A Head-Mounted Device for Recognizing Text in Natural Scenes

摘要

We present a mobile head-mounted device for detecting and tracking text that is encased in an ordinary flat-cap hat. The main parts of the device are an integrated camera and audio webcam together with a simple remote control system, all connected via a USB hub to a laptop. A near to real-time text detection algorithm (around 14 fps for 640 × 480 images) which uses Maximal Stable Extremal Regions (MSERs) for image segmentation is proposed. Comparative text detection results against the ICDAR 2003 text locating competition database along with performance figures are presented.