Efficient video copy detection using simple and effective extraction of color features

Abstract

In the present Multimedia era, the exponential growth of illegal videos and huge piracy issues increased the importance of Content Based video Copy Detection (CBCD) techniques. CBCD systems require compact and computationally efficient descriptors for detecting video copies. In this paper, we propose a simple and efficient video signature scheme using Dominant Color Descriptors of MPEG-7 standard in order to implement the proposed CBCD task. Experimental results show that the proposed approach yields better detection rates when compared to that of existing approaches, against common transformations like Contrast change, Noise addition, Rotation, Zooming, Blurring etc. Further, evaluation results also prove that our scheme is computationally efficient by supporting substantial reduction in the total computational cost up to the extent of 65% when compared to that of existing schemes. © 2011 Springer-Verlag.