Loading...
A recognition system for stamped number on non-smooth metallic surface
Author(s)
Date Issued
January 1, 2009
Type
Conference Paper
Abstract
This paper deals with the application of moments for recognizing of stamped number on metallic surface which contains the stamped number. It is a complicated task to be read automatically. Recently, the method of modeling and analyzing 3D surface data using a laser-scanning instrument has been presented. The advantage of the combined moment method is that it is able to eliminate some of the problem associated with the Gibbs phenomena[1][2]. This method is also applied to real-time modeling of 3D surface encountered to separate the global geometry from periodic undulations and local anomalies. Afterwards, the image of numbers on the steel block is clear enough to perform numbers recognition. In this paper, pattern recognition based on moment function such as Hu's moment invariants, Zernike moments, and Orthogonal Fourier Mellin Moments is used to recognize the stamped number after analyzing 3D surface data of the steel block. The results of classification rate are satisfactory, and it is possible to implement in an industrial environment.
Citation
Asia Simulation Conference 2009 Jsst 2009, 2009