VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruning

Apple-picking robots are increasingly applied in smart agriculture, but their performance is limited by complex orchard conditions such as unstable lighting, occlusion, and weather variations. This study proposes an optimized lightweight detection model, VBP-YOLO-prune, based on YOLOv8n, to enhance...

Full description

Saved in:
Bibliographic Details
Main Authors: Haohai You, Hao Wang, Zhanchen Wei, Chunguang Bi, Lijuan Zhang, Xuefang Li, Yingying Yin
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:Alexandria Engineering Journal
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1110016825008828
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849237525835546624
author Haohai You
Hao Wang
Zhanchen Wei
Chunguang Bi
Lijuan Zhang
Xuefang Li
Yingying Yin
author_facet Haohai You
Hao Wang
Zhanchen Wei
Chunguang Bi
Lijuan Zhang
Xuefang Li
Yingying Yin
author_sort Haohai You
collection DOAJ
description Apple-picking robots are increasingly applied in smart agriculture, but their performance is limited by complex orchard conditions such as unstable lighting, occlusion, and weather variations. This study proposes an optimized lightweight detection model, VBP-YOLO-prune, based on YOLOv8n, to enhance detection accuracy and deployment efficiency on edge devices. The model incorporates a V7 downsampling module, BiFPN feature fusion, and an improved PIOUv2 loss function, aiming to improve multi-scale representation and bounding box regression. A custom apple dataset was augmented with diverse lighting and weather conditions to improve generalization. Experimental results using 10-fold cross-validation show that VBP-YOLO-prune achieves 89.0 % mAP50 and 66.26 % mAP50–95, with Precision of 84.01 % and Recall of 80.52 %. Additionally, it reduces parameters by 79.7 %, FLOPs by 60.9 %, and increases FPS by 29.2 % compared to YOLOv8n.The final model contains only 0.61 M parameters, 3.2 GFLOPs, and runs at 102.6 FPS on NVIDIA Jetson Orin Nano. These results demonstrate that VBP-YOLO-prune provides a practical and efficient solution for real-time fruit detection in complex environments. Future research may extend this approach to other crop types and explore full integration into autonomous harvesting systems.
format Article
id doaj-art-1cb60bed39c84d33a74fa5c267521bfe
institution Kabale University
issn 1110-0168
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series Alexandria Engineering Journal
spelling doaj-art-1cb60bed39c84d33a74fa5c267521bfe2025-08-20T04:01:57ZengElsevierAlexandria Engineering Journal1110-01682025-09-01128992101410.1016/j.aej.2025.08.013VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruningHaohai You0Hao Wang1Zhanchen Wei2Chunguang Bi3Lijuan Zhang4Xuefang Li5Yingying Yin6College of Information Technology, Jilin Agricultural University, Changchun 130118, ChinaFaculty of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, ChinaCollege of Information Technology, Jilin Agricultural University, Changchun 130118, ChinaCollege of Information Technology, Jilin Agricultural University, Changchun 130118, ChinaCollege of Information Technology, Jilin Agricultural University, Changchun 130118, China; College of Internet of Things Engineering, Wuxi University, Wuxi 214105, ChinaCollege of Information Technology, Jilin Agricultural University, Changchun 130118, ChinaCollege of Information Technology, Jilin Agricultural University, Changchun 130118, China; Corresponding author.Apple-picking robots are increasingly applied in smart agriculture, but their performance is limited by complex orchard conditions such as unstable lighting, occlusion, and weather variations. This study proposes an optimized lightweight detection model, VBP-YOLO-prune, based on YOLOv8n, to enhance detection accuracy and deployment efficiency on edge devices. The model incorporates a V7 downsampling module, BiFPN feature fusion, and an improved PIOUv2 loss function, aiming to improve multi-scale representation and bounding box regression. A custom apple dataset was augmented with diverse lighting and weather conditions to improve generalization. Experimental results using 10-fold cross-validation show that VBP-YOLO-prune achieves 89.0 % mAP50 and 66.26 % mAP50–95, with Precision of 84.01 % and Recall of 80.52 %. Additionally, it reduces parameters by 79.7 %, FLOPs by 60.9 %, and increases FPS by 29.2 % compared to YOLOv8n.The final model contains only 0.61 M parameters, 3.2 GFLOPs, and runs at 102.6 FPS on NVIDIA Jetson Orin Nano. These results demonstrate that VBP-YOLO-prune provides a practical and efficient solution for real-time fruit detection in complex environments. Future research may extend this approach to other crop types and explore full integration into autonomous harvesting systems.http://www.sciencedirect.com/science/article/pii/S1110016825008828apple detectionYOLOv8deep learninglightweight deployment
spellingShingle Haohai You
Hao Wang
Zhanchen Wei
Chunguang Bi
Lijuan Zhang
Xuefang Li
Yingying Yin
VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruning
Alexandria Engineering Journal
apple detection
YOLOv8
deep learning
lightweight deployment
title VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruning
title_full VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruning
title_fullStr VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruning
title_full_unstemmed VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruning
title_short VBP-YOLO-prune: Robust apple detection under variable weather via feature-adaptive fusion and efficient YOLO pruning
title_sort vbp yolo prune robust apple detection under variable weather via feature adaptive fusion and efficient yolo pruning
topic apple detection
YOLOv8
deep learning
lightweight deployment
url http://www.sciencedirect.com/science/article/pii/S1110016825008828
work_keys_str_mv AT haohaiyou vbpyoloprunerobustappledetectionundervariableweatherviafeatureadaptivefusionandefficientyolopruning
AT haowang vbpyoloprunerobustappledetectionundervariableweatherviafeatureadaptivefusionandefficientyolopruning
AT zhanchenwei vbpyoloprunerobustappledetectionundervariableweatherviafeatureadaptivefusionandefficientyolopruning
AT chunguangbi vbpyoloprunerobustappledetectionundervariableweatherviafeatureadaptivefusionandefficientyolopruning
AT lijuanzhang vbpyoloprunerobustappledetectionundervariableweatherviafeatureadaptivefusionandefficientyolopruning
AT xuefangli vbpyoloprunerobustappledetectionundervariableweatherviafeatureadaptivefusionandefficientyolopruning
AT yingyingyin vbpyoloprunerobustappledetectionundervariableweatherviafeatureadaptivefusionandefficientyolopruning