Document Type
Thesis
Date of Award
10-31-1992
Degree Name
Master of Science in Computer and Information Science - (M.S.)
Department
Computer and Information Science
First Advisor
David T. Wang
Abstract
To handle noisy and distorted pattern is the use of similarity or distance measures. A similarity or distance measure can be defined between a representation of an unknown pattern and a representation of a prototype pattern. Recognition of the unknown pattern can be carried out on the basis of the maximum-similarity or minimum distance criterion (Bunke 1990) This approach is proposed to recognize the noisy or distorted character image. In this work, a directly string representation of the pattern ( prototype as well as unknown input) using the histogram method, a decision procedure for classification is the well known Levenshtein distance or weighted Levenshtein distance (Wanger 1974; Hall and Dowling 1980), the cost of a transformation is specially estimated, and typical elements of the sample set are chosen and a well known statistical decision--nearest neighbor classification (NN-classification) is applied.(P. A. Devijver and J. Kitler 1982) Experiments show that it is an efficient method and it gives satisfactory results.
Recommended Citation
Ye, Xiaozhi, "Character recognition using string matching" (1992). Theses. 2395.
https://digitalcommons.njit.edu/theses/2395
