A comparison of rule-based, k-nearest neighbor, and neuralnet classifiers for automated industrial inspection

摘要

As classifiers for use in automated industrial inspection, thenrule-based, k-nearest-neighbor, and neural-network approachesnare discussed. These approaches were implemented and tested for labelnverification in a machine vision system for hardwood lumber inspection.nThe test results, together with other considerations, have led to thenselection of neural networks as the preferred method for doing the labelnverification in this machine vision system