DGIST Scholar: Performance Improvement of Camera Lane Detection Network Using Image Pre-processing and Polynomial Regression

Department of Electrical Engineering and Computer Science Theses Master

Cited time in webofscience

Cited time in scopus

Performance Improvement of Camera Lane Detection Network Using Image Pre-processing and Polynomial Regression

Title: Performance Improvement of Camera Lane Detection Network Using Image Pre-processing and Polynomial Regression

Description: Lane detection,Histogram equalization,Color space conversion,K-mean clustering,Polynomial regression,차선 인식,히스토그램 평활화,색상 공간,K-평균 군집화,다항식 회귀

Table Of Contents: I. Introduction 1
II. Related Works 4
2.1. Lane detection algorithm using image pre-processing 4
2.2. Building datasets in low illuminance environments 5
2.2.1. CULane 5
2.2.2. BDD100K 6
2.3. Development of the new network model 7
III. Methodology 8
3.1. Overall structure 8
3.2. Image pre-processing 8
3.3. Polynomial regression 10
IV. Experiments 11
4.1. Lane detection performance improvement using the proposed method 11
4.1.1. Dataset setting 11
4.1.2. Experiment Setting 11
4.2. Generalization of the proposed module 12
4.2.1. Dataset setting 12
4.2.2. Experiment Setting 12
V. Results 13
5.1. Lane detection performance improvement using the proposed method 13
5.1.1 Image pre-processing 13
5.1.2. Polynomial regression 14
5.2. Generalization of the proposed module 16
VI. Conclusion 19
Reference 20
요약문 23