ACKNOWLEDGEMENT
ABSTRACT
摘要
Contents
CHAPTER 1:INTRODUCTION
1.1 Background of the study
1.2 Significance of the study
1.3 Literature Review
1.3.1 The possibility of oil spill detection
1.3.2 Optical properties of oil
1.3.3.Visible properties of oil
1.3.4 Infrared(IR)sensor in oil spill detection
1.3.5 Near-infrared Sensor in oil spill detection
1.3.6 Ultraviolet Sensor in oil spill detection
1.3.7 Microwave sensor in oil spill detection
1.3.8 Laser Fluorescence Sensor(LFS)in oil spill detection
1.4 Oil Spill Detection Models
1.4.1 Supervised and Unsupervised Ciassification
1.4.2 Neural Network
1.4.3 Band combination
1.4.4 Water Extraction
1.4.5 Principal component analysis(PCA)
1.4.6 Review of different models
1.5 Problem Statement
1.6 Objectives
1.7 Key Questions
1.8 Organization of the Thesis
CHAPTER 2 STUDY AREA/PRE-DATA PROCESSING
2.1 Study Area
2.2 Data Processing
2.2.1 Collection of image data
2.2.2 Atmospheric Correction
CHAPTER 3.OIL SPILL EXTRACTION BY IMAGE CLASSIFICATION
3.1 Introduction
3.2 Feature Extraction
3.3 Training
3.4 Supervised Classification
3.4.1 Parallelepiped Classification
3.4.2 Minimum Distance Classifier
3.4.3 Mahalanobis Distance Classifier
3.4.4 Maximum Likelihood
3.4.7 Binary Encoding
3.4.8 Neural Network
3.5 Unsupervised Classification
3.5.1 K-means Clustering
3.5.2 Iterative Self-Organizing Data Analysis(ISODATA)
3.6 Post-classification Filtering
3.7 Unsupervised and Supervised Classification
3.8 Experimental Results
3.9 Performance Evaluation
3.10 Conclusions
CHAPTER 4.OIL SPILL EXTRACTION BY BAND/BAND COMBINATION
4.1 Water extraction
4.2 Band Combinations
4.3 Band Operation
4.4 Multispectral Neural Network
CHAPTER 5.COMPARISON OF THE OIL SPILL EXTRACTION METHOD,DISCUSSION,LIMITATIONS AND RECOMMENDATIONS
5.1 Comparison of methods
5.2 Results
5.3 Discussion
5.4 Limitation
CHAPTER 6.CONCLUSIONS AND FUTURE WORK
REFERENCES