A new recognition algorithm of 2-D objects

Author

Xiaoping, Du ; Lei, SHI ; Deren, GU

Author_Institution

Dept. of Electron. Eng., Chengdu Univ. of Sci. & Technol., China

Volume

2

fYear

1996

Firstpage

1342

Abstract

We present a new algorithm to recognize a two-dimensional object of a general shape which has a closed contour. The algorithm includes two stages. The first stage consists of building a one-dimensional signal representation of the two-dimensional boundary of the object and obtaining the wavelet transform of the representation by using a particular mother wavelet which is the first derivative of a smoothing function. The second stage consists of calculating the rise-fall periods of the wavelet transform and its transfer probability matrix as the feature template, and then completing the matching procedure for objects recognition

Keywords

feature extraction; image matching; matrix algebra; object recognition; probability; signal representation; smoothing methods; wavelet transforms; 2D objects; closed contour; feature template; matching procedure; mother wavelet; object recognition; object shape recognition; one-dimensional signal representation; recognition algorithm; rise-fall periods; smoothing function; transfer probability matrix; two-dimensional boundary; two-dimensional object; wavelet transform; Application software; Computer industry; Computer vision; Continuous wavelet transforms; Discrete wavelet transforms; Object recognition; Pattern recognition; Shape; Smoothing methods; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, 1996., 3rd International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-2912-0

Type

conf

DOI

10.1109/ICSIGP.1996.566546

Filename

566546