Title :
A method of caption location and segmentation in news video
Author :
He Huang ; Ping Shi ; Laiwen Yang
Author_Institution :
Sch. of Inf. Eng., Commun. Univ. of China, Beijing, China
Abstract :
Caption texts in news video highly summarize the related news story and can be used for effective retrieval. This paper presents an effective caption location method based on edge detection and an improved caption segmentation method combined with Otsu and K-means clustering algorithm. Experiment results show that the caption location method performs well in the news video and the caption segmentation method has higher accuracy than Otsu algorithm.
Keywords :
edge detection; image segmentation; pattern clustering; text detection; video signal processing; Otsu algorithm; edge detection; k-means clustering algorithm; news story retrieval; news story summarize; news video; text caption location method; text caption segmentation method; Algorithm design and analysis; Clustering algorithms; Gray-scale; Image color analysis; Image edge detection; Image segmentation; Software algorithms; K-means; Otsu; caption location; caption segmentation; news video;
Conference_Titel :
Image and Signal Processing (CISP), 2014 7th International Congress on
Conference_Location :
Dalian
DOI :
10.1109/CISP.2014.7003807
