HAL-MINES ParisTech
Not a member yet
27348 research outputs found
Sort by
Mapping private urban trees in Marseille with LiDAR and machine learning
No full textInternational audienceUrban trees provide essential ecosystem services, particularly in Mediterranean cities facing recurrent heatwaves, drought, and pollution. While public inventories are increasingly available, vegetation on parcels outside the public right-of-way remains largely undocumented, which hinders city-wide estimates of canopy, cooling capacity, and carbon storage and biases actions toward streets and squares only. To address this evidence gap in Marseille, we assemble a multi-source data fusion that combines airborne LiDAR (IGN), terrestrial mobile LiDAR (Greehill), and IGN BD ORTHO ® RGB/IRC orthophotography with machine learning to map dominant genera and canopy structure across both municipal and parcel contexts. Airborne and terrestrial LiDAR show high consistency for canopy height (ordinary least squares, R 2 = 0.78 on n = 60,583 pairs), and a Random Forest baseline reaches ≈80 % overall accuracy for five common genera (Pinus, Platanus, Celtis, Tilia and Cupressus); we also report complementary agreement metrics (macro-averaged F1, Jaccard/IoU, and Cohen's κ) to ensure robust comparison. Building on these metrics, we quantify and compare structural and taxonomic differences between vegetation managed in the public domain and vegetation on parcels,</div
FOOTPASS: A Multi-Modal Multi-Agent Tactical Context Dataset for Play-by-Play Action Spotting in Soccer Broadcast Videos
No full textSoccer video understanding has motivated the creation of datasets for tasks such as temporal action localization, spatiotemporal action detection (STAD), or multiobject tracking (MOT). The annotation of structured sequences of events (who does what, when, and where) used for soccer analytics requires a holistic approach that integrates both STAD and MOT. However, current action recognition methods remain insufficient for constructing reliable play-by-play data and are typically used to assist rather than fully automate annotation. Parallel research has advanced tactical modeling, trajectory forecasting, and performance analysis, all grounded in game-state and play-by-play data. This motivates leveraging tactical knowledge as a prior to support computer-vision-based predictions, enabling more automated and reliable extraction of play-by-play data.We introduce Footovision Play-by-Play Action Spotting in Soccer Dataset (FOOTPASS), the first benchmark for play-by-play action spotting over entire soccer matches in a multi-modal, multi-agent tactical context. It enables the development of methods for player-centric action spotting that exploit both outputs from computer-vision tasks (e.g., tracking, identification) and prior knowledge of soccer, including its tactical regularities over long time horizons, to generate reliable play-by-play data streams. These streams form an essential input for data-driven sports analytics
Optimal schedule of multi-channel quantum Zeno dragging with application to solving the k-SAT problem
No full textQuantum Zeno dragging enables the preparation of common eigenstates of a set of observables by frequent measurement and adiabatic-like modulation of the measurement basis. In this work, we present a deeper analysis of multi-channel Zeno dragging using generalized measurements, i.e. simultaneously measuring a set of non-commuting observables that vary slowly in time, to drag the state towards a target subspace. For concreteness, we will focus on a measurement-driven approach to solving k-SAT problems as examples. We first compute some analytical upper bounds on the convergence time, including the effect of finite measurement time resolution. We then apply optimal control theory to obtain the optimal dragging schedule that lower bounds the convergence time, for low-dimensional settings. This study provides a theoretical foundation for multi-channel Zeno dragging and its optimization, and also serves as a guide for designing optimal dragging schedules for quantum information tasks including measurement-driven quantum algorithms
Non-perturbative switching rates in bistable open quantum systems: from driven Kerr oscillators to dissipative cat qubits
No full textInternational audienceIn this work, we use path integral techniques to predict the switching rate in a single-mode bistable open quantum system. While analytical expressions are well-known to be accessible for systems subject to Gaussian noise obeying classical detailed balance, we generalize this approach to a class of quantum systems, those which satisfy the recently-introduced notion of hidden time-reversal symmetry [1]. In particular, in the context of quantum computing, we deliver precise estimates of bit-flip error rates in cat-qubit architectures, circumventing the need for costly numerical simulations. Our results open new avenues for exploring switching phenomena in multistable single-and many-body open quantum systems
Investigating the uniaxial compressive mechanics of graded polymer foams via in-situ synchrotron X-ray microtomography
No full textGraded polymer foams are emerging as transformative materials for structural applications, outperforming uniform foams due to their spatially tailored density and microstructural features. However, harnessing their full potential requires a deep understanding of how their macroscopic mechanical behavior relates to their complex microstructure evolution. In this study, we elucidate the uniaxial compressive response of graded foams using in-situ synchrotron X-ray microtomography, complemented by comparative experiments on uniform foams of varying densities. Our findings reveal that graded foams exhibit both qualitatively and quantitatively distinct mechanical behavior, driven by unique microscale deformation mechanisms. We evaluate and discuss their superior energy absorption performance and demonstrate how the density profile evolves under increasing macroscopic strain. Notably, the graded architecture enables precise control over the localization and progression of densification bands, offering unprecedented design flexibility for advanced structural applications
Hybrid method for Automatic Fault Detection and Diagnosis in buildings with limited monitoring data
No full textInternational audienc
Optimization of two-phase expansion nozzles: An exergy based analysis of fluid property influence
