@article{ref1,
title="A study on the implementation of camera sensor for object detection based on deep learning and distance estimation",
journal="Transactions of the Korean Society of Automotive Engineers",
year="2022",
author="Hwang, Seonhee and Choi, Jihun and Park, Hasun and Kim, Jonghyuk",
volume="30",
number="8",
pages="659-665",
abstract="Autonomous Emergency Braking(AEB), one of the core technologies of Advanced Driver Assistance System (ADAS), provides drivers convenience and is effective for safety driving. As the number of AEB-equipped vehicles increases, it is expected that requests for analyzing traffic accidents involving AEB-equipped vehicles will rise. In this study, a deep learning-based camera sensor was developed to implement the function of the AEB in Prescan for reconstruction and analysis of traffic accidents related to AEB. The deep learning algorithm, YOLO(You Only Look Once), was used to detect a front vehicle and to obtain the coordinates of the bounding box. The intrinsic parameters of the camera were calculated through camera calibration. The relative distance to the front vehicle was estimated using the coordinates of the bounding box and the calibrated camera parameter. The intrinsic parameters were compared within the standard error range to find the most similar value to the relative distance measured by the idealized sensor. Based on the results, the optimal parameter was obtained and a more accurate relative distance could be estimated using these values.   키워드: 첨단 운전자 보조 시스템, 긴급 제동 장치, 딥러닝, 딥러닝 객체 검출 알고리즘, 카메라 캘리브레이션, ADAS 시뮬레이션 프로그램<p /> <p>Language: ko</p>",
language="ko",
issn="1225-6382",
doi="10.7467/KSAE.2022.30.8.659",
url="http://dx.doi.org/10.7467/KSAE.2022.30.8.659"
}