Hector Bedon
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Hector Bedon has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Miguel Chicchon
Universidad Tecnológica del Perú, Lima, Peru
4 shared publications - Billy Grados
Universidad de Lima
3 shared publications - Carlos Miguel
Universidad Politécnica de Madrid
2 shared publications - Ramón Alcarria
Universidad Politécnica de Madrid
2 shared publications - A. Fernandez Del Campo
Department of Telematic Systems Engineering, Universidad Politecnica de Madrid, Madrid 28040, Spain
1 shared publication Carlos Miguel Nieto
Telematics Engineering Department, Technical University of Madrid (UPM), Spain
1 shared publication
Latest Publications
- 20 December 2023
- Published by Springer Nature
- p. 165-179
QuinuaSmartApp: A Real-Time Agriculture Precision IoT Cloud Platform to Crops Monitoring
- 20 December 2023
- Published by Springer Nature
- p. 165-179
QuinuaSmartApp: A Real-Time Agriculture Precision IoT Cloud Platform to Crops Monitoring
- Open Access
- 28 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 33652-33665
Semantic Segmentation of Fish and Underwater Environments Using Deep Convolutional Neural Networks and Learned Active Contours
The conservation of marine resources requires constant monitoring of the underwater environment by researchers. For this purpose, visual automated monitoring systems are of great interest, especially those that can describe the environment using semantic segmentation based on deep learning. Although they have been successfully used in several applications, such as biomedical ones, obtaining optimal results in underwater environments is still a challenge due to the heterogeneity of water and lighting conditions, and the scarcity of labeled datasets. Even more, the existing deep learning techniques oriented to semantic segmentation only provide low resolution results, lacking the enough spatial details for a high performance monitoring. To address these challenges, a combined loss function based on the active contour theory and level set methods is proposed to refine the spatial segmentation resolution and quality. To evaluate the method, a new underwater dataset with pixel annotations for three classes (fish, seafloor, and water) was created using images from publicly accessible datasets like SUIM, RockFish, and DeepFish. The performance of architectures of convolutional neural networks (CNNs), such as UNet and DeepLabV3+, trained with different loss functions (cross entropy, dice, and active contours) was compared, finding that the proposed combined loss function improved the segmentation results by around 3%, both in the metric Intercept Over Union (IoU) as in Hausdorff Distance (HD).
- 28 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 33652-33665
Open AccessSemantic Segmentation of Fish and Underwater Environments Using Deep Convolutional Neural Networks and Learned Active Contours
The conservation of marine resources requires constant monitoring of the underwater environment by researchers. For this purpose, visual automated monitoring systems are of great interest, especially those that can describe the environment using semantic segmentation based on deep learning. Although they have been successfully used in several applications, such as biomedical ones, obtaining optimal results in underwater environments is still a challenge due to the heterogeneity of water and lighting conditions, and the scarcity of labeled datasets. Even more, the existing deep learning techniques oriented to semantic segmentation only provide low resolution results, lacking the enough spatial details for a high performance monitoring. To address these challenges, a combined loss function based on the active contour theory and level set methods is proposed to refine the spatial segmentation resolution and quality. To evaluate the method, a new underwater dataset with pixel annotations for three classes (fish, seafloor, and water) was created using images from publicly accessible datasets like SUIM, RockFish, and DeepFish. The performance of architectures of convolutional neural networks (CNNs), such as UNet and DeepLabV3+, trained with different loss functions (cross entropy, dice, and active contours) was compared, finding that the proposed combined loss function improved the segmentation results by around 3%, both in the metric Intercept Over Union (IoU) as in Hausdorff Distance (HD).
