생몰정보
소속
직위
직업
활동분야
주기
서지
국회도서관 서비스 이용에 대한 안내를 해드립니다.
검색결과 (전체 1건)
원문 있는 자료 (1) 열기
원문 아이콘이 없는 경우 국회도서관 방문 시 책자로 이용 가능
목차보기더보기
Title Page
Abstract
Contents
List of Abbreviations 16
Chapter 1. Introduction 18
1.1. Background 18
1.2. Organization of Dissertation 21
Chapter 2. Signal Enhancement for ship detection using compact High Frequency radar 23
2.1. Issues & related works 23
2.2. Signal Model for a compact HF radar 25
2.2.1. Signal power model 27
2.2.2. Antenna response of a compact HF radar 27
2.2.3. Ship signal model 28
2.2.4. Bragg scattering with surface current signal model 30
2.2.5. Random clutter model 32
2.2.6. Background noise model 32
2.3. Component analysis for Signal enhancement 33
2.3.1. Windowing 34
2.3.2. Component Analysis 35
2.3.3. Candidate Detection 41
2.3.4. Candidate Positioning 41
2.4. Experiments 42
2.4.1. Database 42
2.4.2. Experiment results with simulation data 45
2.4.3. Experiment results with practical data 50
2.5. Conclusions 52
Chapter 3. Acoustic Event Classification using Subspace Projection Cepstral Coefficients 53
3.1. Issues & related works 53
3.2. Subspace Projection Cepstral Coefficients (SPCCs) 54
3.2.1. Motivation 54
3.2.2. Expected effect by projecting onto event subspace 56
3.2.3. Establish subspace-bank 56
3.2.4. SPCC extraction 57
3.3. Experiments 58
3.3.1. Database 58
3.3.2. Experimental settings 59
3.3.3. Experiment results 61
3.4. Conclusions 63
Chapter 4. Acoustic Scene Classification for IEEE AASP Challenge DCASE2016 65
4.1. Issues & related works 65
4.2. Investigation of SPCCs 67
4.3. Structure for Acoustic Scene Classification 69
4.3.1. Feature extraction 69
4.3.2. Classification 70
4.3.3. Score fusion 71
4.4. Experiments 72
4.4.1. Database 72
4.4.2. Experimental settings 73
4.4.3. Experiment results 73
4.5. Conclusions 77
Chapter 5. Conclusions and Future Works 78
5.1. Conclusions 78
5.2. Future Works 80
Bibliography 81
Appendix 89
Curriculum Vitae 96
Table 2.1. Parameters for generating random clutter 32
Table 2.2. Information from the database used in the experiment 44
Table 2.3. Information for the target ship 44
Table 3.1. The database for acoustic event classification and its size 59
Table 4.1. Accuracies of experiment using Development dataset; the number of mixture... 75
Figure 1.1. Two objectives of dimensionality reduction based on subspace projection 18
Figure 1.2. Concept of signal enhancement based on subspace projection 21
Figure 2.1. Two types of HF radar 23
Figure 2.2. Compact HF radar antenna configuration and its responses 28
Figure 2.3. Scenario of ship parameter estimates 29
Figure 2.4. Proposed ship detection system structure using a compact HF radar 33
Figure 2.5. Configuration of RDM frames for ship detection 34
Figure 2.6. Sequentially generated RDMs 36
Figure 2.7. Enhanced RDM at time m with RDMs shown in Figure 4.3 40
Figure 2.8. Candidate positioning based on bearing estimation (MUSIC) and distance... 41
Figure 2.9. Radar site map: squares are the locations of CODAR SeaSonde HF radar 43
Figure 2.10. Five types of ship path: the location labeled mark the starting position of... 45
Figure 2.11. The ROC curve of the conventional method according to OS-CFAR... 46
Figure 2.12. The ROC curves of the conventional method and three approaches for... 47
Figure 2.13. The ROC curves of the conventional method and PCA approaches for... 48
Figure 2.14. The ROC curve obtained in an experiment with practical data according to... 50
Figure 2.15. The ROC curve of four types of CFAR according to the conventional... 51
Figure 3.1. Normalized energy about four types of events 55
Figure 3.2. Procedure for training subspace-bank 57
Figure 3.3. Block diagram for extracting SPCC 58
Figure 3.4. Experiment results using 32 mixture Gaussian Mixture Model 61
Figure 3.5. Experiment results using Support Vector Machine or Deep Belief Network -... 62
Figure 4.1. Representative acoustic sounds in several locations; café, office, and bus 66
Figure 4.2. Feature distributions 67
Figure 4.3. Scores according to classes for each feature vector 68
Figure 4.4. System architecture for acoustic scene classification 69
Figure 4.5. Example for explain a concept of Covariance Discriminative Learning 70
Figure 4.6. Confusion matrices 76
초록보기 더보기
