Author :
Munson, John H. ; Duda, Richard O. ; Hart, Peter E.
Abstract :
Abstract—The results of three experiments with Highleyman´s hand-printed characters are reported. Nearest-neighbor classification is used to explain the high error rates (42 to 60 percent) obtained by general statistical procedures. An error rate of 32 percent is obtained by preceding piecewise-linear classification by edge-detecting preprocessing. The minimum human error rate is estimated, and suggested as a performance standard.
Keywords :
Index Terms—Character recognition, classification, feature extraction, human performance, nearest-neighbor classification, pattern recognition, preprocessing.; Character recognition; Decision trees; Error analysis; Feature extraction; Humans; Neural networks; Pattern recognition; Piecewise linear techniques; Testing; Training data; Index Terms—Character recognition, classification, feature extraction, human performance, nearest-neighbor classification, pattern recognition, preprocessing.;
