Connaissance & Intelligence Artificielle Distribuées

Handwritten Text Recognition (HTR) is essential for digitizing historical documents in different kinds of archives. In this study, we introduce a hybrid form archive written in French: the Belfort civil registers of births. The digitization of these historical documents is challenging due to their unique characteristics such as writing style variations, overlapped characters and words, and marginal annotations. The objective of this survey paper is to summarize research on handwritten text documents and provide research directions toward effectively transcribing this French dataset. To achieve this goal, we presented a brief survey of several modern and historical HTR offline systems of different international languages, and the top state-of-the-art contributions reported of the French language specifically. The survey classifies the HTR systems based on techniques employed, datasets used, publication years, and the level of recognition. Furthermore, an analysis of the systems' accuracies is presented, highlighting the best-performing approach. We have also showcased the performance of some HTR commercial systems. In addition, this paper presents a summarization of the HTR datasets that publicly available, especially those identified as benchmark datasets in the International Conference on Document Analysis and Recognition (ICDAR) and the International Conference on Frontiers in Handwriting Recognition (ICFHR) competitions. This paper, therefore, presents updated state-of-the-art research in HTR and highlights new directions in the research field.

The ability to sense and understanding the driving environment is a key technology for ADAS and autonomous driving. Human drivers have to pay more visual attention to important or target elements and ignore unnecessary ones present in their field of sight. A model that computes this visual attention of targets in a specific driving environment is essential and useful in supporting autonomous driving, object-specific tracking & detection, driving training, car collision warning, traffic sign detection, etc. In this paper, we propose a new framework of visual attention that can predict important objects in the driving scene using a conditional generative adversarial network. A large scale Visual Attention Driving Database (VADD) of saliency heat-maps is built from existing driving datasets using a saliency mechanism. The proposed framework model takes its strength from these saliency heat-maps as conditioning label variables. The results show that the proposed approach makes us able to predict heat-maps of most important objects in a driving environment.

Traffic signs play an important role for Advanced Driver Assistance Systems (ADAS) as well as for autonomous driving vehicles. Most of the works done focus on recognizing symbol based signs leaving apart important information provided by other type of signs like complementary panels or text based signs. In this paper, we include detection and classification of both symbol and text based signs focusing on the most common ones found in European urban environments. The system consists of three stages, traffic sign detection, refinement and classification. The detection and refinement is performed using Mask R-CNN while the classification is achieved with a proposed Convolutional Neural Network (CNN) architecture. We introduced the extended version of the German Traffic Sign Detection Benchmark (GTSDB), labeled in a pixel manner (masks) with 164 classes grouped into 8 categories. It is used for the detection and classification steps. Experimental results on German environments show that our proposed system is capable of detecting all categories of traffic signs while at the same time recognizing them with high accuracy achieving comparable performance with the state of the art.

We propose an efficient spatio-temporal model for mobile autonomous robots operating in human populated environments. Our method aims to model periodic temporal patterns of people presence, which are based on peoples' routines and habits. The core idea is to project the time onto a set of wrapped dimensions that represent the periodicities of people presence. Extending a 2D spatial model with this multidimensional representation of time results in a memory efficient spatio-temporal model. This model is capable of long-term predictions of human presence, allowing mobile robots to schedule their services better and to plan their paths. The experimental evaluation, performed over datasets gathered by a robot over a period of several weeks, indicates that the proposed method achieves more accurate predictions than the previous state of the art used in robotics.

In this paper, we present a lifelong-learning multisensor system for pedestrian detection in adverse weather conditions. The proposed method combines two people detection pipelines which process data provided by a lidar and an ultrawideband radar. The outputs of these pipelines are combined not only by means of adaptive sensor fusion, but they can also be used to help one another learn. In particular, the lidar-based detector provides labels to the incoming radar data, efficiently training the radar data classifier. In several experiments, we show that the proposed learning-fusion not only results in a gradual improvement of the system performance during routine operation, but also efficiently deals with lidar detection failures caused by thick fog conditions.

CONTEXT: Computational modeling involves the use of computer simulations and models to study and understand real-world phenomena. Its application is particularly relevant in the study of potential interactions between biological elements. It is a promising approach to understand complex biological processes and predict their behavior under various conditions. METHODOLOGY: This paper is a review of the recent literature on computational modeling of biological systems. Our study focuses on the field of oncology and the use of artificial intelligence (AI) and, in particular, agent-based modeling (ABM), between 2010 and May 2023. RESULTS: Most of the articles studied focus on improving the diagnosis and understanding the behaviors of biological entities, with metaheuristic algorithms being the models most used. Several challenges are highlighted regarding increasing and structuring knowledge about biological systems, developing holistic models that capture multiple scales and levels of organization, reproducing emergent behaviors of biological systems, validating models with experimental data, improving computational performance of models and algorithms, and ensuring privacy and personal data protection are discussed.

