On-line Vision Recognition of Auto Rack Girders Based on ART2 Neural Network and D-S Evidence Theory

For automatic inspecting kinds of hundreds of camion rack girders, this paper introduces an on-line automatic inspecting method which synthesizes machine vision wavelet transform theory ART2 neural network and D-S evidence theory on auto rack girder. Firstly, for the real-time gathered auto rack girders top image on product line, extracting wavelet decomposed coefficient of image with wavelet transform, energy value of wavelet coefficient is used as a character template; top image of auto rack girders which is partitioned to 16 sub-images (4times4), estimating numbers of edge pixels of each region respectively, which is used as a character template; the primary image is partitioned to 16 sub-images(4times4) in the same way, calculate center of gravity position of each region respectively, which is used as a character template. Secondly, in order to gain basal reliability of auto rack girders image, three character templates data which are energy value of wavelet coefficient numbers of edge pixels and center of gravity position are used as inputs of ART2 neural network. Finally, according to composition rule of D-S evidence theory, to gain recognition results. Experiments indicate online maximal recognition rate meets demands of production, based on combination of art2 neural network with D-S evidence theory to recognize kinds of auto rack girder, and possessed advantage of more rapid and more precise recognition etc.