Vision Transformer-Based Unhealthy Tree Crown Detection in Mixed Northeastern US Forests and Evaluation of Annotation Uncertainty

Sábháilte in:
Sonraí bibleagrafaíochta
Foilsithe in:Remote Sensing vol. 17, no. 6 (2025), p. 1066
Príomhchruthaitheoir: Joshi, Durga
Rannpháirtithe: Witharana, Chandi
Foilsithe / Cruthaithe:
MDPI AG
Ábhair:
Rochtain ar líne:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Clibeanna: Cuir clib leis
Níl clibeanna ann, Bí ar an gcéad duine le clib a chur leis an taifead seo!

MARC

LEADER 00000nab a2200000uu 4500
001 3182137532
003 UK-CbPIL
022 |a 2072-4292 
024 7 |a 10.3390/rs17061066  |2 doi 
035 |a 3182137532 
045 2 |b d20250101  |b d20251231 
084 |a 231556  |2 nlm 
100 1 |a Joshi, Durga 
245 1 |a Vision Transformer-Based Unhealthy Tree Crown Detection in Mixed Northeastern US Forests and Evaluation of Annotation Uncertainty 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a Forest health monitoring at scale requires high-spatial-resolution remote sensing images coupled with deep learning image analysis methods. However, high-quality large-scale datasets are costly to acquire. To address this challenge, we explored the potential of freely available National Agricultural Imagery Program (NAIP) imagery. By comparing the performance of traditional convolutional neural network (CNN) models (U-Net and DeepLabv3+) with a state-of-the-art Vision Transformer (SegFormer), we aimed to determine the optimal approach for detecting unhealthy tree crowns (UTC) using a publicly available data source. Additionally, we investigated the impact of different spectral band combinations on model performance to identify the most effective configuration without incurring additional data acquisition costs. We explored various band combinations, including RGB, color infrared (CIR), vegetation indices (VIs), principal components (PC) of texture features (PCA), and spectral band with PC (RGBPC). Furthermore, we analyzed the uncertainty associated with potential subjective crown annotation and its impact on model evaluation. Our results demonstrated that the Vision Transformer-based model, SegFormer, outperforms traditional CNN-based models, particularly when trained on RGB images yielding an F1-score of 0.85. In contrast, DeepLabv3+ achieved F1-score of 0.82. Notably, PCA-based inputs yield reduced performance across all models, with U-Net producing particularly poor results (F1-score as low as 0.03). The uncertainty analysis indicated that the Intersection over Union (IoU) could fluctuate between 14.81% and 57.41%, while F1-scores ranged from 8.57% to 47.14%, reflecting the significant sensitivity of model performance to inconsistencies in ground truth annotations. In summary, this study demonstrates the feasibility of using publicly available NAIP imagery and advanced deep learning techniques to accurately detect unhealthy tree canopies. These findings highlight SegFormer’s superior ability to capture complex spatial patterns, even in relatively low-resolution (60 cm) datasets. Our findings underline the considerable influence of human annotation errors on model performance, emphasizing the need for standardized annotation guidelines and quality control measures. 
651 4 |a United States--US 
653 |a Data acquisition 
653 |a Accuracy 
653 |a Vision 
653 |a Datasets 
653 |a Deep learning 
653 |a Quality control 
653 |a Models 
653 |a Color imagery 
653 |a Artificial neural networks 
653 |a Biodiversity 
653 |a Remote sensing 
653 |a Unmanned aerial vehicles 
653 |a Image processing 
653 |a Annotations 
653 |a Feasibility studies 
653 |a Uncertainty analysis 
653 |a Machine learning 
653 |a Performance evaluation 
653 |a Image 
653 |a Vegetation 
653 |a Image analysis 
653 |a Principal components analysis 
653 |a Vegetation index 
653 |a Ground truth 
653 |a Classification 
653 |a Image acquisition 
653 |a Image quality 
653 |a Neural networks 
653 |a Semantics 
700 1 |a Witharana, Chandi 
773 0 |t Remote Sensing  |g vol. 17, no. 6 (2025), p. 1066 
786 0 |d ProQuest  |t Advanced Technologies & Aerospace Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3182137532/abstract/embedded/6A8EOT78XXH2IG52?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3182137532/fulltextwithgraphics/embedded/6A8EOT78XXH2IG52?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3182137532/fulltextPDF/embedded/6A8EOT78XXH2IG52?source=fedsrch