The background Initialization (BI) problem has attracted the attention of researchers in different image/video processing fields. Recently, a tensor-based technique called spatiotemporal slice-based singular value decomposition (SS-SVD) has been proposed for background initialization. SS-SVD applies the SVD on the tensor slices and estimates the background from low-rank information. Despite its efficiency in background initialization, the performance of SS-SVD requires further improvement in the case of complex sequences with challenges such as stationary foreground objects (SFOs), illumination changes, low frame-rate, and clutter. In this paper, a self-motion-assisted tensor completion method is proposed to overcome the limitations of SS-SVD in complex video sequences and enhance the visual appearance of the initialized background. With the proposed method, the motion information, extracted from the sparse portion of the tensor slices, is incorporated with the low-rank information of SS-SVD to eliminate existing artifacts in the initiated background. Efficient blending schemes between the low-rank (background) and sparse (foreground) information of the tensor slices is developed for scenarios such as SFO removal, lighting variation processing, low frame-rate processing, crowdedness estimation, and best frame selection. The performance of the proposed method on video sequences with complex scenarios is compared with the top-ranked state-of-the-art techniques in the field of background initialization. The results not only validate the improved performance over the majority of the tested challenges but also demonstrate the capability of the proposed method to initialize the background in less computational time.

Convolutional Network (ConvNet), with its strong image representation ability, has achieved significant progress in the computer vision and robotic fields. In this paper, we propose a visual localization approach based on place recognition that combines the powerful ConvNet features and localized image sequence matching. The image distance matrix is constructed based on the cosine distance of extracted ConvNet features, and then a sequence search technique is applied on this distance matrix for the final visual recognition. To speed up the computational efficiency, the locality sensitive hashing (LSH) method is applied to achieve real-time performances with minimal accuracy degradation. We present extensive experiments on four real world data sets to evaluate each of the specific challenges in visual recognition. A comprehensive performance comparison of different ConvNet layers (each defining a level of features) considering both appearance and illumination changes is conducted. Compared with the traditional approaches based on hand-crafted features and single image matching, the proposed method shows good performances even in the presence of appearance and illumination changes.

Developed countries have paved the way for the implementation of intelligent transport systems to improve the safety, efficiency, and environmental impact of transport. With developing countries entering the fray, the question is: Is ITS as implemented in developed countries relevant to developing countries? This research question is discussed in this paper for the case of sub-Saharan countries that are among the poorest countries worldwide. To this end, the paper outlines the main differences in transportation scenarios between developed and sub-Saharan countries, and then considering the main constraints of sub-Saharan countries, a guide towards an affordable intelligent transportation system in these countries is discussed. Finally, the paper proposes two main novel ideas for the deployment of intelligent transport systems on dirt roads.

Intersections are at the core of congestion in urban areas. After the end of the Second World War, the problem of intersection management has benefited from a growing body of advances to address the optimization of the traffic lights' phase splits, timing, and offset. These contributions have significantly improved traffic safety and efficiency in urban areas. However, with the growth of transportation demand and motorization, traffic lights show their limits. At the end of the 1990s, the perspective of autonomous and connected driving systems motivated researchers to introduce a paradigm shift for controlling intersections. This new paradigm is well known today as autonomous intersection management (AIM). It harnesses the self-organization ability of future vehicles to provide more accurate control approaches that use the smallest available time window to reach unprecedented traffic performances. This is achieved by optimizing two main points of the interaction of connected and autonomous vehicles at intersections: the motion control of vehicles and the schedule of their accesses. Considering the great potential of AIM and the complexity of the problem, the proposed approaches are very different, starting from various assumptions. With the increasing popularity of AIM, this paper provides readers with a comprehensive vision of noticeable advances toward enhancing traffic efficiency. It shows that it is possible to tailor vehicles' speed and schedule according to the traffic demand by using distributed particle swarm optimization. Moreover, it brings the most relevant contributions in the light of traffic engineering, where flow-speed diagrams are used to measure the impact of the proposed optimizations. Finally, this paper presents the current challenging issues to be addressed.

