An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry Diseases
This paper proposes an improved lightweight YOLOv5 model for the real-time detection of strawberry diseases. The ghost convolution (GhostConv) module is incorporated into the YOLOv5 network, reducing the parameter numbers and floating-point operations (FLOPs) for extracting feature information using...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10143201/ |
| _version_ | 1797808459743232000 |
|---|---|
| author | Shunlong Chen Yinghua Liao Feng Lin Bo Huang |
| author_facet | Shunlong Chen Yinghua Liao Feng Lin Bo Huang |
| author_sort | Shunlong Chen |
| collection | DOAJ |
| description | This paper proposes an improved lightweight YOLOv5 model for the real-time detection of strawberry diseases. The ghost convolution (GhostConv) module is incorporated into the YOLOv5 network, reducing the parameter numbers and floating-point operations (FLOPs) for extracting feature information using the backbone network. An involution operator is utilized in the backbone network to expand the receptive field, enhance the spatial information on strawberry disease characteristics, and reduce the number of FLOPs in the model. A convolutional block attention module (CBAM) is incorporated into the backbone network to enhance the network’s ability to extract strawberry disease features and suppress non-critical information. The upsampling module is replaced by a lightweight upsampling operator called Content-Aware ReAssembly of Features (CARAFE), which extracts feature map information and enhances the ability to focus on strawberry disease features. The experimental results on an open-source strawberry disease dataset show that the model achieves mean average precision (mAP)@0.5 of 94.7% with 3.9 M parameters and 3.6 G FLOPs. The improved model has higher detection precision than the original one and lower hardware requirements, providing a new strategy for strawberry disease identification and control. |
| first_indexed | 2024-03-13T06:37:49Z |
| format | Article |
| id | doaj.art-25c110387c7246878c047b64ccfe91e6 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-13T06:37:49Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-25c110387c7246878c047b64ccfe91e62023-06-08T23:01:18ZengIEEEIEEE Access2169-35362023-01-0111540805409210.1109/ACCESS.2023.328230910143201An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry DiseasesShunlong Chen0https://orcid.org/0000-0002-2977-7222Yinghua Liao1https://orcid.org/0000-0003-1821-3699Feng Lin2Bo Huang3School of Mechanical Engineering, Sichuan University of Science and Engineering, Yibin, ChinaSchool of Mechanical Engineering, Sichuan University of Science and Engineering, Yibin, ChinaSchool of Mechanical Engineering, Sichuan University of Science and Engineering, Yibin, ChinaSchool of Mechanical Engineering, Sichuan University of Science and Engineering, Yibin, ChinaThis paper proposes an improved lightweight YOLOv5 model for the real-time detection of strawberry diseases. The ghost convolution (GhostConv) module is incorporated into the YOLOv5 network, reducing the parameter numbers and floating-point operations (FLOPs) for extracting feature information using the backbone network. An involution operator is utilized in the backbone network to expand the receptive field, enhance the spatial information on strawberry disease characteristics, and reduce the number of FLOPs in the model. A convolutional block attention module (CBAM) is incorporated into the backbone network to enhance the network’s ability to extract strawberry disease features and suppress non-critical information. The upsampling module is replaced by a lightweight upsampling operator called Content-Aware ReAssembly of Features (CARAFE), which extracts feature map information and enhances the ability to focus on strawberry disease features. The experimental results on an open-source strawberry disease dataset show that the model achieves mean average precision (mAP)@0.5 of 94.7% with 3.9 M parameters and 3.6 G FLOPs. The improved model has higher detection precision than the original one and lower hardware requirements, providing a new strategy for strawberry disease identification and control.https://ieeexplore.ieee.org/document/10143201/Computer visionimage classificationlightweight networkYOLOv5 |
| spellingShingle | Shunlong Chen Yinghua Liao Feng Lin Bo Huang An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry Diseases IEEE Access Computer vision image classification lightweight network YOLOv5 |
| title | An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry Diseases |
| title_full | An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry Diseases |
| title_fullStr | An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry Diseases |
| title_full_unstemmed | An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry Diseases |
| title_short | An Improved Lightweight YOLOv5 Algorithm for Detecting Strawberry Diseases |
| title_sort | improved lightweight yolov5 algorithm for detecting strawberry diseases |
| topic | Computer vision image classification lightweight network YOLOv5 |
| url | https://ieeexplore.ieee.org/document/10143201/ |
| work_keys_str_mv | AT shunlongchen animprovedlightweightyolov5algorithmfordetectingstrawberrydiseases AT yinghualiao animprovedlightweightyolov5algorithmfordetectingstrawberrydiseases AT fenglin animprovedlightweightyolov5algorithmfordetectingstrawberrydiseases AT bohuang animprovedlightweightyolov5algorithmfordetectingstrawberrydiseases AT shunlongchen improvedlightweightyolov5algorithmfordetectingstrawberrydiseases AT yinghualiao improvedlightweightyolov5algorithmfordetectingstrawberrydiseases AT fenglin improvedlightweightyolov5algorithmfordetectingstrawberrydiseases AT bohuang improvedlightweightyolov5algorithmfordetectingstrawberrydiseases |