A Wavelet Optimized Video Copy Detection Using Content Fingerprinting

The video analysis technique facilitates automatic understanding of visual content and semantic information transmitted by video streams. Video copy detection is enhanced with the comprehensive usage of video analysis techniques and fingerprinting. Numerous literary works utilize content-based fingerprinting. Extensive conversion of video content from the spatial domain to the frequency domain forms the basis of this technique. Most works have utilized the Discrete Cosine Transform (DCT), its variants, and recently, wavelets have also been used. Wavelet is commonly used for general image processing. The popularly known Daubechies wavelet forms the basis of the majority of current approaches. To improve the video copy detection, optimization of the wavelet filter bank aids in boosting the wavelet performance. A wavelet design method that employs polyphase representation and devises the problem with Gravitational Search Algorithm (GSA) technique has been proposed in this paper. Simulation outcomes demonstrate this proposed technique's superior performance compared to other techniques.