Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field Landscapes

Rice (<i>Oryza sativa</i> L.) has long been threatened by the brown planthopper (BPH, <i>Nilaparvata lugens</i>) and white-backed planthopper (WBPH, <i>Sogatella furcifera</i>). It is difficult to detect and count rice planthoppers from RGB images, and there are a...

Full description

Saved in:
Bibliographic Details
Main Authors: Xusheng Ji, Jiaxin Li, Xiaoxu Cai, Xinhai Ye, Mostafa Gouda, Yong He, Gongyin Ye, Xiaoli Li
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Agriculture
Subjects:
Online Access:https://www.mdpi.com/2077-0472/15/13/1366
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849429358606811136
author Xusheng Ji
Jiaxin Li
Xiaoxu Cai
Xinhai Ye
Mostafa Gouda
Yong He
Gongyin Ye
Xiaoli Li
author_facet Xusheng Ji
Jiaxin Li
Xiaoxu Cai
Xinhai Ye
Mostafa Gouda
Yong He
Gongyin Ye
Xiaoli Li
author_sort Xusheng Ji
collection DOAJ
description Rice (<i>Oryza sativa</i> L.) has long been threatened by the brown planthopper (BPH, <i>Nilaparvata lugens</i>) and white-backed planthopper (WBPH, <i>Sogatella furcifera</i>). It is difficult to detect and count rice planthoppers from RGB images, and there are a limited number of publicly available datasets for agricultural pests. This study publishes a publicly available planthopper dataset, explores the potential of YOLOv8-p2 and proposes an efficient improvement strategy, designated SwinT YOLOv8-p2, for detecting and counting BPH and WBPH from RGB images. The Swin Transformer was incorporated into the YOLOv8-p2 in the strategy. Additionally, the Spatial and Channel Reconstruction Convolution (SCConv) was applied, replacing Convolution (Conv) in the C2f module of YOLOv8. The dataset contains diverse pest small targets, and it is easily available to the public. YOLOv8-p2 can accurately detect different pests, with mAP50, mAP50:95, F1-score, Recall, Precision and FPS up to 0.847, 0.835, 0.899, 0.985, 0.826 and 16.69, respectively. The performance of rice planthopper detection was significantly improved by SwinT YOLOv8-p2, with increases in mAP50 and mAP50:95 ranging from 1.9% to 61.8%. Furthermore, the correlation relationship between the manually counted and detected insects was strong for SwinT YOLOv8-p2, with an R<sup>2</sup> above 0.85, and RMSE and MAE below 0.64 and 0.11. Our results suggest that SwinT YOLOv8-p2 can efficiently detect and count rice planthoppers.
format Article
id doaj-art-6a77b64f4e094f648d13f215d9baaa98
institution Kabale University
issn 2077-0472
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Agriculture
spelling doaj-art-6a77b64f4e094f648d13f215d9baaa982025-08-20T03:28:24ZengMDPI AGAgriculture2077-04722025-06-011513136610.3390/agriculture15131366Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field LandscapesXusheng Ji0Jiaxin Li1Xiaoxu Cai2Xinhai Ye3Mostafa Gouda4Yong He5Gongyin Ye6Xiaoli Li7College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, ChinaCollege of Agriculture and Biotechnology, Zhejiang University, Hangzhou 310058, ChinaThe Rural Development Academy, Zhejiang University, Hangzhou 310058, ChinaCollect of Advanced Agricultural Sciences, Zhejiang A&F University, Hangzhou 311300, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, ChinaCollege of Agriculture and Biotechnology, Zhejiang University, Hangzhou 310058, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, ChinaRice (<i>Oryza sativa</i> L.) has long been threatened by the brown planthopper (BPH, <i>Nilaparvata lugens</i>) and white-backed planthopper (WBPH, <i>Sogatella furcifera</i>). It is difficult to detect and count rice planthoppers from RGB images, and there are a limited number of publicly available datasets for agricultural pests. This study publishes a publicly available planthopper dataset, explores the potential of YOLOv8-p2 and proposes an efficient improvement strategy, designated SwinT YOLOv8-p2, for detecting and counting BPH and WBPH from RGB images. The Swin Transformer was incorporated into the YOLOv8-p2 in the strategy. Additionally, the Spatial and Channel Reconstruction Convolution (SCConv) was applied, replacing Convolution (Conv) in the C2f module of YOLOv8. The dataset contains diverse pest small targets, and it is easily available to the public. YOLOv8-p2 can accurately detect different pests, with mAP50, mAP50:95, F1-score, Recall, Precision and FPS up to 0.847, 0.835, 0.899, 0.985, 0.826 and 16.69, respectively. The performance of rice planthopper detection was significantly improved by SwinT YOLOv8-p2, with increases in mAP50 and mAP50:95 ranging from 1.9% to 61.8%. Furthermore, the correlation relationship between the manually counted and detected insects was strong for SwinT YOLOv8-p2, with an R<sup>2</sup> above 0.85, and RMSE and MAE below 0.64 and 0.11. Our results suggest that SwinT YOLOv8-p2 can efficiently detect and count rice planthoppers.https://www.mdpi.com/2077-0472/15/13/1366publicly available datasetrice planthopperdetection and countingSwin Transformer-based moduleYOLOv8-p2 architecturefield landscapes
spellingShingle Xusheng Ji
Jiaxin Li
Xiaoxu Cai
Xinhai Ye
Mostafa Gouda
Yong He
Gongyin Ye
Xiaoli Li
Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field Landscapes
Agriculture
publicly available dataset
rice planthopper
detection and counting
Swin Transformer-based module
YOLOv8-p2 architecture
field landscapes
title Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field Landscapes
title_full Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field Landscapes
title_fullStr Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field Landscapes
title_full_unstemmed Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field Landscapes
title_short Driving by a Publicly Available RGB Image Dataset for Rice Planthopper Detection and Counting by Fusing Swin Transformer and YOLOv8-p2 Architectures in Field Landscapes
title_sort driving by a publicly available rgb image dataset for rice planthopper detection and counting by fusing swin transformer and yolov8 p2 architectures in field landscapes
topic publicly available dataset
rice planthopper
detection and counting
Swin Transformer-based module
YOLOv8-p2 architecture
field landscapes
url https://www.mdpi.com/2077-0472/15/13/1366
work_keys_str_mv AT xushengji drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes
AT jiaxinli drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes
AT xiaoxucai drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes
AT xinhaiye drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes
AT mostafagouda drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes
AT yonghe drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes
AT gongyinye drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes
AT xiaolili drivingbyapubliclyavailablergbimagedatasetforriceplanthopperdetectionandcountingbyfusingswintransformerandyolov8p2architecturesinfieldlandscapes