12PrincipalComponentsAnalysisLectureNotes - 1 Geo 4037c...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Geo 4037c Digital Image Analysis The Principal Components Analysis Principal components analysis (PCA)is a multivariate statistical technique that can be used to dramatically reduce the size of the data set. Suppose a study area overlapped with four TM images. Each image contains 7 bands and consumes approximately half a gigabyte of storage space. PCA can be used to reduce the file size and consequently processing time by removing redundant information. The way that PCA works is that the bands are plotted in multidimensional spectral space, and a regression line (or vector, in this case, called an eigenvector) is regressed to the best fit for the data. Optimum coefficients are derived for each band. These coefficients find the best fitting of the eigenvector to the spread of data. Thus, they maximize the ability of the regression eigenvector to explain the spread of data points. A single file is created, called PC1, and this new file is an image of eigenvector 1, which typically accounts for over 90% of the variability in the data set. PC1 is now an image that contains, say, 91% of all the information of the 7 band scene.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/21/2010 for the course GIS 4037c taught by Professor Roberts during the Fall '10 term at FAU.

Page1 / 3

12PrincipalComponentsAnalysisLectureNotes - 1 Geo 4037c...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online