Improved collaborative representation model with multitask learning using spatial support for target detection in hyperspectral imagery

Chunhui Zhao; Wei Li; G. Arturo Sanchez-Azofeifa; Bin Qi; Bing Cui

doi:10.1117/1.JRS.10.016009

10 February 2016 Improved collaborative representation model with multitask learning using spatial support for target detection in hyperspectral imagery

Chunhui Zhao, Wei Li, G. Arturo Sanchez-Azofeifa, Bin Qi, Bing Cui

Author Affiliations +

Journal of Applied Remote Sensing, Vol. 10, Issue 1, 016009 (February 2016). https://doi.org/10.1117/1.JRS.10.016009

Abstract

We propose an improved collaborative representation model with multitask learning using spatial support (ICRTD-MTL) for target detection (TD) in hyperspectral imagery. The proposed model consists of the following aspects. First, multiple features are extracted from the hyperspectral image to represent pixels from different perspectives. Next, we apply these features into the unified CRTD-MTL to acquire a collaborative vector for each feature. To adjust the contribution of each feature, a weight coefficient is included in the optimization problem. Once the collaborative vector is obtained, the class of the test sample can be determined by the characteristics of the collaborative vector on reconstruction. Finally, the spatial correlation and spectral similarity of adjacent neighboring pixels are incorporated into each feature to improve the detection accuracy. The experimental results suggest that the proposed algorithm obtains an excellent performance.

Citation Download Citation

Chunhui Zhao, Wei Li, G. Arturo Sanchez-Azofeifa, Bin Qi, and Bing Cui "Improved collaborative representation model with multitask learning using spatial support for target detection in hyperspectral imagery," Journal of Applied Remote Sensing 10(1), 016009 (10 February 2016). https://doi.org/10.1117/1.JRS.10.016009

Published: 10 February 2016

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