An Improved Scene Text Extraction Method Using Conditional Random Field and Optical Character Recognition.
Hongwei ZhangChangsong LiuCheng YangXiaoqing DingKongqiao WangPublished in: ICDAR (2011)
Keyphrases
- conditional random fields
- optical character recognition
- text regions
- character recognition
- text recognition
- graphical models
- text detection
- document images
- probabilistic model
- information extraction
- higher order
- generative model
- hidden markov models
- pairwise
- text lines
- text extraction
- markov random field
- superpixels
- machine vision
- face recognition
- complex background
- hough transform