Text/Non-text Ink Stroke Classification in Japanese Handwriting Based on Markov Random Fields

In this paper, we present an approach for separat- ing text and non-text ink strokes in online handwritten Japanese documents based on Markov random fields (MRFs), which effectively utilize the spatial relation- ship between strokes. Support vector machine (SVM) classifiers are trained for individual stroke and stroke pair classification, and on converting the SVM outputs to probabilities, the likelihood clique potentials of MRF are derived. In experiments on the TUAT Kon- date database, the proposed MRF approach yield su- perior performance compared to individual stroke classification and sequence classification based on hidden Markov models (HMMs).