Mosaicing-by-recognition for recognizing texts captured in multiple video frames

Seiichi Uchida, Hiromitsu Miyazaki, Hiroaki Sakoe

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Text recognition in video frames is promising because of its following superiorities over text recognition in a still camera image: (1) it is possible to recognize longer texts by concatenating the frames, and (2) it is also possible to improve the quality of the text image by integrating the frames. In this paper, a mosaicing-by-recognition technique is proposed where video mosaicing and text recognition are simultaneously and collaboratively performed in a one-step manner by a dynamic programming-based optimization algorithm. In this optimization algorithm, rotation, scaling, vertical shift, and speed fluctuation of camera motion are efficiently compensated. The results of experiments to evaluate not only the accuracy of text recognition but also that of video mosaicing indicates that the proposed technique is practical and can provide reasonable results in most cases.

Original languageEnglish
Title of host publicationProceedings of the 1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005
Pages3-9
Number of pages7
Publication statusPublished - 2005
Event1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005 - Seoul, Korea, Republic of
Duration: Aug 29 2005Aug 29 2005

Other

Other1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005
CountryKorea, Republic of
CitySeoul
Period8/29/058/29/05

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Software

Fingerprint Dive into the research topics of 'Mosaicing-by-recognition for recognizing texts captured in multiple video frames'. Together they form a unique fingerprint.

Cite this