Analysis of forest classification accuracy.
The following study was proposed with these objectives: to develop a computer system that implements an iterative proportional fitting technique to 'normalize' the coefficients within classification error matrices; to develop hierarchical models for testing the significance of several factors (e.g., image date, classification algorithm, the analyst, etc.) on the resulting classification accuracy, and to test the above techniques for classification accuracy. -from US Govt Reports Announcements, 11, 1982
Natural Resources and the Environment
Remote Sensing Research Report
Congalton, R. G., R. A. Mead, R. G. Oderwald, and J. Heinen. 1981. Analysis of forest classification accuracy. Remote Sensing Research Report 81-1. Agristars Report RR-U1-04066, JSC 17123. 85 pp.
Work of the US Gov. Public Use Permitted