Title :
Textfinder: an automatic system to detect and recognize text in images
Author :
Wu, Victor ; Manmatha, Raghavan ; Riseman, Edward M.
Author_Institution :
Dept. of Comput. Sci., Massachusetts Univ., Amherst, MA, USA
fDate :
11/1/1999 12:00:00 AM
Abstract :
A robust system is proposed to automatically detect and extract text in images from different sources, including video, newspapers, advertisements, stock certificates, photographs, and checks. Text is first detected using multiscale texture segmentation and spatial cohesion constraints, then cleaned up and extracted using a histogram-based binarization algorithm. An automatic performance evaluation scheme is also proposed
Keywords :
document image processing; image segmentation; image texture; Textfinder; advertisements; checks; cheques; histogram-based binarization algorithm; images; multiscale texture segmentation; newspapers; photographs; spatial cohesion constraints; stock certificates; text detection; text extraction; text recognition; video; Character recognition; Data mining; Image recognition; Image segmentation; Image storage; Neural networks; Optical character recognition software; Optical filters; Strontium; Text recognition;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on