SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV Scenes
Object detection in images captured by unmanned aerial vehicles (UAVs) holds great potential in various domains, including civilian applications, urban planning, and disaster response. However, it faces several challenges, such as multi-scale variations, dense scenes, complex backgrounds, and tiny-s...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/18/4580 |
| _version_ | 1797577198016659456 |
|---|---|
| author | Yuming Wang Hua Zou Ming Yin Xining Zhang |
| author_facet | Yuming Wang Hua Zou Ming Yin Xining Zhang |
| author_sort | Yuming Wang |
| collection | DOAJ |
| description | Object detection in images captured by unmanned aerial vehicles (UAVs) holds great potential in various domains, including civilian applications, urban planning, and disaster response. However, it faces several challenges, such as multi-scale variations, dense scenes, complex backgrounds, and tiny-sized objects. In this paper, we present a novel scale-adaptive YOLO framework called SMFF-YOLO, which addresses these challenges through a multi-level feature fusion approach. To improve the detection accuracy of small objects, our framework incorporates the ELAN-SW object detection prediction head. This newly designed head effectively utilizes both global contextual information and local features, enhancing the detection accuracy of tiny objects. Additionally, the proposed bidirectional feature fusion pyramid (BFFP) module tackles the issue of scale variations in object sizes by aggregating multi-scale features. To handle complex backgrounds, we introduce the adaptive atrous spatial pyramid pooling (AASPP) module, which enables adaptive feature fusion and alleviates the negative impact of cluttered scenes. Moreover, we adopt the Wise-IoU(WIoU) bounding box regression loss to enhance the competitiveness of different quality anchor boxes, which offers the framework a more informed gradient allocation strategy. We validate the effectiveness of SMFF-YOLO using the VisDrone and UAVDT datasets. Experimental results demonstrate that our model achieves higher detection accuracy, with AP50 reaching 54.3% for VisDrone and 42.4% for UAVDT datasets. Visual comparative experiments with other YOLO-based methods further illustrate the robustness and adaptability of our approach. |
| first_indexed | 2024-03-10T22:05:42Z |
| format | Article |
| id | doaj.art-b7dc26eaef584136bfc341b23816d031 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T22:05:42Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-b7dc26eaef584136bfc341b23816d0312023-11-19T12:49:44ZengMDPI AGRemote Sensing2072-42922023-09-011518458010.3390/rs15184580SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV ScenesYuming Wang0Hua Zou1Ming Yin2Xining Zhang3School of Computer Science, Wuhan University, Wuhan 430072, ChinaSchool of Computer Science, Wuhan University, Wuhan 430072, ChinaSchool of Electronic and Electrical Engineering, Wuhan Textile University, Wuhan 430077, ChinaSchool of Computer Science, Wuhan University, Wuhan 430072, ChinaObject detection in images captured by unmanned aerial vehicles (UAVs) holds great potential in various domains, including civilian applications, urban planning, and disaster response. However, it faces several challenges, such as multi-scale variations, dense scenes, complex backgrounds, and tiny-sized objects. In this paper, we present a novel scale-adaptive YOLO framework called SMFF-YOLO, which addresses these challenges through a multi-level feature fusion approach. To improve the detection accuracy of small objects, our framework incorporates the ELAN-SW object detection prediction head. This newly designed head effectively utilizes both global contextual information and local features, enhancing the detection accuracy of tiny objects. Additionally, the proposed bidirectional feature fusion pyramid (BFFP) module tackles the issue of scale variations in object sizes by aggregating multi-scale features. To handle complex backgrounds, we introduce the adaptive atrous spatial pyramid pooling (AASPP) module, which enables adaptive feature fusion and alleviates the negative impact of cluttered scenes. Moreover, we adopt the Wise-IoU(WIoU) bounding box regression loss to enhance the competitiveness of different quality anchor boxes, which offers the framework a more informed gradient allocation strategy. We validate the effectiveness of SMFF-YOLO using the VisDrone and UAVDT datasets. Experimental results demonstrate that our model achieves higher detection accuracy, with AP50 reaching 54.3% for VisDrone and 42.4% for UAVDT datasets. Visual comparative experiments with other YOLO-based methods further illustrate the robustness and adaptability of our approach.https://www.mdpi.com/2072-4292/15/18/4580object detectionunmanned aerial vehiclestiny objectscomplex scenariosmulti-level feature information fusion |
| spellingShingle | Yuming Wang Hua Zou Ming Yin Xining Zhang SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV Scenes Remote Sensing object detection unmanned aerial vehicles tiny objects complex scenarios multi-level feature information fusion |
| title | SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV Scenes |
| title_full | SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV Scenes |
| title_fullStr | SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV Scenes |
| title_full_unstemmed | SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV Scenes |
| title_short | SMFF-YOLO: A Scale-Adaptive YOLO Algorithm with Multi-Level Feature Fusion for Object Detection in UAV Scenes |
| title_sort | smff yolo a scale adaptive yolo algorithm with multi level feature fusion for object detection in uav scenes |
| topic | object detection unmanned aerial vehicles tiny objects complex scenarios multi-level feature information fusion |
| url | https://www.mdpi.com/2072-4292/15/18/4580 |
| work_keys_str_mv | AT yumingwang smffyoloascaleadaptiveyoloalgorithmwithmultilevelfeaturefusionforobjectdetectioninuavscenes AT huazou smffyoloascaleadaptiveyoloalgorithmwithmultilevelfeaturefusionforobjectdetectioninuavscenes AT mingyin smffyoloascaleadaptiveyoloalgorithmwithmultilevelfeaturefusionforobjectdetectioninuavscenes AT xiningzhang smffyoloascaleadaptiveyoloalgorithmwithmultilevelfeaturefusionforobjectdetectioninuavscenes |