A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization
Military object detection from Unmanned Aerial Vehicle (UAV) reconnaissance images faces challenges, including lack of image data, images with poor quality, and small objects. In this work, we simulate UAV low-altitude reconnaissance and construct the UAV reconnaissance image tank database UAVT-3. T...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/23/12236 |
| _version_ | 1797463592433352704 |
|---|---|
| author | Huanhua Liu Yonghao Yu Shengzong Liu Wei Wang |
| author_facet | Huanhua Liu Yonghao Yu Shengzong Liu Wei Wang |
| author_sort | Huanhua Liu |
| collection | DOAJ |
| description | Military object detection from Unmanned Aerial Vehicle (UAV) reconnaissance images faces challenges, including lack of image data, images with poor quality, and small objects. In this work, we simulate UAV low-altitude reconnaissance and construct the UAV reconnaissance image tank database UAVT-3. Then, we improve YOLOv5 and propose UAVT-YOLOv5 for object detection of UAV images. First, data augmentation of blurred images is introduced to improve the accuracy of fog and motion-blurred images. Secondly, a large-scale feature map together with multi-scale feedback is added to improve the recognition ability of small objects. Thirdly, we optimize the loss function by increasing the loss penalty of small objects and classes with fewer samples. Finally, the anchor boxes are optimized by clustering the ground truth object box of UAVT-3. The feature visualization technique Class Action Mapping (CAM) is introduced to explore the mechanisms of the proposed model. The experimental results of the improved model evaluated on UAVT-3 show that the mAP reaches 99.2%, an increase of 2.1% compared with YOLOv5, the detection speed is 40 frames per second, and data augmentation of blurred images yields an mAP increase of 20.4% and 26.6% for fog and motion blur images detection. The class action maps show the discriminant region of the tanks is the turret for UAVT-YOLOv5. |
| first_indexed | 2024-03-09T17:52:56Z |
| format | Article |
| id | doaj.art-17d099581cfe4f279e25b2bc9ce6a15b |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T17:52:56Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-17d099581cfe4f279e25b2bc9ce6a15b2023-11-24T10:33:14ZengMDPI AGApplied Sciences2076-34172022-11-0112231223610.3390/app122312236A Military Object Detection Model of UAV Reconnaissance Image and Feature VisualizationHuanhua Liu0Yonghao Yu1Shengzong Liu2Wei Wang3School of Information Technology and Management, Hunan University of Finance and Economics, Changsha 410205, ChinaSchool of Information Technology and Management, Hunan University of Finance and Economics, Changsha 410205, ChinaSchool of Information Technology and Management, Hunan University of Finance and Economics, Changsha 410205, ChinaSchool of Information Technology and Management, Hunan University of Finance and Economics, Changsha 410205, ChinaMilitary object detection from Unmanned Aerial Vehicle (UAV) reconnaissance images faces challenges, including lack of image data, images with poor quality, and small objects. In this work, we simulate UAV low-altitude reconnaissance and construct the UAV reconnaissance image tank database UAVT-3. Then, we improve YOLOv5 and propose UAVT-YOLOv5 for object detection of UAV images. First, data augmentation of blurred images is introduced to improve the accuracy of fog and motion-blurred images. Secondly, a large-scale feature map together with multi-scale feedback is added to improve the recognition ability of small objects. Thirdly, we optimize the loss function by increasing the loss penalty of small objects and classes with fewer samples. Finally, the anchor boxes are optimized by clustering the ground truth object box of UAVT-3. The feature visualization technique Class Action Mapping (CAM) is introduced to explore the mechanisms of the proposed model. The experimental results of the improved model evaluated on UAVT-3 show that the mAP reaches 99.2%, an increase of 2.1% compared with YOLOv5, the detection speed is 40 frames per second, and data augmentation of blurred images yields an mAP increase of 20.4% and 26.6% for fog and motion blur images detection. The class action maps show the discriminant region of the tanks is the turret for UAVT-YOLOv5.https://www.mdpi.com/2076-3417/12/23/12236object detectionfeature visualizationunmanned aerial vehicle imageYOLOv5data augmentation |
| spellingShingle | Huanhua Liu Yonghao Yu Shengzong Liu Wei Wang A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization Applied Sciences object detection feature visualization unmanned aerial vehicle image YOLOv5 data augmentation |
| title | A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization |
| title_full | A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization |
| title_fullStr | A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization |
| title_full_unstemmed | A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization |
| title_short | A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization |
| title_sort | military object detection model of uav reconnaissance image and feature visualization |
| topic | object detection feature visualization unmanned aerial vehicle image YOLOv5 data augmentation |
| url | https://www.mdpi.com/2076-3417/12/23/12236 |
| work_keys_str_mv | AT huanhualiu amilitaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization AT yonghaoyu amilitaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization AT shengzongliu amilitaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization AT weiwang amilitaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization AT huanhualiu militaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization AT yonghaoyu militaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization AT shengzongliu militaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization AT weiwang militaryobjectdetectionmodelofuavreconnaissanceimageandfeaturevisualization |