Vision-based vehicle detection is an important issue for advanced driver assistance
systems. In this paper, we presented an improved multi-vehicle detection and
tracking method using cascade Adaboost and Adaptive Kalman filter(AKF) with
target identity awareness. A cascade Adaboost classifier using Haar-like features
was built for vehicle detection, followed by a more comprehensive verification
process which could refine the vehicle hypothesis in terms of both location and
dimension. In vehicle tracking, each vehicle was tracked with independent identity
by an Adaptive Kalman filter in collaboration with a data association approach.
The AKF adaptively adjusted the measurement and process noise covariance
through on-line stochastic modelling to compensate the dynamics changes. The
data association correctly assigned different detections with tracks using global
nearest neighbour(GNN) algorithm while considering the local validation. During
tracking, a temporal context based track management was proposed to decide
whether to initiate, maintain or terminate the tracks of different objects, thus
suppressing the sparse false alarms and compensating the temporary detection
failures. Finally, the proposed method was tested on various challenging real
roads, and the experimental results showed that the vehicle detection performance
was greatly improved with higher accuracy and robustness.
제 1편 : MATLAB 기본편
1. MATLAB 기본사용편 ···················· 003
1.1 MATLAB 시작하기 ·························· 003
명령창(command Window)에서의 입력 005
도움말(Help)의 이용 ······························· 007
1.2 입력 오류의 수정 ····························· 008
계산의 중지 ·············································· 009
MATLAB 종료하기 ································· 009
1.3 연산과 변수의 할당 ·························· 009
연산자 우선순위 ······································· 011
내장함수 ···················································· 012
1.4 데이터의 표현 ··································· 013
1.5 변수의 처리 ······································· 015
변수 이름 ·················································· 015
clear 명령어 ············································· 016
특수변수와 정수 ······································· 017
whos 명령어 ············································ 017
1.6 벡터와 행렬 ······································· 018
벡터 ··························································· 018
행렬 ·························································· 023
스크린 출력과 억제 ································· 024
1.7 랜덤(Random)수와 복소수 ·············· 025
랜덤 수 ····················································· 025
복소수 ······················································· 027
1.8 기호를 이용한 연산 ·························· 028
기호식에서의 치환 ··································· 029
1.9 코드 파일 ·········································· 030
스크립트 코드 파일 ································· 030
코멘트의 추가 ·········································· 032
함수 코드 파일 ···································· 033
사용자 정의함수 ······································ 036
1.10 간단한 그래프의 생성 ····················· 037
ezplot을 이용한 그래프 ·························· 037
plot을 이용한 그래프 ·························· 039
3차원 그래프 ··········································· 042
1.11 MATLAB과 엑셀(Excel)의 접속 043 엑셀 데이터 불러오기 ····························· 043
데이터 가져오기 옵션 ························· 046
스크립트 생성 옵션 ································· 049
함수 생성 옵션 ········································ 049
생성된 데이터를 엑셀파일로 저장하기 ·· 050
제 2편 : 연구논문
Extended-Kalman-filter-based dynamic mode decomposition for
simultaneous system identification and denoising
1. Introduction 51
2. Cascade Adaboost for vehicle detection 54
3. Hypothesis generation 56
4. Multi-vehicle tracking with identity awareness 57
5. Adaptive Kalman filter 58
6. Experiments 63
7. Multi-vehicle tracking performance 65
8. References 70
