Fast Text Line Detection Algorithm from Complex Document Image
Qin Mu-ting · Science Technology and Engineering · 2008
Text embedded in images contains large quantities of useful semantic information which can be used to fully understand images. Detection and extraction of text line in images have been used in many applications,such as mobile robot navigation, vehicle license detection and recognition, object identification, document retrieving, page segmentation, ect. a fast text extraction algorithm based on run length analysis is proposed, which fill white spaces inside and between characters in same text line and cut strokes that extend to inter space between text lines, so that it can automatically detect and extract text line in complex images . The proposed method is proved to be robust to font style, size, and language, it can filter Logo and bar code which are similar to text line spatially.