Mathieu Varin
Research Subjects
Mathieu Varin has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Bilel Chalghaf
Centre d’enseignement et de recherche en foresterie de Sainte-Foy (CERFO), Quebec City, Canada
2 shared publications François Milord
2 shared publications- Jade Savage
Department of Biology and Biochemistry, Bishop’s University, Sherbrooke, Québec, Canada
2 shared publications - Jérôme Théau
Université de Sherbrooke
2 shared publications - Kirsten E. Crandall
Affiliation Unité Maladies Infectieuses en communauté, Direction des risques biologiques, Institut national de santé publique du Québec, Quebec
2 shared publications - Virginie Millien
Department of Biology, McGill University, Montreal, Canada
2 shared publications
Latest Publications
- Open Access
- 25 June 2026
- Published by Wiley in Remote Sensing in Ecology and Conservation
Fusion or Confusion? Assessing the Impact of Visible‐Thermal Image Fusion for Automated Wildlife Detection
Efficient wildlife monitoring methods are necessary for biodiversity conservation and management. The combination of remote sensing, aerial imagery, and deep learning offers promising opportunities to renew or improve existing survey methods. The complementary use of visible (VIS) and thermal infrared (TIR) imagery can add information compared to a single‐source image and improve results in an automated detection context. However, the alignment and fusion process can be challenging, especially since visible and thermal images usually have different fields of view (FOV) and spatial resolutions. This research presents a case study on the great blue heron ( Ardea herodias ) to evaluate the performances of synchronous aerial VIS and TIR imagery to automatically detect individuals and nests using a YOLO11n model. Two VIS–TIR fusion methods were tested and compared: an early fusion approach and a late fusion approach to determine if the addition of the TIR image gives any added value compared to a VIS‐only model. VIS and TIR images were automatically aligned using a deep learning model. A principal component analysis fusion method was applied to VIS–TIR image pairs to form the early fusion dataset. A classification and regression tree was used to process the late fusion dataset based on the detection from the VIS‐only and TIR‐only trained models. Across all classes, both late and early fusion improved the F1 score compared to the VIS‐only model. For the main class, occupied nest, the late fusion improved the F1 score from 90.2 (VIS‐only) to 93.0%. This model was also able to identify false positives from both sources with 90% recall. Although fusion methods seem to give better results, this approach comes with a limiting TIR FOV and alignment constraints that eliminate data. Using an aircraft‐mounted very high‐resolution visible sensor (wide FOV) could be an interesting compromise for operationalizing surveys.
- 25 June 2026
- Published by Wiley in Remote Sensing in Ecology and Conservation
Open AccessFusion or Confusion? Assessing the Impact of Visible‐Thermal Image Fusion for Automated Wildlife Detection
Efficient wildlife monitoring methods are necessary for biodiversity conservation and management. The combination of remote sensing, aerial imagery, and deep learning offers promising opportunities to renew or improve existing survey methods. The complementary use of visible (VIS) and thermal infrared (TIR) imagery can add information compared to a single‐source image and improve results in an automated detection context. However, the alignment and fusion process can be challenging, especially since visible and thermal images usually have different fields of view (FOV) and spatial resolutions. This research presents a case study on the great blue heron ( Ardea herodias ) to evaluate the performances of synchronous aerial VIS and TIR imagery to automatically detect individuals and nests using a YOLO11n model. Two VIS–TIR fusion methods were tested and compared: an early fusion approach and a late fusion approach to determine if the addition of the TIR image gives any added value compared to a VIS‐only model. VIS and TIR images were automatically aligned using a deep learning model. A principal component analysis fusion method was applied to VIS–TIR image pairs to form the early fusion dataset. A classification and regression tree was used to process the late fusion dataset based on the detection from the VIS‐only and TIR‐only trained models. Across all classes, both late and early fusion improved the F1 score compared to the VIS‐only model. For the main class, occupied nest, the late fusion improved the F1 score from 90.2 (VIS‐only) to 93.0%. This model was also able to identify false positives from both sources with 90% recall. Although fusion methods seem to give better results, this approach comes with a limiting TIR FOV and alignment constraints that eliminate data. Using an aircraft‐mounted very high‐resolution visible sensor (wide FOV) could be an interesting compromise for operationalizing surveys.
