An Efficient Video Text Recognition System

In this paper, we present efficient schemes to utilize multiple frames that contain the same text to get every clear word from these frames. Firstly, we use multiple frame traction to reduce text detection false alarms. And then detect and joint every clear text block from those frames to form a clearer integration frame. Later we uses Otsu threshold method to generate a binary text image. Finally, the binarized frames are sent to OCR engine for recognition. The experimental results show our method can improve the performance of text recognition in video significantly.