DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATION
Forest monitoring and tree species categorization has a vital importance in terms of biodiversity conservation, ecosystem health assessment, climate change mitigation, and sustainable resource management. Due to large-scale coverage of forest areas, remote sensing technology plays a crucial role in...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2023-08-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://isprs-archives.copernicus.org/articles/XLVIII-M-1-2023/471/2023/isprs-archives-XLVIII-M-1-2023-471-2023.pdf |
| _version_ | 1797743067276509184 |
|---|---|
| author | O. C. Bayrak F. Erdem M. Uzar |
| author_facet | O. C. Bayrak F. Erdem M. Uzar |
| author_sort | O. C. Bayrak |
| collection | DOAJ |
| description | Forest monitoring and tree species categorization has a vital importance in terms of biodiversity conservation, ecosystem health assessment, climate change mitigation, and sustainable resource management. Due to large-scale coverage of forest areas, remote sensing technology plays a crucial role in the monitoring of forest areas by timely and regular data acquisition, multi-spectral and multi-temporal analysis, non-invasive data collection, accessibility and cost-effectiveness. High-resolution satellite and airborne remote sensing technologies have supplied image data with rich spatial, color, and texture information. Nowadays, deep learning models are commonly utilized in image classification, object recognition, and semantic segmentation applications in remote sensing and forest monitoring as well. We, in this study, selected a popular CNN and object detection algorithm YOLOv8 variants for tree species classification from aerial images of TreeSatAI benchmark. Our results showed that YOLOv8-l outperformed benchmark’s initial release results, and other YOLOv8 variants with 71,55% and 72,70% for weighted and micro averaging scores, respectively. |
| first_indexed | 2024-03-12T14:50:11Z |
| format | Article |
| id | doaj.art-d8c6e2e5495841ac9815982d9e693862 |
| institution | Directory Open Access Journal |
| issn | 1682-1750 2194-9034 |
| language | English |
| last_indexed | 2024-03-12T14:50:11Z |
| publishDate | 2023-08-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-d8c6e2e5495841ac9815982d9e6938622023-08-15T16:29:13ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342023-08-01XLVIII-M-1-202347147610.5194/isprs-archives-XLVIII-M-1-2023-471-2023DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATIONO. C. Bayrak0F. Erdem1M. Uzar2Dept. of Geomatics Engineering, Faculty of Civil Engineering, 34220 Esenler, Istanbul, TürkiyeInstitude of Earth and Space Sciences, Eskisehir Technical University, 26555 Tepebasi, Eskisehir, TürkiyeDept. of Geomatics Engineering, Faculty of Civil Engineering, 34220 Esenler, Istanbul, TürkiyeForest monitoring and tree species categorization has a vital importance in terms of biodiversity conservation, ecosystem health assessment, climate change mitigation, and sustainable resource management. Due to large-scale coverage of forest areas, remote sensing technology plays a crucial role in the monitoring of forest areas by timely and regular data acquisition, multi-spectral and multi-temporal analysis, non-invasive data collection, accessibility and cost-effectiveness. High-resolution satellite and airborne remote sensing technologies have supplied image data with rich spatial, color, and texture information. Nowadays, deep learning models are commonly utilized in image classification, object recognition, and semantic segmentation applications in remote sensing and forest monitoring as well. We, in this study, selected a popular CNN and object detection algorithm YOLOv8 variants for tree species classification from aerial images of TreeSatAI benchmark. Our results showed that YOLOv8-l outperformed benchmark’s initial release results, and other YOLOv8 variants with 71,55% and 72,70% for weighted and micro averaging scores, respectively.https://isprs-archives.copernicus.org/articles/XLVIII-M-1-2023/471/2023/isprs-archives-XLVIII-M-1-2023-471-2023.pdf |
| spellingShingle | O. C. Bayrak F. Erdem M. Uzar DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATION The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATION |
| title_full | DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATION |
| title_fullStr | DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATION |
| title_full_unstemmed | DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATION |
| title_short | DEEP LEARNING BASED AERIAL IMAGERY CLASSIFICATION FOR TREE SPECIES IDENTIFICATION |
| title_sort | deep learning based aerial imagery classification for tree species identification |
| url | https://isprs-archives.copernicus.org/articles/XLVIII-M-1-2023/471/2023/isprs-archives-XLVIII-M-1-2023-471-2023.pdf |
| work_keys_str_mv | AT ocbayrak deeplearningbasedaerialimageryclassificationfortreespeciesidentification AT ferdem deeplearningbasedaerialimageryclassificationfortreespeciesidentification AT muzar deeplearningbasedaerialimageryclassificationfortreespeciesidentification |