Deep learning as a tool for ecology and evolution

Methods in Ecology and Evolution, Journal Year: 2022, Volume and Issue: 13(8), P. 1640 - 1660

Published: May 30, 2022

Abstract Deep learning is driving recent advances behind many everyday technologies, including speech and image recognition, natural language processing autonomous driving. It also gaining popularity in biology, where it has been used for automated species identification, environmental monitoring, ecological modelling, behavioural studies, DNA sequencing population genetics phylogenetics, among other applications. relies on artificial neural networks predictive modelling excels at recognizing complex patterns. In this review we synthesize 818 studies using deep the context of ecology evolution to give a discipline‐wide perspective necessary promote rethinking inference approaches field. We provide an introduction machine contrast with mechanistic inference, followed by gentle primer learning. applications discuss its limitations efforts overcome them. practical biologists interested their toolkit identify possible future find that being rapidly adopted evolution, 589 (64%) published since beginning 2019. Most use convolutional (496 studies) supervised identification but tasks molecular data, sounds, data or video as input. More sophisticated uses biology are appear. Operating within paradigm, can be viewed alternative modelling. desirable properties good performance scaling increasing complexity, while posing unique challenges such sensitivity bias input data. expect rapid adoption will continue, especially automation biodiversity monitoring discovery from genetic Increased unsupervised visualization clusters gaps, simplification multi‐step analysis pipelines, integration into graduate postgraduate training all likely near future.

Language: Английский

---

Status, advancements and prospects of deep learning methods applied in forest studies

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2024, Volume and Issue: 131, P. 103938 - 103938

Published: June 4, 2024

Language: Английский

---

Automatic Detection of Ditches and Natural Streams from Digital Elevation Models Using Deep Learning

Computers & Geosciences, Journal Year: 2025, Volume and Issue: unknown, P. 105875 - 105875

Published: Jan. 1, 2025

Language: Английский

---

Mapping Tree Species Using Advanced Remote Sensing Technologies: A State-of-the-Art Review and Perspective

Journal of Remote Sensing, Journal Year: 2021, Volume and Issue: 2021

Published: Jan. 1, 2021

Timely and accurate information on tree species (TS) is crucial for developing strategies sustainable management conservation of artificial natural forests. Over the last four decades, advances in remote sensing technologies have made TS classification possible. Since many studies topic been conducted their comprehensive results novel findings published literature, it necessary to conduct an updated review status, trends, potentials, challenges recommend future directions. The will provide overview various optical light detection ranging (LiDAR) sensors; present assess current techniques/methods for, a general trend method development in, classification; identify limitations In this review, several concluding remarks were made. They include following: (1) A large group using high-resolution satellite, airborne multi-/hyperspectral imagery, LiDAR data. (2) “multiple” was observed. (3) Machine learning methods including deep models demonstrated be significant improving accuracy. (4) Recently, unmanned aerial vehicle- (UAV-) based sensors caught interest researchers practitioners topic-related research applications. addition, three directions recommended, refining categories methods, data fusion algorithms or processing chains, exploring new spectral unmixing automatically extract map from satellite hyperspectral

Language: Английский

---

Detecting and mapping tree crowns based on convolutional neural network and Google Earth images

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2022, Volume and Issue: 108, P. 102764 - 102764

Published: April 1, 2022

Mapping tree crown is critical for estimating the functional and spatial distribution of ecosystem services. However, accurate up-to-date urban mapping remains a challenge due to time-consuming nature field sampling heterogeneity. Another data cost, which always concern low-cost processing forest maps on large scales. Here, we developed novel working framework by integrating an advanced deep learning technology, Mask Region-based Convolutional Neural Network (Mask R-CNN) model with Google Earth images detect cover in New York's Central Park, typical testbed area highly heterogeneous cover. The results indicated that number detection rate estimated R-CNN was 82.8% 81.8% entire study area. detected isolated trees closed areas recall 87.5% 81.6% numbers, respectively. analysis indicates could accurately crowns under complex environments demonstrates great potential map covers.