Light detection and ranging (LiDAR) limitations in adverse weather (e.g., rain, fog, and snow) prevent adopting high-level autonomous vehicles in all weather conditions. Furthermore, collecting and annotating these sparse point clouds in adverse weather is often cumbersome, inefficient, and expensive. In this article, we propose a data-driven approach to statistically model the performance of a popular near-infrared (NIR) time-of-flight (ToF) LiDAR in fog, with noisy point clouds collected in a well-controlled artificial fog chamber. Given manually defined visibility describing the levels of fog, our models can directly forecast a probability distribution of a laser’s noisy range measurement. Moreover, the real road data collected in clear weather is utilized to produce auto-labeled noisy point clouds using a LiDAR performance simulator, which is then used to train a semantic segmentation network to recognize point cloud noise in the real world in adverse weather. Qualitative and quantitative experimental results verify the applicability of our LiDAR performance models in fog and show how our Sim2Real strategy of the denoising algorithm can be applied to noisy point clouds under various weather conditions. The developed robot operating system (ROS) packages are publicly available at: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/cavayangtao/lanoise_pp</uri> .

With the rapid increase of the world's urban population, the infrastructure of the constantly expanding metropolitan areas is undergoing an immense pressure. To meet the growing demands of sustainable urban environments and improve the quality of life for citizens, municipalities will increasingly rely on novel transport solutions. In particular, Unmanned Aerial Vehicles (UAVs) are expected to have a crucial role in the future smart cities thanks to their interesting features such as autonomy, flexibility, mobility, adaptive altitude, and small dimensions. However, densely populated megalopolises of the future are administrated by several municipals, governmental and civil society actors, where vivid economic activities involving a multitude of individual stakeholders take place. In such megalopolises, the use of agents for UAVs is gaining more interest especially in complex application scenarios where coordination and cooperation are necessary. This paper sketches a visionary view of the UAVs' role in the transport domain of future smart cities. Additionally, four challenging research directions are highlighted including problems related to autonomy, explainability, security and validation &amp; verification of the UAVs behavior.

We present a human-centric spatiotemporal model for service robots operating in densely populated environments for long time periods. The method integrates observations of pedestrians performed by a mobile robot at different locations and times into a memory efficient model, that represents the spatial layout of natural pedestrian flows and how they change over time. To represent temporal variations of the observed flows, our method does not model the time in a linear fashion, but by several dimensions wrapped into themselves. This representation of time can capture long-term (i.e. days to weeks) periodic patterns of peoples' routines and habits. Knowledge of these patterns allows making long-term predictions of future human presence and walking directions, which can support mobile robot navigation in human-populated environments. Using datasets gathered for several weeks, we compare the model to state-of-the-art methods for pedestrian flow modelling.

This paper proposes a framework to extract dynamic trip flows and travel demand patterns from large-scale 2 G and 3 G cellular signaling data. Novel data pre-processing techniques based on cell phone activity metrics are presented. The trip extraction method relies on the detection of stationary activities to form trip sequences related to resident users. A probabilistic solution is introduced to estimate the trip starting time, allowing to aggregate trips by time of the day and reconstruct hourly travel flows. To better characterize these flows, a spatial clustering process combined with land-use data is proposed based on the temporal demand profile of each zone. Empirical comparisons have been performed showing that the resulting dynamic travel demand patterns are consistent with those obtained from travel survey data with high correlation coefficients of about 0.9. The results prove the potential of signaling data to generate low-cost valuable information for large-scale travel demand modeling.

The Codebook model is one of the popular real-time models for background subtraction. In this paper, we first extend it from traditional Red-Green-Blue (RGB) color model to multispectral sequences. A self-adaptive mechanism is then designed based on the statistical information extracted from the data themselves, with which the performance has been improved, in addition to saving time and effort to search for the appropriate parameters. Furthermore, the Spectral Information Divergence is introduced to evaluate the spectral distance between the current and reference vectors, together with the Brightness and Spectral Distortion. Experiments on five multispectral sequences with different challenges have shown that the multispectral self-adaptive Codebook model is more capable of detecting moving objects than the corresponding RGB sequences. The proposed research framework opens a door for future works for applying multispectral sequences in moving object detection.

