A video text detection method based on key text points

  • Authors:

  • Zhi Li;Guizhong Liu;Xueming Qian;Chen Wang;Yana Ma;Yang Yang

  • Affiliations:

  • School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, China;School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, China;School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, China;School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, China;School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, China;School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, China

  • Venue:
  • PCM'10 Proceedings of the 11th Pacific Rim conference on Advances in multimedia information processing: Part I
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a novel video text detection method based on the key text points. For text detection, the keyframes is decomposed by wavelet transform. The key text points (KTPs) are determined by three resulting high-frequency subbands, and merged by the morphological operations. The antitexture-direction-projection method is proposed for text line localization and verification. A fast text tracking scheme is proposed, in which text detection is only implemented on the first keyframe of an identical text line in the duration. The appearing (disappearing) frame is determined by a fast search method. Experimental results show that the proposed text detection method is robust to the font size, style, color and alignment of texts. The proposed text tracking greatly speeds up the text detection.