HOB-CNN: Hallucination of Occluded Branches with a Convolutional Neural Network for 2D Fruit Trees
Authors: Zijue Chen, Keenan Granland, Rhys Newbury and Chao Chen
Publication Under Review
Abstract Orchard automation has attracted the attention of researchers recently due to the shortage of global labor force. To automate tasks in orchards such as pruning, thinning, and harvesting, a detailed understanding of the tree structure is required. However, occlusions from foliage and fruits can make it challenging to predict the position of occluded trunks and branches. This work proposes a regression-based deep learning model, Hallucination of Occluded Branch Convolutional Neural Network (HOB-CNN), for tree branch position prediction in varying occluded conditions. We formulate tree branch position prediction as a regression problem towards the horizontal locations of the branch along the vertical direction or vice versa. We present comparative experiments on Y-shaped trees with two state-of-the-art baselines, representing common approaches to the problem. Experiments show that HOB-CNN outperformed the baselines at predicting branch position and showed robustness against varying levels of occlusion. We further validated HOB-CNN against two different types of 2D trees, and HOB-CNN showed generalization across different trees and robustness under different occluded conditions.
Dataset: This dataset includes 435 RGB-D Y-shaped Pink Lady apple trees with three types of label: Position Annotation, Visible Branch Annotation and Whole Branch Annotation. 103 images are from summer (heavily occluded), 89 images are from autumn (medium occluded) and 243 images are from winter (no occlusion).
Personnel - Ms Zijue Chen (PhD student), Dr Chao Chen (contact person)
Acknowledgments
