Min-Young Hwang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Min-Young Hwang has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Ki-Dong Park
Department of Smart Agriculture Engineering, Sunchon National University, Suncheon 57922, Republic of Korea
3 shared publications - Ryu-Gap Lim
Department of Convergent Biosystems Engineering, Sunchon National University, Suncheon 57922, Republic of Korea
3 shared publications - Seung-Yong Seo
Department of Smart Agriculture Engineering, Sunchon National University, Suncheon 57922, Republic of Korea
2 shared publications - Deok-Hyeon Ko
Department of AI Mechanical and Control Engineering, Kyung Hee Cyber University, Seoul 02447, Republic of Korea
1 shared publication - Ho-Ah Lee
Department of Agriculture Life Bio System, Korea Polytechnic College, Kimje 54352, Republic of Korea
1 shared publication - Jae-Min Lee
Department of Convergence Biosystems Machinery Engineering, Sunchon National University, Suncheon 57922, Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 12 November 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (22) , 2484
Distance-Adaptive Depth Camera–LiDAR Fusion for Object Distance Estimation on a Moving Tractor: A Field Evaluation Against RTK-GNSS
Agricultural machinery has extensive blind zones, and reliable proximity alerts need distance information that no single sensor provides across the working range. We fused a depth camera and a 16-channel LiDAR by distance-adaptive inverse-variance weighting, with weights derived from nominal, specification-based error models that cross at 3.87 m. YOLO11m, trained on five classes, detected objects, and the pipeline ran at 26 frames per second on an edge computing platform. Distances to Person and Tractor targets were evaluated against real-time kinematic GNSS in 2521 samples from front and rear tractor approaches at nominal speeds of 3, 5, and 7 km/h. Fusion reduced the root mean square error (RMSE) to 0.211 m, 30.9% below that of the depth camera alone, with the largest reduction (35.8%) at 7–12 m and no change within 7 m. The LiDAR alone was more accurate (0.167 m) but valid in only 82.7% of frames. With both sensors valid, the fused RMSE (0.141 m) was lower than that of either sensor. For detected targets, alert levels agreed with the reference in 99.0% of frames, but a Person recall of 0.72 limits overall reliability. Target geometry and LiDAR resolution strongly affect the benefit of fusion, which may guide sensor selection for other agricultural machines.
- 12 November 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (22) , 2484
Open AccessDistance-Adaptive Depth Camera–LiDAR Fusion for Object Distance Estimation on a Moving Tractor: A Field Evaluation Against RTK-GNSS
Agricultural machinery has extensive blind zones, and reliable proximity alerts need distance information that no single sensor provides across the working range. We fused a depth camera and a 16-channel LiDAR by distance-adaptive inverse-variance weighting, with weights derived from nominal, specification-based error models that cross at 3.87 m. YOLO11m, trained on five classes, detected objects, and the pipeline ran at 26 frames per second on an edge computing platform. Distances to Person and Tractor targets were evaluated against real-time kinematic GNSS in 2521 samples from front and rear tractor approaches at nominal speeds of 3, 5, and 7 km/h. Fusion reduced the root mean square error (RMSE) to 0.211 m, 30.9% below that of the depth camera alone, with the largest reduction (35.8%) at 7–12 m and no change within 7 m. The LiDAR alone was more accurate (0.167 m) but valid in only 82.7% of frames. With both sensors valid, the fused RMSE (0.141 m) was lower than that of either sensor. For detected targets, alert levels agreed with the reference in 99.0% of frames, but a Person recall of 0.72 limits overall reliability. Target geometry and LiDAR resolution strongly affect the benefit of fusion, which may guide sensor selection for other agricultural machines.
