Generation of video hash
An apparatus and method are providing for generating a hash in video in which a sample series of temporal difference are sampled in an image order. A temporal averaging is performed and a rate of change is detected to identify as distinctive events regions of high rate of change. Images having a distinctive event are labelled as distinctive images. For each image, the temporal spacing in images is calculated between that image and other distinctive images to provide a set of temporal spacings for that image; and a hash is derived for that image from that set of temporal spacings.
1. An apparatus for determining a distinctive event in a sequence of video images, comprising:
a first filter configured to receive a series of difference sample values in at least one video image of the sequence of video images, with each difference sample value representing a difference in respective pixel values in the at least one video image of the sequence of video images, with the first filter configured to perform an averaging of the series of difference sample values;
a second filter configured to determine a rate of change based on a magnitude corresponding to a difference between the averaged series of difference sample values of the at least one video image with an averaged sample series of at least one other video image in the sequence of video images, and further configured to detect a distinctive event based on the respective magnitudes indicating the rate of change for the difference sample values between the respective video images; and
a buffer arrangement configured to store the detected distinctive event with a corresponding temporal location for a respective video image in the sequence of video images.
2. The apparatus of claim 1 , wherein the second filter is further configured to detect the distinctive event when a peak of each respective magnitude is above a predetermined threshold.
3. The apparatus of claim 1 , further comprising a hash generator configured to derive a hash for the respective video image based on at least one temporal spacing between the respective video image and the at least one other video image being in a temporal neighborhood of the respective video image having associated therewith the detected distinctive event.
4. The apparatus of claim 3 , wherein, for a location of the at least one video image in the sequence of video images, the hash generator is further configured to calculate a temporal spacing between the respective location and each of a plurality of distinctive events in a temporal neighborhood of the respective location to provide a set of temporal spacings for the respective location.
5. The apparatus of claim 4 , wherein the hash generator is further configured to derive the hash for the respective location from the set of temporal spacings for the respective location.
6. The apparatus of claim 1 , wherein the difference sample values are temporal difference sample values in an image order of the sequence of video images, with each temporal difference sample value representing a difference in respective pixel values between the at least one video image and an adjoining video image in the image order.
7. The apparatus of claim 6 , wherein the respective temporal difference sample value for each video image in the image order is a single number for each respective video image.
8. An apparatus for determining a distinctive event in a sequence of video images, comprising:
a first filter configured to perform a temporal averaging of a series of temporal difference samples in the sequence of video images, with each temporal difference sample representing a difference in respective pixel values between a video image and an adjoining video image in the sequence of video images;
a second filter configured to determine a rate of change based on a magnitude corresponding to a difference between the temporally averaged series of the video image with the temporally averaged sample series of the adjoining video image, and further configured to detect a distinctive event based on the respective magnitudes indicating the rate of change for the temporal difference samples; and
a buffer arrangement configured to store the detected distinctive event with a corresponding temporal location in the sequence of video images.
9. The apparatus of claim 8 , further comprising a hash generator configured to derive a hash for the respective video image based on at least one temporal spacing between the respective video image and at least one other video image being in a temporal neighborhood of the respective video image having associated therewith the detected distinctive event.
10. The apparatus of claim 9 , wherein, for a location of the video image in the sequence of video images, the hash generator is further configured to calculate a temporal spacing between the respective location and each of a plurality of distinctive events in a temporal neighborhood of the respective location to provide a set of temporal spacings for the respective location.
11. The apparatus of claim 10 , wherein the hash generator is further configured to derive the hash for the respective location from the set of temporal spacings for the respective location.
12. The apparatus of claim 8 , wherein the respective temporal difference samples for each video image in the image order is a single number for each respective video image.
13. The apparatus of claim 8 , wherein the second filter is further configured to detect the distinctive event when a peak of each respective magnitude is above a predetermined threshold.
14. An apparatus for determining a distinctive event in a sequence of video images, comprising:
means for averaging a series of difference sample values in at least one video image of the sequence of video images, with each difference sample value representing a difference in respective pixel values in the at least one video image of the sequence of video images;
means for determining a rate of change based on a magnitude corresponding to a difference between the averaged series of difference sample values of the at least one video image with an averaged sample series of at least one other video image in the sequence of video images;
means for detecting a distinctive event in the sequence of video images based on the respective magnitudes indicating the rate of change for the difference sample values between the respective video images; and
means for storing in a buffer the detected distinctive event with a corresponding temporal location for a respective video image in the sequence of video images.
15. The apparatus of claim 14 , further comprising means for deriving a hash for the respective video image based on at least one temporal spacing between the respective video image and the at least one other video image being in a temporal neighborhood of the respective video image having associated therewith the detected distinctive event.
16. The apparatus of claim 15 , further comprising means for calculating a temporal spacing between a location of the at least one video image in the sequence of video images and each of a plurality of distinctive events in a temporal neighborhood of the respective location to provide a set of temporal spacings for the respective location.
17. The apparatus of claim 16 , further comprising means for deriving the hash for the respective location from the set of temporal spacings for the respective location.
18. The apparatus of claim 14 , wherein the difference sample values are temporal difference sample values in an image order of the sequence of video images, with each temporal difference sample value representing a difference in respective pixel values between the at least one video image and an adjoining video image in the image order.
19. The apparatus of claim 18 , wherein the respective temporal difference sample value for each video image in the image order is a single number for each respective video image.
20. The apparatus of claim 14 , further comprising means for detecting the distinctive event when a peak of each respective magnitude is above a predetermined threshold.