Please use this identifier to cite or link to this item:
|Title:||Image-based cloud detection, tracking, and analysis||Authors:||Ting, Guan Zhao.||Keywords:||DRNTU::Engineering::Electrical and electronic engineering::Satellite telecommunication
DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
|Issue Date:||2013||Abstract:||This FYP Final Report seeks to provide a full account of the author’s contributions in the development of MATLAB codes to achieve image-based cloud detection capabilities. The report will discuss the works of the author during the past two semesters. The objective of this project is to study the effects of clouds on signal strength in satellite communications. This can be achieved by studying image-based techniques to detect clouds and track cloud movements through a series of images of clouds taken with a whole sky imager. The author will be studying image processing techniques to detect clouds in images, which will then be correlated with signal strength in satellite communications. The report will discuss on the various analysis techniques that were implemented during Phase 1 of project development. Techniques such as grayscale analysis and histogram analysis will provide insights into an image’s intensity value and how it is related to creating a classification threshold. Color histogram analysis will ultimately lead to the implementation of R/B ratio threshold technique, which is a technique that utilizes an image’s red and blue color plane to achieve successful cloud and sky differentiation. The report will then discuss on Phase 2 of project development, which was mainly focused on statistical image analysis that was applied to a database of training images. Mean R/B ratios and standard deviation for respective regions, sky and cloud, would be computed using a set of equations, and the values would be used for graphical analysis in the attempt to determine a relationship between the mean R/B ratio and an image’s properties such as aperture and shutter speed. MATLAB code enhancements such as automated imaging loading and cropping will also be discussed. These enhancements will complement the analysis of a large database of images. The report will end with a conclusion to the project and also the author’s recommendations on various areas of interest that can be explored for the future development of this project.||URI:||http://hdl.handle.net/10356/53421||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Student Reports (FYP/IA/PA/PI)|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.