Issue No.04 - April (1968 vol.17)
J.H. Munson , IEEE
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.
Index Terms?Character recognition, classification, feature extraction, human performance, nearest-neighbor classification, pattern recognition, preprocessing.
J.H. Munson, R.O. Duda, P.E. Hart, "Experiments with Highleyman's Data", IEEE Transactions on Computers, vol.17, no. 4, pp. 399-401, April 1968, doi:10.1109/TC.1968.229391