메뉴 건너뛰기
.. 내서재 .. 알림
소속 기관/학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
로그인 회원가입 고객센터 ENG
주제분류

추천
검색

논문 기본 정보

자료유형
학위논문
저자정보

권용현, Kwon, Yong Hyun (충북대학교, 충북대학교 대학원)

지도교수
김형원
발행연도
2019
저작권
충북대학교 논문은 저작권에 의해 보호받습니다.

이용수23

표지
AI에게 요청하기
추천
검색

이 논문의 연구 히스토리 (2)

초록· 키워드

오류제보하기
There has been a lot of research on autonomous driving lately. Various applications for self-driving cars carry out object recognition through cameras or various sensors in advance. It is also important to recognize the roadway in which the host vehicle drives during this prior recognition. This is because many applications of connected cars or self-driving cars are based on recognition of driving Road lanes. Methods for determining the driving road lane are as follows. Locate the yellow lane line after detecting the lane marker based on the characteristics of the lane marker. Locate the load lane of the host vehicle depending on the location of the yellow lane found. However, we have identified that it is difficult to find a driving lane on roads above three lanes, and that intersections without lane markers are also difficult to identify. This paper presents new ways to utilize V2X and Deep Learning technologies on roads above the three-lane. A Deep Learning model was designed to identify the driving lanes of a vehicle in question using the front image characteristics received from the vehicle within the communication range from the RSU installed on the side of the road. Through this, the predicted accuracy of 90% or more was verified in the three-lane road and intersection environment.

목차

Ⅰ. Introduction 1
Ⅱ. Related work 3
2.1 V2X Communication application 3
2.2 Lane Detection algorithm 4
Ⅲ. Driving road lane prediction technique in 3-lane roads or below 5
3.1 Image resize and cropping 6
3.2 Image filtering 7
3.2.1 Color filtering 7
3.2.2 Histogram Equalization 10
3.2.3 Top Hat filtering 13
3.3 Region Of Interest Setting 17
3.4 Lane Detection 19
3.4.1 Contour detection 20
3.4.2 Hough transform 21
3.4.3 Best Line detection 23
3.5 Center line detection and driving road lane prediction 27
Ⅳ. Driving road lane prediction technique in multi-lane roads and at intersections 31
4.1 Technical Overview 31
4.2 CNN classification model 32
4.3 Road lane Predictive model using V2X and CNN 37
4.4 Semantic Segmentation network 38
4.5 Predictive model for multi-lane roads using Segnet 42
4.5.1 CNN-based Classification model configuration 43
4.5.2 Image Alignment Correction 45
Ⅴ. Simulation result 48
5.1 Simulation condition 48
5.2 Verification methods and results in road environments below three lanes 48
5.3 Method and Results of Verification of Prediction Model Using CNN 51
5.3.1 Road environment with three or more Lane 51
5.3.2 Two lane intersection environment without lane markers 55
Ⅵ. Conclusion 58
References 61

최근 본 자료

전체보기

댓글(0)

0