Language: Английский

---

Multi-Scale Research on Blasting Damage of Rock Based on Fractal Theory

Rock Mechanics and Rock Engineering, Journal Year: 2024, Volume and Issue: 57(8), P. 5899 - 5911

Published: May 22, 2024

Language: Английский

---

Deep Convolutional Neural Network for Large-Scale Date Palm Tree Mapping from UAV-Based Images

Remote Sensing, Journal Year: 2021, Volume and Issue: 13(14), P. 2787 - 2787

Published: July 15, 2021

Large-scale mapping of date palm trees is vital for their consistent monitoring and sustainable management, considering substantial commercial, environmental, cultural value. This study presents an automatic approach the large-scale from very-high-spatial-resolution (VHSR) unmanned aerial vehicle (UAV) datasets, based on a deep learning approach. A U-Shape convolutional neural network (U-Net), residual framework, was developed semantic segmentation trees. comprehensive set labeled data established to enable training evaluation proposed model increase its generalization capability. The performance compared with those various state-of-the-art fully networks (FCNs) different encoder architectures, including U-Net (based VGG-16 backbone), pyramid scene parsing network, two variants DeepLab V3+. Experimental results showed that outperformed other FCNs in validation testing datasets. generalizability complex dataset exhibited higher classification accuracy could be automatically mapped VHSR UAV images F-score, mean intersection over union, precision, recall 91%, 85%, 0.91, 0.92, respectively. provides efficient architecture UAV-based images.

Language: Английский

---

A Review of General Methods for Quantifying and Estimating Urban Trees and Biomass

Forests, Journal Year: 2022, Volume and Issue: 13(4), P. 616 - 616

Published: April 15, 2022

Understanding the biomass, characteristics, and carbon sequestration of urban forests is crucial for maintaining improving quality life ensuring sustainable planning. Approaches to forest management have been incorporated into interdisciplinary, multifunctional, technical efforts. In this review, we evaluate recent developments in research methods, compare accuracy efficiency different identify emerging themes assessment. This review focuses on biomass estimation individual tree feature detection, showing that rapid development remote sensing technology applications years has greatly benefited study dynamics. Included are light detection ranging-based techniques estimating deep learning algorithms can extract crowns species, methods measuring large canopies using unmanned aerial vehicles estimate structure, approaches capturing street information view images. Conventional based field measurements highly beneficial accurately recording species-specific characteristics. There an urgent need combine multi-scale spatiotemporal improve at scales.

Language: Английский

---

Development of Semantic Maps of Vegetation Cover from UAV Images to Support Planning and Management in Fine-Grained Fire-Prone Landscapes

Remote Sensing, Journal Year: 2022, Volume and Issue: 14(5), P. 1262 - 1262

Published: March 4, 2022

In Mediterranean landscapes, the encroachment of pyrophytic shrubs is a driver more frequent and larger wildfires. The high-resolution mapping vegetation cover essential for sustainable land planning management wildfire prevention. Here, we propose methods to simplify automate segmentation shrub in RGB images acquired by UAVs. main contribution systematic exploration best practices train convolutional neural network (CNN) with architecture (U-Net) detect heterogeneous landscapes. Several semantic models were trained tested partitions provided data alternative augmentation, patch cropping, rescaling hyperparameter tuning (the number filters, dropout rate batch size). most effective cropping rescaling. developed classification model achieved an average F1 score 0.72 on three separate test datasets even though it was relatively small training dataset. This study demonstrates ability state-of-the-art CNNs map fine-grained patterns from remote sensing data. Because performance affected quality labeling, optimal selection pre-processing requisite improve results.

Language: Английский

---

Deep Semantic Segmentation of Trees Using Multispectral Images

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal Year: 2022, Volume and Issue: 15, P. 7589 - 7604

Published: Jan. 1, 2022

Forests can be efficiently monitored by automatic semantic segmentation of trees using satellite and/or aerial images. Still, several challenges make the problem difficult, including varying spectral signature different trees, lack sufficient labelled data, and geometrical occlusions. In this paper, we address tree multispectral imagery. While carry out large-scale experiments on deep learning architectures various input combinations, also attempt to explore whether hand-crafted vegetation indices improve performance models in trees. Our include benchmarking a variety remote sensing image sets, architectures, bands as inputs, number indices. From our experiments, draw useful conclusions. One particularly important conclusion is that, with no additional computation burden, combining categories indices, such NVDI, ARVI, SAVI, within single three-channel input, state-of-the-art accuracy improved under certain conditions, compared high-resolution visible nearinfrared input.

Language: Английский

---