Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart Agriculture

Rapid and accurate detection of rice foliar diseases is essential for yield prediction and food security. This study proposes a multi-size rice leaf disease detection model, YOLOv7-tiny, for fast and accurate detection of rice leaf diseases. The MobileNetV3 lightweight network is introduced to repla...

Full description

Saved in:
Bibliographic Details
Main Authors: Fuxu Guo, Jing Li, Xingcheng Liu, Sinuo Chen, Hongze Zhang, Yingli Cao, Songhong Wei
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Agronomy
Subjects:
Online Access:https://www.mdpi.com/2073-4395/14/12/2796
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846106387038339072
author Fuxu Guo
Jing Li
Xingcheng Liu
Sinuo Chen
Hongze Zhang
Yingli Cao
Songhong Wei
author_facet Fuxu Guo
Jing Li
Xingcheng Liu
Sinuo Chen
Hongze Zhang
Yingli Cao
Songhong Wei
author_sort Fuxu Guo
collection DOAJ
description Rapid and accurate detection of rice foliar diseases is essential for yield prediction and food security. This study proposes a multi-size rice leaf disease detection model, YOLOv7-tiny, for fast and accurate detection of rice leaf diseases. The MobileNetV3 lightweight network is introduced to replace the backbone network of YOLOv7-tiny, which reduces the size of the model parameters and improves the extraction capability of features of different sizes; the RCS-OSA is used to replace the original ELAN-1 module, which improves the extraction capability of interlayer features; the TSCODE detector head is designed to enhance the extraction capability of the model for small targets; and the MPDIoU loss function is used to improve the model’s convergence speed and effect. The experimental results show that the average accuracy of ofYOLOv7-TMRTM is 97.9%, and compared with the baseline YOLOv7-tiny model, the accuracy of leaf spot detection is improved for different sizes and types of small target detection results, the YOLOv7-TMRTM model improves mAP0.5 by 4.4%, recall by 4.7% and precision by 8.8% compared to YOLOv7-tiny. The comparison with Faster RCNN, SSD, YOLOv4, YOLOv5s, YOLOv8s, and other mainstream target detection models shows that this method greatly solves the field environment. The problem of small spots and fuzzy edges of photographed rice diseases provides a basis for intelligent management of diseases in the field, which in turn promotes food security in China.
format Article
id doaj-art-fcf226d8fa71479486ea5e487e14931d
institution Kabale University
issn 2073-4395
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Agronomy
spelling doaj-art-fcf226d8fa71479486ea5e487e14931d2024-12-27T14:04:00ZengMDPI AGAgronomy2073-43952024-11-011412279610.3390/agronomy14122796Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart AgricultureFuxu Guo0Jing Li1Xingcheng Liu2Sinuo Chen3Hongze Zhang4Yingli Cao5Songhong Wei6College of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Plant Protection, Shenyang Agricultural University, Shenyang 110866, ChinaRapid and accurate detection of rice foliar diseases is essential for yield prediction and food security. This study proposes a multi-size rice leaf disease detection model, YOLOv7-tiny, for fast and accurate detection of rice leaf diseases. The MobileNetV3 lightweight network is introduced to replace the backbone network of YOLOv7-tiny, which reduces the size of the model parameters and improves the extraction capability of features of different sizes; the RCS-OSA is used to replace the original ELAN-1 module, which improves the extraction capability of interlayer features; the TSCODE detector head is designed to enhance the extraction capability of the model for small targets; and the MPDIoU loss function is used to improve the model’s convergence speed and effect. The experimental results show that the average accuracy of ofYOLOv7-TMRTM is 97.9%, and compared with the baseline YOLOv7-tiny model, the accuracy of leaf spot detection is improved for different sizes and types of small target detection results, the YOLOv7-TMRTM model improves mAP0.5 by 4.4%, recall by 4.7% and precision by 8.8% compared to YOLOv7-tiny. The comparison with Faster RCNN, SSD, YOLOv4, YOLOv5s, YOLOv8s, and other mainstream target detection models shows that this method greatly solves the field environment. The problem of small spots and fuzzy edges of photographed rice diseases provides a basis for intelligent management of diseases in the field, which in turn promotes food security in China.https://www.mdpi.com/2073-4395/14/12/2796rice diseasedigital imagingfeature fusionsmall target detection
spellingShingle Fuxu Guo
Jing Li
Xingcheng Liu
Sinuo Chen
Hongze Zhang
Yingli Cao
Songhong Wei
Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart Agriculture
Agronomy
rice disease
digital imaging
feature fusion
small target detection
title Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart Agriculture
title_full Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart Agriculture
title_fullStr Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart Agriculture
title_full_unstemmed Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart Agriculture
title_short Improved YOLOv7-Tiny for the Detection of Common Rice Leaf Diseases in Smart Agriculture
title_sort improved yolov7 tiny for the detection of common rice leaf diseases in smart agriculture
topic rice disease
digital imaging
feature fusion
small target detection
url https://www.mdpi.com/2073-4395/14/12/2796
work_keys_str_mv AT fuxuguo improvedyolov7tinyforthedetectionofcommonriceleafdiseasesinsmartagriculture
AT jingli improvedyolov7tinyforthedetectionofcommonriceleafdiseasesinsmartagriculture
AT xingchengliu improvedyolov7tinyforthedetectionofcommonriceleafdiseasesinsmartagriculture
AT sinuochen improvedyolov7tinyforthedetectionofcommonriceleafdiseasesinsmartagriculture
AT hongzezhang improvedyolov7tinyforthedetectionofcommonriceleafdiseasesinsmartagriculture
AT yinglicao improvedyolov7tinyforthedetectionofcommonriceleafdiseasesinsmartagriculture
AT songhongwei improvedyolov7tinyforthedetectionofcommonriceleafdiseasesinsmartagriculture