Deep Learning-Based Multiple Droplet Contamination Detector for Vision Systems Using a You Only Look Once Algorithm

被引:1
作者
Kim, Youngkwang [1 ]
Kim, Woochan [1 ]
Yoon, Jungwoo [1 ]
Chung, Sangkug [1 ]
Kim, Daegeun [2 ]
机构
[1] Myongji Univ, Dept Mech Engn, Yongin 17058, South Korea
[2] Microsyst Inc, Yongin 17058, South Korea
关键词
object detection; classification; contamination detection; autonomous driving systems; machine leaning; NEURAL-NETWORKS;
D O I
10.3390/info15030134
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper presents a practical contamination detection system for camera lenses using image analysis with deep learning. The proposed system can detect contamination in camera digital images through contamination learning utilizing deep learning, and it aims to prevent performance degradation of intelligent vision systems due to lens contamination in cameras. This system is based on the object detection algorithm YOLO (v5n, v5s, v5m, v5l, and v5x), which is trained with 4000 images captured under different lighting and background conditions. The trained models showed that the average precision improves as the algorithm size increases, especially for YOLOv5x, which showed excellent efficiency in detecting droplet contamination within 23 ms. They also achieved an average precision (mAP@0.5) of 87.46%, recall (mAP@0.5:0.95) of 51.90%, precision of 90.28%, recall of 81.47%, and F1 score of 85.64%. As a proof of concept, we demonstrated the identification and removal of contamination on camera lenses by integrating a contamination detection system and a transparent heater-based cleaning system. The proposed system is anticipated to be applied to autonomous driving systems, public safety surveillance cameras, environmental monitoring drones, etc., to increase operational safety and reliability.
引用
收藏
页数:14
相关论文
共 50 条
  • [41] PANet: Few-Shot Image Semantic Segmentation with Prototype Alignment
    Wang, Kaixin
    Liew, Jun Hao
    Zou, Yingtian
    Zhou, Daquan
    Feng, Jiashi
    [J]. 2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 9196 - 9205
  • [42] A Smoke Detection Model Based on Improved YOLOv5
    Wang, Zhong
    Wu, Lei
    Li, Tong
    Shi, Peibei
    [J]. MATHEMATICS, 2022, 10 (07)
  • [43] Wentao Liu, 2021, IOP Conference Series: Earth and Environmental Science, V769, DOI 10.1088/1755-1315/769/4/042069
  • [44] A Forest Fire Detection System Based on Ensemble Learning
    Xu, Renjie
    Lin, Haifeng
    Lu, Kangjie
    Cao, Lin
    Liu, Yunfei
    [J]. FORESTS, 2021, 12 (02): : 1 - 17
  • [45] A Small Target Forest Fire Detection Model Based on YOLOv5 Improvement
    Xue, Zhenyang
    Lin, Haifeng
    Wang, Fang
    [J]. FORESTS, 2022, 13 (08):
  • [46] A Real-Time Apple Targets Detection Method for Picking Robot Based on Improved YOLOv5
    Yan, Bin
    Fan, Pan
    Lei, Xiaoyan
    Liu, Zhijie
    Yang, Fuzeng
    [J]. REMOTE SENSING, 2021, 13 (09)
  • [47] Sensor and Sensor Fusion Technology in Autonomous Vehicles: A Review
    Yeong, De Jong
    Velasco-Hernandez, Gustavo
    Barry, John
    Walsh, Joseph
    [J]. SENSORS, 2021, 21 (06) : 1 - 37
  • [48] THE IMPACT OF ADVERSE WEATHER CONDITIONS ON AUTONOMOUS VEHICLES How Rain, Snow, Fog, and Hail Affect the Performance of a Self-Driving Car
    Zang, Shizhe
    Ding, Ming
    Smith, David
    Tyler, Paul
    Rakotoarivelo, Thierry
    Kaafar, Mohamed Ali
    [J]. IEEE VEHICULAR TECHNOLOGY MAGAZINE, 2019, 14 (02): : 103 - 111
  • [49] Self-detection of optical contamination or occlusion in vehicle vision systems
    Zhang, Yi
    Yang, Jie
    Liu, Kun
    Zhang, Xiang
    [J]. OPTICAL ENGINEERING, 2008, 47 (06)
  • [50] A Survey of ADAS Technologies for the Future Perspective of Sensor Fusion
    Ziebinski, Adam
    Cupek, Rafal
    Erdogan, Hueseyin
    Waechter, Sonja
    [J]. COMPUTATIONAL COLLECTIVE INTELLIGENCE, ICCCI 2016, PT II, 2016, 9876 : 135 - 146