A5038522461- Mosquito-borne diseases and control
- Insect behavior and control techniques
- Forensic Entomology and Diptera Studies
- 3D Surveying and Cultural Heritage
- Advanced Memory and Neural Computing
- Diptera species taxonomy and behavior
- Remote Sensing and LiDAR Applications
- Cell Image Analysis Techniques
- CCD and CMOS Imaging Sensors
- Insect and Pesticide Research
- Robotics and Sensor-Based Localization
- Medical Image Segmentation Techniques
- Neural dynamics and brain function
- Malaria Research and Control
- Conservation Techniques and Studies
- Advanced Vision and Imaging
- Visual Attention and Saliency Detection
- French Urban and Social Studies
- Cultural Insights and Digital Impacts
- Vehicle License Plate Recognition
- Digital Media and Visual Art
- Advanced Neural Network Applications
- Vector-borne infectious diseases
- Bird parasitology and diseases
- Insect Resistance and Genetics
École Nationale Supérieure de l'Électronique et de ses Applications
2019-2025
Equipes Traitement de l'Information et Systèmes
2022-2025
Centre National de la Recherche Scientifique
2019-2024
CY Cergy Paris Université
2018-2024
Les Afriques dans le Monde
2023
Hôpital Avicenne
2023
Assistance Publique – Hôpitaux de Paris
2023
Laboratoire d’Etudes du Rayonnement et de la Matière en Astrophysique et Atmosphères
2023
Université Paris-Seine
2018-2019
Institut de la Vision
2018
Abstract Several Diptera species are known to transmit pathogens of medical and veterinary interest. However, identifying these using conventional methods can be time-consuming, labor-intensive, or expensive. A computer vision-based system that uses Wing interferential patterns (WIPs) identify insects could solve this problem. This study introduces a dataset for training evaluating recognition dipteran importance WIPs. The includes pictures Culicidae, Calliphoridae, Muscidae, Tabanidae,...
<title>Abstract</title> In this paper, we test the possibility of using Wing Interference Patterns (WIPs) and deep learning (DL) for identification Culex mosquitoes species to evaluate extent which a generic method could be developed surveying Dipteran insects major importance human health. Previous applications WIPs DL have successfully demonstrated their utility in identifying Anopheles, Aedes, sandflies, tsetse flies, providing rationale extending approach Culex. Accurate these is crucial...
We present a new and innovative identification method based on deep learning of the wing interferential patterns carried by mosquitoes Anopheles genus to classify assign 20 species, including 13 malaria vectors. provide additional evidence that this approach can identify spp. with an accuracy up 100% for ten out species. Although, was moderate (> 65%) or weak (50%) three seven The process discriminate cryptic sibling species is also assessed belonging Gambiae complex. Strikingly, An....
Bio-inspired Event-Based (EB) cameras are a promising new technology that outperforms standard frame-based in extreme lighted and fast moving scenes. Already, number of EB corner detection techniques have been developed; however, the performance these detectors has only evaluated based on few author-selected criteria rather than unified common basis, as proposed here. Moreover, their experimental conditions mainly limited to less interesting operational regions camera (on which can also...
Abstract A simple method for accurately identifying Glossina spp in the field is a challenge to sustain future elimination of Human African Trypanosomiasis (HAT) as public health scourge, well sustainable management Animal (AAT). Current methods species identification heavily rely on few well-trained experts. Methodologies that molecular methodologies like DNA barcoding or mass spectrometry protein profiling (MALDI TOFF) haven’t been thoroughly investigated sp. Nevertheless, because they are...
Hematophagous insects belonging to the Aedes genus are proven vectors of viral and filarial pathogens medical interest. albopictus is an increasingly important vector because its rapid worldwide expansion. In context global climate change emergence zoonotic infectious diseases, identification tools with field application required strengthen efforts in entomological survey arthropods Large scales proactive surveys mosquitoes need skilled technicians and/or costly technical equipment, further...
