Multimedia Text Extraction for Healthcare Applications

Abstract

Image, audio, and video are rapidly evolving multimedia data, creating a platform for the creation, exchange, and storage of information in the modern era. Multiple techniques are available for content-based image, video, and audio retrieval using image processing methods to extract text features of the data. These methods produce similar text but not the same as per the user expectation. In this chapter, the authors offer an approach that uses edge map generation and Google tesseract-based text detection in image as well as in video. The accuracy of our method is 98% for speed video extraction, 100% for image/video/audio data.