- Open Access
- 1 January 2026
- Published by Institut de recherche Robert-Sauvé en santé et en sécurité du travail
Modélisation spatiotemporelle de la maladie de Lyme : évaluation du risque d'exposition chez les travailleurs en plein air au Québec
La maladie de Lyme, transmise par la tique à pattes noires, devient un problème de plus en plus important au Québec. Depuis 2009, les cas humains ont fortement augmenté, surtout dans le sud de la province. Les changements climatiques et les transformations du territoire issues des activités humaines favorisent la présence des tiques et de leurs hôtes, ce qui augmente le risque pour les personnes qui travaillent à l’extérieur. Ce projet avait pour but de mieux comprendre où et quand ce risque est le plus élevé afin d’aider à protéger les travailleurs. Pour y arriver, nous avons combiné : des données de surveillance des tiques (collectées par des experts et par le public); des informations sur l’environnement (climat, utilisation du sol, paysages); et des données sur les comportements humains (connaissances et mesures de protection). Nous avons utilisé des modèles mathématiques pour prédire la présence des tiques, de leurs hôtes et de la bactérie responsable de la maladie de Lyme. Ces prédictions ont été faites pour aujourd’hui et pour 2050, en tenant compte de deux scénarios d’émission de gaz à effet de serre. Les résultats ont permis de créer des cartes très détaillées qui montrent les zones les plus à risque pour les travailleurs. Ce qu’il faut retenir : Le nombre de cas augmente rapidement au Québec. Les zones les plus à risque se trouvent dans le sud (Montérégie, Estrie, Centre-du-Québec), mais pourraient s’étendre vers le nord et vers l’est. Une carte interactive a été créée pour aider à planifier les mesures de prévention. Pourquoi est-ce important ? Ces outils permettent d’anticiper les risques et de mieux protéger les travailleurs en milieu naturel.
- 1 January 2026
- Published by Institut de recherche Robert-Sauvé en santé et en sécurité du travail
Open AccessModélisation spatiotemporelle de la maladie de Lyme : évaluation du risque d'exposition chez les travailleurs en plein air au Québec
La maladie de Lyme, transmise par la tique à pattes noires, devient un problème de plus en plus important au Québec. Depuis 2009, les cas humains ont fortement augmenté, surtout dans le sud de la province. Les changements climatiques et les transformations du territoire issues des activités humaines favorisent la présence des tiques et de leurs hôtes, ce qui augmente le risque pour les personnes qui travaillent à l’extérieur. Ce projet avait pour but de mieux comprendre où et quand ce risque est le plus élevé afin d’aider à protéger les travailleurs. Pour y arriver, nous avons combiné : des données de surveillance des tiques (collectées par des experts et par le public); des informations sur l’environnement (climat, utilisation du sol, paysages); et des données sur les comportements humains (connaissances et mesures de protection). Nous avons utilisé des modèles mathématiques pour prédire la présence des tiques, de leurs hôtes et de la bactérie responsable de la maladie de Lyme. Ces prédictions ont été faites pour aujourd’hui et pour 2050, en tenant compte de deux scénarios d’émission de gaz à effet de serre. Les résultats ont permis de créer des cartes très détaillées qui montrent les zones les plus à risque pour les travailleurs. Ce qu’il faut retenir : Le nombre de cas augmente rapidement au Québec. Les zones les plus à risque se trouvent dans le sud (Montérégie, Estrie, Centre-du-Québec), mais pourraient s’étendre vers le nord et vers l’est. Une carte interactive a été créée pour aider à planifier les mesures de prévention. Pourquoi est-ce important ? Ces outils permettent d’anticiper les risques et de mieux protéger les travailleurs en milieu naturel.
- Open Access
WITHDRAWN: Monitoring Lyme disease cases in the context of rapid emergence in Quebec, Canada
Withdrawal Statement: The authors have withdrawn this manuscript because there is a funder embargo on the results presented in this work. Therefore, the authors do not wish this work to be cited as reference for the project. If you have any questions, please contact the corresponding author.
Open AccessWITHDRAWN: Monitoring Lyme disease cases in the context of rapid emergence in Quebec, Canada
Withdrawal Statement: The authors have withdrawn this manuscript because there is a funder embargo on the results presented in this work. Therefore, the authors do not wish this work to be cited as reference for the project. If you have any questions, please contact the corresponding author.
