Method for identifying motion video/audio content
OwnerJi Zhanginventor
Life of this patent
Estimating the term…
Abstract
A method for identifying motion video/audio content, by means of comparing a video A to a registered video B so as to determine if they are originally the same as each other, wherein said method at least comprises the steps of extracting a fingerprint A from the video A; and searching from a fingerprint database for a pre-extracted and registered fingerprint B of the video B by means of comparison of fingerprint A with a sliding window of a possible fingerprint B, so as to determine that the video A is visually identical to the video B if a match is found. According to the present invention, the method for extracting a fingerprint data from video/audio signals facilitates the automatic identification, archiving and search of video content, and can be of without the need for human visual inspections.
Claims7
A method for identifying motion video/audio content, by means of comparing a video A to a registered video B so as to determine if they are originally the same as each other, wherein said method at least comprises the steps of
- extracting a fingerprint A from the video A;
- and
- searching from a fingerprint database for a pre-extracted and registered fingerprint B of the video B by means of comparison of fingerprint A with a sliding window of a possible fingerprint B, so as to determine that the video A is identical to the video B if a match is found, the comparison steps further comprise the steps of
- Calculating the sum of absolute difference (SAD) between the fingerprint A and possible fingerprint B, in which it is assumed that the fingerprint A has fewer number of rows than fingerprint B, each row contains the samples for a single video image, if N samples obtained from the first video frame of the fingerprint A are denoted as (A1, A2, A3, A4, A5, . . . , AN), N samples obtained from the first video frame of the fingerprint B within a sliding window are denoted as (B1, B2, B3, B4, B5, . . . , BN), and ((A1 and B1), (A2 and B2), . . . , (AN and BN)) are on the same pixel sample location of a frame from video A and a frame from video B respectively, then a SAD operation between these two frames of the fingerprint A and possible fingerprint B is given by
- SAD
-
- (
- A
- ,
- B
- )
- =
- ∑
- i
- =
- 1
- N
-
-
-
- Ai
- -
- Bi
-
- where | . . . | is the absolute value operation, the SAD operation is performed repeatedly between the samples from the second frame from video A and the second frame from video B so that for every video frames of the fingerprint A and possible fingerprint B, until all of the frames in video A are included in the calculation, the results are added together to form a single sum of SAD (SSAD) number,
- the same process is then repeated by shifting the position of fingerprint B relative to A by one frame, each such shift results in a new SSAD value generated, so a series of SSAD values are generated and saved as a SSAD array, and
- a fingerprint match is identified by a percentage drop below a given threshold in the SSAD values just before the match and a percentage increase above a given threshold in SSAD values just after the match.
Dependencies are inferred from the claim text.