Title :
Effective Dissolve Detection Based on Accumulating Histogram Difference and the Support Point
Author :
Ji, Qing-Ge ; Feng, Jian-Wei ; Zhao, Jie ; Lu, Zhe-Ming
Author_Institution :
Sch. of Inf. Sci. & Technol., Sun Yat-sen Univ., Guangzhou, China
Abstract :
This paper presents a novel dissolve detection method based on accumulating histogram difference (AHD) and support points. The proposed algorithm can detect fades and dissolves and avoid the false detection caused by the flashlight. In terms of recall, precision, computational complexity per frame and computational complexity per pixel, we compare our method with the Twin algorithm and the Two Measures Two Thresholds (TMTT) algorithm for fade detection and dissolve detection. The experimental results demonstrate the superiority of our algorithm.
Keywords :
computational complexity; image segmentation; object detection; video signal processing; accumulating histogram difference; computational complexity; computational complexity per frame; computational complexity per pixel; dissolve detection method; false detection; shot boundary detection; support points; twin algorithm; two measures two thresholds algorithm; video temporal segmentation; Algorithm design and analysis; Computational complexity; Histograms; Image color analysis; Pixel; Signal processing algorithms; Streaming media; accumulating histogram difference; dissolve transition; fade transition; gradual transitions; shot boundary detection;
Conference_Titel :
Pervasive Computing Signal Processing and Applications (PCSPA), 2010 First International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4244-8043-2
Electronic_ISBN :
978-0-7695-4180-8
DOI :
10.1109/PCSPA.2010.73