Uncertainty-Guided Attention and Entropy-Weighted Loss for Precise Plant Seedling Segmentation

2026-04-12

AI Summary

This study addresses the challenge of insufficient segmentation accuracy for fine structures of plant seedlings—particularly leaf margins—in complex backgrounds. To this end, we propose UGDA-Net, which innovatively integrates an uncertainty-guided dual attention mechanism, an entropy-weighted hybrid loss function, and deep supervision applied to intermediate encoder layers. Our method quantifies uncertainty via channel-wise variance to selectively enhance attention on high-uncertainty boundary regions, thereby improving detail-preserving segmentation. Implemented within both U-Net and LinkNet architectures, the approach was evaluated on a dataset of 432 high-resolution seedling images. Experimental results demonstrate a 9.3% improvement in Dice coefficient over baseline models and a 13.2% reduction in result variance for LinkNet, significantly mitigating leaf boundary misclassifications and enhancing the accuracy of automated phenotypic analysis.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Plant seedling segmentation supports automated phenotyping in precision agriculture. Standard segmentation models face difficulties due to intricate background images and fine structures in leaves. We introduce UGDA-Net (Uncertainty-Guided Dual Attention Network with Entropy-Weighted Loss and Deep Supervision). Three novel components make up UGDA-Net. The first component is Uncertainty-Guided Dual Attention (UGDA). UGDA uses channel variance to modulate feature maps. The second component is an entropy-weighted hybrid loss function. This loss function focuses on high-uncertainty boundary pixels. The third component employs deep supervision for intermediate encoder layers. We performed a comprehensive systematic ablation study. This study focuses on two widely-used architectures, U-Net and LinkNet. It analyzes five incremental configurations: Baseline, Loss-only, Attention-only, Deep Supervision, and UGDA-Net. We trained UGDA-net using a high-resolution plant seedling image dataset containing 432 images. We demonstrate improved segmentation performance and accuracy. With an increase in Dice coefficient of 9.3% above baseline. LinkNet's variance is 13.2% above baseline. Overlays that are qualitative in nature show the reduced false positives at the leaf boundary. Uncertainty heatmaps are consistent with the complex morphology. UGDA-Net aids in the segmentation of delicate structures in plants and provides a high-def solution. The results showed that uncertainty-guided attention and uncertainty-weighted loss are two complementing systems.

Problem

• plant seedling segmentation
• complex background
• fine structures
• uncertainty
• precision agriculture

Innovation

• Uncertainty-Guided Attention
• Entropy-Weighted Loss
• Deep Supervision
• Plant Seedling Segmentation
• Dual Attention