Combination of multiple classifiers by minimizing the upper bound of bayes error rate for unconstrained handwritten numeral recognition.
Hee-Joong KangSeong-Whan LeePublished in: Int. J. Pattern Recognit. Artif. Intell. (2005)
Keyphrases
- handwritten numeral recognition
- upper bound
- lower bound
- bayes error rate
- combination of multiple classifiers
- pattern recognition problems
- handwritten numerals
- conditional entropy
- error rate
- generalization error
- image processing
- information theory
- multiple classifier systems
- semi supervised
- pattern recognition
- optimal solution
- objective function
- computer vision