Abstract The evolving global landscape of packaging regulations, driven by heightened public concern over single‐use packaging and plastic waste, signals a significant shift toward circular economy principles. These regulatory changes, including the United Nations Environmental Programme's endorsement of a legally binding international agreement on plastic waste management, are expected to have far‐reaching impacts on product safety, packaging design, and global trade. The newly adopted and emerging policies prioritize waste reduction, reuse, recycling, and the integration of recycled materials. However, the rapid introduction of environmentally driven packaging regulations—especially those targeting sensitive sectors like food and cosmetics—presents potential risks for product/packaging safety and food waste, alongside challenges for trade harmonization across regions with divergent regulatory goals. This forum article reviews key regulatory frameworks from Asia, Europe, and the Americas, detailing material bans, reuse and refill targets, recycling mandates, and extended producer responsibility initiatives. It examines the potential impact of these regulations on global trade dynamics and their implications for universally accepted safety standards. Despite the well‐meaning intent behind these laws, gaps in understanding the protective role of packaging and its broader environmental impact may lead to misconceptions in developing sustainable product‐packaging systems. Current policies primarily target waste reduction, but an integrated approach is needed – one that considers how packaging extends food shelf life, lowers resource use, and reduces greenhouse gas emissions. Consumer education and transparent communication from regulators and industry will guide informed, sustainable choices, ultimately supporting a more transparent shift toward a circular economy.

This paper presents a robust monocular visual teach-and-repeat (VT&R) navigation system for long-term operation in outdoor environments. The approach leverages deep-learned descriptors to deal with the high illumination variance of the real world. In particular, a tailored self-supervised descriptor, DarkPoint, is proposed for autonomous navigation in outdoor environments. We seamlessly integrate the localisation with control, in which proportional–integral control is used to eliminate the visual error with the pitfall of the unknown depth. Consequently, our approach achieves day-to-night navigation using a single-experience map and is able to repeat complex and fast manoeuvres. To verify our approach, we performed a vast array of navigation experiments in various outdoor environments, where both navigation accuracy and robustness of the proposed system are investigated. The experimental results show that our approach is superior to the baseline method with regards to accuracy and robustness.

In this paper, we propose a method that removes raindrops with light field image using image inpainting. We first use the depth map generated from light field image to detect raindrop regions which are then expressed as a binary mask. The original image with raindrops is improved by refocusing on the far regions and filtering by a high-pass filter. With the binary mask and the enhanced image, image inpainting is then utilized to eliminate raindrops from the original image. We compare pre-trained models of several deep learning based image inpainting methods. A light field raindrop dataset is released to verify our method. Image quality analysis is performed to evaluate the proposed image restoration method. The recovered images are further applied to object detection and visual localization tasks.

Background: Despite the advancement in eXplainable Artificial Intelligence, the explanations provided by model-agnostic predictors still call for improvements (i.e., lack of accurate descriptions of predictors’ behaviors). Contribution: We present a tool for Deep Explanations and Rule Extraction (DEXiRE) to approximate rules for Deep Learning models with any number of hidden layers. Methodology: DEXiRE proposes the binarization of neural networks to induce Boolean functions in the hidden layers, generating as many intermediate rule sets. A rule set is inducted between the first hidden layer and the input layer. Finally, the complete rule set is obtained using inverse substitution on intermediate rule sets and first-layer rules. Statistical tests and satisfiability algorithms reduce the final rule set’s size and complexity (filtering redundant, inconsistent, and non-frequent rules). DEXiRE has been tested in binary and multiclass classifications with six datasets having different structures and models. Results: The performance is consistent (in terms of accuracy, fidelity, and rule length) with respect to the state-of-the-art rule extractors (i.e., ECLAIRE). Moreover, compared with ECLAIRE, DEXiRE has generated shorter rules (i.e., up to 74% fewer terms) and has shortened the execution time (improving up to 197% in the best-case scenario). Conclusions: DEXiRE can be applied for binary and multiclass classification of deep learning predictors with any number of hidden layers. Moreover, DEXiRE can identify the activation pattern per class and use it to reduce the search space for rule extractors (pruning irrelevant/redundant neurons)—shorter rules and execution times with respect to ECLAIRE.

Quantum computing offers a new approach to the problem modeling and solving. This paper deals with the quantum modeling of reactive agents. It also proposes a quantum algorithm to implement the subsumption architecture, widely used by reactive agents, particularly in robotics. This work shows the contribution of the formalism proposed by quantum mechanics to the modeling and the proof of certain properties of the agent behavior. After, the definition of the reactive agent state modeling, the paper suggests a behavior modeling approach based on two steps for subsumption architecture. The first one models the preset behavior that links each action to the perception states. The second one determines, among several actuated actions, the one that the robot must achieve. The subsumption architecture raises the challenge of modeling hierarchical priority of actions. To this end, a multipartite entanglement is used in the second step. More precisely, the paper proposes and generalizes a W-state circuit in order to be used for modeling hierarchical priority actions and controlling the robot accordingly. The result of both steps provides a formal model that links the robot’s perception (input) to the actions (output), with respect to the subsumption architecture. The proposed model of agent is simulated using IBM quantum computer. The simulation shows that the model can either be served as a control unit of the robot (CU) to obtain the suitable action or to simulate the robot behavior.