Identifying species from the air: UAVs and the very high resolution challenge for plant conservation

Baena, Susana; Moat, Justin; Whaley, Oliver Q.; Boyd, Doreen S.

doi:10.1371/journal.pone.0188714

Cited by 117 publications

(106 citation statements)

References 74 publications

Supporting

Mentioning

100

Contrasting

Unclassified

Order By: Relevance

“…These studies did not considered the elimination of shadows prior to the allocation of spectral values to the segmented clumps (e.g. Alvarez-Taboada et al 2017;Baena et al 2017;Cao et al 2018). Integrating both sunlit and shaded canopy reflectance into the segments could hamper the success of classification tasks if shadows are considered as noise.…”

Section: Variable Importancementioning

confidence: 99%

“…This is important for early and accurate prediction of invasive species occurrences (e.g. Baena et al 2017;Cao et al 2018). Such UAV approaches are especially suitable to: (1) understand the invasion dynamics and processes at local scale through repetitive acquisitions, and (2) to derive reference data for large-scale satellite-based mapping of the invasions (Kattenborn et al submitted).General benefits of UAV-based sensing include the possibility of optical data acquisition under cloudy conditions (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…Meanwhile, disadvantages of UAV-based sensing include their relatively small area cover and the relatively low radiometrical quality of the sensors (Hruska et al 2012). UAV-based invasive species mapping has yielded high accuracies using different data types, like RGB or VNIR information (Michez et al 2016;Alvarez-Taboada et al 2017;Baena et al 2017;Mafanya et al 2017;Cao et al 2018;de S a et al 2018), hyperspectral data (Cao et al 2018) or 2D textural (Michez et al 2016;Cao et al 2018) and 3D structural (Kattenborn et al 2014;Franklin et al 2017) information derived from photogrammetric algorithms.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

How canopy shadow affects invasive plant species classification in high spatial resolution remote sensing

Lopatin

Dološ

Kattenborn

2019

Remote Sens Ecol Conserv

View full text Add to dashboard Cite

Plant invasions can result in serious threats for biodiversity and ecosystem functioning. Reliable maps at very‐high spatial resolution are needed to assess invasions dynamics. Field sampling approaches could be replaced by unmanned aerial vehicles (UAVs) to derive such maps. However, pixel‐based species classification at high spatial resolution is highly affected by within‐canopy variation caused by shadows. Here, we studied the effect of shadows on mapping the occurrence of invasive species using UAV‐based data. MaxEnt one‐class classifications were applied to map Acacia dealbata, Ulex europaeus and Pinus radiata in central‐south Chile using combinations of UAV‐based spectral (RGB and hyperspectral), 2D textural and 3D structural variables including and excluding shaded canopy pixels during model calibration. The model accuracies in terms of area under the curve (AUC), Cohen's Kappa, sensitivity (true positive rate) and specificity (true negative rate) were examined in sunlit and shaded canopies separately. Bootstrapping was used for validation and to assess statistical differences between models. Our results show that shadows significantly affect the accuracies obtained with all types of variables. The predictions in shaded areas were generally inaccurate, leading to misclassification rates between 65% and 100% even when shadows were included during model calibration. The exclusion of shaded areas from model calibrations increased the predictive accuracies (especially in terms of sensitivity), decreasing false positives. Spectral and 2D textural information showed generally higher performances and improvements when excluding shadows from the analysis. Shadows significantly affected the model results obtained with any of the variables used, hence the exclusion of shadows is recommended prior to model calibration. This relatively easy pre‐processing step enhances models for classifying species occurrences using high‐resolution spectral imagery and derived products. Finally, a shadow simulation showed differences in the ideal acquisition window for each species, which is important to plan revisit campaigns.

show abstract

Section: Variable Importancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

How canopy shadow affects invasive plant species classification in high spatial resolution remote sensing

Lopatin

Dološ

Kattenborn

2019

Remote Sens Ecol Conserv

View full text Add to dashboard Cite

show abstract

“…Additionally, rapid advances in the miniaturization of hyperspectral and LiDAR sensors increasingly enable them to be mounted on UAVs (Sankey, Donager, McVay, & Sankey, ) alongside RGB cameras and predicted future price drops will increase their accessibility. While RGB images allow visual species identification (Baena, Boyd, & Moat, ; Getzin, Wiegand, & Schöning, ), multispectral or hyperspectral sensors may allow this to be automated (Baena, Moat, Whaley, & Boyd, ; Sankey et al., ), further increasing the speed and ease of liana identification. Future work could test whether a suitably equipped UAV could automate mapping of liana infestation and changes in infestation similar to the approach adopted by Marvin et al.…”

Section: Discussionmentioning

confidence: 99%

A view from above: Unmanned aerial vehicles (UAVs) provide a new tool for assessing liana infestation in tropical forest canopies

