PENGESTIMASIAN IMPERVIOUS SURFACE AREA DATA LANDSAT DENGAN MULTIPLE LINEAR REGRESSION

Abstract

This application uses multiple linear regression for its method. This method useful for making regression model for making estimation appliances which will be used in the testing process. The way how to making the model will be helped with the matrix method of approach by Norman Drapper.

This research uses Landsat image for band 1, 4, and 5, and also the high resolution image of Geo-Eye. The use of Landsat imagery is to obtain the pixel value for each pixel which is its characteristic of an image. And the reason for using the Geo-Eye imagery is for determine the value of impervious surface area of the Landsat image which we have gotten, for training process. Then from the made regression model is used for estimating an image, then the image will be differed which is considered to be impervious surface area and which is not.

Key words

Impervious surface area, multiple linear regression, matrix approach Norman Drapper