We present a technique for the multi-sensor registration of featureless datasets based on photogrammetric tracking acquisition systems in use. This method is developed situ study cultural heritage objects and tested by digitizing small canvas successively with 3D digitization system multispectral camera while simultaneously four cameras using cubic target frame side length 500 mm. The achieved accuracy better than 0.03 mm spatially 0.150 mrad angularly. allows us to seamlessly register...
La vidéocapsule endoscopique du grêle (VCE-G) est un examen de première ligne dans l'exploration non-invasive l'intestin grêle. Avec nombre moyen 50000 images par examen, l'interprétation d'un enregistrement reste fastidieux et chronophage (en moyenne 30 à 40 minutes vidéo), source potentielle lésions manquées. Aussi, le développement programmes détection assisté ordinateur (DAO) enjeu majeur recherche en VCE-G. création d'une banque d'images vidéo VCE-G, ouvertes la communauté des...
This paper introduces a color asynchronous neuromorphic event-based camera and methodology to process output from the device perform segmentation tracking at native temporal resolution of sensor (down one microsecond). Our vision prototype is combination three Asynchronous Time-based Image Sensors, sensitive absolute information. We devise processing algorithm leveraging this It designed be computationally cheap, thus showing how low level benefits acquisition high data. The resulting...
The asynchronous time-based neuromorphic image sensor ATIS is an array of autonomously operating pixels able to encode luminance information with exceptionally high dynamic range (143 dB). This paper introduces event-based methodology display data from this type imagers, taking into account the large and temporal accuracy that go beyond available mainstream technologies. We introduce tone mapping for asynchronously acquired time encoded gray-level data. A global a local operator are...
Luminescence multispectral imaging is a developing and promising technique in the fields of conservation science cultural heritage studies. In this article, we present new methodology for recording spatially resolved luminescence properties objects. This relies on development lab-made camera setup optimized to collect low-yield images. addition classic data preprocessing procedure reduce noise data, an innovative method, based neural network algorithm, that allows us obtain radiometrically...
To characterize the growth of brain organoids (BOs), cultures that replicate some early physiological or pathological developments human are usually manually extracted. Due to their novelty, only small datasets these images available, but segmenting organoid shape automatically with deep learning (DL) tools requires a larger number images. Light U-Net segmentation architectures, which reduce training time while increasing sensitivity under input datasets, have recently emerged. We further...
Sandflies (Diptera; Psychodidae) are medical and veterinary vectors that transmit diverse parasitic, viral, bacterial pathogens. Their identification has always been challenging, particularly at the specific sub-specific levels, because it relies on examining minute mostly internal structures. Here, to circumvent such limitations, we have evaluated accuracy reliability of Wing Interferential Patterns (WIPs) generated surface sandfly wings in conjunction with deep learning (DL) procedures...
Real-time monitoring of hematophagous diptera (such as mosquitoes) populations in the field is a crucial challenge to foresee vaccination campaigns and restrain potential diseases spreading. However, current methods heavily rely on costly DNA extraction which destructive, costly, time consuming requires experts. The contributions this work are: 1) usage new type imaging, named Wing Interference Patterns (WIPs), non-destructive easier produce during experiments; 2) deep learning architecture...
Selective attention is an essential mechanism to filter sensory input and select only its most important components, allowing the capacity-limited cognitive structures of brain process them in detail. The saliency map model, originally developed understand selective primate visual system, has also been extensively used computer vision. Due wide-spread use frame-based video, this how dynamic from non-stationary scenes commonly implemented maps. However, temporal structure modality very...
Event cameras offer unparalleled advantages for real-time perception in dynamic environments, thanks to their microsecond-level temporal resolution and asynchronous operation. Existing event-based object detection methods, however, are limited by fixed-frequency paradigms fail fully exploit the high-temporal adaptability of event cameras. To address these limitations, we propose FlexEvent, a novel camera framework that enables at arbitrary frequencies. Our approach consists two key...