Drought Stress Classification using 3D Plant Models

TL;DR

This paper presents a novel deep learning pipeline for 3D plant modeling and drought stress classification, improving accuracy over traditional methods by leveraging shape priors and feature aggregation.

Contribution

It introduces an end-to-end deep neural network approach for 3D plant reconstruction and drought stress analysis, incorporating shape priors to handle occlusions and self-similarities.

Findings

Deep network-based features outperform conventional descriptors.

The pipeline accurately reconstructs 3D plant models despite occlusions.

Enhanced drought stress classification accuracy achieved.

Abstract

Quantification of physiological changes in plants can capture different drought mechanisms and assist in selection of tolerant varieties in a high throughput manner. In this context, an accurate 3D model of plant canopy provides a reliable representation for drought stress characterization in contrast to using 2D images. In this paper, we propose a novel end-to-end pipeline including 3D reconstruction, segmentation and feature extraction, leveraging deep neural networks at various stages, for drought stress study. To overcome the high degree of self-similarities and self-occlusions in plant canopy, prior knowledge of leaf shape based on features from deep siamese network are used to construct an accurate 3D model using structure from motion on wheat plants. The drought stress is characterized with a deep network based feature aggregation. We compare the proposed methodology on several descriptors, and show that the network outperforms conventional methods.