The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-3/W12-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W12-2020, 355–360, 2020
https://doi.org/10.5194/isprs-archives-XLII-3-W12-2020-355-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W12-2020, 355–360, 2020
https://doi.org/10.5194/isprs-archives-XLII-3-W12-2020-355-2020

  06 Nov 2020

06 Nov 2020

SEMANTIC SEGMENTATION OF ENDANGERED TREE SPECIES IN BRAZILIAN SAVANNA USING DEEPLABV3+ VARIANTS

D. L. Torres1, R. Q. Feitosa1, L. E. C. La Rosa1, P. N. Happ1, J. Marcato Junior2, W. N. Gonçalves2, J. Martins2, and V. Liesenberg3 D. L. Torres et al.
  • 1Dept. of Electrical Engineering, Pontifical Catholic University of Rio de Janeiro, Brazil
  • 2Faculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do Sul, Brazil
  • 3Department of Forest Engineering, Santa Catarina State University, Lages, Brazil

Keywords: Fully Convolution Neural Networks, Unmanned Aerial Vehicles (UAVs), Deep Learning, Semantic Segmentation

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.