1 AIT Asian Institute of Technology

Land use classification based on Dempster-Shafer probability model

AuthorValera, Novaliza V.
Call NumberAIT Thesis no. CS-92-31
Subject(s)Land use--Remote sensing
Image processing

NoteA thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Engineering and Technology
PublisherAsian Institute of Technology
AbstractIn many circumstances, image classification in remote sensing is in the state of uncertainty. This stems from either a lack of knowledge and information, or from the incompleteness or unreliability of information at disposal. In order to make meaningful judgements, it is !mportĀ·1nt to have a satisfactory way to deal with uncertainty. The most important aspect of any treatment of uncertainty is the rule of combination for updating the degrees of uncertainty by combining more than one body of evidence. In this study, the body of evidence referred to are two images of the same scene taken at different date (also known as multi-temporal images) by Marine Observation Satellite. Two classification methods for combining multi-temporal MOS-1 image data are evaluated. They are named as a Likelihood Addition Classifier (LAC), which is considered as the existing combination method, and a Dempster-Shafer Classifier (DSC), which is based on Dempster-Shafer theory of evidence, particularly the Dempster Rule of Combination. The basic strategies using these methods are to calculate the likelihoods for each temporal data and to combine obtained likelihoods for final classification. Descriptions of the different algorithms are provided and experiments are given to illustrate the potential of the methods for combining output from image classifier, which is the Maximum Likelihood Classifier (MLC).
Year1992
TypeThesis
SchoolSchool of Engineering and Technology (SET)
DepartmentDepartment of Information and Communications Technologies (DICT)
Academic Program/FoSComputer Science (CS)
Chairperson(s)Hosomura, Tsukasa
Examination Committee(s)Huynh, Ngoc Phien ;Vilas Wuwongse
Scholarship Donor(s)Canadian International Development Agency (CIDA) ;
DegreeThesis (M.Sc.) - Asian Institute of Technology, 1992


Usage Metrics
View Detail0
Read PDF0
Download PDF0