Novel Text Extraction from Video Image Using Multispectral Fusion

Image Processing play on vital role an enhanced image or to extract some useful information from it. It is a type of signal dispensation in which input is image like video frame or photograph. Text character and strings in natural scene can provide valuable information for more applications. Retrieving text directly from natural scene image or video is challenging task because of diverse text patterns and variant background interferences. In our existing system two modules are examined, first describes the discriminative character descriptor by combining several state of the art future. Second describes the modeling of a character structure. In this proposed System a method of scene text recognition from both natural image and videos. And also to convert the text into voice using specified algorithm and also provide explanation for recognized text. In future, this project is enhanced to translate the converted text into user defined specified languages.