Combination of statistical and neural classifiers for a high-accuracy recognition of large character sets.
Yoshimasa KimuraToru WakaharaAkira TomonoPublished in: Systems and Computers in Japan (2005)
Keyphrases
- high accuracy
- handwritten characters
- printed characters
- classifier combination
- neural classifier
- recognition accuracy
- recognition rate
- license plate
- network architecture
- handwritten digit recognition
- neural network
- character recognition
- combining classifiers
- improve recognition accuracy
- automatic recognition
- pattern recognition
- feature extraction
- object recognition
- printed documents
- statistical analysis
- decision trees
- recognition process
- multiple classifier systems
- classification decisions
- test set
- training data
- statistical tests
- linear classifiers
- recognition algorithm
- machine learning algorithms
- naive bayes
- hand written
- support vector
- improve the recognition accuracy
- feature selection
- majority voting
- document analysis
- statistical methods
- training set
- multiple classifiers
- optical character recognition
- handwritten words
- learning rules
- combining multiple
- document images
- svm classifier