Deep learning method for cucumber disease detection in complex environments for new agricultural productivity

Abstract Cucumber disease detection under complex agricultural conditions faces significant challenges due to multi-scale variation, background clutter, and hardware limitations. This study proposes YOLO-Cucumber, an improved lightweight detection algorithm based on YOLOv11n, incorporating four key...

Full description

Saved in:
Bibliographic Details
Main Authors: Jun Liu, Xuewei Wang, Qian Chen, Peng Yan, Xin Liu
Format: Article
Language:English
Published: BMC 2025-07-01
Series:BMC Plant Biology
Subjects:
Online Access:https://doi.org/10.1186/s12870-025-06841-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849344368125673472
author Jun Liu
Xuewei Wang
Qian Chen
Peng Yan
Xin Liu
author_facet Jun Liu
Xuewei Wang
Qian Chen
Peng Yan
Xin Liu
author_sort Jun Liu
collection DOAJ
description Abstract Cucumber disease detection under complex agricultural conditions faces significant challenges due to multi-scale variation, background clutter, and hardware limitations. This study proposes YOLO-Cucumber, an improved lightweight detection algorithm based on YOLOv11n, incorporating four key innovations: (1) Deformable Convolutional Networks (DCN) for enhanced feature extraction of irregular targets, (2) a P2 prediction layer for fine-grained detection of early-stage lesions, (3) a Target-aware Loss (TAL) function addressing class imbalance, and (4) Channel Pruning via Batch Normalization (CPBN) for model compression. Experiments on our cucumber disease dataset demonstrate that YOLO-Cucumber achieves a 6.5% improvement in mAP@50 (93.8%), while reducing model size by 3.87 MB and increasing inference speed to 218 FPS. The model effectively handles symptom variability and complex detection scenarios, outperforming mainstream detection algorithms in accuracy, speed, and compactness, making it ideal for embedded agricultural applications.
format Article
id doaj-art-df026757f87f446b887339fbc9b2d413
institution Kabale University
issn 1471-2229
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series BMC Plant Biology
spelling doaj-art-df026757f87f446b887339fbc9b2d4132025-08-20T03:42:40ZengBMCBMC Plant Biology1471-22292025-07-0125111810.1186/s12870-025-06841-yDeep learning method for cucumber disease detection in complex environments for new agricultural productivityJun Liu0Xuewei Wang1Qian Chen2Peng Yan3Xin Liu4Shandong Engineering Research Center of Green and High-value Marine Fine Chemical, Weifang University of Science and TechnologyShandong Engineering Research Center of Green and High-value Marine Fine Chemical, Weifang University of Science and TechnologySchool of Computer, Sichuan Technology and Business UniversityThe Industry-Education Integration Office, Sichuan Technology and Business UniversityShandong Engineering Research Center of Green and High-value Marine Fine Chemical, Weifang University of Science and TechnologyAbstract Cucumber disease detection under complex agricultural conditions faces significant challenges due to multi-scale variation, background clutter, and hardware limitations. This study proposes YOLO-Cucumber, an improved lightweight detection algorithm based on YOLOv11n, incorporating four key innovations: (1) Deformable Convolutional Networks (DCN) for enhanced feature extraction of irregular targets, (2) a P2 prediction layer for fine-grained detection of early-stage lesions, (3) a Target-aware Loss (TAL) function addressing class imbalance, and (4) Channel Pruning via Batch Normalization (CPBN) for model compression. Experiments on our cucumber disease dataset demonstrate that YOLO-Cucumber achieves a 6.5% improvement in mAP@50 (93.8%), while reducing model size by 3.87 MB and increasing inference speed to 218 FPS. The model effectively handles symptom variability and complex detection scenarios, outperforming mainstream detection algorithms in accuracy, speed, and compactness, making it ideal for embedded agricultural applications.https://doi.org/10.1186/s12870-025-06841-yCucumber disease detectionYOLOv11nDeformable Convolution Networks (DCN)Target-aware LossChannel pruningEmbedded deployment
spellingShingle Jun Liu
Xuewei Wang
Qian Chen
Peng Yan
Xin Liu
Deep learning method for cucumber disease detection in complex environments for new agricultural productivity
BMC Plant Biology
Cucumber disease detection
YOLOv11n
Deformable Convolution Networks (DCN)
Target-aware Loss
Channel pruning
Embedded deployment
title Deep learning method for cucumber disease detection in complex environments for new agricultural productivity
title_full Deep learning method for cucumber disease detection in complex environments for new agricultural productivity
title_fullStr Deep learning method for cucumber disease detection in complex environments for new agricultural productivity
title_full_unstemmed Deep learning method for cucumber disease detection in complex environments for new agricultural productivity
title_short Deep learning method for cucumber disease detection in complex environments for new agricultural productivity
title_sort deep learning method for cucumber disease detection in complex environments for new agricultural productivity
topic Cucumber disease detection
YOLOv11n
Deformable Convolution Networks (DCN)
Target-aware Loss
Channel pruning
Embedded deployment
url https://doi.org/10.1186/s12870-025-06841-y
work_keys_str_mv AT junliu deeplearningmethodforcucumberdiseasedetectionincomplexenvironmentsfornewagriculturalproductivity
AT xueweiwang deeplearningmethodforcucumberdiseasedetectionincomplexenvironmentsfornewagriculturalproductivity
AT qianchen deeplearningmethodforcucumberdiseasedetectionincomplexenvironmentsfornewagriculturalproductivity
AT pengyan deeplearningmethodforcucumberdiseasedetectionincomplexenvironmentsfornewagriculturalproductivity
AT xinliu deeplearningmethodforcucumberdiseasedetectionincomplexenvironmentsfornewagriculturalproductivity