Experiments with Highleyman's Data
John H. Munson, Richard O. Duda, Peter E. Hart · IEEE Transactions on Computers · 1968
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.
Cited by 22 works. doi:10.1109/tc.1968.229391
Authors
References
- Nearest neighbor pattern classification
- Linear Decision Functions, with Application to Pattern Recognition
- GRAPHICAL-DATA-PROCESSING RESEARCH STUDY AND EXPERIMENTAL INVESTIGATION
- Pattern Classification by Iteratively Determined Linear and Piecewise Linear Discriminant Functions
- A Recognition Method Using Neighbor Dependence
- AN ASYMPTOTIC ANALYSIS OF THE NEAREST-NEIGHBOR DECISION RULE.
- An Analog Method for Character Recognition
- Comments on a Character Recognition Method of Bledsoe and Browning
- A note on human recognition of hand-printed characters
- Further Results on the N-tuple Pattern Recognition Method