An automated system for document recognition

Wei Chung Lin*, Yu Jen Eugene Feng

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

In this paper, an automated document recognition system is presented. The distinctive feature of this system is that it recognizes a document based on the invariant horizontal line segments instead of extracting the key words in the document. In other words, it avoids the more complicated character-recognition or text-understanding techniques. The system consists of four major components: digitizer, preprocessor, feature-extractor, and line-pattern classifier. The digitizer converts input documents to digital bit-map images. The preprocessor scales the digitized image to a suitable level of resolution and eliminates artifacts produced by the digitizer. The feature-extractor locates line segments in the scaled image and adjusts them to minimize skew and shift effects. The line-pattern classifier takes the adjusted line segments as input and checks the model database to decide whether the unknown document can be classified as one of the prestored model classes. Experimental results are given to demonstrate the performance of the system.

Original languageEnglish (US)
Pages (from-to)120-130
Number of pages11
JournalEngineering Applications of Artificial Intelligence
Volume2
Issue number2
DOIs
StatePublished - Jun 1989

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'An automated system for document recognition'. Together they form a unique fingerprint.

Cite this