With the new “video gene†technology, even if some scenes of the video are deleted or their color and resolution are changed, the “DNA†of the video clip can be sequenced using bioinformatics analysis technology just like biological DNA sequences. The technology invented by researchers at the Israel Institute of Technology Engineering can effectively prevent the spread of pirated video on the Internet.
The researchers explained the principle of video genes. The characteristics of the video are converted into information strings as if the genes were expressed by DNA nucleotide sequences. The video gene consists of a set of frames and shapes, and the search algorithm uses a set of words to find similar texts. The researchers also pointed out that ordinary video operations will not change the video gene. Subsequently, the characteristic frequency of each frame of image needs to be plotted and expressed in 64-bit data blocks. When a video file is played locally or over a network, biometric information analysis techniques can be used to compare the video clip's gene and video gene database. When an unauthorized video file is determined to match the copyrighted video content in the video gene database, the video may be filtered or deleted.
According to researchers, this technology can be used to monitor pirated content on sites such as YouTube. In theory, thousands of hours of video sequences can be processed in a matter of days with an accuracy of 99%.
The researchers explained the principle of video genes. The characteristics of the video are converted into information strings as if the genes were expressed by DNA nucleotide sequences. The video gene consists of a set of frames and shapes, and the search algorithm uses a set of words to find similar texts. The researchers also pointed out that ordinary video operations will not change the video gene. Subsequently, the characteristic frequency of each frame of image needs to be plotted and expressed in 64-bit data blocks. When a video file is played locally or over a network, biometric information analysis techniques can be used to compare the video clip's gene and video gene database. When an unauthorized video file is determined to match the copyrighted video content in the video gene database, the video may be filtered or deleted.
According to researchers, this technology can be used to monitor pirated content on sites such as YouTube. In theory, thousands of hours of video sequences can be processed in a matter of days with an accuracy of 99%.