System and method for digital fingerprinting of media content
Systems, methods, and computer-readable storage media for digital fingerprinting of media content. The system does this by normalizing and pre-processing the media content into a standard format, then generating fingerprints for image, audio, and/or video aspects of the media. These fingerprints are mathematical vectors generated using one or more techniques, such as perceptual hashes coupled with machine learning. Once the fingerprints for new content is generated, the system can compare those fingerprints to fingerprints of known content to determine if the new content contains prohibited content.
1. A method comprising:
receiving, at a computer system, a media file comprising video content;
converting, via at least one processor of the computer system, the video content to greyscale, resulting in greyscaled video content;
segmenting, via the at least one processor of the computer system, the greyscaled video content using a rolling window of predetermined length, resulting in a downsampled plurality of processed segments;
calculating, via the at least one processor for each segment within the plurality of downsampled processed segments, an average of pixel values, resulting in an average grey pixel value for each frame;
hashing, via the at least one processor, the average grey pixel value for each segment within the plurality of downsampled processed segments, resulting in a plurality of hashed segment pixel values; and
generating, via the at least one processor by combining at least two hashed segment pixel values from the plurality of hashed segment pixel values, a fingerprint of at least a portion of the media file.
2. The method of claim 1 , wherein the rolling window causes each segment to have an amount of overlap with a previous segment and a subsequent segment.
3. The method of claim 1 , wherein segments of the media file corresponding to the at least two hashed segment pixel values are consecutive within the media file.
4. The method of claim 1 , further comprising:
comparing, via the at least one processor, the fingerprint against at least one known fingerprint, resulting in at least one percentage of similarity.
5. The method of claim 4 , further comprising:
determining, via the at least one processor, that the media file comprises prohibited content based on the at least one percentage of similarity.
6. The method of claim 4 , further comprising:
sending, via the at least one processor, the media file and the at least one percentage of similarity to a compliance review team to determine if the media file comprises prohibited content.
7. The method of claim 4 , wherein the comparing of the fingerprint against the at least one known fingerprint identifies a distinction within a temporal order of the media file and at least one piece of known content.
8. A system comprising:
at least one processor; and
a non-transitory computer-readable storage medium having instructions stored which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
receiving a media file comprising video content;
converting the video content to greyscale, resulting in greyscaled video content;
segmenting the greyscaled video content using a rolling window of predetermined length, resulting in a downsampled plurality of processed segments;
calculating, for each segment within the plurality of downsampled processed segments, an average of pixel values, resulting in an average grey pixel value for each frame;
hashing the average grey pixel value for each segment within the plurality of downsampled processed segments, resulting in a plurality of hashed segment pixel values; and
generating, by combining at least two hashed segment pixel values from the plurality of hashed segment pixel values, a fingerprint of at least a portion of the media file.
9. The system of claim 8 , wherein the rolling window causes each segment to have an amount of overlap with a previous segment and a subsequent segment.
10. The system of claim 8 , wherein segments of the media file corresponding to the at least two hashed segment pixel values are consecutive within the media file.
11. The system of claim 8 , the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
comparing the fingerprint against at least one known fingerprint, resulting in at least one percentage of similarity.
12. The system of claim 11 , the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
determining that the media file comprises prohibited content based on the at least one percentage of similarity.
13. The system of claim 11 , the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the at least one processor, cause the processor to perform operations comprising:
sending the media file and the at least one percentage of similarity to a compliance review team to determine if the media file comprises prohibited content.
14. The system of claim 11 , wherein the comparing of the fingerprint against the at least one known fingerprint identifies a distinction within a temporal order of the media file and at least one piece of known content.
15. A non-transitory computer-readable storage medium having instructions stored which, when executed by at least one processor, cause the at least one processor to perform operations comprising:
receiving a media file comprising video content;
converting the video content to greyscale, resulting in greyscaled video content;
segmenting the greyscaled video content using a rolling window of predetermined length, resulting in a downsampled plurality of processed segments;
calculating, for each segment within the plurality of downsampled processed segments, an average of pixel values, resulting in an average grey pixel value for each frame;
hashing the average grey pixel value for each segment within the plurality of downsampled processed segments, resulting in a plurality of hashed segment pixel values; and
generating, by combining at least two hashed segment pixel values from the plurality of hashed segment pixel values, a fingerprint of at least a portion of the media file.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the rolling window causes each segment to have an amount of overlap with a previous segment and a subsequent segment.
17. The non-transitory computer-readable storage medium of claim 15 , wherein segments of the media file corresponding to the at least two hashed segment pixel values are consecutive within the media file.
18. The non-transitory computer-readable storage medium of claim 15 , having additional instructions stored which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
comparing the fingerprint against at least one known fingerprint, resulting in at least one percentage of similarity.
19. The non-transitory computer-readable storage medium of claim 18 , having additional instructions stored which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
determining that the media file comprises prohibited content based on the at least one percentage of similarity.
20. The non-transitory computer-readable storage medium of claim 18 , having additional instructions stored which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
sending the media file and the at least one percentage of similarity to a compliance review team to determine if the media file comprises prohibited content.