Common matrix approach-based multispectral image fusion and its application to edge detection

Şahin Isik; Kemal Özkan

doi:10.1117/1.JRS.13.016515

7 February 2019 Common matrix approach-based multispectral image fusion and its application to edge detection

Şahin Isik, Kemal Özkan

Author Affiliations +

Journal of Applied Remote Sensing, Vol. 13, Issue 1, 016515 (February 2019). https://doi.org/10.1117/1.JRS.13.016515

Abstract

We propose an information fusion algorithm—common matrix approach-based edge detection—for edge detection in multispectral images based on a tensor projection approach. In order to merge the edge information situated on different spectral bands, the concept of the common matrix approach is enhanced for each band to project a tensor of gradients onto a derived basis. Once a single-gradient map is acquired from the tensor, a smart thinning and thresholding methodology is utilized at the edge extraction stage in terms of getting rid of spots and discontinuous edge pixels. Upon inspecting the experiments’ results on datasets including the Computer Vision Laboratory at Columbia University, Scene, scanning electron microscopy using the energy-dispersive x-ray microanalysis, and particular hyperspectral remote sensing datasets, it is observed that the proposed method achieves pleasing results in terms of subjective and objective measures in comparison to some recently proposed edge detectors on multispectral and hyperspectral images.

Citation Download Citation

Şahin Isik and Kemal Özkan "Common matrix approach-based multispectral image fusion and its application to edge detection," Journal of Applied Remote Sensing 13(1), 016515 (7 February 2019). https://doi.org/10.1117/1.JRS.13.016515

Received: 11 September 2018; Accepted: 15 January 2019; Published: 7 February 2019

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