Kannada handwritten word conversion to electronic textual format using HMM model

Sushma, A.; Veena, G

doi:10.1109/csitss.2016.7779380

Cited by 9 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar approach was applied by Ma et al [21] for the OCR of a Tibetan script consisting of 562 characters and applied a three-layers DBM initialized using a RBM. SIFT local image descriptors were applied by Sushma and Veena [22] to encode and detect the characters of the Kannada language that uses 49 phonemic characters, where different characters can be composed to encode a single symbol. A Hidden Markov Model (HMM) was applied to improve the decoding of the language symbols by training the HMM using texts.…”

Section: Related Workmentioning

confidence: 99%

Auto-ML Deep Learning for Rashi Scripts OCR

Mahpod,

Keller

2018

Preprint

View full text Add to dashboard Cite

In this work we propose an OCR scheme for manuscripts printed in Rashi font that is an ancient Hebrew font and corresponding dialect used in religious Jewish literature, for more than 600 years. The proposed scheme utilizes a convolution neural network (CNN) for visual inference and Long-Short Term Memory (LSTM) to learn the Rashi scripts dialect. In particular, we derive an AutoML scheme to optimize the CNN architecture, and a book-specific CNN training to improve the OCR accuracy. The proposed scheme achieved an accuracy of more than 99.8% using a dataset of more than 3M annotated letters from the Responsa Project dataset.

show abstract

Section: Related Workmentioning

confidence: 99%