Image Text Detection and Conversion into Text Form

We are purposing a system by which we can detect the image text and convert into the text form for which we firstly apply MSER (Maximally Stable External Region) which is used as a method of blob detection in images, or to detect character candidates. After that some text features by which the text can be recognized. To recognize the text feature we apply some geometric filtration by which we can easily identified the character from the image. This is why all the character is formed by combination of geometric figures. After recognizing the text, reject the false positives i.e. background, figures, and many things expect character. Performing all these we apply OCR function by which Segmenting out the text from a cluttered scene helps with related tasks. Lastly after finding all these we have the text which is in text form which is easily modified and editable.