Abstract
EXTRACTION OF TEXT FROM COMPLEX IMAGES FOR MULTIMEDIA APPLICATION
Harshitha Urs T R, Chaitra B S.* and Santhosh B.
ABSTRACT
Large amount of information are embedded in natural scenes which are often required to be automatically recognized and processed. For the process the automatic identification segmentation and detection of visual text entities in usual scene images are required. Scene text could be any textual part of the scene images such as street signs, name plates or text appearing on t-shirts. The extracted text can be displayed on a palm size PDA or synthesized as voice output message over the ear phone. Text in image contains useful information which can be used to fully understand images. The various methods have been proposed in the past of detection and localization of text in images and videos. The properties related to the text in an image such as color, intensity and edges are used to distinguish text regions from the background and other region within the image. In this paper we have proposed a system that detect and extract the text from the complex images using a MATLAB tool.
[Full Text Article] [Download Certificate]