Text Detection in Videos Using Super Resolution

Abstract

Text in images and videos contain useful information for automatic annotation, indexing, and structuring of images and videos. Extraction of this information involves detection, localization, tracking, extraction, enhancement, and recognition of the text from a given image and videos. However, variations of text due to differences in size, style, orientation, and alignment, as well as low image contrast and complex background make the problem of automatic text extraction extremely challenging.In the complete process of text extraction, text detection is the primary and fundamental step. We can increase the performance of the text detection by giving high-resolution images or video frames instead of low-resolution images or video frames. Here, Super-resolution can be used to produce the high-resolution image or video frame. In high-resolution images or video frames, there is more pixel density and thus provides finer details of the image or scene. In this report, we talk about how we can increase the range of different size text detected from videos using super resolution. More detected text from a video will imply performance increase in text recognition.