- Open Access
- 5 July 2024
- Published by Wiley in Zoonoses and Public Health
- Vol. 71 (7) , 855-867
Outdoor worker knowledge of ticks and Lyme disease in Québec
Background and Aims: Lyme disease is a well‐known occupational risk across North America caused by exposure to Borrelia burgdorferi via blacklegged ticks (Ixodes scapularis). As the geographic range of B. burgdorferi advances with the increasing distribution of blacklegged ticks, more outdoor workers are at risk of contracting Lyme disease. In this study, we examined the demography and personal perceptions of outdoor workers within one framework to better determine the overall risk for those working outdoors.Methods and Results: We analysed outdoor worker knowledge of ticks and of behaviours that can prevent tick bites and Lyme disease. We then compared these risk perceptions of individuals across age, sex, education, and industry, as well as time spent outdoors. We tested the hypothesis that the risk perception of an individual and their knowledge of Lyme disease transmission was dependent on their demographics, experience in their job, and the region in which they spend time outdoors. We estimated a knowledge‐based risk score based on individuals' answers to a questionnaire on risk perception given to voluntary participants who work outdoors. Those who had higher risk scores were more at risk. We found that knowledge‐based risk scores were correlated with geographic risk levels and with the number of hours per day spent outdoors. Those who work longer hours and who work in areas with mid‐level risk had higher risk scores. Those who spend more time outdoors recreationally had lower risk scores.Conclusions: Further examination and acknowledgment of the reasoning behind why these factors are affecting workers' risks must be considered to recognize that it is not necessarily demographics or geographically high‐risk areas that affect an individual's risk. Workers' knowledge of these risks is affected by several variables that should be taken into consideration when implementing safety measures and awareness programs.
- 5 July 2024
- Published by Wiley in Zoonoses and Public Health
- Vol. 71 (7) , 855-867
Open AccessOutdoor worker knowledge of ticks and Lyme disease in Québec
Background and Aims: Lyme disease is a well‐known occupational risk across North America caused by exposure to Borrelia burgdorferi via blacklegged ticks (Ixodes scapularis). As the geographic range of B. burgdorferi advances with the increasing distribution of blacklegged ticks, more outdoor workers are at risk of contracting Lyme disease. In this study, we examined the demography and personal perceptions of outdoor workers within one framework to better determine the overall risk for those working outdoors.Methods and Results: We analysed outdoor worker knowledge of ticks and of behaviours that can prevent tick bites and Lyme disease. We then compared these risk perceptions of individuals across age, sex, education, and industry, as well as time spent outdoors. We tested the hypothesis that the risk perception of an individual and their knowledge of Lyme disease transmission was dependent on their demographics, experience in their job, and the region in which they spend time outdoors. We estimated a knowledge‐based risk score based on individuals' answers to a questionnaire on risk perception given to voluntary participants who work outdoors. Those who had higher risk scores were more at risk. We found that knowledge‐based risk scores were correlated with geographic risk levels and with the number of hours per day spent outdoors. Those who work longer hours and who work in areas with mid‐level risk had higher risk scores. Those who spend more time outdoors recreationally had lower risk scores.Conclusions: Further examination and acknowledgment of the reasoning behind why these factors are affecting workers' risks must be considered to recognize that it is not necessarily demographics or geographically high‐risk areas that affect an individual's risk. Workers' knowledge of these risks is affected by several variables that should be taken into consideration when implementing safety measures and awareness programs.
