time video denoising for embedded systems
- Author
-
PETRETO, Andrea, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Sorbonne Université, Lionel Lacassagne, Professeur, LIP6 - Sorbonne Université, STAR, ABES, Laboratoire d'Informatique de Paris 6 (LIP6), Sorbonne Université, UPMC University of Paris 6, and Lionel Lacassagne
- Subjects
Débruitage vidéo ,Traitement d'images ,Embedded systems ,Fast optical flow ,Optical flow ,Video denoising ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,Débruitgae vidéo ,Optimisation CPU ,Systèmes embarqués ,SIMD ,[INFO.INFO-ES] Computer Science [cs]/Embedded Systems ,Fast video denoising ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,Vision par ordinateur ,Flot optique ,Architecture des processeurs ,[INFO.INFO-ES]Computer Science [cs]/Embedded Systems ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,Optimisations ,CPU optimization - Abstract
In many applications, noisy video can be a major problem. There are denoising methods with highly effective denoising capabilities but at the cost of a very high computational complexity. Other faster methods are limited in their applications since they does not handle high levels of noise correctly. For many applications, it is however very important to preserve a good image quality in every situation with sometimes strong embedding constraints. In this work, the goal is to propose an embedded solution for live video denoising. The method needs to remain efficient with even under high level of noise. We limit our work to embedded CPU under 30W of power consumption. This work led to a new video denoising algorithm called RTE-VD: Real-Time Embedded Video Denoising. RTE-VD is composed of 3 steps: stabilization, movement compensation by dense optical flow estimation and spatio-temporal filtering. On an embedded CPU (Jetson AGX), RTE-VD runs at 30 frame per seconds on qHD videos (960x580 pixels). In order to achieve such performance, many compromises and optimizations had to be done. We compare RTE-VD to other state-of-the-art methods in both terms of denoising capabilities and processing time. We show that RTE-VD brings a new relevant tradeoff between quality and speed., Pour beaucoup d'applications, la présence de bruit dans les vidéos peut être un problème majeur. Des algorithmes de débruitage existent, la plupart sont qualitativement très efficaces mais au prix d'un temps de traitement trop important pour pouvoir envisager une implémentation temps réel embarquée. D'autres méthodes, plus rares, peuvent être exécutées en temps réel mais gèrent mal les forts niveaux de bruit. Pour bon nombre d'applications, il est cependant important de conserver une bonne qualité d'images en toutes conditions avec parfois d'importantes contraintes d'embarquabilité. Dans ces travaux, nous cherchons à proposer une solution de débruitage vidéo qui permet un traitement en direct sur des systèmes embarqués. La solution proposée doit rester efficace même pour de forts niveaux de bruit. Ici, nous nous limitons à l'utilisation de CPU embarqués d'une consommation inférieure à 30W. Les travaux menés dans le cadre de cette thèse ont permis la mise en place d'une chaîne de débruitage nommée RTE-VD (Real-Time Embedded Video Denoising). RTE-VD se décompose en trois étapes : stabilisation, recalage temporel par flot optique et filtrage spatio-temporel. Sur un CPU embarqué (Jetson AGX), RTE-VD permet de traiter à une cadence de 30 images par seconde, des vidéos au format qHD (960x540 pixels). Afin de pouvoir atteindre ces performances, de nombreux compromis et optimisations ont dû être faits. Nous comparons RTE-VD à d'autres méthodes de référence de la littérature à la fois en termes de qualité du débruitage et de vitesse d'exécution. Nous montrons que RTE-VD apporte un nouveau positionnement pertinent au niveau du rapport qualité/vitesse.