- Open Access
- 25 October 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (21) , 2329
GIS-Based Optimal Site Selection of Agricultural By-Product Resource Recovery Facilities for Smart Agricultural and Livestock Systems: A Case Study of Naju-si and Goheung-gun, Jeollanam-do, South Korea
The importance of producing bioenergy using agricultural by-products is increasing in response to energy depletion and climate change caused by dependence on fossil fuels. In particular, rice straw is one of the most abundant agricultural by-products generated in Korea and has high potential for utilization in various bioenergy resource recovery applications. The objective of this study was to estimate rice straw production and select optimal locations for agricultural by-product resource recovery facilities based on resource availability, facility processing capacity, and collection and transportation costs. Field measurements of soil properties were conducted to analyze their relationships with rice straw production, while rice cultivation and production statistics were obtained from the Korean Statistical Information Service (KOSIS), and digital map data were obtained from the National Geographic Information Institute (NGII). The soil-property data were statistically analyzed to evaluate their relationships with rice straw production, while the digital map data were processed using ArcGIS Pro to delineate paddy field areas, apply correction factors based on statistical rice cultivation areas, generate representative centroid locations, identify candidate sites using spatial exclusion criteria, and calculate road-network-based transportation distances and costs. However, since the economic feasibility of resource recovery facilities is greatly affected by the stable supply of raw materials and the efficiency of collection costs, optimal site selection considering resource generation and transportation characteristics is necessary. The suitability of rice cultivation was evaluated by investigating the soil moisture content, electrical conductivity (EC), penetration resistance, and soil texture of the study area, and the resource supply was estimated based on the rice cultivation area and rice straw generation in 2024. In addition, ArcGIS Pro 3.6.4 (Esri, Redlands, CA, USA) was used to derive candidate sites by considering conditions such as roads, slope, streams, rivers, lakes, and wetlands, and the optimal locations by facility scale were evaluated by analyzing the total logistics costs, including collection and transportation costs. The results showed that both study areas exhibited soil properties suitable for rice cultivation and were capable of providing a stable supply of rice straw. For the 30,000 ton processing facility, two optimal locations were selected in each of Naju-si and Goheung-gun, while one optimal location was selected in each region for the 60,000 ton processing facility. The GIS-based site selection method proposed in this study is expected to be applicable not only to rice straw resource recovery facilities but also to the planning and supply chain development of various agricultural biomass resource recovery facilities. It is also expected to contribute to the development of efficient systems for achieving sustainable agriculture and carbon neutrality.
- 25 October 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (21) , 2329
Open AccessGIS-Based Optimal Site Selection of Agricultural By-Product Resource Recovery Facilities for Smart Agricultural and Livestock Systems: A Case Study of Naju-si and Goheung-gun, Jeollanam-do, South Korea
The importance of producing bioenergy using agricultural by-products is increasing in response to energy depletion and climate change caused by dependence on fossil fuels. In particular, rice straw is one of the most abundant agricultural by-products generated in Korea and has high potential for utilization in various bioenergy resource recovery applications. The objective of this study was to estimate rice straw production and select optimal locations for agricultural by-product resource recovery facilities based on resource availability, facility processing capacity, and collection and transportation costs. Field measurements of soil properties were conducted to analyze their relationships with rice straw production, while rice cultivation and production statistics were obtained from the Korean Statistical Information Service (KOSIS), and digital map data were obtained from the National Geographic Information Institute (NGII). The soil-property data were statistically analyzed to evaluate their relationships with rice straw production, while the digital map data were processed using ArcGIS Pro to delineate paddy field areas, apply correction factors based on statistical rice cultivation areas, generate representative centroid locations, identify candidate sites using spatial exclusion criteria, and calculate road-network-based transportation distances and costs. However, since the economic feasibility of resource recovery facilities is greatly affected by the stable supply of raw materials and the efficiency of collection costs, optimal site selection considering resource generation and transportation characteristics is necessary. The suitability of rice cultivation was evaluated by investigating the soil moisture content, electrical conductivity (EC), penetration resistance, and soil texture of the study area, and the resource supply was estimated based on the rice cultivation area and rice straw generation in 2024. In addition, ArcGIS Pro 3.6.4 (Esri, Redlands, CA, USA) was used to derive candidate sites by considering conditions such as roads, slope, streams, rivers, lakes, and wetlands, and the optimal locations by facility scale were evaluated by analyzing the total logistics costs, including collection and transportation costs. The results showed that both study areas exhibited soil properties suitable for rice cultivation and were capable of providing a stable supply of rice straw. For the 30,000 ton processing facility, two optimal locations were selected in each of Naju-si and Goheung-gun, while one optimal location was selected in each region for the 60,000 ton processing facility. The GIS-based site selection method proposed in this study is expected to be applicable not only to rice straw resource recovery facilities but also to the planning and supply chain development of various agricultural biomass resource recovery facilities. It is also expected to contribute to the development of efficient systems for achieving sustainable agriculture and carbon neutrality.
