Home // ICAS 2011, The Seventh International Conference on Autonomic and Autonomous Systems // View article


Modular Structure of Neural Networks for Classification of Wooden Surfaces with PLC Industrial Implementation

Authors:
Irina Topalova
Alexander Tzokev

Keywords: texture classification; machine vision; automation; PLC

Abstract:
This paper presents development and research results, applying new approaches and means to the design of Modular Structure of Neural Network for classification of wooden balks (MSNN). It is based on machine vision, unified recognition algorithm and modular neural network structure for real time operation in a standard Programmable Logic Controller (PLC). MSNN is modular, which provides possibility for constructing different structures using in parallel many neural network function blocks with different topologies. It includes development of decision making method for obtaining high recognition accuracy in texture classification using histograms as input data. The method simplicity combined with the modular performance contributes to fast computations and high flexibility of the proposed system. The modular MSNN, containing standard functional blocks, can find application in different applied science fields.

Pages: 13 to 17

Copyright: Copyright (c) IARIA, 2011

Publication date: May 22, 2011

Published in: conference

ISSN: 2308-3913

ISBN: 978-1-61208-134-2

Location: Venice/Mestre, Italy

Dates: from May 22, 2011 to May 27, 2011