Convolutional Neural Networks accurately predict cover fractions of plant species and communities in Unmanned Aerial Vehicle imagery

Kattenborn, Teja; Eichel, Jana; Wiser, Susan K.; Burrows, Larry; Fassnacht, Fabian Ewald; Schmidtlein, Sebastian

doi:10.1002/rse2.146

Cited by 92 publications

(59 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among them are instance segmentation problems, e.g., tree counting problem, tree species detection, identification of small canopy gaps. On the other hand, the spatial extents at which unmanned aerial vehicles data are commonly acquired for vegetation mapping are generally limited to not more than a few hectares or square kilometers [47].…”

Section: Discussionmentioning

confidence: 99%

Automatic Windthrow Detection Using Very-High-Resolution Satellite Imagery and Deep Learning

Kislov¹,

Korznikov²

2020

Remote Sensing

View full text Add to dashboard Cite

Wind disturbances are significant phenomena in forest spatial structure and succession dynamics. They cause changes in biodiversity, impact on forest ecosystems at different spatial scales, and have a strong influence on economics and human beings. The reliable recognition and mapping of windthrow areas are of high importance from the perspective of forest management and nature conservation. Recent research in artificial intelligence and computer vision has demonstrated the incredible potential of neural networks in addressing image classification problems. The most efficient algorithms are based on artificial neural networks of nested and complex architecture (e.g., convolutional neural networks (CNNs)), which are usually referred to by a common term—deep learning. Deep learning provides powerful algorithms for the precise segmentation of remote sensing data. We developed an algorithm based on a U-Net-like CNN, which was trained to recognize windthrow areas in Kunashir Island, Russia. We used satellite imagery of very-high spatial resolution (0.5 m/pixel) as source data. We performed a grid search among 216 parameter combinations defining different U-Net-like architectures. The best parameter combination allowed us to achieve an overall accuracy for recognition of windthrow sites of up to 94% for forested landscapes by coniferous and mixed coniferous forests. We found that the false-positive decisions of our algorithm correspond to either seashore logs, which may look similar to fallen tree trunks, or leafless forest stands. While the former can be rectified by applying a forest mask, the latter requires the usage of additional information, which is not always provided by satellite imagery.

show abstract

Section: Discussionmentioning

confidence: 99%

Automatic Windthrow Detection Using Very-High-Resolution Satellite Imagery and Deep Learning

Kislov¹,

Korznikov²

2020

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Vegetation per se is also considered in efforts to detect, monitor, and respond to invasions. For the first two, research evaluates the effectiveness of remote sensing methods for these applications, and ranges from spatial and radiometric coarse-scaled satellite imagery to finer scales even including drones, and their combination (Campbell et al, 2020; Kattenborn et al, 2020; Kopeć et al, 2020; Rivas-Torres et al, 2018). Responses also include the development and use of integrated pest management plans that incorporate a diversity of tactics to remove invasive species, minimize herbicide resistance (Owen et al, 2014), and reduce potential impacts to property, people, and the environment (US Environmental Protection Agency, 2019).…”

Section: Vegetation Ecology and Anthropogenic Driversmentioning

confidence: 99%

Vegetation ecology with anthropic drivers and consequences

Malanson

Talal

Pansing

et al. 2021

Progress in Physical Geography: Earth and Environment

View full text Add to dashboard Cite

Current research on vegetation makes a difference in people’s lives. Plant community classification is a backbone of land management, plant communities are changing in response to anthropogenic drivers, and the processes of change have impacts on ecosystem services. In the following progress report, we summarize the status of classification and recent research on vegetation responses to pollution, especially nitrogen deposition, invasive species, climate change, and land use and direct exploitation. Two areas with human feedbacks are underscored: fire ecology and urban ecology. Prominent questions at the current research frontier are highlighted with attention to new perspectives.

show abstract

“…In the last 10 years, great advances have been achieved in the field of deep neural networks [44,45]. Recently, this development has been further enhanced due to easier access to algorithms via open-source machine learning libraries, such as scikit-learn [46], Pytorch [47], and Keras-Tensorflow [48,49].…”

Section: Introductionmentioning

confidence: 99%

Monitoring Invasive Plant Species Using Hyperspectral Remote Sensing Data

Papp

Leeuwen

Szilassi

et al. 2021

Land

View full text Add to dashboard Cite

The species richness and biodiversity of vegetation in Hungary are increasingly threatened by invasive plant species brought in from other continents and foreign ecosystems. These invasive plant species have spread aggressively in the natural and semi-natural habitats of Europe. Common milkweed (Asclepias syriaca) is one of the species that pose the greatest ecological menace. Therefore, the primary purpose of the present study is to map and monitor the spread of common milkweed, the most common invasive plant species in Europe. Furthermore, the possibilities to detect and validate this special invasive plant by analyzing hyperspectral remote sensing data were investigated. In combination with field reference data, high-resolution hyperspectral aerial images acquired by an unmanned aerial vehicle (UAV) platform in 138 spectral bands in areas infected by common milkweed were examined. Then, support vector machine (SVM) and artificial neural network (ANN) classification algorithms were applied to the highly accurate field reference data. As a result, common milkweed individuals were distinguished in hyperspectral images, achieving an overall accuracy of 92.95% in the case of supervised SVM classification. Using the ANN model, an overall accuracy of 99.61% was achieved. To evaluate the proposed approach, two experimental tests were conducted, and in both cases, we managed to distinguish the individual specimens within the large variety of spreading invasive species in a study area of 2 ha, based on centimeter spatial resolution hyperspectral UAV imagery.

show abstract

Convolutional Neural Networks accurately predict cover fractions of plant species and communities in Unmanned Aerial Vehicle imagery

Cited by 92 publications

References 73 publications

Automatic Windthrow Detection Using Very-High-Resolution Satellite Imagery and Deep Learning

Automatic Windthrow Detection Using Very-High-Resolution Satellite Imagery and Deep Learning

Vegetation ecology with anthropic drivers and consequences

Monitoring Invasive Plant Species Using Hyperspectral Remote Sensing Data

Contact Info

Product

Resources

About