Video analysis method and apparatus and computer program
Provided is a video analysis (VA) method comprises calculating, by a VA apparatus, a size of each of a plurality of frames by summing sizes of a plurality of packets constituting each of the plurality of frames, analyzing, by the VA apparatus, the size of each of the plurality of frames constituting a video to determine a pattern of the sizes of the plurality of frames and determining, by the VA apparatus, whether there is a motion in the video based on the size pattern.
1. A video analysis (VA) method comprising:
calculating, by a VA apparatus, a size of each of a plurality of frames by summing sizes of a plurality of packets constituting each of the plurality of frames;
analyzing, by the VA apparatus, the size of each of the plurality of frames constituting a video to determine a size pattern of the plurality of frames; and
determining, by the VA apparatus, whether there is a motion in the video based on the size pattern,
wherein the analyzing comprises calculating three or more among an average, a deviation, a minimum, a maximum, and a period of the plurality of frames, and
wherein the determining comprises determining that there is no motion in the video in response to the minimum and the maximum periodically repeating, and determining that there is the motion in the video in response to the minimum and the maximum not periodically repeating.
2. The VA method of claim 1 , wherein the calculating the size of each of the plurality of frames comprises detecting for each of the plurality of frames, when packets of the video have not been encrypted, a packet including marker information indicating a start of a frame to a packet including marker information indicating an end of the frame as the plurality of packets constituting the frame.
3. The VA method of claim 1 , wherein the calculating the size of each of the plurality of frames comprises detecting, when packets of the video have been encrypted, packets corresponding to a number of packets per frame, the number of packets per frame being previously machine-learned according to a codec of the video.
4. The VA method of claim 1 , wherein the determining further comprises determining that there is no motion in the video in response to the average being within a preset threshold value from an average of the minimum and the maximum.
5. The VA method of claim 1 , wherein the determining further comprises determining that there is no motion in the video in response to the deviation being less than a preset threshold value.
6. The VA method of claim 1 , wherein the determining further comprises determining that there is the motion in the video in response to the average being within a preset threshold value from the minimum.
7. The VA method of claim 1 , wherein the determining further comprises determining that there is the motion in the video in response to the deviation being equal to or greater than a preset threshold value.
8. The VA method of claim 1 , wherein the determining further comprises quantifying a degree of the motion based on a decision tree previously machine-learned according to a codec of the video in response to the determining indicating that there is the motion in the video, and
the decision tree has an average and a deviation of the plurality of frames as input variables and has the quantified degree of the motion as a target variable.
9. The VA method of claim 1 , further comprising dropping frames with no motion from the video.
10. The VA method of claim 9 , further comprising transmitting the video from which the frames with no motion have been dropped via a network.
11. The VA method of claim 9 , further comprising storing the video from which the frames with no motion have been dropped in a storage.
12. A video analysis (VA) apparatus comprising:
a memory configured to store a computer program; and
at least one processor configured to read and execute the computer program,
wherein the computer program comprises:
frame size calculating instructions configured to cause the at least one processor to calculate a size of a frame by summing sizes of a plurality of packets constituting the frame; and
motion detection instructions configured to cause the at least one processor to analyze a pattern of sizes of a plurality of frames constituting a video by analyzing the size of each of the plurality of frames, and determine whether there is a motion in the video based on the pattern,
wherein the analyzing comprises calculating three or more among an average, a deviation, a minimum, a maximum, and a period of the plurality of frames, and
wherein the determining comprises determining that there is no motion in the video in response to the minimum and the maximum periodically repeating, and determining that there is the motion in the video in response to the minimum and the maximum not periodically repeating.
13. A non-transitory computer-readable medium having stored thereon a computer program, which when executed by a processor of a video analysis (VA) apparatus, causes the VA apparatus to:
calculate a size of a frame by summing a size of each of a plurality of packets constituting the frame;
analyze a pattern of sizes of a plurality of frames constituting a video by analyzing the size of each of the plurality of frames; and
determine whether there is a motion in the video based on the pattern,
wherein the analyzing comprises calculating three or more among an average, a deviation, a minimum, a maximum, and a period of the plurality of frames, and
wherein the determining comprises determining that there is no motion in the video in response to the minimum and the maximum periodically repeating, and determining that there is the motion in the video in response to the minimum and the maximum not periodically repeating.