Choreo: Empowering the Future of Enterprise Software Engineering
Real time video copy detection based on hadoop
1. Real-time Video Copy Detection Based on
Hadoop
Hardik Parmar
Sanket Thakur
Pranav Sangam
Sachin Tripathi
2. ABSTRACT:
With the development of multimedia technology and Internet, the amount of videos in the Internet is
increasing quickly.
Among the large amount of videos in the Internet, a considerable number of them are copies of original
videos, which are simply revised versions of the original ones.
3. Introduction
• Introduction to Video Copy Detection
Due to rapid development of multimedia hardware and software technologies, the cost of image
and video data collection, creation, and storage is becoming low.
Among these huge volumes of videos, there exist large numbers of copies.
4. Introduction to Hadoop Platform
• Hadoop was developed by the Apache Foundation. IT consists of map reduce model.
• MapReduce is the programming model of Hadoop which includes Map function and Reduce
function.
5. Proposed System
We propose a video copy detection using method based on Brightness sequence and the method
based on TIRI-DCT algorithm.
High accuracy in locating copies.
Very reliable for detecting copied videos.
6. Advantages Of Proposed System
1. The performance on detecting copies from large data set is satisfactory and Hadoop platform
can significantly improve the efficiency of video copy detection.
2. The proposed system high fault tolerance, high throughput, easy scalability and etc.
3. The measurement of copy detections performance System making result is faster than
existing system.
4. The proposed system video hashing algorithm methods has strong robustness, high
distinction, high compactness and low complexity.
7. Flow Chart
:
Yes
NO
YES
NO
Start
FFMPEG Transcoding
Close
Upload/ Querying Video Process
Calculate images hash value
Convert video into
pictures in the form of
frames
Create hash value library
Calculate distance between hash values
Match hash value
Enter Username
and Password
FFMPEG Transcoding
Training Videos
Calculate images hash value
Convert video into
pictures in the form of
frames
Stored in HDFS
Create hash value library
Database
Distance value
< Threshold
value
Detection of video copy & Reject video
Upload into database
8. Hardware and Software requirements
• Hardware:
Processor: Pentium 4
RAM: 4GB or more
Hard disk: 16 GB or more
• Software Specification:
Windows Operating System.
Eclipse
NetBeans
Java
Apache Tomcat Server
MySQL
Hadoop
9. Applications
• Using the same scenario of system, we can implement following mobile
applications:
1. For Government Services Advertisement Video.
2. For Online Shopping System Advertisement Video.
3. For Video Upload in private own cloud.
10. CONCLUSION
• In this paper, two video copy detection methods, the method based on brightness sequence and the
method based on TIRI-DCT are implemented and the recalls and precisions of the two methods with
different video numbers and different thresholds are analyzed.
• The algorithms are implemented on Hadoop distributed computing platform and the efficiencies are
compared in different video amounts and different map amounts.