- Open Access
- 26 April 2022
- Published by Universidad de Lima in Ingeniería Industrial
- p. 297-330
Proyección climática para el periodo 2006-2075 para el valle de Jauja, simulada por la intercomparación de modelos acoplados CSIRO Mk 3.0, MIROC-ESM y CNRM de fase 5 (CMIP5)
Se evaluaron los datos climáticos para el valle de Jauja, región Junín, sierra central del Perú para el siglo xxi, como lo simulan los modelos climáticos utilizados en la fase 5 por la intercomparación de modelos acoplados (CMP5). Los modelos utilizados fueron tres: CSIRO Mk 3.6, MIROC-ESM y CNRM-CM5, respectivamente, a una resolución espacial de 1,875 × 1,86°, 2,81 × 1,87° y 1,41 × 1,40°, construidos con data meteorológica observada en el valle de Jauja durante el periodo 1975-2005, empleando el escenario RCP2.6 y 8.5 del IPCC. El objetivo del presente trabajo es proporcionar proyecciones climáticas locales para esta área, generando una primera base de datos climáticos futura de la región, como herramienta de toma de decisiones por parte de agricultores y otros usuarios de la cuenca. Las proyecciones climáticas muestran un calentamiento significativo desde 2,0 grados de temperatura para el escenario RCP2.6 hasta 3,5 grados de temperatura para el escenario RCP8.5, esperado en toda el área evaluada del valle de Jauja para los próximos cincuenta años, junto con una disminución de la precipitación. Las proyecciones de precipitación dependen de la resolución horizontal, lo que sugiere la necesidad de simulaciones adicionales con una resolución más alta, especialmente para una representación adecuada de eventos climáticos extremos.
- 26 April 2022
- Published by Universidad de Lima in Ingeniería Industrial
- p. 297-330
Open AccessProyección climática para el periodo 2006-2075 para el valle de Jauja, simulada por la intercomparación de modelos acoplados CSIRO Mk 3.0, MIROC-ESM y CNRM de fase 5 (CMIP5)
Se evaluaron los datos climáticos para el valle de Jauja, región Junín, sierra central del Perú para el siglo xxi, como lo simulan los modelos climáticos utilizados en la fase 5 por la intercomparación de modelos acoplados (CMP5). Los modelos utilizados fueron tres: CSIRO Mk 3.6, MIROC-ESM y CNRM-CM5, respectivamente, a una resolución espacial de 1,875 × 1,86°, 2,81 × 1,87° y 1,41 × 1,40°, construidos con data meteorológica observada en el valle de Jauja durante el periodo 1975-2005, empleando el escenario RCP2.6 y 8.5 del IPCC. El objetivo del presente trabajo es proporcionar proyecciones climáticas locales para esta área, generando una primera base de datos climáticos futura de la región, como herramienta de toma de decisiones por parte de agricultores y otros usuarios de la cuenca. Las proyecciones climáticas muestran un calentamiento significativo desde 2,0 grados de temperatura para el escenario RCP2.6 hasta 3,5 grados de temperatura para el escenario RCP8.5, esperado en toda el área evaluada del valle de Jauja para los próximos cincuenta años, junto con una disminución de la precipitación. Las proyecciones de precipitación dependen de la resolución horizontal, lo que sugiere la necesidad de simulaciones adicionales con una resolución más alta, especialmente para una representación adecuada de eventos climáticos extremos.
- 26 October 2021
- Published by Springer Nature
- p. 301-309
Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning
- 26 October 2021
- Published by Springer Nature
- p. 301-309
Semantic Segmentation of Underwater Environments Using DeepLabv3+ and Transfer Learning
- 26 October 2021
- Published by Springer Nature
- p. 229-237
Control System to International Parental Child Abduction, Case: Colombia
- 26 October 2021
- Published by Springer Nature
- p. 229-237
Control System to International Parental Child Abduction, Case: Colombia
- 1 April 2021
- Published by Springer Nature
- p. 272-286
Is It Intelligent? A Systematic Review of Intelligence in the Most Cited Papers in IoT
- 1 April 2021
- Published by Springer Nature
- p. 272-286
Is It Intelligent? A Systematic Review of Intelligence in the Most Cited Papers in IoT
- 1 April 2021
- Published by Springer Nature
- p. 3-17
Marine Delay and Disruption Tolerant Networks (MaDTN): Application for Artisanal Fisheries
- 1 April 2021
- Published by Springer Nature
- p. 3-17
Marine Delay and Disruption Tolerant Networks (MaDTN): Application for Artisanal Fisheries
- 3 March 2020
- Published by Springer Nature
- p. 473-485
Semantic Segmentation of Weeds and Crops in Multispectral Images by Using a Convolutional Neural Networks Based on U-Net
- 3 March 2020
- Published by Springer Nature
- p. 473-485
Semantic Segmentation of Weeds and Crops in Multispectral Images by Using a Convolutional Neural Networks Based on U-Net
- 1 July 2016
- Published by Elsevier in Ad Hoc Networks
- Vol. 44, 76-89
Message fragmentation assessment in DTN nanosatellite-based sensor networks
- 1 July 2016
- Published by Elsevier in Ad Hoc Networks
- Vol. 44, 76-89
Message fragmentation assessment in DTN nanosatellite-based sensor networks