Discriminative Bernoulli HMMs for isolated handwritten word recognition
نویسندگان
چکیده
منابع مشابه
Discriminative Bernoulli HMMs for isolated handwritten word recognition
Bernoulli HMMs (BHMMs) have been successfully applied to handwritten text recognition (HTR) tasks such as continuous and isolated handwritten words. BHMMs belong to the generative model family and, hence, are usually trained by (joint) maximum likelihood estimation (MLE) by means of the Baum-Welch algorithm. Despite the good properties of the MLE criterion, there are better training criteria su...
متن کاملArabic Handwritten Word Recognition based on Bernoulli Mixture HMMs
This thesis presents new approaches in off-line Arabic Handwriting Recognition based on conventional Bernoulli Hidden Markov models. Until now, the off-line handwriting recognition, in particular, the Arabic handwriting recognition is still far away form being perfect. Hidden Markov Models (HMMs) are now widely used for off-line handwriting recognition in many languages and, in particular, in A...
متن کاملHandwriting word recognition using windowed Bernoulli HMMs
Hidden Markov Models (HMMs) are now widely used for off-line handwriting recognition in many languages. As in speech recognition, they are usually built from shared, embedded HMMs at symbol level, where state-conditional probability density functions in each HMM are modeled with Gaussian mixtures. In contrast to speech recognition, however, it is unclear which kind of features should be used an...
متن کاملWord Level Discriminative Training for Handwritten Word Recognition
Word level training refers to the process of learning the parameters of a word recognition system based on word level criteria functions. Previously, researchers trained lexicon-driven handwritten word recognition systems at the character level individually. These systems generally use statistical or neural based character recognizers to produce character level confidence scores. In the case of...
متن کاملMixture of Experts for Persian handwritten word recognition
This paper presents the results of Persian handwritten word recognition based on Mixture of Experts technique. In the basic form of ME the problem space is automatically divided into several subspaces for the experts, and the outputs of experts are combined by a gating network. In our proposed model, we used Mixture of Experts Multi Layered Perceptrons with Momentum term, in the classification ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Pattern Recognition Letters
سال: 2014
ISSN: 0167-8655
DOI: 10.1016/j.patrec.2013.05.016