Related Experiment Video
Updated: Jun 30, 2026
12:23
Analysis of Yersinia enterocolitica Effector Translocation into Host Cells Using Beta-lactamase Effector Fusions
Published on: October 13, 2015
Research on Apple Surface Disease Detection Method Based on Improved YOLOv11s.
Dongliang Liu1, Yan Li2, Xiaona Song1
1School of Mechanical Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450045, China.
Foods (Basel, Switzerland)
|May 13, 2026
Summary
This study introduces an improved YOLOv11s model for efficient apple surface disease detection. The enhanced model significantly boosts detection accuracy, improving apple quality and yield prediction.
Area of Science:
- Agricultural Science
- Computer Vision
- Machine Learning
Background:
- Apple surface diseases significantly impact crop quality and yield.
- Traditional manual inspection methods are inefficient and lack real-time capabilities.
Purpose of the Study:
- To develop an advanced apple surface disease detection system.
- To improve the accuracy and efficiency of disease identification in apples.
Main Methods:
- An improved YOLOv11s model incorporating GAM attention mechanisms, Haar-based feature downsampling, and a WFU module.
- Integration of the PIOUv2 loss function for optimized bounding box regression.
- Application of data augmentation techniques for small datasets to prevent overfitting.
Main Results:
- The proposed model achieved a 4.2% increase in F1-score and a 2.4% boost in mAP@50:95.
- Demonstrated superior detection performance compared to existing models.
- Effectively identified multi-scale defect features and tiny defect spots.
Conclusions:
- The improved YOLOv11s model offers a highly effective and superior solution for apple surface disease detection.
- The method enhances feature fusion, detail retention, and multi-scale defect recognition.
- The approach contributes to better apple quality assessment and yield management.
