A SMART IMAGE PROCESSING ALGORITHM FOR TEXT RECOGNITION, INFORMATION EXTRACTION AND VOCALIZATION FOR THE VISUALLY CHALLENGED

This paper proposes a smart algorithm for image processing by means of recognition of text, extraction of information and vocalization for the visually challenged. The system uses LattePanda Alpha system on board that processes the scanned images. The image is categorized into its equivalent alphanumeric characters following pre-processing, segmentation, extraction of features and post-processing of the scanned or image based information. Further, a text to speech synthesizer is used for vocalization processed content. In converting handwritten scripts, the system offers an accuracy of 97% in conversion. This also depends on the legibility of the data. The time delay for the entire conversion process is also analysed and the efficiency of the system is estimated.