Project Title: MULTISCALE EDGE-BASED TEXT EXTRACTION
Text that seems in pictures contains crucial and information that is useful. Detection and extraction of text in pictures have been utilized in numerous applications. In this project, a multiscale text that is edge-based algorithm, which could automatically identify and extract text in complex pictures is proposed. The proposed method is just a text that is general-purpose and extraction algorithm, which can deal perhaps not only with printed document pictures but additionally with scene text. It's robust with respect to the font size, design, color, orientation, and alignment of text and can be utilized in a number that is big of areas, such as for example mobile robot navigation, automobile license detection and recognition, object identification, document retrieving, page segmentation, etc.
* Database of indoor/ outdoor scene images