For dimensionality reduction task, hence, mapping from high-dimensional vector space to low-dimensional subspace, two objectives can be considered; signal representation and signal classification. Firstly, the dimensionality reduction for signal representation objective can be considered for representing the original data onto a low-dimensional space in order to save computational load or to extract signal component from noisy observations. This is typically achieved by minimizing discrepancy between the original data and that of the recovered data from compressed version in low-dimensional space, or that of the distorted data in noisy environment. Signal classification is achieved by estimating a subspace that best partitions samples of different classes by maximizing interclass separation while minimizing intra-class distance. Many algorithms have been developed for dimensionality reduction and their performances can be interpreted from these two objectives.
This dissertation addresses signal representation objective of dimensionality reduction, which is aimed at extracting signal component from noisy observation as there are many signal processing applications in real environment. In particular, signal representation is proposed by this dissertation by means of subspace projection toward achieving two following goals; signal enhancement from noisy signal and feature extraction for classification. The specific approaches are developed to achieve these two goals and several applications are presented such as ship detection using a high frequency radar, noise robust acoustic event recognition, and acoustic scene classification.
Firstly, an effective signal enhancement method for improving ship detection performance is developed and applied to High Frequency radar system which has been primarily optimized for observing surface radial current velocities. Previously proposed ship detection algorithms using High Frequency radar systems have been vulnerable to error sources such as environmental noise and clutter when they are applied in compact high frequency radar optimized for observing surface current. To overcome this problem, a signal enhancement method is proposed in this dissertation, for compact HF radar, which projects range-Doppler map data onto a noise suppressed signal subspace obtained by eigenvalue decomposition. The proposed method is then validated by comparing it to the conventional ship detection method in terms of detection and false alarm rates. The experimental results confirm that the proposed method show superior performance in both simulated and practical environments.
Secondly, a novel feature devised for noise robust acoustic event classification is developed. For obtaining the new feature, a subspace-bank is firstly trained via target event analysis in complex vector space. Then, by projecting observation vectors onto a subspace-bank, noise effect is reduced while generating discriminant characters originated from differing event subspaces. To demonstrate robustness of the proposed feature, experiments with several classifiers are conducted with varying SNR cases under four noisy environments. According to the experimental results, the proposed method shows superior performance over prominent conventional methods in terms of accuracy.
Thirdly, the subspace-bank based feature described above is investigated for effectiveness in acoustic scene classification. Even in same space, various overlapping sounds may be emitted depending on the presence of people, objects, and/or their behaviors. This overlapping sound is a critical issue in acoustic scene classification task. This dissertation proposes to resolve the issue by applying the new feature to the acoustic scene classification task. The proposed acoustic scene classification method by exploiting the new feature achieved 4th ranking result in the IEEE AASP challenge of Detection and Classification of Acoustic Scenes and Events 2016. The acoustic scene classification method is demonstrated and validated by the experiments conducted using the Challenge database.
원문구축 및 2018년 이후 자료는 524호에서 직접 열람하십시요.
도서위치안내: / 서가번호:
우편복사 목록담기를 완료하였습니다.
* 표시는 필수사항 입니다.
* 주의: 국회도서관 이용자 모두에게 공유서재로 서비스 됩니다.
저장 되었습니다.
로그인을 하시려면 아이디와 비밀번호를 입력해주세요. 모바일 간편 열람증으로 입실한 경우 회원가입을 해야합니다.
공용 PC이므로 한번 더 로그인 해 주시기 바랍니다.
아이디 또는 비밀번호를 확인해주세요