- 26 February 2021
- Published by Springer Nature in Wetlands
- Vol. 41 (3) , 1-15
Mapping Vernal Pools Using LiDAR Data and Multitemporal Satellite Imagery
- 26 February 2021
- Published by Springer Nature in Wetlands
- Vol. 41 (3) , 1-15
Mapping Vernal Pools Using LiDAR Data and Multitemporal Satellite Imagery
- Open Access
- 21 September 2020
- Published by MDPI AG in Remote Sensing
- Vol. 12 (18) , 3092
Object-Based Approach Using Very High Spatial Resolution 16-Band WorldView-3 and LiDAR Data for Tree Species Classification in a Broadleaf Forest in Quebec, Canada
Species identification in Quebec, Canada, is usually performed with photo-interpretation at the stand level, and often results in a lack of precision which affects forest management. Very high spatial resolution imagery, such as WorldView-3 and Light Detection and Ranging have the potential to overcome this issue. The main objective of this study is to map 11 tree species at the tree level using an object-based approach. For modeling, 240 variables were derived from WorldView-3 with pixel-based and arithmetic feature calculation techniques. A global approach (11 species) was compared to a hierarchical approach at two levels: (1) tree type (broadleaf/conifer) and (2) individual broadleaf (five) and conifer (six) species. Five different model techniques were compared: support vector machine, classification and regression tree, random forest (RF), k-nearest neighbors, and linear discriminant analysis. Each model was assessed using 16-band or first 8-band derived variables, with the results indicating higher precision for the RF technique. Higher accuracies were found using 16-band instead of 8-band derived variables for the global approach (overall accuracy (OA): 75% vs. 71%, Kappa index of agreement (KIA): 0.72 vs. 0.67) and tree type level (OA: 99% vs. 97%, KIA: 0.97 vs. 0.95). For broadleaf individual species, higher accuracy was found using first 8-band derived variables (OA: 70% vs. 68%, KIA: 0.63 vs. 0.60). No distinction was found for individual conifer species (OA: 94%, KIA: 0.93). This paper demonstrates that a hierarchical classification approach gives better results for conifer species and that using an 8-band WorldView-3 instead of a 16-band is sufficient.
- 21 September 2020
- Published by MDPI AG in Remote Sensing
- Vol. 12 (18) , 3092
Open AccessObject-Based Approach Using Very High Spatial Resolution 16-Band WorldView-3 and LiDAR Data for Tree Species Classification in a Broadleaf Forest in Quebec, Canada
Species identification in Quebec, Canada, is usually performed with photo-interpretation at the stand level, and often results in a lack of precision which affects forest management. Very high spatial resolution imagery, such as WorldView-3 and Light Detection and Ranging have the potential to overcome this issue. The main objective of this study is to map 11 tree species at the tree level using an object-based approach. For modeling, 240 variables were derived from WorldView-3 with pixel-based and arithmetic feature calculation techniques. A global approach (11 species) was compared to a hierarchical approach at two levels: (1) tree type (broadleaf/conifer) and (2) individual broadleaf (five) and conifer (six) species. Five different model techniques were compared: support vector machine, classification and regression tree, random forest (RF), k-nearest neighbors, and linear discriminant analysis. Each model was assessed using 16-band or first 8-band derived variables, with the results indicating higher precision for the RF technique. Higher accuracies were found using 16-band instead of 8-band derived variables for the global approach (overall accuracy (OA): 75% vs. 71%, Kappa index of agreement (KIA): 0.72 vs. 0.67) and tree type level (OA: 99% vs. 97%, KIA: 0.97 vs. 0.95). For broadleaf individual species, higher accuracy was found using first 8-band derived variables (OA: 70% vs. 68%, KIA: 0.63 vs. 0.60). No distinction was found for individual conifer species (OA: 94%, KIA: 0.93). This paper demonstrates that a hierarchical classification approach gives better results for conifer species and that using an 8-band WorldView-3 instead of a 16-band is sufficient.
- 1 September 2019
- Published by Elsevier in Journal of Environmental Management
- Vol. 246, 334-344
Mapping ecosystem services provided by wetlands at multiple spatiotemporal scales: A case study in Quebec, Canada
- 1 September 2019
- Published by Elsevier in Journal of Environmental Management
- Vol. 246, 334-344
Mapping ecosystem services provided by wetlands at multiple spatiotemporal scales: A case study in Quebec, Canada
- 27 February 2015
- Published by Springer Nature in Wetlands Ecology and Management
- Vol. 23 (4) , 707-725
Meta-analysis for the transfer of economic benefits of ecosystem services provided by wetlands within two watersheds in Quebec, Canada
- 27 February 2015
- Published by Springer Nature in Wetlands Ecology and Management
- Vol. 23 (4) , 707-725
Meta-analysis for the transfer of economic benefits of ecosystem services provided by wetlands within two watersheds in Quebec, Canada