Satellite imagery and change detection for land cover

• Main Author: Liu, Lu
• Other Authors: Lu Yilong
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/71753
• Institution: Nanyang Technological University

Satellite imagery is widely used for studying the surface of earth. The optical images which contain multi-spectral bands provide a great amount of information which cannot be observed by human eyes directly. As large amount of satellite images is produced, processing the images to gather useful data becomes critical. Change detection is one of the common topic in satellite image processing. The applications of change detection include agriculture, urban planning, ocean study, etc. Proposed by Twenty First Century Aerospace Technology (Asia) Pte Ltd (21AT), this project conducts a research on potential change detection algorithms to find out the efficient and accurate method of change detection which is suitable for their cases. During this project, a MATLAB image change detection toolbox is built containing five change detection methods, including Image Differencing, Normalized Difference Vegetation Index, Principal Component Analysis, K-means Clustering and Maximum Likelihood Classification. In this report, five change detection methods will be compared through three case studies in Singapore, India and China respectively. All methods show their advantages and disadvantages. Among them, Image Differencing and Principal Component Analysis are recommended for similar cases.