Algorithm and architecture for feature extraction in image recognition

Author

Starzyk, Janusz A. ; Jan, Ying-Wei

Author_Institution

Dept. of Electr. & Comput. Eng., Ohio Univ., Athens, OH, USA

fYear

1994

fDate

20-22 Mar 1994

Firstpage

448

Lastpage

452

Abstract

The authors present a method of measuring the degree of similarity of handwritten characters. This method is based on the segments of handwritten characters. Each handwritten image is decomposed into several segments. Then they compare the segments which belong to different handwritten images using their distance fields. The result is distance of two handwritten images. These distances can be used in training a neural network for the task of handwritten character recognition. In order to make clear segments from handwritten characters, before the measurement, a hardware oriented thinning algorithm is introduced. Every pixel in a handwritten image can be processed in parallel to save time

Keywords

character recognition; feature extraction; image recognition; image segmentation; neural nets; parallel processing; architecture; feature extraction; handwritten character recognition; handwritten image; handwritten images; image recognition; image segments; neural network training; parallel processing; pixel; similarity measurement; thinning algorithm; Feature extraction; Field programmable gate arrays; Hardware; Humans; Image recognition; Iterative algorithms; Parallel processing; Pixel; Programmable logic arrays; Writing;

fLanguage

English

Publisher

ieee

Conference_Titel

System Theory, 1994., Proceedings of the 26th Southeastern Symposium on

Conference_Location

Athens, OH

ISSN

0094-2898

Print_ISBN

0-8186-5320-5

Type

conf

DOI

10.1109/SSST.1994.287835

Filename

287835