Automation and speed-up of the RSC N-FINDR algorithm for Endmember extraction

Alex Jivin, Stanley R. Rotman

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Endmember extraction (EE) is an important step in hyperspectral data unmixing. There are many algorithms for Endmember extraction, one of the most common methods is the N-FINDR. The N-FINDR is the base for numerous EE algorithms including the Successive N-FINDR (SC N-FINDR) and the Random N-FINDR (RSC N-FINDR). The main issues of N-FINDR based algorithms is the enormous computation time, inconsistent final endmember selection and the fact that the amount of Endmembers is chosen in advance. This choice is made without knowing the amount of true Endmembers in the hyperspectral data. In this paper, a single run of the k-means++ algorithm is suggested in order to automate the process. The solution of the immense computation time problem consists in the use orthogonal projection of the found Endmember on the rest of the pixels, and using SAM method in order to eliminate similar pixels from the image.

Original languageEnglish
Title of host publication2016 IEEE International Conference on the Science of Electrical Engineering, ICSEE 2016
PublisherInstitute of Electrical and Electronics Engineers
ISBN (Electronic)9781509021529
DOIs
StatePublished - 4 Jan 2017
Event2016 IEEE International Conference on the Science of Electrical Engineering, ICSEE 2016 - Eilat, Israel
Duration: 16 Nov 201618 Nov 2016

Publication series

Name2016 IEEE International Conference on the Science of Electrical Engineering, ICSEE 2016

Conference

Conference2016 IEEE International Conference on the Science of Electrical Engineering, ICSEE 2016
Country/TerritoryIsrael
CityEilat
Period16/11/1618/11/16

Keywords

  • Endmember
  • Hyperspectral
  • K-means++
  • N-Finder
  • OSP
  • Orthogonal projection
  • Pure-pixel
  • Random Successive N-FINDR (RSC-N FINDR)
  • SAM algorithm
  • Simplex
  • Successive N-FINDR (SCN-FINDR)
  • Target detection

ASJC Scopus subject areas

  • Computer Science Applications
  • Hardware and Architecture
  • Artificial Intelligence
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Automation and speed-up of the RSC N-FINDR algorithm for Endmember extraction'. Together they form a unique fingerprint.

Cite this