YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenarios
Background Small object detection via unmanned Aerial vehicle (UAV) is crucial for smart agriculture, enhancing yield and efficiency. Methods This study addresses the issue of missed detections in crowded environments by developing an efficient algorithm tailored for precise, real-time small object...
| Autors principals: | , , , |
|---|---|
| Format: | Article |
| Idioma: | English |
| Publicat: |
PeerJ Inc.
2024-12-01
|
| Col·lecció: | PeerJ Computer Science |
| Matèries: | |
| Accés en línia: | https://peerj.com/articles/cs-2477.pdf |
| _version_ | 1826928229578964992 |
|---|---|
| author | Yihang Li WenZhong Yang Zhifeng Lu Houwang Shi |
| author_facet | Yihang Li WenZhong Yang Zhifeng Lu Houwang Shi |
| author_sort | Yihang Li |
| collection | DOAJ |
| description | Background Small object detection via unmanned Aerial vehicle (UAV) is crucial for smart agriculture, enhancing yield and efficiency. Methods This study addresses the issue of missed detections in crowded environments by developing an efficient algorithm tailored for precise, real-time small object detection. The proposed Yield Health Robust Transformer-YOLO (YH-RTYO) model incorporates several key innovations to advance conventional convolutional models. The model features an efficient convolutional expansion module that captures additional feature information through extended branches while maintaining parameter efficiency by consolidating features into a single convolution during validation. It also includes a local feature pyramid module designed to suppress background interference during feature interaction. Furthermore, the loss function is optimized to accommodate various object scales in different scenes by adjusting the regression box size and incorporating angle factors. These enhancements collectively contribute to improved detection performance and address the limitations of traditional methods. Result Compared to YOLOv8-L, the YH-RTYO model achieves superior performance in all key accuracy metrics, with a 13% reduction in the scale of model. Experimental results demonstrate that the YH-RTYO model outperforms others in key detection metrics. The model reduces the number of parameters by 13%, facilitating deployment while maintaining accuracy. On the OilPalmUAV dataset, it achieves a 3.97% improvement in average precision (AP). Additionally, the model shows strong generalization on the RFRB dataset, with AP50 and AP values exceeding those of the YOLOv8 baseline by 3.8% and 2.7%, respectively. |
| first_indexed | 2025-02-17T15:45:55Z |
| format | Article |
| id | doaj.art-b4fe624f901b49de864d03e5df093125 |
| institution | Directory Open Access Journal |
| issn | 2376-5992 |
| language | English |
| last_indexed | 2025-02-17T15:45:55Z |
| publishDate | 2024-12-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ Computer Science |
| spelling | doaj.art-b4fe624f901b49de864d03e5df0931252024-12-18T15:05:11ZengPeerJ Inc.PeerJ Computer Science2376-59922024-12-0110e247710.7717/peerj-cs.2477YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenariosYihang Li0WenZhong Yang1Zhifeng Lu2Houwang Shi3College of Computer Science and Technology, Xinjiang University, Urumchi, ChinaXinjiang Key Laboratory of Multilingual Information Technology, Xinjiang University, Urumqi, Xinjiang Uygur Autonomous Regions, ChinaSchool of Information Science and Technology, Xinjiang Teacher’s College, Urumchi, ChinaCollege of Computer Science and Technology, Xinjiang University, Urumchi, ChinaBackground Small object detection via unmanned Aerial vehicle (UAV) is crucial for smart agriculture, enhancing yield and efficiency. Methods This study addresses the issue of missed detections in crowded environments by developing an efficient algorithm tailored for precise, real-time small object detection. The proposed Yield Health Robust Transformer-YOLO (YH-RTYO) model incorporates several key innovations to advance conventional convolutional models. The model features an efficient convolutional expansion module that captures additional feature information through extended branches while maintaining parameter efficiency by consolidating features into a single convolution during validation. It also includes a local feature pyramid module designed to suppress background interference during feature interaction. Furthermore, the loss function is optimized to accommodate various object scales in different scenes by adjusting the regression box size and incorporating angle factors. These enhancements collectively contribute to improved detection performance and address the limitations of traditional methods. Result Compared to YOLOv8-L, the YH-RTYO model achieves superior performance in all key accuracy metrics, with a 13% reduction in the scale of model. Experimental results demonstrate that the YH-RTYO model outperforms others in key detection metrics. The model reduces the number of parameters by 13%, facilitating deployment while maintaining accuracy. On the OilPalmUAV dataset, it achieves a 3.97% improvement in average precision (AP). Additionally, the model shows strong generalization on the RFRB dataset, with AP50 and AP values exceeding those of the YOLOv8 baseline by 3.8% and 2.7%, respectively.https://peerj.com/articles/cs-2477.pdfGrowth conditionsTarget detectionConvolutional modelTransformerUAV scenes |
| spellingShingle | Yihang Li WenZhong Yang Zhifeng Lu Houwang Shi YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenarios PeerJ Computer Science Growth conditions Target detection Convolutional model Transformer UAV scenes |
| title | YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenarios |
| title_full | YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenarios |
| title_fullStr | YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenarios |
| title_full_unstemmed | YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenarios |
| title_short | YH-RTYO: an end-to-end object detection method for crop growth anomaly detection in UAV scenarios |
| title_sort | yh rtyo an end to end object detection method for crop growth anomaly detection in uav scenarios |
| topic | Growth conditions Target detection Convolutional model Transformer UAV scenes |
| url | https://peerj.com/articles/cs-2477.pdf |
| work_keys_str_mv | AT yihangli yhrtyoanendtoendobjectdetectionmethodforcropgrowthanomalydetectioninuavscenarios AT wenzhongyang yhrtyoanendtoendobjectdetectionmethodforcropgrowthanomalydetectioninuavscenarios AT zhifenglu yhrtyoanendtoendobjectdetectionmethodforcropgrowthanomalydetectioninuavscenarios AT houwangshi yhrtyoanendtoendobjectdetectionmethodforcropgrowthanomalydetectioninuavscenarios |