Waite

Heijden

Field

et al. 2019

Journal of Applied Ecology

Self Cite

View full text Add to dashboard Cite

Tropical forests store and sequester large quantities of carbon, mitigating climate change. Lianas (woody vines) are important tropical forest components, most conspicuous in the canopy. Lianas reduce forest carbon uptake and their recent increase may, therefore, limit forest carbon storage with global consequences for climate change. Liana infestation of tree crowns is traditionally assessed from the ground, which is labour intensive and difficult, particularly for upper canopy layers. We used a lightweight unmanned aerial vehicle (UAV) to assess liana infestation of tree canopies from above. It was a commercially available quadcopter UAV with an integrated, standard three‐waveband camera to collect aerial image data for 150 ha of tropical forest canopy. By visually interpreting the images, we assessed the degree of liana infestation for 14.15 ha of forest for which ground‐based estimates were collected simultaneously. We compared the UAV liana infestation estimates with those from the ground to determine the validity, strengths, and weaknesses of using UAVs as a new method for assessing liana infestation of tree canopies. Estimates of liana infestation from the UAV correlated strongly with ground‐based surveys at individual tree and plot level, and across multiple forest types and spatial resolutions, improving liana infestation assessment for upper canopy layers. Importantly, UAV‐based surveys, including the image collection, processing, and visual interpretation, were considerably faster and more cost‐efficient than ground‐based surveys. Synthesis and applications. Unmanned aerial vehicle (UAV) image data of tree canopies can be easily captured and used to assess liana infestation at least as accurately as traditional ground data. This novel method promotes reproducibility of results and quality control, and enables additional variables to be derived from the image data. It is more cost‐effective, time‐efficient and covers larger geographical extents than traditional ground surveys, enabling more comprehensive monitoring of changes in liana infestation over space and time. This is important for assessing liana impacts on the global carbon balance, and particularly useful for forest management where knowledge of the location and change in liana infestation can be used for tailored, targeted, and effective management of tropical forests for enhanced carbon sequestration (e.g., REDD+ projects), timber concessions, and forest restoration.

show abstract

“…Unmanned aerial vehicles (UAVs) have become a reliable observation platform for environmental remote sensing applications, including wildfire mapping [1,2], atmospheric studies [3][4][5], precision agriculture [6][7][8] and plant identification [9][10][11]. Compared to the traditional remote sensing platforms such as satellites and manned aircrafts, which have low spatiotemporal resolutions and high operational costs [12], light-weight UAVs are less costly and more flexible.…”

Section: Introductionmentioning

confidence: 99%

Four-band Thermal Mosaicking: A New Method to Process Infrared Thermal Imagery of Urban Landscapes from UAV Flights

Yang

Lee

2019

Remote Sensing

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) support a large array of technological applications and scientific studies due to their ability to collect high-resolution image data. The processing of UAV data requires the use of mosaicking technology, such as structure-from-motion, which combines multiple photos to form a single image mosaic and to construct a 3-D digital model of the measurement target. However, the mosaicking of thermal images is challenging due to low lens resolution and weak contrast in the single thermal band. In this study, a novel method, referred to as four-band thermal mosaicking (FTM), was developed in order to process thermal images. The method stacks the thermal band obtained by a thermal camera onto the RGB bands acquired on the same flight by an RGB camera and mosaics the four bands simultaneously. An object-based calibration method is then used to eliminate inter-band positional errors. A UAV flight over a natural park was carried out in order to test the method. The results demonstrated that with the assistance of the high-resolution RGB bands, the method enabled successful and efficient thermal mosaicking. Transect analysis revealed an inter-band accuracy of 0.39 m or 0.68 times the ground pixel size of the thermal camera. A cluster analysis validated that the thermal mosaic captured the expected contrast of thermal properties between different surfaces within the scene.3 of 17 register, with sufficient positional accuracy, the thermal mosaic to the RGB imagery of the same target for proper interpretation of the former [29].Motivated by these challenges and the importance of thermal mosaicking, this work aims to design a novel method, referred to as the four-band thermal mosaicking (FTM), in order to process thermal images acquired by a UAV, with three goals in mind:

show abstract

Identifying species from the air: UAVs and the very high resolution challenge for plant conservation

Cited by 117 publications

References 74 publications

How canopy shadow affects invasive plant species classification in high spatial resolution remote sensing

How canopy shadow affects invasive plant species classification in high spatial resolution remote sensing

A view from above: Unmanned aerial vehicles (UAVs) provide a new tool for assessing liana infestation in tropical forest canopies

Four-band Thermal Mosaicking: A New Method to Process Infrared Thermal Imagery of Urban Landscapes from UAV Flights

Contact Info

Product

Resources

About