A Model Selection Criterion for Classification: Application to HMM Topology Optimization
ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
A Deterministic Method for Initializing K-Means Clustering
ICTAI '04 Proceedings of the 16th IEEE International Conference on Tools with Artificial Intelligence
Handwritten-Word Spotting Using Biologically Inspired Features
IEEE Transactions on Pattern Analysis and Machine Intelligence
Selecting hidden Markov model state number with cross-validated likelihood
Computational Statistics
SVM-based hierarchical architectures for handwritten Bangla character recognition
International Journal on Document Analysis and Recognition
Separability versus prototypicality in handwritten word-image retrieval
Pattern Recognition
Hi-index | 0.00 |
This paper investigates the performance of hidden Markov models (HMMs) for handwriting recognition. The Segmental K-Means algorithm is used for updating the transition and observation probabilities, instead of the Baum-Welch algorithm. Observation probabilities are modelled as multi-variate Gaussian mixture distributions. A deterministic clustering technique is used to estimate the initial parameters of an HMM. Bayesian information criterion (BIC) is used to select the topology of the model. The wavelet transform is used to extract features from a grey-scale image, and avoids binarization of the image.