HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 Network
Weeds significantly impact crop yields and quality, necessitating strict control. Effective weed identification is essential to precision weeding in the field. Existing detection methods struggle with the inconsistent size scales of weed targets and the issue of small targets, making it difficult to...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/15/1/57 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589422036516864 |
|---|---|
| author | Long Deng Zhonghua Miao Xueguan Zhao Shuo Yang Yuanyuan Gao Changyuan Zhai Chunjiang Zhao |
| author_facet | Long Deng Zhonghua Miao Xueguan Zhao Shuo Yang Yuanyuan Gao Changyuan Zhai Chunjiang Zhao |
| author_sort | Long Deng |
| collection | DOAJ |
| description | Weeds significantly impact crop yields and quality, necessitating strict control. Effective weed identification is essential to precision weeding in the field. Existing detection methods struggle with the inconsistent size scales of weed targets and the issue of small targets, making it difficult to achieve efficient detection, and they are unable to satisfy both the speed and accuracy requirements for detection at the same time. Therefore, this study, focusing on three common types of weeds in the field—<i>Amaranthus retroflexus</i>, <i>Eleusine indica</i>, and <i>Chenopodium</i>—proposes the HAD-YOLO model. With the purpose of improving the model’s capacity to extract features and making it more lightweight, this algorithm employs the HGNetV2 as its backbone network. The Scale Sequence Feature Fusion Module (SSFF) and Triple Feature Encoding Module (TFE) from the ASF-YOLO are introduced to improve the model’s capacity to extract features across various scales, and on this basis, to improve the model’s capacity to detect small targets, a P2 feature layer is included. Finally, a target detection head with an attention mechanism, Dynamic head (Dyhead), is utilized to improve the detection head’s capacity for representation. Experimental results show that on the dataset collected in the greenhouse, the <i>mAP</i> for weed detection is 94.2%; using this as the pre-trained weight, on the dataset collected in the field environment, the <i>mAP</i> for weed detection is 96.2%, and the detection FPS is 30.6. Overall, the HAD-YOLO model has effectively addressed the requirements for accurate weed identification, offering both theoretical and technical backing for automatic weed control. Future efforts will involve collecting more weed data from various agricultural field scenarios to validate and enhance the generalization capabilities of the HAD-YOLO model. |
| format | Article |
| id | doaj-art-a75f0d289cfc4afe81a39fbe6faebaf9 |
| institution | Kabale University |
| issn | 2073-4395 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj-art-a75f0d289cfc4afe81a39fbe6faebaf92025-01-24T13:16:31ZengMDPI AGAgronomy2073-43952024-12-011515710.3390/agronomy15010057HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 NetworkLong Deng0Zhonghua Miao1Xueguan Zhao2Shuo Yang3Yuanyuan Gao4Changyuan Zhai5Chunjiang Zhao6School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, ChinaSchool of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, ChinaInformation Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaInformation Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaSchool of Agricultural Engineering, Jiangsu University, Zhenjiang 212013, ChinaInformation Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaInformation Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaWeeds significantly impact crop yields and quality, necessitating strict control. Effective weed identification is essential to precision weeding in the field. Existing detection methods struggle with the inconsistent size scales of weed targets and the issue of small targets, making it difficult to achieve efficient detection, and they are unable to satisfy both the speed and accuracy requirements for detection at the same time. Therefore, this study, focusing on three common types of weeds in the field—<i>Amaranthus retroflexus</i>, <i>Eleusine indica</i>, and <i>Chenopodium</i>—proposes the HAD-YOLO model. With the purpose of improving the model’s capacity to extract features and making it more lightweight, this algorithm employs the HGNetV2 as its backbone network. The Scale Sequence Feature Fusion Module (SSFF) and Triple Feature Encoding Module (TFE) from the ASF-YOLO are introduced to improve the model’s capacity to extract features across various scales, and on this basis, to improve the model’s capacity to detect small targets, a P2 feature layer is included. Finally, a target detection head with an attention mechanism, Dynamic head (Dyhead), is utilized to improve the detection head’s capacity for representation. Experimental results show that on the dataset collected in the greenhouse, the <i>mAP</i> for weed detection is 94.2%; using this as the pre-trained weight, on the dataset collected in the field environment, the <i>mAP</i> for weed detection is 96.2%, and the detection FPS is 30.6. Overall, the HAD-YOLO model has effectively addressed the requirements for accurate weed identification, offering both theoretical and technical backing for automatic weed control. Future efforts will involve collecting more weed data from various agricultural field scenarios to validate and enhance the generalization capabilities of the HAD-YOLO model.https://www.mdpi.com/2073-4395/15/1/57weed identificationYOLOV5HAD-YOLOdeep learningsmall target detectionmulti-scale feature fusion |
| spellingShingle | Long Deng Zhonghua Miao Xueguan Zhao Shuo Yang Yuanyuan Gao Changyuan Zhai Chunjiang Zhao HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 Network Agronomy weed identification YOLOV5 HAD-YOLO deep learning small target detection multi-scale feature fusion |
| title | HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 Network |
| title_full | HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 Network |
| title_fullStr | HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 Network |
| title_full_unstemmed | HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 Network |
| title_short | HAD-YOLO: An Accurate and Effective Weed Detection Model Based on Improved YOLOV5 Network |
| title_sort | had yolo an accurate and effective weed detection model based on improved yolov5 network |
| topic | weed identification YOLOV5 HAD-YOLO deep learning small target detection multi-scale feature fusion |
| url | https://www.mdpi.com/2073-4395/15/1/57 |
| work_keys_str_mv | AT longdeng hadyoloanaccurateandeffectiveweeddetectionmodelbasedonimprovedyolov5network AT zhonghuamiao hadyoloanaccurateandeffectiveweeddetectionmodelbasedonimprovedyolov5network AT xueguanzhao hadyoloanaccurateandeffectiveweeddetectionmodelbasedonimprovedyolov5network AT shuoyang hadyoloanaccurateandeffectiveweeddetectionmodelbasedonimprovedyolov5network AT yuanyuangao hadyoloanaccurateandeffectiveweeddetectionmodelbasedonimprovedyolov5network AT changyuanzhai hadyoloanaccurateandeffectiveweeddetectionmodelbasedonimprovedyolov5network AT chunjiangzhao hadyoloanaccurateandeffectiveweeddetectionmodelbasedonimprovedyolov5network |