Video fingerprint method based on contourlet transformation model

Abstract

The invention discloses a video fingerprint method based on a contourlet transformation model, which is characterized by comprising the following steps of: standardizing a piece of video and segmenting each image; transforming each image block through a multi-scale and multi-directional contourlet, and using a hidden Markov tree technology to extract model parameters; decomposing singular values of a standard variance matrix of each block, extracting the maximum singular value, and cascading and normalizing the singular value to construct a video fingerprint vector; using a two-step search strategy in the fingerprint matching stage; and randomly selecting the fingerprint of each frame in the video to be checked, rapidly determining candidate video, and then measuring the similarity of the video to be tested and the candidate video according to the Euclidean distance. The stability, identification efficiency and search efficiency of video fingerprints can be improved, and the method can be widely applied in copy detection, retrieval, identification and copyright protection of video files.

Description

technical field [0001] The invention belongs to the technical field of multimedia information processing, and in particular relates to a video fingerprint method that can be used for copy detection, video identification, video retrieval, copyright protection and piracy tracking. Background technique [0002] In the 12th issue of Information Forensics and Security Transactions of the Institute of Electrical and Electronics Engineering (IEEE) in 2007, it was pointed out that with the popularization of digital devices and the diversification of communication methods, thousands of video files are uploaded and shared to the Internet every day. Illegal copying and processing of video files is a very common phenomenon in TV and mobile communication devices. The widespread copyright infringement requires the development of fast and accurate content-based video copy detection methods. The video fingerprint technology used to deal with this challenge, the fingerprint described here is...

AI Technical Summary

Problems solved by technology

[0004] The existing fingerprint methods at home and abroad mainly include the method based on the centroid of the gradient direction proposed in the 7th issue of the Video Technology Transactions of the Institute of Electrical and Electronics Engineering (IEEE) Circuits and Systems in 2008. The features extracted by this method cannot effectively express the image. At the same time, the amount of data is large, and the operation robustness to various contents is not high; the method based on discrete wavelet transform and PCA decomposition proposed in the third issue of Information Forensics and Security Transactions of the Institute of Electrical and Electronics Engineering (IEEE) in 2010 , the correct rate of recognition in video recognition and retrieval applications is not high, and the probability of false alarms is relatively high under a certain detection probability; DCT was proposed in the first issue of Information Forensics and Security Transactions of the Institute of Electrical and Electronics Engineering (IEEE) in 2011. The time information of the domain represents the method of image technology. This method adopts a global search strategy in the matching process with slow search speed, poor real-time performance, and low database search efficiency.

Images