- Open Access
- 24 October 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (21) , 2321
Combining Public and On-Farm Datasets for Deep Learning-Based Broiler Detection and Counting in a Commercial Broiler House
The high-density rearing environment in the domestic broiler industry causes birds to overlap and obscure one another, which is a major factor leading to reduced performance in object-recognition-based monitoring; consequently, the need for video-based automated analysis technology is growing. Accordingly, this study established an integrated video monitoring pipeline—ranging from object detection to individual counting—by combining public datasets with on-site CCTV datasets to ensure field adaptability, and by integrating heterogeneous detector fusion with multi-object tracking. A training dataset was constructed by combining the AI-Hub Intelligent Smart Poultry House Broiler Video Dataset with CCTV footage collected from a commercial broiler farm. The convolutional neural network-based YOLOv8m and Transformer-based RT-DETR-L were trained independently, and their predictions were subsequently fused using Weighted Boxes Fusion (WBF). ByteTrack was applied for tracking to maintain object identity across frames, and a virtual line crossing counting algorithm was implemented to automatically determine the points where tracking trajectories intersect with virtual lines set within the video, thereby calculating the cumulative number of objects by direction. The experimental results showed that, after training with field data, the model achieved a mean average precision at an intersection over union (IoU) threshold of 0.5 (mAP@0.5) of 0.9880 based on YOLOv8m, and the WBF fusion improved the mAP@0.5 by 26.4% compared to the best single model among the pairs trained solely on public data. The YOLOv8m model trained using combined data achieved an average absolute error of 6.33 birds and an accuracy of 97.60% in frame-level object counting, and recorded an F1-score of 0.856 in virtual line crossing counting. This study confirmed that combined training using public and field data is effective in mitigating domain gaps and that frame-level object counting can be utilized as an indicator for flock monitoring.
- 24 October 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (21) , 2321
Open AccessCombining Public and On-Farm Datasets for Deep Learning-Based Broiler Detection and Counting in a Commercial Broiler House
The high-density rearing environment in the domestic broiler industry causes birds to overlap and obscure one another, which is a major factor leading to reduced performance in object-recognition-based monitoring; consequently, the need for video-based automated analysis technology is growing. Accordingly, this study established an integrated video monitoring pipeline—ranging from object detection to individual counting—by combining public datasets with on-site CCTV datasets to ensure field adaptability, and by integrating heterogeneous detector fusion with multi-object tracking. A training dataset was constructed by combining the AI-Hub Intelligent Smart Poultry House Broiler Video Dataset with CCTV footage collected from a commercial broiler farm. The convolutional neural network-based YOLOv8m and Transformer-based RT-DETR-L were trained independently, and their predictions were subsequently fused using Weighted Boxes Fusion (WBF). ByteTrack was applied for tracking to maintain object identity across frames, and a virtual line crossing counting algorithm was implemented to automatically determine the points where tracking trajectories intersect with virtual lines set within the video, thereby calculating the cumulative number of objects by direction. The experimental results showed that, after training with field data, the model achieved a mean average precision at an intersection over union (IoU) threshold of 0.5 (mAP@0.5) of 0.9880 based on YOLOv8m, and the WBF fusion improved the mAP@0.5 by 26.4% compared to the best single model among the pairs trained solely on public data. The YOLOv8m model trained using combined data achieved an average absolute error of 6.33 birds and an accuracy of 97.60% in frame-level object counting, and recorded an F1-score of 0.856 in virtual line crossing counting. This study confirmed that combined training using public and field data is effective in mitigating domain gaps and that frame-level object counting can be utilized as an indicator for flock monitoring.