Improving Faster R-CNN Framework for Multiscale Chinese Character Detection and Localization

Minseong KIM  Hyun-Chul CHOI  

IEICE TRANSACTIONS on Information and Systems   Vol.E103-D   No.7   pp.1777-1781
Publication Date: 2020/07/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2019EDL8217
Type of Manuscript: LETTER
Category: Pattern Recognition
Chinese character localization,  multiscale object detection,  deep learning,  

Full Text: PDF(1.6MB)>>
Buy this Article

Faster R-CNN uses a region proposal network which consists of a single scale convolution filter and fully connected networks to localize detected regions. However, using a single scale filter is not enough to detect full regions of characters. In this letter, we propose a simple but effective way, i.e., utilizing variously sized convolution filters, to accurately detect Chinese characters of multiple scales in documents. We experimentally verified that our method improved IoU by 4% and detection rate by 3% than the previous single scale Faster R-CNN method.