Machine learning-assisted wood materials: Applications and future prospects

Extreme Mechanics Letters, Journal Year: 2024, Volume and Issue: 71, P. 102209 - 102209

Published: July 19, 2024

Language: Английский

---

Evaluation of Selected Digital Elevation Models over a Tropical Rainforest: A Case Study at Brunei Darussalam’s Tropical Rainforest

Journal of the Indian Society of Remote Sensing, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 2, 2025

Language: Английский

---

Estimation of Tree Vitality Reduced by Pine Needle Disease Using Multispectral Drone Images

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(2), P. 271 - 271

Published: Jan. 14, 2025

Multispectral imagery from unmanned aerial vehicles (UAVs) can provide high-resolution data to map tree mortality caused by pests or diseases. Although many studies have investigated UAV-imagery-based methods detect trees under acute stress followed mortality, few tested the feasibility and accuracy of detecting chronic stress. This study aims develop test how well UAV-based multispectral pine needle disease long before mortality. images were acquired four times through growing season in an area with infected pathogens. Vegetation indices (VIs) used quantify decline vitality, which was verified retention (%) estimated ground. Results showed that several VIs had strong correlations level identify severely defoliated (<75% retention) 0.71 overall classification accuracy, while slightly (>75% very low. The results one also implied more defoliation observed UAV (top view) than ground (bottom view). We conclude using efficiently needle-cast pathogens, thus assisting forest health monitoring.

Language: Английский

---

Canopy extraction of mango trees in hilly and plain orchards using UAV images: Performance of machine learning vs deep learning

Ecological Informatics, Journal Year: 2025, Volume and Issue: unknown, P. 103101 - 103101

Published: March 1, 2025

Language: Английский

---

Using Citizen Science Data as Pre-Training for Semantic Segmentation of High-Resolution UAV Images for Natural Forests Post-Disturbance Assessment

Forests, Journal Year: 2025, Volume and Issue: 16(4), P. 616 - 616

Published: March 31, 2025

The ability to monitor forest areas after disturbances is key ensure their regrowth. Problematic situations that are detected can then be addressed with targeted regeneration efforts. However, achieving this automated photo interpretation problematic, as training such systems requires large amounts of labeled data. To effect, we leverage citizen science data (iNaturalist) alleviate issue. More precisely, seek generate pre-training from a classifier trained on selected exemplars. This accomplished by using moving-window approach carefully gathered low-altitude images an Unmanned Aerial Vehicle (UAV), WilDReF-Q (Wild Drone Regrowth Forest—Quebec) dataset, high-quality pseudo-labels. accurate pseudo-labels, the predictions our for each window integrated majority voting approach. Our results indicate semantic segmentation network over 140,000 auto-labeled yields F1 score 43.74% 24 different classes, separate ground truth dataset. In comparison, only 32.45%, while fine-tuning pre-trained marginal improvements (46.76%). Importantly, demonstrate able benefit more unlabeled images, opening door learning at scale. We also optimized hyperparameters pseudo-labeling, including number assigned pixel in process. Overall, demonstrates auto-labeling greatly reduce development cost plant identification regions, based UAV imagery.

Language: Английский

---

Preliminary Machine Learning-Based Classification of Ink Disease in Chestnut Orchards Using High-Resolution Multispectral Imagery from Unmanned Aerial Vehicles: A Comparison of Vegetation Indices and Classifiers

Forests, Journal Year: 2025, Volume and Issue: 16(5), P. 754 - 754

Published: April 28, 2025

Ink disease, primarily caused by the pathogen Phytophthora xcambivora, significantly threatens health and productivity of sweet chestnut (Castanea sativa Mill.) orchards, highlighting need for accurate detection methods. This study investigates efficacy machine learning (ML) classifiers combined with high-resolution multispectral imagery acquired via unmanned aerial vehicles (UAVs) to assess tree at a site in Tuscany, Italy. Three algorithms—support vector machines (SVMs), Gaussian Naive Bayes (GNB), logistic regression (Log)—were evaluated against eight vegetation indices (VIs), including NDVI, GnDVI, RdNDVI, classify crowns as either symptomatic or asymptomatic. High-resolution images were processed derive that effectively captured subtle spectral variations indicative disease presence. Ground-truthing involved visual assessments performed expert forest pathologists, subsequently validated through leaf area index (LAI) measurements. Correlation analysis confirmed significant associations between LAI most VIs, supporting robust physiological metric validating assessments. GnDVI RdNDVI SVM GNB achieved highest classification accuracy (95.2%), demonstrating their superior sensitivity discriminating from asymptomatic trees. Indices such MCARI SAVI showed limited discriminative power, underscoring importance selecting appropriate VIs are tailored specific symptoms. highlights potential integrating UAV-derived techniques, LAI, an effective approach ink enabling precision forestry practices informed orchard management strategies.

Language: Английский

---

Temporal generalization in evergreen leaf type classification using tailored sentinel-2 composites

Ecological Informatics, Journal Year: 2025, Volume and Issue: unknown, P. 103167 - 103167

Published: May 1, 2025

Language: Английский

---

Potential of Lightweight Drones and Object-Oriented Image Segmentation in Forest Plantation Assessment

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(9), P. 1554 - 1554

Published: April 27, 2024

Forests play a vital role in maintaining ecological balance and provide numerous benefits. The monitoring managing of large-scale forest plantations can be challenging expensive. In recent years, advancements remote sensing technologies, such as lightweight drones object-oriented image analysis, have opened up new possibilities for efficient accurate plantation monitoring. This study aimed to explore the utility cost-effective method mapping characteristics two 50 ha plots Nayla Range, Jaipur. By combining aerial photographs collected by drone with photogrammetry limited ground survey data, well topography edaphic variables, this examined relative contribution drone-derived canopy information. results demonstrate immense potential analysis providing valuable insights optimizing silvicultural operations planting trees complex environments.

Language: Английский

---

Species-specific machine learning models for UAV-based forest health monitoring: Revealing the importance of the BNDVI

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2024, Volume and Issue: 135, P. 104257 - 104257

Published: Nov. 10, 2024

Language: Английский

---

Design of a multi-component system-based fixed-wing unmanned aerial vehicle maintenance policy and its case study

Computers & Industrial Engineering, Journal Year: 2024, Volume and Issue: 198, P. 110701 - 110701

Published: Nov. 8, 2024

Language: Английский

---