eduzhai > Applied Sciences > Engineering >

Identification of Tree Species in Japanese Forests based on Aerial Photography and Deep Learning

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 15 pages

Abstract: Natural forests are complex ecosystems whose tree species distribution andtheir ecosystem functions are still not well understood. Sustainable managementof these forests is of high importance because of their significant role inclimate regulation, biodiversity, soil erosion and disaster prevention amongmany other ecosystem services they provide. In Japan particularly, naturalforests are mainly located in steep mountains, hence the use of aerial imageryin combination with computer vision are important modern tools that can beapplied to forest research. Thus, this study constitutes a preliminary researchin this field, aiming at classifying tree species in Japanese mixed forestsusing UAV images and deep learning in two different mixed forest types: a blackpine (Pinus thunbergii)-black locust (Robinia pseudoacacia) and a larch (Larixkaempferi)-oak (Quercus mongolica) mixed forest. Our results indicate that itis possible to identify black locust trees with 62.6 True Positives (TP) and98.1 True Negatives (TN), while lower precision was reached for larch trees(37.4 TP and 97.7 TN).

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...