loading page

Injecting Domain Knowledge Into Deep Neural Networks for Tree Crown Delineation
  • +6
  • Ira Harmon ,
  • Sergio Marconi ,
  • Ben Weinstein ,
  • Sarah Graves ,
  • Daisy Zhe Wang ,
  • Stephanie Bohlman ,
  • Alina Zare ,
  • Aditya Singh ,
  • Ethan White
Ira Harmon
University of Florida

Corresponding Author:[email protected]

Author Profile
Sergio Marconi
Author Profile
Ben Weinstein
Author Profile
Sarah Graves
Author Profile
Daisy Zhe Wang
Author Profile
Stephanie Bohlman
Author Profile
Alina Zare
Author Profile
Aditya Singh
Author Profile
Ethan White
Author Profile


Automated individual tree crown (ITC) delineation plays an important role in forest remote sensing. Accurate ITC delineation benefits biomass estimation, allometry estimation, and species classification among other forest related tasks, all of which are used to monitor forest health and make important decisions in forest management.  In this paper, we introduce Neuro-Symbolic DeepForest,  a convolutional neural network (CNN) based ITC delineation algorithm that uses a neuro-symbolic framework to inject domain knowledge (represented as rules written in probabilistic soft logic) into a CNN.  We create rules that encode concepts for competition, allometry, constrained growth, mean ITC area, and crown color.  Our results show that the delineation model learns from the annotated training data as well as the rules and that under some conditions, the injection of rules improves model performance and affects model bias.  We then analyze the effects of each rule on its related aspects of model performance.
2022Published in IEEE Transactions on Geoscience and Remote Sensing volume 60 on pages 1-19. 10.1109/TGRS.2022.3216622