Using entropy for drawing reliable templates

Author

De Stefano, C. ; Tortorella, F. ; Vento, M.

Author_Institution

Dipartimento di Inf. e Sistemistica, Univ. degli Studi di Napoli, Italy

fYear

1993

fDate

20-22 Oct 1993

Firstpage

345

Lastpage

348

Abstract

The method presented allows the drawing of templates that are reliable in locating distorted occurrences of the symbol to recognize and robust against the noise and the false alarms present on the background. A learning by showing technique is used to draw the template by considering samples of the symbol coming from a training set. The design of the template involves the evaluation of the reliablity of each pixel, obtained by estimating the entropy which characterizes the pixel. To be considered reliable, a pixel must show an entropy less than a threshold H_o set so as to maximize a match performance figure: the obtained template is constituted only by the pixels meeting this requirement

Keywords

document handling; document image processing; image matching; learning (artificial intelligence); distorted occurrences; entropy; false alarms; learning by showing technique; match performance figure; pixel; reliable templates; template matching; training set; Computational efficiency; Entropy; Filtering; Image recognition; Matched filters; Noise robustness; Noise shaping; Pixel; Shape; Text analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis and Recognition, 1993., Proceedings of the Second International Conference on

Conference_Location

Tsukuba Science City

Print_ISBN

0-8186-4960-7

Type

conf

DOI

10.1109/ICDAR.1993.395718

Filename

395718