Towards a Robust OCR System for Indic Scripts

Krishnan, Praveen; Sankaran, Naveen; Singh, Ajeet; Jawahar, C. V.

doi:10.1109/das.2014.74

Cited by 23 publications

(18 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In all these cases, our feature learning method consistently improves the accuracy over profile features [10] which is a typical example of a hand crafted feature for the word recognition. Other advantages include that the learnt features will be of lower dimension and therefore are compact, efficient and takes smaller training time.…”

Section: Introductionmentioning

confidence: 78%

Unsupervised feature learning for optical character recognition

Sahu

Jawahar

2015

2015 13th International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

Abstract-Most of the popular optical character recognition (OCR) architectures use a set of handcrafted features and a powerful classifier for isolated character classification. Success of these methods often depend on the suitability of these features for the language of interest. In recent years, whole word recognition based on Recurrent Neural Networks (RNN) has gained popularity. These methods use simple features such as raw pixel values or profiles. Success of these methods depend on the learning capabilities of these networks to encode the script and language information. In this work, we investigate the possibility of learning an appropriate set of features for designing OCR for a specific language. We learn the language specific features from the data with no supervision. This enables the seamless adaptation of the architecture across languages. In this work, we learn features using a stacked Restricted Boltzman Machines (RBM) and use it with the RNN based recognition solution. We validate our method on five different languages. In addition, these novel features also result in better convergence rate of the RNNs.

show abstract

Section: Introductionmentioning

confidence: 78%

Unsupervised feature learning for optical character recognition

Sahu

Jawahar

2015

2015 13th International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

show abstract

“…These networks have been used in the past for printed text [2] and handwritten text recognition [3]. This network consists of two LSTM networks in which one network takes the input from beginning to end while other network takes the input from end to beginning.…”

Section: Rnn For Script and Language Identificationmentioning

confidence: 99%

“…For this, we use the popular profile features [2,17], which can be used to represent the lines and words as a feature sequence. In this work, we calculate six profile features from every word and image.…”

Section: A Representation Of Words and Linesmentioning

confidence: 99%

See 1 more Smart Citation

Can RNNs reliably separate script and language at word and line level?

Singh

Jawahar

2015

2015 13th International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

View full text Add to dashboard Cite

Abstract-In this work, we investigate the utility of Recurrent Neural Networks (RNNs) for script and langauge identification. Both these problems have been attempted in the past with representations computed from the distribution of connected components or characters (e.g. texture, n-gram). Often these features are computed from a larger segment (a paragraph or a page). We argue that one can predict the script or language with minimal evidence (e.g. given only a word or a line) very accurately with the help of a pre-trained RNN. We propose a simple and generic solution for the task of script and language identification which do not require any special tuning. Our method represents the word images as a sequence of feature vectors, and employ the RNNs for the identification. We verify the method on a large corpus of more than 15.03M words from 55K document images comprising 15 scripts and languages. We report an accurate script and language identification at word and line level.

show abstract

“…Many methods ( [1]; [4]; [5]; [15]) are developed in the past for recognizing Indian scripts, which generally focused on plain background images and used connected component analysis for recognition of scripts. The scope of the existing methods is limited to a particular script but not for multiple scripts because these methods exploited the knowledge of the scripts for script recognizing.…”

Section: 0introductionmentioning

confidence: 99%