Systems and methods of determining retention of video surveillance data
This disclosure articulates methods for calculating a metric based on actual time period of retention (ρ α ) in a given video recording system. The measurement is performed on a per-camera-stream basis for each camera stream being recorded on a system. The metric is related to Video Retention Compliance (VRC), which is another key performance indicators of video surveillance applications, as it is tied to a business and, sometimes, regulatory requirement of an organization. This disclosure describes methods of calculating retention for each camera stream recording on a server, and doing so in an efficient enough way so as to be practical for ongoing operations.
1. A computer-implemented method of intelligently measuring actual video data retention in a video surveillance infrastructure by an independent audit of video data, where incoming video data streams received from a plurality of video capturing devices are processed by a video management software (VMS) application running on a video processing server, and processed outgoing video data streams are directed to one or more storage devices, the method comprising:
for each video data path,
obtaining usage profile of video data on a storage device for a particular time period;
obtaining usage profile of video data on the storage device for a time period immediately preceding the particular time period;
calculating by a computer processor, an actual time period of retention of video data using a difference between the usage profiles of the successive time periods;
comparing the actual time period of retention with a desired time period of retention associated with an individual video data stream from a video capturing device to a storage device to calculate a key performance indicator (KPI);
using the KPI to determine quality of video data retention in the video surveillance infrastructure.
2. The method of claim 1 , wherein the KPI is Video Retention Compliance (VRC).
3. The method of claim 2 , wherein a video data retention window is adjusted based on the calculated VRC.
4. The method of claim 1 , wherein the desired time period of retention varies depending on special data retention policies and jurisdictions.
5. The method of claim 1 , wherein the actual time period of retention depends on degree of motion detected in a video stream.
6. The method of claim 1 , wherein the actual time period of retention data is extracted based on age of a stored file.
7. The method of claim 1 , wherein the time period is a day and successive measurements are taken at the same time on each day of a plurality of days.
8. The method of claim 6 , wherein an algorithm builds a distribution of the age of the stored file in a master filesystem table (MTF).
9. The method of claim 8 , wherein a subset of data is used from the MTF, including one or more of: file identification numeric value, filename, filepath, parent directory identification, file creation date, last modification date, and filesize.
10. The method of claim 8 , wherein specific knowledge of the VMS is used to expedite the algorithm.
11. The method of claim 8 , wherein extracted versions of the MTFs are saved for each measurement, and the differential between MTFs of successive measurements are utilized to refine the algorithm.
12. The method of claim 2 , wherein VRCs calculated for each video path may be analyzed and aggregated to monitor overall health of the video surveillance infrastructure.
13. The method of claim 12 , wherein VRC is used in conjunction with other KPIs while assessing the overall health of the video surveillance infrastructure.
14. The method of claim 1 , wherein the time period is chosen from: a day, a plurality of days, an hour, a plurality of hours, or, a fraction of an hour.
15. The method of claim 14 , wherein the usage profile of video data is a histogram depicting video data usage over the chosen time period.
16. The method of claim 1 , wherein the method enables detection of premature deletion of video data.
17. The method of claim 16 , wherein the premature deletion of video data is caused by malicious action or accidental action.