# SEAFEC: a spatial–edge adaptive convolution for multi-scale and boundary-aware plant disease and weed imagery

## TL;DR

SEAFEC is a new deep learning module that improves the accuracy of plant disease and weed detection by better handling scale variations and blurred boundaries in images.

## Contribution

SEAFEC introduces a dual-branch convolutional module that jointly enhances scale adaptivity and boundary precision in agricultural imagery.

## Key findings

- SEAFEC improved accuracy by 1.8% in plant disease classification.
- It achieved +2.5% mAP in corn leaf disease detection and +3.4% mIoU in sugarcane-weed segmentation.
- The module showed notable gains in boundary-sensitive agricultural image analysis tasks.

## Abstract

Plant diseases and weeds are among the leading biological threats to global crop production. While deep learning has advanced automated analysis, existing approaches often fail under challenges like large multi-scale variations and blurred boundaries.

To address this, we propose SEAFEC (Spatial-Edge Adaptive Feature Enhancement Convolution), a novel convolutional module that jointly enhances scale adaptivity and boundary precision. SEAFEC employs a dual-branch design: the SCARF branch dynamically adjusts receptive fields, while the MEFE branch explicitly strengthens edge features.

Across three representative tasks—plant disease classification, corn leaf disease detection, and sugarcane-weed segmentation—SEAFEC achieved consistent improvements (+1.8% accuracy, +2.5% mAP, +3.4% mIoU), with notable gains in boundary-sensitive cases.

These results highlight SEAFEC as a general-purpose enhancement module, providing a unified solution for tackling scale-boundary challenges in agricultural imagery to support reliable disease diagnosis and precision weed management.