A hierarchical labeled object classification system

This paper describes the design of a labeled object classification system to be used for product classification at the final inspection stage of an IBM personal computer manufacturing line. The classification problem is broken down into pattern extraction, feature extraction, and feature classification levels. Pattern extraction and normalization are performed by image processing. Normalized images of the labels so obtained are compressed using an autoassociative network. Features extracted in this manner are used as inputs to a second learning vector quantization (LVQ) network trained to classify the labels. The system so designed is shown to satisfy the primary requirements of a typical industrial classification system