Statistical pattern recognition of low resolution pictures

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

From the features of the connected picture and its edge map of an object to be recognized feature parameters can be obtained. For each object an error-distance is calculated from the feature parameters. These error-distances also have statistical distribution for each object. From these statistics the most probable object and the sureness of recognition decision can be calculated.

Original languageEnglish
Pages (from-to)221-228
Number of pages8
JournalPattern Recognition Letters
Volume8
Issue number4
DOIs
Publication statusPublished - Nov 1988

Keywords

  • Pattern recognition
  • contour-following
  • decision-sureness
  • feature parameters
  • filtering
  • low resolution
  • normal distribution
  • picture-connecting
  • picture-noises
  • pyroelectric sensors
  • robot stations

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Statistical pattern recognition of low resolution pictures'. Together they form a unique fingerprint.

  • Cite this