A Method for Detecting Subtitle Regions in Videos Using Video Text Candidate Images and Color Segmentation Images

In this paper, a method for detecting text regions in digital videos with telop, such as drama, movie and news programming, is proposed. The typical characteristics of telop are that it does not move, and that its edges are strong. This method takes advantage of these characteristics to produce video text candidate images. Then, this method produces the video text region images from both the video text candidate images and the color segmentation images. The video text region images and the original image are used to identify the color of the telop. Finally, text regions are detected by increasing neighboring pixels of the identied color. The experiment results show that the precision of this method was 80.36% and the recall was 77.55%, whereas the precision of the traditional method was 40.22% with the recall 75.48%. Higher accuracy was achieved by using this new method.

[1]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.

[2]  Michael J. Witbrock,et al.  Informedia News-On Demand: Using Speech Recognition to Create a Digital Video Library , 1998 .

[3]  Yihong Gong,et al.  Lessons Learned from Building a Terabyte Digital Video Library , 1999, Computer.

[4]  Nozha Boujemaa,et al.  Object-based queries using color points of interest , 2001, Proceedings IEEE Workshop on Content-Based Access of Image and Video Libraries (CBAIVL 2001).