Movie Piracy Detection Based on Audio Features Using Mel-Frequency Cepstral Coefficients and Vector Quantization
B. Srinivas1, K.Venkata Rao2, P. Suresh Varma3

1B.Srinivas, Computer Science and Engineering Department, M.V.G.R College of Engineering, Vizianagaram, India.
2Prof. K.Venkata Rao, Head of The Department, Computer Science and Engineering,Vignan IIT, Vishakapatnam, India.
3Prof. P.Suresh Varma, Principal, Professor, Computer Science and Engineering, University College Adikavi Nannaya University, Rajahmundry, Andhra Pradesh, India
Manuscript received on September 01, 2012. | Revised Manuscript received on September 02, 2012. | Manuscript published on September 05, 2012. | PP: 27-31 | Volume-2 Issue-4, September 2012. | Retrieval Number: D0856072412/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: Along with the increase in the advancement of technology in movie industry over internet, there is also an increase in the movie piracy via internet which affects factors like economy and repudiation of movie industry. Internet movie piracy is the most common means for pirates as well as downloader’s to break copyright laws by anonymous illegal uploads/downloads. In this paper we proposed an automated internet movie piracy detection mechanism based on audio fingerprint, which implements two famous algorithms, one is Mel-Frequency Cepstral Coefficients (MFCC) for feature extraction and the other is Vector Quantization (VQ) for classification. Our trained system initially looks for the sites which offer illegitimate copies of movies and if there is any suspicion based on a particular movie which is similar to the database of copyrighted movies that are registered with our trained system, it simply compares the fingerprints that are generated by implementing the above specified algorithms for both the trained and suspected movies. We collected various audio samples of different movies and we also extracted audio samples of pirated movies via internet with and without noises and trained and tested with our system. Finally, our system rendered efficient results with few error rates. We collected 52 audio samples without noise and 48 samples with noise and the resulted success classification is 96% and 92% respectively
Keywords: Classification, Code Book, Movie Piracy,