Abstract
Abstract. Knowing the spatial distribution of endangered tree species in a forest ecosystem or forest remnants is a valuable information to support environmental conservation practices. The use of Unmanned Aerial Vehicles (UAVs) offers a suitable alternative for this task, providing very high-resolution images at low costs. In parallel, recent advances in the computer vision field have led to the development of effective deep learning techniques for end-to-end semantic image segmentation. In this scenario, the DeepLabv3+ is well established as the state-of-the-art deep learning method for semantic segmentation tasks. The present paper proposes and assesses the use of DeepLabv3+ for mapping the threatened Dipteryx alata Vogel tree, popularly also known as cumbaru. We also compare two backbone networks for feature extraction in the DeepLabv3+ architecture: the Xception and MobileNetv2. Experiments carried out on a dataset consisting of 225 UAV/RGB images of an urban area in Midwest Brazil demonstrated that DeepLabv3+ was able to achieve in mean overall accuracy and F1-score above 90%, and IoU above 80%. The experimental analysis also pointed out that the MobileNetv2 backbone overcame its counterpart by a wide margin due to its comparatively simpler architecture in view of the available training data.
Highlights
Over the years, many remote sensing techniques like LiDAR, hyperspectral, optical and SAR imaging, have been widely used for performing large-scale analysis of forest systems (Jeronimo et al, 2018, Laurin et al, 2013, Alonzo et al, 2014)
Advances in unmanned aerial vehicles (UAVs) technology offered a suitable alternative to standard remote sensing solutions, providing high-resolution images at lower costs (Mohan et al, 2017)
Studies focused on individual tree level mapping through UAV images, suggest its potential for the detection and delineation of tree crowns, and subsequently estimate parameters of its morphology (Lim et al, 2015, Grznarovaet al., 2019, Tang, Shao, 2015)
Summary
Many remote sensing techniques like LiDAR (light detection and ranging), hyperspectral, optical and SAR (synthetic aperture radar) imaging, have been widely used for performing large-scale analysis of forest systems (Jeronimo et al, 2018, Laurin et al, 2013, Alonzo et al, 2014). Within the larger field of mapping forest trends, monitoring of endangered species populations has received increasing attention (Wang et al, 2016, Santos et al, 2019) In this context, for single tree detection, it is fundamental to understand crown morphology. Advances in unmanned aerial vehicles (UAVs) technology offered a suitable alternative to standard remote sensing solutions, providing high-resolution images at lower costs (Mohan et al, 2017). In this way, UAV images have aroused great interest within the remote sensing community and have been used for a wide variety of subjects (Honkavaara et al, 2013, Lizarazo et al, 2017, Rauhala et al, 2017, Chenari et al, 2017). Studies focused on individual tree level mapping through UAV images, suggest its potential for the detection and delineation of tree crowns, and subsequently estimate parameters of its morphology (Lim et al, 2015, Grznarovaet al., 2019, Tang, Shao, 2015)
Sign-in/Register to view highlights & summary 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 callMore From: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
