Character Recognition Using Correlation Techniques.
Abstract
This thesis presents an algorithm for recognition of letters of the English alphabet. The letters may be single, isolated letters; or concatenated, groups of letters. The algorithm essentially consists of a two pass correlation in which the size and energy of the unknown characters are set equal to a known template. Originator supplied keywords: Character Recognition, Correlation Techniques.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1984
- Accession Number
- ADA152006
Entities
People
- L. Shum
Organizations
- Air Force Institute of Technology