Neural Network Techniques for Object Orientation Detection: Solution by Optimal Feedforward Network and Learning Vector Quantization Approaches
Issue No. 11 - November (1990 vol. 12)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/34.61712
<p>The computer-vision problem of determining object orientation from the consensus of orientations of individual symbols or marks is examined. The problem arises in automatic inspection where orientation can be detected from printed text but there is no knowledge of the content of the text. This is a high-dimensional classification problem, and there is a requirement for highly accurate detection and rapid processing. The typical multilayer threshold networks are seen as unsuitable, and the optimal Bayesian detector is derived and found to have the highly parallel structure of a feedforward network. The learning vector quantization neural network method of T. Kohonen (1988) is also applied. Experimental results, comparisons, and a complete implementation are described.</p>
symbol orientation; mark orientation; pattern recognition; character recognition; object orientation detection; optimal feedforward network; learning vector quantization; computer-vision; automatic inspection; high-dimensional classification problem; optimal Bayesian detector; parallel structure; learning vector quantization neural network; automatic optical inspection; computer vision; computerised pattern recognition; learning systems; neural nets
L. Rubin, R. Morris and H. Tirri, "Neural Network Techniques for Object Orientation Detection: Solution by Optimal Feedforward Network and Learning Vector Quantization Approaches," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 12, no. , pp. 1107-1115, 1990.