An on-line Japanese character recognition method using length-based stroke correspondence algorithm.
Yojiro TonouchiAkinori KawamuraPublished in: ICDAR (1997)
Keyphrases
- high accuracy
- character recognition
- significant improvement
- preprocessing
- computational cost
- objective function
- computational complexity
- dynamic programming
- optimization algorithm
- detection method
- k means
- cost function
- recognition algorithm
- matching algorithm
- clustering method
- chinese characters
- convergence rate
- segmentation method
- input data
- template matching
- image matching
- handwritten digits
- alphanumeric characters
- probabilistic model
- similarity measure
- segmentation algorithm
- optimal solution
- energy function
- machine vision
- learning algorithm
- gray scale
- expectation maximization
- edge detection
- automatic recognition
- rigid transformation
- hand written
- character segmentation