Contribution to a conference proceedings/Contribution to a book

FZJ-2019-03517

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png A Unified Approach to the Processing of Hyperspectral Images

Burgeth, B. (Corresponding author) ; Didas, S. ; Kleefeld, A.FZJ*

2019
Springer International Publishing Cham
ISBN: 978-3-030-20866-0 (print), 978-3-030-20867-7 (electronic)

This record in other databases:  

Please use a persistent id in citations: http://hdl.handle.net/2128/22385  doi:10.1007/978-3-030-20867-7_16

Abstract: Since vector fields, such as RGB-color, multispectral or hyperspectral images, possess only limited algebraic and ordering structures they do not lend themselves easily to image processing methods. However, for fields of symmetric matrices a sufficiently elaborate calculus, that includes, for example, suitable notions of multiplication, supremum/infimum and concatenation with real functions, is available. In this article a vector field is coded as a matrix field, which is then processed by means of the matrix valued counterparts of image processing methods. An approximate decoding step transforms a processed matrix field back into a vector field. Here we focus on proposing suitable notions of a pseudo-supremum/infimum of two vectors/colors and a PDE-based dilation/erosion process of color images as a proof-of-concept. In principle there is no restriction on the dimension of the vectors considered. Experiments, mainly on RGB-images for presentation reasons, will reveal the merits and the shortcomings of the proposed methods.

---

Contributing Institute(s):

1. Jülich Supercomputing Center (JSC)

Research Program(s):

1. 511 - Computational Science and Mathematical Methods (POF3-511) (POF3-511)

Appears in the scientific report 2019

---

Database coverage:
; Nationallizenz ; SCOPUS

---