End-to-End Approach for Recognition of Historical Digit Strings

04/28/2021
by   Mengqiao Zhao, et al.
0

The plethora of digitalised historical document datasets released in recent years has rekindled interest in advancing the field of handwriting pattern recognition. In the same vein, a recently published data set, known as ARDIS, presents handwritten digits manually cropped from 15.000 scanned documents of Swedish church books and exhibiting various handwriting styles. To this end, we propose an end-to-end segmentation-free deep learning approach to handle this challenging ancient handwriting style of dates present in the ARDIS dataset (4-digits long strings). We show that with slight modifications in the VGG-16 deep model, the framework can achieve a recognition rate of 93.2 a feasible solution free of heuristic methods, segmentation, and fusion methods. Moreover, the proposed approach outperforms the well-known CRNN method (a model widely applied in handwriting recognition tasks).

READ FULL TEXT

page 5

page 11

research
04/24/2018

Segmentation-Free Approaches for Handwritten Numeral String Recognition

This paper presents segmentation-free strategies for the recognition of ...
research
09/22/2020

Whole page recognition of historical handwriting

Historical handwritten documents guard an important part of human knowle...
research
07/19/2020

Survey on Deep Learning-based Kuzushiji Recognition

Owing to the overwhelming accuracy of the deep learning method demonstra...
research
04/21/2016

Evaluation of the Effect of Improper Segmentation on Word Spotting

Word spotting is an important recognition task in historical document an...
research
06/07/2018

Writing Style Invariant Deep Learning Model for Historical Manuscripts Alignment

Historical manuscript alignment is a widely known problem in document an...
research
04/20/2016

Local Binary Pattern for Word Spotting in Handwritten Historical Document

Digital libraries store images which can be highly degraded and to index...
research
04/12/2016

Scan, Attend and Read: End-to-End Handwritten Paragraph Recognition with MDLSTM Attention

We present an attention-based model for end-to-end handwriting recogniti...

Please sign up or login with your details

Forgot password? Click here to reset