YOLOv10과 RGB 이미지를 이용한 수목 탐지 가능성 검토
Individual Tree Detection in Various Vegetation Areas Using YOLOv10 and RGB Images
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This study utilizes the YOLOv10 model, a deeplearning-based object detection algorithm, to evaluate the performance of individual tree detection using RGB images in different vegetation types. Forests and orchards play an important role in ecosystems, and appropriate monitoring of individual trees is essential for forest management. The YOLOv10, an AI model for real-time object detection, was used to perform individual tree detection in a variety of environments in high-resolution aerial images. Data augmentation techniques and transfer learning were applied to improve the model performance. In our experiments, we achieved a mean average precision (mAP) of 0.737, precision of 0.777, and recall of 0.638 on the test dataset. However, the results suggest the difficulty of detecting individual trees in dense vegetation. This study shows the potential of YOLOv10 and the limitation of the current AI datasets for tree detection. The need for further algorithmic improvements and data diversity for real-world applications should be emphasized.
Abstract
This study utilizes the YOLOv10 model, a deeplearning-based object detection algorithm, to evaluate the performance of individual tree detection using RGB images in different vegetation types. Forests and orchards play an important role in ecosystems, and appropriate monitoring of individual trees is essential for forest management. The YOLOv10, an AI model for real-time object detection, was used to perform individual tree detection in a variety of environments in high-resolution aerial images. Data augmentation techniques and transfer learning were applied to improve the model performance. In our experiments, we achieved a mean average precision (mAP) of 0.737, precision of 0.777, and recall of 0.638 on the test dataset. However, the results suggest the difficulty of detecting individual trees in dense vegetation. This study shows the potential of YOLOv10 and the limitation of the current AI datasets for tree detection. The need for further algorithmic improvements and data diversity for real-world applications should be emphasized.
참고문헌 (24)
- Culman, M., "Individual palm tree detection using deep learning on RGB imagery to support tree inventory", Remote Sensing, 12(21), pp.3476~ (2020)
- Diwan, T., "Object detection using YOLO : challenges, architectural successors, datasets and applications", Multimedia Tools and Applications, 82(6), pp.9243~9275 (2023)
- Hwang, S. R., "Detection of individual trees and estimation of mean tree height using airborne LIDAR data", Spatial Information Research, 20(3), pp.27~38 (2012)
- Jaskierniak, D., "Individual tree detection and crown delineation from Unmanned Aircraft System(UAS)LiDAR in structurally complex mixed species eucalypt forests", ISPRS Journal of Photogrammetry and Remote Sensing, 171, pp.171~187 (2021)
- Jiang, P., "A review of Yolo algorithm developments", Procedia Computer Science, 199, pp.1066~1073 (2022)
- Kanerva, H., "Estimating tree health decline caused by Ips typographus L. from UAS RGB images using a deep one-stage object detection neural network", Remote Sensing, 14(24), pp.6257~ (2022)
- Lausch, A., "Understanding forest health with remote sensing, part III : requirements for a scalable multi-source forest health monitoring network based on data science approaches", Remote sensing, 10(7), pp.1120~ (2018)
- Liu, S., "Path aggregation network for instance segmentation", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.8759~8768 (2018)
- Ok, A. O., "Detection of citrus trees from UAV DSMs, ISPRS Annals of the Photogrammetry", Remote Sensing and Spatial Information Sciences, 4, pp.27~34 (2017)
- Rodríguez-Puerta, F., "UAV-based LiDAR scanning for individual tree detection and height measurement in young forest permanent trials", Remote Sensing, 14(1), pp.170~ (2021)
- (2023)[URL]
- Solares-Canal, A., "Automatic tree detection and attribute characterization using portable terrestrial lidar", Trees, 37(3), pp.963~979 (2023)
- Treder, W., "Relationships between yield, crop density coefficient and average fruit weight in ‘Lobo’ apple trees under various planting systems and irrigation treatments", HortTechnology, 11(2), pp.248~254 (2001)
- Trumbore, S., "Forest health and global change", Science, 349(6250), pp.814~818 (2015)
- Ventura, J., "Individual tree detection in large-scale urban environments using high-resolution multispectral imagery", International Journal of Applied Earth Observation and Geoinformation, 130, pp.103848~ (2024)
- Vijayakumar, A., "Yolo-based object detection models: a review and its applications", Multimedia Tools and Applications (2024)[DOI]
- Weinstein, B. G., "A benchmark dataset for canopy crown detection and delineation in co-registered airborne RGB, LiDAR and hyperspectral imagery from the National Ecological Observation Network", PLoS computational biology, 17(7), pp.e1009180~ (2021)
- Weinstein, B. G., "Cross-site learning in deep learning RGB tree crown detection", Ecological Informatics, 56, pp.101061~ (2020)
- Xia, L., "Evaluation of deep learning segmentation models for detection of pine wilt disease in unmanned aerial vehicle images", Remote Sensing, 13(18), pp.3594~ (2021)
- Zhong, H., "Individual tree species identification for complex coniferous and broad-leaved mixed forests based on deep learning combined with UAV LiDAR data and RGB images", Forests, 15(2), pp.293~ (2024)