No full textInternational audienceTwo-phase expansion energy-recovery systems, such as ejectors and two-phase turbines, can substantially improve the efficiency of air conditioning, heating, liquefaction or power generation thermodynamic processes. One of their key component is the stator, or nozzle, that converts fluid enthalpy into kinetic energy.The choice of fluid depends on considerations such as cycle efficiency. However, estimating the efficiency of two-phase expansion is not trivial and can be highly dependent on the working fluid.This paper analyses the effects of the geometric design and fluid properties on the efficiency of two-phase nozzles. The aim is to identify trends in the properties that most significantly affect performance so the fluid hydrodynamic properties are included in the selection process.To this end, a tool for the optimal 1D bi-fluid design of the nozzle is developed, followed by a exergy analysis to determine the thermophysical properties responsible for the losses.The optimization tool is based on an Euler-Euler flow solver with advanced closure terms and a genetic algorithm profile optimizer.The calibration of the physical model is based on experimental data from the literature on R744, R290 and R134a fluids used in commercial refrigeration operating conditions. Optimization and exergetic analysis are performed for similar conditions. A nozzle per fluid was designed.The results show an increase in nozzle efficiency of around 10 % associated with the optimization.Once the nozzles are optimized, comparative exergy analysis indicates that the key parameters affecting losses are liquid density, liquid-vapour density ratio and liquid heat capacity to characteristic length ratio.</div
HEAT RECOVERY FROM BUSES FOR DISTRICT HEATING NETWORK; EVALUATION FOR A FRENCH EXAMPLE
No full textInternational audienceThe fight against climate change has become a public policy priority today. The transport sector is responsible for a third of France's CO2 emissions, making it a priority target for decarbonizing the economy. Therefore, companies in this field are committing to initiatives, like the RATP (Autonomous Parisian Transportation Authority), by transitioning their bus fleets to more environmentally friendly vehicles. In particular, this will involve replacing part of the diesel fleet with biogas buses. As part of a city-wide approach to global energy optimization, this article presents a preliminary analysis that looks at the possibilities for recovering waste heat from internal combustion bus fleets in district heating networks. The aim was to develop a tool to harness the waste heat from buses in the Île-de-France region and redistribute it within the region's heating networks. Using simulations of bus velocities based on their routes, their consumption and available thermal energy per route were calculated. The considered routes were the 15 lines located at the Massy-Palaiseau station in Ile-de-France region, which is close to a district heating network (DHN). A preliminary study on the storage medium was done with high temperature oil to estimate the required storage volumes. The analysis was done using exhaust gas temperature data from diesel buses.The study yields a total potential estimation of 5.8 GWh per year (workdays, excluding July-August).This would reduce the share of carbon-based energy in the nearest DHN by 7.5%. As a first approximation, the gross revenue from the resale of the collected heat is equivalent to approximately 30% of the cost of the fuel consumed by the buses. Next, we need to optimize the on-board storage system, the collection system and the connection to the DHN. It will then be necessary to define the LCOE of this heat source.</div
Contextual LCIA without the overhead: an exchange-based framework for flexible impact assessment
No full textInternational audienceAbstract Purpose Life Cycle Impact Assessment (LCIA) has traditionally applied characterization factors (CFs) to elementary flows in isolation, treating impacts as fixed properties of substances, regardless of where and how they occur within the life cycle product system. While recent advances have introduced regionalized LCIA methods and GIS-based spatial modeling frameworks, these remain difficult to operationalize in routine assessments and are often limited to differentiation at the elementary flow level. This paper presents a methodological advancement in LCIA: the application of CFs at the level of exchanges—the resolved biosphere and technosphere flows between processes. Methods Shifting the CF unit from elementary flows to exchanges enables CFs to reflect the full context of each exchange, including the geographic origin and destination, the identity of the emitting process and environmental recipient, and prospective, scenario-dependent parameters. The method offers a flexible, intermediate solution between traditional elementary flow-based LCIA and full spatially explicit models, supporting national and subnational regionalization without requiring high-resolution GIS integration. Results The approach is implemented in the open-source Python library edges , which extends the Brightway LCA framework to support exchange-specific and symbolic CFs. We illustrate its capabilities through four applications: 1) Regionalized LCIA, using the AWARE water scarcity method with dynamic handling of region aggregation and disaggregation; 2) Technosphere-based LCIA, via a new implementation of the GeoPolRisk indicator, which assigns CFs based on country-to-country commodity trade relationships; and scenario-sensitive prospective LCIAs, enabling alignment with climate scenarios, where CFs are defined by symbolic expressions that depend on scenario-specific variables, with application 3) focusing on global warming potential based on atmospheric gas concentration, and application 4) addressing fossil resource scarcity through dynamic fossil fuels extraction rates. Conclusions Together, these examples demonstrate how exchange-resolved LCIA expands the methodological space of impact modeling, offering a scalable, exchange-aware framework for regional, relational, and future-oriented life cycle assessments
Nicknames for Group Signatures
No full textInternational audienceNicknames for Group Signatures (NGS) is a new signature scheme that extends Group Signatures (GS) with Signatures with Flexible Public Keys (SFPK). Via GS, each member of a group can sign messages on behalf of the group without revealing his identity, except to a designated auditor. Via SFPK, anyone can create new identities for a particular user, enabling anonymous transfers with only the intended recipient able to trace these new identities. To prevent the potential abuses that this anonymity brings, NGS integrates flexible public keys into the GS framework to support auditable transfers. In addition to introducing NGS, we describe its security model and provide a mathematical construction proved secure in the Random Oracle Model. As a practical NGS use case, we build NickHat, a blockchain-based token-exchange prototype system on top of Ethereum