Estimation of proportions of objects and determination of training sample-size in a remote sensing applicationA multichannel scanning device may fail to observe objects because of obstructions blocking the view, or different categories of objects may make up a resolution element giving rise to a single observation. Ground truth will be required on any such categories of objects in order to estimate their expected proportions associated with various classes represented in the remote sensing data. Considering the classes to be distributed as multivariate normal with different mean vectors and common covariance, maximum likelihood estimates are given for the expected proportions of objects associated with different classes, using the Bayes procedure for classification of individuals obtained from these classes. An approximate solution for simultaneous confidence intervals on these proportions is given, and thereby a sample-size needed to achieve a desired amount of accuracy for the estimates is determined.
Document ID
19740034827
Acquisition Source
Legacy CDMS
Document Type
Conference Proceedings
Authors
Chhikara, R. S.
Odell, P. L. (Texas, University Dallas, Tex., United States)
Date Acquired
August 7, 2013
Publication Date
January 1, 1973
Subject Category
Geophysics
Meeting Information
Meeting: Conference on Machine processing of remotely sensed data