Please use this identifier to cite or link to this item:
Type: Artigo de evento
Title: Parts Classification In Assembly Lines Using Multilayer Feedforward Neural Networks
Author: Ferreira Costa Jose Alfredo
de Andrade Netto Marcio Luiz
Abstract: This paper describes a low cost system for a position, scale, and rotation invariant classification of mechanical parts in assembly lines using multilayer feedforward neural networks. After image acquisition, moment invariants are calculated for each significant region in the input image. Different network sizes were tested for classifying these features and we compare these results with the traditional k-nearest neighbor (k-NN), for different k values. Hybrid strategies were adopted for training the networks. We used deterministic methods, such as conjugate gradient and the Levenberg-Marquardt algorithms, combined with a stochastic method, simulated annealing. The system deals with digital images with unknown number of unoccluded object types and poses. Results show that, in our case, artificial neural networks performed best generalization capability than k-NN, despite geometrical transformations and other degradations over the images. The system runs on low cost personal computers, therefore it can be easily adapted for being used even by small factories.
Editor: IEEE, Piscataway, NJ, United States
Rights: fechado
Identifier DOI: 
Date Issue: 1997
Appears in Collections:Unicamp - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
2-s2.0-0031352165.pdf636.96 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.