Title :
Texture-based approach for text detection in images using support vector machines and continuously adaptive mean shift algorithm
Author :
Kim, Kwang In ; Jung, Keechul ; Kim, Jin Hyung
Author_Institution :
Dept. of Comput. Sci., Korea Adv. Inst. of Sci. & Technol., Taejon, South Korea
Abstract :
The current paper presents a novel texture-based method for detecting texts in images. A support vector machine (SVM) is used to analyze the textural properties of texts. No external texture feature extraction module is used, but rather the intensities of the raw pixels that make up the textural pattern are fed directly to the SVM, which works well even in high-dimensional spaces. Next, text regions are identified by applying a continuously adaptive mean shift algorithm (CAMSHIFT) to the results of the texture analysis. The combination of CAMSHIFT and SVMs produces both robust and efficient text detection, as time-consuming texture analyses for less relevant pixels are restricted, leaving only a small part of the input image to be texture-analyzed.
Keywords :
character recognition; document image processing; feature extraction; image texture; indexing; support vector machines; text analysis; SVM; continuously adaptive mean shift algorithm; feature extraction; high dimensional spaces; input image; pixels; support vector machines; text detection; textural pattern; textural properties; texture based approach; time consuming texture analyses; Carbon capture and storage; Degradation; Feature extraction; Image analysis; Image texture analysis; Indexing; Neural networks; Pixel; Robustness; Support vector machines;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2003.1251157