4.7 Article

Injecting Domain Knowledge Into Deep Neural Networks for Tree Crown Delineation

Journal

Publisher

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TGRS.2022.3216622

Keywords

Convolutional neural network (CNN); forest ecology; neuro-symbolics; remote sensing; tree crown delineation

Funding

  1. NSF [1926542]
  2. Division Of Environmental Biology
  3. Direct For Biological Sciences [1926542] Funding Source: National Science Foundation

Ask authors/readers for more resources

This article introduces a convolutional neural network-based ITC delineation algorithm that uses a neuro-symbolic framework to inject domain knowledge. The results show that the injection of rules improves model performance and affects model bias. The addition of domain data positively impacts the accuracy of the model and reduces errors.
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 article, 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. We find that the addition of domain data can improve F1 by as much as four F1 points, reduce the Kullback-Leibler divergence (KL-divergence) between ground-truth and predicted area distributions, and reduce the aggregate error in area between ground-truth and predicted delineations.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available