Abstract
Woody plant encroachment into grasslands ecosystems causes significantly ecological destruction and economic losses. Effective and efficient management largely benefits from accurate and timely detection of encroaching species at an early development stage. Recent advances in unmanned aircraft systems (UAS) enabled easier access to ultra-high spatial resolution images at a centimeter level, together with the latest machine learning based image segmentation algorithms, making it possible to detect small-sized individuals of target species at early development stage and identify them when mixed with other species. However, few studies have investigated the optimal practical spatial resolution of early encroaching species detection. Hence, we investigated the performance of four popular semantic segmentation algorithms (decision tree, DT; random forest, RF; AlexNet; and ResNet) on a multi-species forest classification case with UAS-collected RGB images in original and down-sampled coarser spatial resolutions. The objective of this study was to explore the optimal segmentation algorithm and spatial resolution for eastern redcedar (Juniperus virginiana, ERC) early detection and its classification within a multi-species forest context. To be specific, firstly, we implemented and compared the performance of the four semantic segmentation algorithms with images in the original spatial resolution (0.694 cm). The highest overall accuracy was 0.918 achieved by ResNet with a mean interaction over union at 85.0%. Secondly, we evaluated the performance of ResNet algorithm with images in down-sampled spatial resolutions (1 cm to 5 cm with 0.5 cm interval). When applied on the down-sampled images, ERC segmentation performance decreased with decreasing spatial resolution, especially for those images coarser than 3 cm spatial resolution. The UAS together with the state-of-the-art semantic segmentation algorithms provides a promising tool for early-stage detection and localization of ERC and the development of effective management strategies for mixed-species forest management.
Highlights
Among the four algorithms, convolutional neural network (CNN) (i.e., AlexNet and ResNet) gained higher performance than classical machine learning algorithms for this case study with complex multi-species forest environment settings
With the highest performance algorithm in this study (ResNet), we identified a threshold of 3 cm as the lowest image spatial resolution required for a desired multi-species forest classification result using only the RGB imagery
We investigated the potential of unmanned aircraft systems (UAS) based RGB remote sensing and four semantic segmentation algorithms on the early detection of ERCs and the classification and delineation of ERCs in a multi-species forest context
Summary
Woody plant encroachment affects ecological stability and causes significant economic loss worldwide [1,2,3], especially in semi-arid and sub-humid grasslands and savannas [4,5]. Species composition and community, and ecosystem services including biodiversity, biogeochemical and hydrological cycles can shift at the Remote Sens. Encroaching species are currently being managed using mechanical (e.g., prescribed fire, grazing and felling) or chemical approaches (e.g., herbicide) [9,10,11], which are costly and sometimes lead to severe ecological destructions [11]. Managing ERC in encroached sites can become costly over time with an increase in density and size of the species, which lead to ecological and economic consequences (e.g., changing ecosystem functioning and reducing livestock production) [20]. Early detection and control of woody plant encroachment are critical for ecological and cost-effective management practices
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
