Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle Detection
Aiming to solve the problems of false detection, missed detection, and insufficient detection ability of infrared vehicle images, an infrared vehicle target detection algorithm based on the improved YOLOv5 is proposed. The article analyzes the image characteristics of infrared vehicle detection, and...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-07-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/11/15/2344 |
_version_ | 1797442427402846208 |
---|---|
author | Youchen Fan Qianlong Qiu Shunhu Hou Yuhai Li Jiaxuan Xie Mingyu Qin Feihuang Chu |
author_facet | Youchen Fan Qianlong Qiu Shunhu Hou Yuhai Li Jiaxuan Xie Mingyu Qin Feihuang Chu |
author_sort | Youchen Fan |
collection | DOAJ |
description | Aiming to solve the problems of false detection, missed detection, and insufficient detection ability of infrared vehicle images, an infrared vehicle target detection algorithm based on the improved YOLOv5 is proposed. The article analyzes the image characteristics of infrared vehicle detection, and then discusses the improved YOLOv5 algorithm in detail. The algorithm uses the DenseBlock module to increase the ability of shallow feature extraction. The Ghost convolution layer is used to replace the ordinary convolution layer, which increases the redundant feature graph based on linear calculation, improves the network feature extraction ability, and increases the amount of information from the original image. The detection accuracy of the whole network is enhanced by adding a channel attention mechanism and modifying loss function. Finally, the improved performance and comprehensive improved performance of each module are compared with common algorithms. Experimental results show that the detection accuracy of the DenseBlock and EIOU module added alone are improved by 2.5% and 3% compared with the original YOLOv5 algorithm, respectively, and the addition of the Ghost convolution module and SE module alone does not increase significantly. By using the EIOU module as the loss function, the three modules of DenseBlock, Ghost convolution and SE Layer are added to the YOLOv5 algorithm for comparative analysis, of which the combination of DenseBlock and Ghost convolution has the best effect. When adding three modules at the same time, the mAP fluctuation is smaller, which can reach 73.1%, which is 4.6% higher than the original YOLOv5 algorithm. |
first_indexed | 2024-03-09T12:41:42Z |
format | Article |
id | doaj.art-3d4c9b14958747159257638c08b58916 |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-09T12:41:42Z |
publishDate | 2022-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-3d4c9b14958747159257638c08b589162023-11-30T22:17:12ZengMDPI AGElectronics2079-92922022-07-011115234410.3390/electronics11152344Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle DetectionYouchen Fan0Qianlong Qiu1Shunhu Hou2Yuhai Li3Jiaxuan Xie4Mingyu Qin5Feihuang Chu6School of Space Information, Space Engineering University, Beijing 101416, ChinaSchool of Space Information, Space Engineering University, Beijing 101416, ChinaSchool of Space Information, Space Engineering University, Beijing 101416, ChinaSchool of Space Information, Space Engineering University, Beijing 101416, ChinaSchool of Space Information, Space Engineering University, Beijing 101416, ChinaGraduate School, Department of Electronic and Optical Engineering, Space Engineering University, Beijing 101416, ChinaSchool of Space Information, Space Engineering University, Beijing 101416, ChinaAiming to solve the problems of false detection, missed detection, and insufficient detection ability of infrared vehicle images, an infrared vehicle target detection algorithm based on the improved YOLOv5 is proposed. The article analyzes the image characteristics of infrared vehicle detection, and then discusses the improved YOLOv5 algorithm in detail. The algorithm uses the DenseBlock module to increase the ability of shallow feature extraction. The Ghost convolution layer is used to replace the ordinary convolution layer, which increases the redundant feature graph based on linear calculation, improves the network feature extraction ability, and increases the amount of information from the original image. The detection accuracy of the whole network is enhanced by adding a channel attention mechanism and modifying loss function. Finally, the improved performance and comprehensive improved performance of each module are compared with common algorithms. Experimental results show that the detection accuracy of the DenseBlock and EIOU module added alone are improved by 2.5% and 3% compared with the original YOLOv5 algorithm, respectively, and the addition of the Ghost convolution module and SE module alone does not increase significantly. By using the EIOU module as the loss function, the three modules of DenseBlock, Ghost convolution and SE Layer are added to the YOLOv5 algorithm for comparative analysis, of which the combination of DenseBlock and Ghost convolution has the best effect. When adding three modules at the same time, the mAP fluctuation is smaller, which can reach 73.1%, which is 4.6% higher than the original YOLOv5 algorithm.https://www.mdpi.com/2079-9292/11/15/2344target detectioninfrareddeep learningYOLOv5 algorithm |
spellingShingle | Youchen Fan Qianlong Qiu Shunhu Hou Yuhai Li Jiaxuan Xie Mingyu Qin Feihuang Chu Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle Detection Electronics target detection infrared deep learning YOLOv5 algorithm |
title | Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle Detection |
title_full | Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle Detection |
title_fullStr | Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle Detection |
title_full_unstemmed | Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle Detection |
title_short | Application of Improved YOLOv5 in Aerial Photographing Infrared Vehicle Detection |
title_sort | application of improved yolov5 in aerial photographing infrared vehicle detection |
topic | target detection infrared deep learning YOLOv5 algorithm |
url | https://www.mdpi.com/2079-9292/11/15/2344 |
work_keys_str_mv | AT youchenfan applicationofimprovedyolov5inaerialphotographinginfraredvehicledetection AT qianlongqiu applicationofimprovedyolov5inaerialphotographinginfraredvehicledetection AT shunhuhou applicationofimprovedyolov5inaerialphotographinginfraredvehicledetection AT yuhaili applicationofimprovedyolov5inaerialphotographinginfraredvehicledetection AT jiaxuanxie applicationofimprovedyolov5inaerialphotographinginfraredvehicledetection AT mingyuqin applicationofimprovedyolov5inaerialphotographinginfraredvehicledetection AT feihuangchu applicationofimprovedyolov5inaerialphotographinginfraredvehicledetection |