Method and system for producing multimedia fingerprint based on quantum hashing
View Patent ↗Disclosed are a method and system for producing a multimedia fingerprint based on quantum hashing. The method includes receiving an input of a multimedia file, extracting a quantum hash type fingerprint from the input multimedia file, calculating similarity between the extracted quantum hash type fingerprint and a binary fingerprint stored in a database, and selecting, as a calculation result, data having a fingerprint calculated as having the highest similarity.
1. A method of producing a multimedia fingerprint based on quantum hashing, comprising:
receiving an input of a multimedia file;
extracting a quantum hash type fingerprint from the input multimedia file;
calculating similarity between the extracted quantum hash type fingerprint and a binary fingerprint stored in a binary fingerprint database; and
selecting, as a calculation result, data having a fingerprint calculated as having a highest similarity;
wherein extracting the quantum hash type fingerprint from the input multimedia file includes extracting one or more intermediate hashes from the input multimedia file, and extracting quantum hashes, which belong to two-dimensional Hilbert space on an orthonormal basis, among the one or more intermediate hashes.
2. The method according to claim 1 , wherein the multimedia file comprises a distorted signal.
3. The method according to claim 2 , wherein if the multimedia file is an audio file, the distorted signal is generated when the audio file is passed through an equalizer, when noise is added to the audio file, or when an audio file of a compact disc format is encoded into an MP3 format.
4. The method according to claim 2 , wherein if the multimedia file is a video file, the distorted signal is generated when the video file is captured in a movie theater using a camcorder and released as a movie, or when the video file is subjected to image brightness change, size change or rotation.
5. A system for producing a multimedia fingerprint based on quantum hashing, comprising:
an input unit receiving an input of a multimedia file;
a fingerprint extraction unit extracting a quantum hash type fingerprint from the input multimedia file;
a binary fingerprint database storing a binary type fingerprint;
a database search unit calculating similarity between the extracted quantum hash type fingerprint and the binary type fingerprint stored in the binary fingerprint database; and
a fingerprint matching unit selecting and outputting, as a calculation result, data having a fingerprint calculated as having a highest similarity;
wherein when extracting the quantum hash type fingerprint, the fingerprint extraction unit extracts one or more intermediate hashes from the input multimedia file and extracts quantum hashes, which belong to two-dimensional Hilbert space on an orthonormal basis, among the one or more intermediate hashes.