Enhancing Leaf Area Segmentation by Using Attention Gates and Knowledge Distillation in UNet Architecture

Abstract

Accurate segmentation of leaf regions plays a vital role in plant phenotyping and agricultural analysis. This paper presents AKDUNet, a lightweight UNet-based architecture that integrates attention gates and knowledge distillation to improve segmentation performance while minimizing computational complexity. The architecture replaces traditional skip connections with attention gates to focus on salient spatial features and employs a two-stage training pipeline, where a compact student model learns from a deeper teacher model using a tailored distillation loss function. AKDUNet is evaluated on two benchmark datasets (CWFID and Sunflower) and outperforms a range of state-of-the-art models, including UNet++, Inception UNet, VGG-based UNets, SDUNet, INSCA UNet, and SegFormer. Ablation studies confirm the advantages of attention modules, and qualitative analyses using Grad-CAM visualizations reveal the model's ability to effectively focus on crucial leaf structures. The results demonstrate that AKDUNet is not only computationally efficient but also highly accurate, making it suitable for real-time deployment in resource-constrained agricultural environments.

Plant phenotyping relevance

植物の葉領域を抽出する画像セグメンテーション手法の開発とベンチマーク評価が中心であり、葉面積などの表現型取得に直接利用できる。

Code and data availability

The paper evaluates AKDUNet leaf segmentation on the public CWFID dataset (and a Sunflower dataset), and the acknowledgments explicitly state the datasets are publicly available at the authors' cited repository URL. No author analysis code or trained model checkpoints are released.

No evidence-backed public reproduction asset is currently recorded.