User I ndependent Activity Recognition on Triaxial Accelerometer Data using Neural Network Classifier with Genetic Algorithm F eature S election
- 발행기관 아주대학교
- 지도교수 Seok-Won Lee
- 발행년도 2014
- 학위수여년월 2014. 6
- 학위명 석사
- 학과 및 전공 Graduate School 컴퓨터공학과
- 실제URI http://www.dcollection.net/handler/ajou/000000017497
- 본문언어 영어
- 저작권 아주대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
Advancement in wireless sensor networks gave birth to applications that can provide friendly and intelligent services based on the recognition of human activities. Although the technology supports monitoring activity patterns, enabling applications to recognize activities user independently is still a main concern. Achieving this goal is touch for two reasons: Firstly, different people exhibit different physical patterns for the same activity due to their different behavior. Secondly, different activities performed by the same person could have different underlying models. Therefore, it is unwise to recognize different activities using the same features. This work presents a solution to this problem. The proposed system uses simple time domain features with a single neural network and a three -stage genetic algorithm-based feature selection method for accurate user independent activity recognition. System evaluation is carried out for six activities in a user independent setting using 27 subjects. Recognition performance is also compared with well-known existing methods. Average accuracy of 93% in these experiments shows the feasibility of using our method for subject independent human activity recognition.
more목차
ACKNOWLEDGEMENTS i
ABSTRACT iii
TABLE OF CONTENTS iv
LIST OF FIGURES vi
LIST OF TABLES viii
CHAPTER 1. Introduction 1
1.1 Motivation 1
1.2 Challenges 2
1.3 Contribution 3
CHAPTER 2. Related Work 4
2.1 Human Activity Recognition 4
2.2 Data Collection 6
2.2.1 Noise Reduction 7
2.2.2 Windowing Technique 7
2.3 Feature Extraction 8
2.4 Feature Selection 9
2.5 Activity Classification 10
CHAPTER 3. Proposed Methodology 12
CHAPTER 4. Data Collection 15
4.1 Data Collection Application 15
4.2 Dataset 16
4.3 Noise Reduction 16
4.4 Windowing Technique 17
CHAPTER 5. Features of Signal 18
5.1 Simple Time Domain Features 18
5.2 Time Series Modeling Feature 18
5.2.1 Autoregressive 20
5.2.2 Moving Average 21
5.2.3 Autoregressive-Moving Average 21
5.3 Feature Analysis 22
CHAPTER 6. Features Selection 25
6.1 Genetic Algorithm Overview 25
6.2 Proposed Feature Selection 26
6.2.1 First Stage Feature Selection 26
6.2.2 Second Stage Feature Selection 27
6.2.3 Third Stage Feature Selection 28
CHAPTER 7. Activity Classification 30
7.1 Neural Network Overview 30
7.2 Network Setting 31
7.3 Classification Process 32
CHAPTER 8. Performance Evaluation and Comparison 33
8.1 Experimental Design 33
8.1.1 Study Questions 33
8.1.2 Study Proposition 33
8.1.3 Unit of Analysis 34
8.1.4 Linking Data 35
8.1.5 Criteria to Interpret The Finding 36
8.2 Experimental Result 36
CHAPTER 9. Conclusion 44
REFERENCES 45