- Published
- 2020
13. Real-time embedded video denoiser prototype
- Author
-
Petreto, Andrea, Romera, Thomas, Lemaitre, Florian, Bouyer, Manuel, Gaillard, Boris, Menard, Patrice, Meunier, Quentin, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Lhéritier - Alcen, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Lacassagne, Lionel
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-AR] Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing ,Embedded systems ,[INFO.INFO-SE] Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,[INFO.INFO-DS] Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[SPI.AUTO]Engineering Sciences [physics]/Automatic ,[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,Image processing ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-RB] Computer Science [cs]/Robotics [cs.RO] ,Video denoising ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-DM] Computer Science [cs]/Discrete Mathematics [cs.DM] ,[SPI.AUTO] Engineering Sciences [physics]/Automatic ,[INFO.INFO-TI] Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-AO] Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,Real-time ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing - Abstract
International audience; Low light or other poor visibility conditions often generate noise on any vision system. However, video denoising requires a lot of computational effort and most of the state-of-the-art algorithms cannot be run in real-time at camera framerate. Noisy video is thus a major issue especially for embedded systems that provide low computational power. This article presents a new real-time video denoising algorithm for embedded platforms called RTE-VD [1]. We first compare its denoising capabilities with other online and offline algorithms. We show that RTE-VD can achieve real-time performance (25 frames per second) for qHD video (960x540 pixels) on embedded CPUs with an output image quality comparable to state-of-the-art algorithms. In order to reach real-time denoising, we applied several high-level transforms and optimizations. We study the relation between computation time and power consumption on several embedded CPUs and show that it is possible to determine find out frequency and core configurations in order to minimize either the computation time or the energy. Finally, we introduce VIRTANS our embedded real-time video denoiser based on RTE-VD.
- Published
- 2020
14. A New Real-Time Embedded Video Denoising Algorithm
- Author
-
Petreto, Andrea, primary, Romera, Thomas, additional, Lemaitre, Florian, additional, Masliah, Ian, additional, Gaillard, Boris, additional, Bouyer, Manuel, additional, Meunier, Quentin L., additional, and Lacassagne, Lionel, additional
- Published
- 2019
- Full Text
- View/download PDF
15. Comparaison de la consommation énergétique et du temps d'exécution d'un algorithme de traitement d'images optimisé sur des architectures SIMD et GPU
- Author
-
Petreto, Andrea, Hennequin, Arthur, Koehler, Thomas, Romera, Thomas, Fargeix, Yohan, Gaillard, Boris, Bouyer, Manuel, Meunier, Quentin, Lacassagne, Lionel, Lacassagne, Lionel, Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Systèmes Electroniques (SYEL), and Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-AR] Computer Science [cs]/Hardware Architecture [cs.AR] ,[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-SE] Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-RB] Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[INFO.INFO-DS] Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[SPI.AUTO]Engineering Sciences [physics]/Automatic ,[INFO.INFO-DM] Computer Science [cs]/Discrete Mathematics [cs.DM] ,[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] ,[SPI.AUTO] Engineering Sciences [physics]/Automatic ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-TI] Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,[INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO] ,[INFO.INFO-AO] Computer Science [cs]/Computer Arithmetic ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing - Abstract
International audience; Cet article présente et compare les implémentations optimisées d'un algorithme de flot optique sur des cartes embarquées à base de processeurs SIMD multicoeurs et de GPU. La comparaison est effectuée à la fois en termes de vitesse de calcul – pour atteindre une cadence de traitement temps réel – et en termes d'énergie. Les résultats obtenus montrent que les GPU sont les plus efficaces à la fois en termes de vitesse et de consommation, pouvant traiter dans la meilleure configuration 25 images de 8M pixels par secondes pour 0.35 joule par image. Mots-clés : flot optique, efficacité énergétique, optimisation de code, architectures multi-coeur SIMD/GPU, exécution temps réel.
- Published
- 2018
16. Energy and Execution Time Comparison of Optical Flow Algorithms on SIMD and GPU Architectures
- Author
-
Petreto, Andrea, primary, Hennequin, Arthur, additional, Koehler, Thomas, additional, Romera, Thomas, additional, Fargeix, Yohan, additional, Gaillard, Boris, additional, Bouyer, Manuel, additional, Meunier, Quentin L., additional, and Lacassagne, Lionel, additional
- Published
- 2018
- Full Text
- View/download PDF
17. A new SIMD iterative connected component labeling algorithm
- Author
-
Lacassagne, Lionel, primary, Cabaret, Laurent, additional, Etiemble, Daniel, additional, Hebache, Farouk, additional, and Petreto, Andrea, additional
- Published
- 2016
- Full Text
- View/download PDF
18. A new SIMD iterative connected component labeling algorithm.
- Author
-
Lacassagne, Lionel, Cabaret, Laurent, Etiemble, Daniel, Hebache, Farouk, and Petreto, Andrea
- Published
- 2016
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.