Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter Head

Accurate citrus disease identification is essential for targeted orchard pesticide application. Current models struggle with accuracy and efficiency due to diverse leaf lesion patterns and complex orchard environments. This study presents YOLOv8n-DE, an improved lightweight YOLOv8-based model for en...

Full description

Saved in:
Bibliographic Details
Main Authors: Xu Guo, Xingmeng Wang, Wenhao Zhu, Simon X. Yang, Lepeng Song, Ping Li, Qinzheng Li
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/7/1971
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849730247129300992
author Xu Guo
Xingmeng Wang
Wenhao Zhu
Simon X. Yang
Lepeng Song
Ping Li
Qinzheng Li
author_facet Xu Guo
Xingmeng Wang
Wenhao Zhu
Simon X. Yang
Lepeng Song
Ping Li
Qinzheng Li
author_sort Xu Guo
collection DOAJ
description Accurate citrus disease identification is essential for targeted orchard pesticide application. Current models struggle with accuracy and efficiency due to diverse leaf lesion patterns and complex orchard environments. This study presents YOLOv8n-DE, an improved lightweight YOLOv8-based model for enhanced citrus disease detection. It introduces the DR module structure for effective feature enhancement and the Detect_Shared architecture for parameter efficiency. Evaluated on public and orchard-collected datasets, YOLOv8n-DE achieves 97.6% classification accuracy, 91.8% recall, and 97.3% mAP, with a 90.4% mAP for challenging diseases. Compared to the original YOLOv8, it reduces parameters by 48.17%, computational load by 59.26%, and model size by 41.94%, while significantly decreasing classification and regression errors, and false positives/negatives. YOLOv8n-DE offers outstanding performance and lightweight advantages for citrus disease detection, supporting precision agriculture development in orchards.
format Article
id doaj-art-422470bed9df4deb8db9df24f0331388
institution DOAJ
issn 1424-8220
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-422470bed9df4deb8db9df24f03313882025-08-20T03:08:56ZengMDPI AGSensors1424-82202025-03-01257197110.3390/s25071971Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter HeadXu Guo0Xingmeng Wang1Wenhao Zhu2Simon X. Yang3Lepeng Song4Ping Li5Qinzheng Li6School of Big Data and Automation, Chongqing Chemical Industry Vocational College, Chongqing 401228, ChinaSchool of Electronic and Electrical Engineering, Chongqing University of Science & Technology, Chongqing 401331, ChinaSchool of Electronic and Electrical Engineering, Chongqing University of Science & Technology, Chongqing 401331, ChinaAdvanced Robotics and Intelligent Systems Laboratory, School of Engineering, University of Guelph, Guelph, ON N1G 2W1, CanadaSchool of Electronic and Electrical Engineering, Chongqing University of Science & Technology, Chongqing 401331, ChinaChongqing Academy of Agricultural Sciences, Chongqing 400039, ChinaSchool of Electronic and Electrical Engineering, Chongqing University of Science & Technology, Chongqing 401331, ChinaAccurate citrus disease identification is essential for targeted orchard pesticide application. Current models struggle with accuracy and efficiency due to diverse leaf lesion patterns and complex orchard environments. This study presents YOLOv8n-DE, an improved lightweight YOLOv8-based model for enhanced citrus disease detection. It introduces the DR module structure for effective feature enhancement and the Detect_Shared architecture for parameter efficiency. Evaluated on public and orchard-collected datasets, YOLOv8n-DE achieves 97.6% classification accuracy, 91.8% recall, and 97.3% mAP, with a 90.4% mAP for challenging diseases. Compared to the original YOLOv8, it reduces parameters by 48.17%, computational load by 59.26%, and model size by 41.94%, while significantly decreasing classification and regression errors, and false positives/negatives. YOLOv8n-DE offers outstanding performance and lightweight advantages for citrus disease detection, supporting precision agriculture development in orchards.https://www.mdpi.com/1424-8220/25/7/1971YOLOv8n-DEdilated reparam feature enhancementshared parameter headcitrus disease detectionreal-time detection
spellingShingle Xu Guo
Xingmeng Wang
Wenhao Zhu
Simon X. Yang
Lepeng Song
Ping Li
Qinzheng Li
Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter Head
Sensors
YOLOv8n-DE
dilated reparam feature enhancement
shared parameter head
citrus disease detection
real-time detection
title Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter Head
title_full Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter Head
title_fullStr Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter Head
title_full_unstemmed Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter Head
title_short Citrus Disease Detection Based on Dilated Reparam Feature Enhancement and Shared Parameter Head
title_sort citrus disease detection based on dilated reparam feature enhancement and shared parameter head
topic YOLOv8n-DE
dilated reparam feature enhancement
shared parameter head
citrus disease detection
real-time detection
url https://www.mdpi.com/1424-8220/25/7/1971
work_keys_str_mv AT xuguo citrusdiseasedetectionbasedondilatedreparamfeatureenhancementandsharedparameterhead
AT xingmengwang citrusdiseasedetectionbasedondilatedreparamfeatureenhancementandsharedparameterhead
AT wenhaozhu citrusdiseasedetectionbasedondilatedreparamfeatureenhancementandsharedparameterhead
AT simonxyang citrusdiseasedetectionbasedondilatedreparamfeatureenhancementandsharedparameterhead
AT lepengsong citrusdiseasedetectionbasedondilatedreparamfeatureenhancementandsharedparameterhead
AT pingli citrusdiseasedetectionbasedondilatedreparamfeatureenhancementandsharedparameterhead
AT qinzhengli citrusdiseasedetectionbasedondilatedreparamfeatureenhancementandsharedparameterhead