Application of ART-2 Neural Network and Invariant Moment in Image Pattern Recognition

Article Preview

Abstract:

This paper mainly uses image pre-processing and feature extraction to calculate the invariant moment of image, and ultimately realizes the image pattern recognition based on ART-2 neural network. Experimental results show that ART-2 neural network has high recognition rate. It also solves the contradiction between network's plasticity and stability, when new recognition model appears.

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 433-440)

Pages:

4014-4019

Citation:

Online since:

January 2012

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2012 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] Barbara Zitova, etal. Robust detection of significant points in mufti- frame image[J]. Pattern Recognition Letters, 1999, 20(2): 199-206.

Google Scholar

[2] Sherstinsky. A, Picard. R. W. On the efficiency of the orthogonal least square straining method for radial basis function networks[J]. IEEE Trans. Neural Networks, 1996, 7(1): 195-200.

DOI: 10.1109/72.478404

Google Scholar

[3] Ao. G,Akazawa. H, lzumi. M, Fukunaga. K. A method of model-based object recognition[C]. Japan/USA Symposium on Flexible Automation, 1996, Vol. 2: 905-912. TABLE I. data and identify the results of moment invariants Graph Invariant moment Recognition results Similarity Connector.

Google Scholar

15985161831054 0. 00005025294886 0. 00000058889709 00.

Google Scholar

99999996470383 Screw.

Google Scholar

21761412731229 0. 01106272671171 0. 00489215041824 01.

Google Scholar

99994752367603 Nut.

Google Scholar

41110313033402 0. 00021366605113 0. 00000109955088 10.

Google Scholar

99998126111532 Gasket.

Google Scholar

16719891793747 0. 00001631257643 0. 00000091024859 11.

Google Scholar

99999996422654.

Google Scholar