Multivariate change detection based on canonical transformation

Jie Yang; Deren Li; Pan Zhu; Wen Yang

doi:10.1117/12.441380

25 September 2001 Multivariate change detection based on canonical transformation

Jie Yang, Deren Li, Pan Zhu, Wen Yang

Proceedings Volume 4548, Multispectral and Hyperspectral Image Acquisition and Processing; (2001) https://doi.org/10.1117/12.441380
Event: Multispectral Image Processing and Pattern Recognition, 2001, Wuhan, China

Abstract

This article introduces the multivariate change detection which is based on the established canonical correlation analysis. It also proposes using post processing of the change detected by the multivariate change detection variables using maximum autocorrelation factor analysis. Differing from traditional schemes, the strategy takes two multivariate satellite images covering the same geographic area acquired at different points in time as a random whole sample and transforms two sets of random variables into one set of new random multivariate by using the so-called canonical transformation introduced in the paper. In doing so, the correlation between spectral bands in the same image and in the two different images is removed out as much as possible that the actual changes in all channels simultaneously can be accurately detected. The strategy is invariant to linear scaling. Therefore, it is insensitive to differences in gaining settings in a measuring device, or to linear radiometric and atmosphere correction schemes. The experimental results show the fact that the presented method is exactly creditable and effective on multivariate change detection of remote sensing satellite data.

Citation Download Citation

Jie Yang, Deren Li, Pan Zhu, and Wen Yang "Multivariate change detection based on canonical transformation", Proc. SPIE 4548, Multispectral and Hyperspectral Image Acquisition and Processing, (25 September 2001); https://doi.org/10.1117/12.441380

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available