GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array Lidar
For the recognition of armored targets in complex battlefield environments, how to reduce missed and false alarms while achieving real-time is an urgent issue. To this end, the GCD-YOLOv5 algorithm is innovatively proposed. Firstly, array lidar is used to acquire the armor target data. Secondly, the...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Photonics Journal |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9803045/ |
| _version_ | 1828204815783559168 |
|---|---|
| author | Jian Dai Xu Zhao Lian Peng Li Xiao Fei Ma |
| author_facet | Jian Dai Xu Zhao Lian Peng Li Xiao Fei Ma |
| author_sort | Jian Dai |
| collection | DOAJ |
| description | For the recognition of armored targets in complex battlefield environments, how to reduce missed and false alarms while achieving real-time is an urgent issue. To this end, the GCD-YOLOv5 algorithm is innovatively proposed. Firstly, array lidar is used to acquire the armor target data. Secondly, the armor target data is expanded with an improved GAN(Generative Adversarial Network) to increase the diversity of training data. Afterward, the expanded dataset is fed into the GCD-YOLv5(You Only Look Once) for training. And the GCD-YOLOv5 is reflected in the following aspects. Firstly, the CBAM(Convolutional Block Attention Module) and the multi-scale feature fusion are added to improve the feature extraction capability and detection efficiency, increasing the recognition capability of small and obscured targets. Secondly, combining with DETR(Detection Transformer) to lighten YOLOv5 to achieve the real-time requirement. Thirdly, the YOLOv5 loss function and prediction box filtering method are improved to increase the detection accuracy and the confidence of the detection boxes. The experimental results show that the GCD-YOLOv5 algorithm has higher accuracy and real-time, the mAP(mean Average Precision) can reach 99.7%, and fps is 68.56% higher compared to YOLOv5, which significantly improves the recognition capability of armored targets in complex battlefield environments. |
| first_indexed | 2024-04-12T12:29:56Z |
| format | Article |
| id | doaj.art-9323a559c8c04f288da16808f63a7c67 |
| institution | Directory Open Access Journal |
| issn | 1943-0655 |
| language | English |
| last_indexed | 2024-04-12T12:29:56Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Photonics Journal |
| spelling | doaj.art-9323a559c8c04f288da16808f63a7c672022-12-22T03:33:03ZengIEEEIEEE Photonics Journal1943-06552022-01-0114411110.1109/JPHOT.2022.31853049803045GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array LidarJian Dai0https://orcid.org/0000-0002-4102-2332Xu Zhao1https://orcid.org/0000-0002-1712-6588Lian Peng Li2https://orcid.org/0000-0002-3595-4897Xiao Fei Ma3Beijing Information Science and Technology University, Beijing, ChinaBeijing Information Science and Technology University, Beijing, ChinaBeijing Information Science and Technology University, Beijing, ChinaBeijing Institute of Control and Electronic Technology, Beijing, ChinaFor the recognition of armored targets in complex battlefield environments, how to reduce missed and false alarms while achieving real-time is an urgent issue. To this end, the GCD-YOLOv5 algorithm is innovatively proposed. Firstly, array lidar is used to acquire the armor target data. Secondly, the armor target data is expanded with an improved GAN(Generative Adversarial Network) to increase the diversity of training data. Afterward, the expanded dataset is fed into the GCD-YOLv5(You Only Look Once) for training. And the GCD-YOLOv5 is reflected in the following aspects. Firstly, the CBAM(Convolutional Block Attention Module) and the multi-scale feature fusion are added to improve the feature extraction capability and detection efficiency, increasing the recognition capability of small and obscured targets. Secondly, combining with DETR(Detection Transformer) to lighten YOLOv5 to achieve the real-time requirement. Thirdly, the YOLOv5 loss function and prediction box filtering method are improved to increase the detection accuracy and the confidence of the detection boxes. The experimental results show that the GCD-YOLOv5 algorithm has higher accuracy and real-time, the mAP(mean Average Precision) can reach 99.7%, and fps is 68.56% higher compared to YOLOv5, which significantly improves the recognition capability of armored targets in complex battlefield environments.https://ieeexplore.ieee.org/document/9803045/Armor targettarget recognitionGANCBAMDETRYOLOv5 |
| spellingShingle | Jian Dai Xu Zhao Lian Peng Li Xiao Fei Ma GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array Lidar IEEE Photonics Journal Armor target target recognition GAN CBAM DETR YOLOv5 |
| title | GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array Lidar |
| title_full | GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array Lidar |
| title_fullStr | GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array Lidar |
| title_full_unstemmed | GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array Lidar |
| title_short | GCD-YOLOv5: An Armored Target Recognition Algorithm in Complex Environments Based on Array Lidar |
| title_sort | gcd yolov5 an armored target recognition algorithm in complex environments based on array lidar |
| topic | Armor target target recognition GAN CBAM DETR YOLOv5 |
| url | https://ieeexplore.ieee.org/document/9803045/ |
| work_keys_str_mv | AT jiandai gcdyolov5anarmoredtargetrecognitionalgorithmincomplexenvironmentsbasedonarraylidar AT xuzhao gcdyolov5anarmoredtargetrecognitionalgorithmincomplexenvironmentsbasedonarraylidar AT lianpengli gcdyolov5anarmoredtargetrecognitionalgorithmincomplexenvironmentsbasedonarraylidar AT xiaofeima gcdyolov5anarmoredtargetrecognitionalgorithmincomplexenvironmentsbasedonarraylidar |