ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume IV-3/W2-2020
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-3/W2-2020, 65–70, 2020
https://doi.org/10.5194/isprs-annals-IV-3-W2-2020-65-2020
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-3/W2-2020, 65–70, 2020
https://doi.org/10.5194/isprs-annals-IV-3-W2-2020-65-2020

  29 Oct 2020

29 Oct 2020

FCRN-BASED MULTI-TASK LEARNING FOR AUTOMATIC CITRUS TREE DETECTION FROM UAV IMAGES

L. E. C. La Rosa1,3, M. Zortea1, B. H. Gemignani2, D. A. B. Oliveira1, and R. Q. Feitosa3 L. E. C. La Rosa et al.
  • 1IBM Research Av. Paster, 146, Rio de Janeiro, Brazil
  • 23DGEO
  • 3Dept. of Electrical Engineering, Pontifical Catholic University of Rio de Janeiro, Brazil

Keywords: Citrus Trees, Fully Convolutional Regression Network, Multi-Task Learning, Unmanned Aerial Vehicle

Abstract. Citrus producers need to monitor orchards frequently, and would benefit greatly from having automated tools to analyze aerial images acquired by drones over the plantations. However, analysing large aerial data sets to enable producers to take management decisions that would optimize productivity and sustainability over time and space remains challenging. Motivated by the success of deep learning in computer vision, this work proposes a novel approach based on Fully Convolutional Regression Networks and Multi-Task Learning to detect individual full-grown trees, tree seedlings, and tree gaps in citrus orchards for inventory tracking. We show that the proposal can identify eight-year-old orange trees with accuracy between 95–99% in high-density commercial plantations where adjacent crowns overlap. This quality of detection was achieved on RGB orthomosaics with a pixel size of about 9.5 cm and requires the nominal spacing between adjacent trees as a priori information. Our results also highlight that detecting tree seedlings and tree gaps remains a challenge. For these two categories, classification sensitivity (recall) was between 59–100% and 63–94%, respectively.