Vehicle Detection at Night Based on Style Transfer Image Enhancement
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Most vehicle detection methods have poor vehicle feature extraction performance at night, and their robustnessis reduced; hence, this study proposes a night vehicle detection method based on style transfer image enhancement. First, a style transfer model is constructed using cycle generative adversarial networks (cycleGANs). Thedaytime data in the BDD100K dataset were converted into nighttime data to form a style dataset. The datasetwas then divided using its labels. Finally, based on a YOLOv5s network, a nighttime vehicle image is detectedfor the reliable recognition of vehicle information in a complex environment. The experimental results of theproposed method based on the BDD100K dataset show that the transferred night vehicle images are clear andmeet the requirements. The precision, recall, mAP@.5, and mAP@.5:.95 reached 0.696, 0.292, 0.761, and0.454, respectively.
Abstract
Most vehicle detection methods have poor vehicle feature extraction performance at night, and their robustnessis reduced; hence, this study proposes a night vehicle detection method based on style transfer image enhancement. First, a style transfer model is constructed using cycle generative adversarial networks (cycleGANs). Thedaytime data in the BDD100K dataset were converted into nighttime data to form a style dataset. The datasetwas then divided using its labels. Finally, based on a YOLOv5s network, a nighttime vehicle image is detectedfor the reliable recognition of vehicle information in a complex environment. The experimental results of theproposed method based on the BDD100K dataset show that the transferred night vehicle images are clear andmeet the requirements. The precision, recall, mAP@.5, and mAP@.5:.95 reached 0.696, 0.292, 0.761, and0.454, respectively.
참고문헌 (10)
- RYOTARO MATSUO, "The Importance of Screening for Sleep Disordered Breathing in the Prevention of Motor Vehicle Crashes: The Toon Health Study", Juntendo Medical Journal, 66(6), pp.476~477 (2020), The Jutendo Medical Journal[DOI]
- A. D. Puzanau, "Synthesis of algorithm of unmanned aerial vehicle detection by acoustic noise", Doklady BGUIR, 19(2), pp.65~73 (2021), Belarusian State University of Informatics and Radioelectronics[DOI]
- Junfeng Lei, "Tiny moving vehicle detection in satellite video with constraints of multiple prior information", International Journal of Remote Sensing, 42(11), pp.4110~4125 (2021), Informa UK Limited[DOI]
- Shahnaj Parvin, "Vision-based On-Road Nighttime Vehicle Detection and Tracking Using Taillight and Headlight Features", Journal of Computer and Communications, 09(03), pp.29~53 (2021), Scientific Research Publishing, Inc.[DOI]
- Meng Shu, "Small moving vehicle detection via local enhancement fusion for satellite video", International Journal of Remote Sensing, 42(19), pp.7189~7214 (2021), Informa UK Limited[DOI]
- Thathupara Subramanyan Kavya, "Vehicle Detection and Tracking from a Video Captured by Moving Host", Indian Journal of Computer Science and Engineering, 11(3), pp.226~235 (2020), ENGG Journals Publications[DOI]
- Latha Anuj , Et. al., "V-DaT: A Robust method for Vehicle Detection and Tracking", Turkish Journal of Computer and Mathematics Education (TURCOMAT), 12(2), pp.2492~2505 (2021), Auricle Technologies, Pvt., Ltd.[DOI]
- Jie Feng, "Cross-frame keypoint-based and spatial motion information-guided networks for moving vehicle detection and tracking in satellite videos", ISPRS Journal of Photogrammetry and Remote Sensing, 177, pp.116~130 (2021), Elsevier BV[DOI]
- Vijay Paidi, "Deep learning‐based vehicle occupancy detection in an open parking lot using thermal camera", IET Intelligent Transport Systems, 14(10), pp.1295~1302 (2020), Institution of Engineering and Technology (IET)[DOI]
- Ji-qing Luo, "Multi-scale traffic vehicle detection based on faster R–CNN with NAS optimization and feature enrichment", Defence Technology, 17(4), pp.1542~1554 (2021), Elsevier BV[DOI]