Methods and systems for excluding individuals from retail analytics
Systems and methods are described for excluding individuals from retail analytics. Pedestrian traffic within a predetermined area is monitored via a video sensor to determine a video entrance. The video entrance may correspond to entrances of the predetermined area by a first set of individuals. The pedestrian traffic may also be monitored via a plurality of wireless sensors to determine a wireless entrance. The wireless entrance may correspond to entrances by a second set of individuals. The video entrance and the wireless entrance are compared to determine if any of the entrances associated with a video timestamp match any of the entrances with a wireless timestamp. An overall entrance count corresponding to the number of customers entering the predetermined area is determined. Retail analytics excluding employees may subsequently be calculated using the overall entrance count.
1. A method comprising:
monitoring, by a video sensor, pedestrian traffic to determine
video paths for a first plurality of individuals within a first predetermined area over a first predetermined period of time and video paths for a second plurality of individuals within a second predetermined area over a second predetermined period of time, and to identify for each individual, a corresponding path, comprising a plurality of locations at a plurality of timestamps, each timestamp being associated with a corresponding location;
monitoring, by at least three wireless sensors, the pedestrian traffic within the plurality of predetermined areas over the first and second predetermined periods of time
to determine a wireless path of an individual, the wireless path comprising a plurality of locations corresponding to a wireless timestamp, the timestamp being associated with a time window during which the at least three wireless sensors detect movement of the individual in a specific direction;
comparing the wireless path and the video paths for the first and second plurality of individuals to see if a match exists;
excluding a video path matching the wireless path from a compiled set of customer path data comprising the video paths for the first and second plurality of individuals; and
calculating retail analytics using the compiled set of customer path data, the retail analytics outputting data being related to the number of customers in the first and second predetermined areas during the first and second predetermined periods of time.
2. The method of claim 1 , wherein a first sensor being located near the predetermined area, a second sensor being located inside a retail space, and a third sensor being located outside of the retail space.
3. The method of claim 1 wherein the comparing includes comparing the video paths with the wireless path and identifying a match when a direction of the wireless path and a direction of the video paths are substantially the same.
4. The method of claim 1 wherein comparing involves comparing positional data obtained both wirelessly and by video at the same time points to see if there is a match.
5. The method of claim 1 , wherein there is a threshold for matching.
6. The method of claim 1 , wherein the comparing includes using an error correction model.
7. The method of claim 1 , the video sensor and the plurality of wireless sensors being situated at a specified location deep inside the retail space.
8. The method of claim 1 , the video sensor and the plurality of wireless sensors being situated at an entryway to the retail space.
9. The method of claim 1 , further comprising computing employee retail analytics using a compiled set of employee path data comprising a plurality of video paths, including the video path matching the wireless path, from the video paths for the plurality of individuals over the first and second predetermined periods of time, the employee retail analytics outputting data that is related to the number of employees in the first and second predetermined areas during the first and second predetermined periods of time.
10. The method of claim 1 , where the wireless path of the individual is determined using one of trilateration, Markov localization, and nearest neighbor localization to determine the location of the individual at each timestamp.
11. The method of claim 1 , the comparing further comprising a trajectory-based comparison of the wireless path with each of the video paths.
12. The method of claim 1 , wherein the first predetermined period of time is the same as the second predetermined period of time.
13. The method of claim 1 , wherein the first predetermined period of time is different than the second predetermined period of time.
14. A method comprising:
monitoring, by a video sensor, pedestrian traffic within a predetermined area over a predetermined period of time, to determine video paths for a plurality of individuals, each video path comprising a plurality of locations at a plurality of timestamps, each timestamp being associated with a corresponding location, and to determine whether any of the video paths include a dwell, the dwell being a portion of a path where the location is the same for a dwell period comprising a predetermined number of timestamps;
monitoring, by at least three wireless sensors, the pedestrian traffic within the predetermined area over the predetermined period of time to determine, by the processor, a wireless path of an individual, the wireless path comprising a plurality of locations corresponding to a plurality of wireless timestamps, the timestamps being associated with a time window during which the at least three wireless sensors detect movement of the individual in a specific direction;
comparing, by the processor, the wireless path and the video paths for the plurality of individuals to see if a match exists;
excluding, by the processor, from a compiled set of customer path data comprising the video paths for the plurality of individuals over the predetermined period of time, a video path matching the wireless path and each video path identified as having a dwell; and
calculating retail analytics using the compiled set of customer path data, the retail analytics outputting data that is related to: i) the number of customers in the predetermined area during the predetermined period of time, and ii) the location of the dwell for each individual whose corresponding video path includes a dwell.
15. The method of claim 14 , wherein a first sensor being located near the predetermined area, a second sensor being located inside a retail space, and a third sensor being located outside of the retail space.
16. The method of claim 14 , wherein the comparing includes comparing the video paths with the wireless path and identifying a match when a direction of the wireless path and a direction of the video paths are substantially the same.
17. The method of claim 14 , wherein comparing involves comparing positional data obtained both wirelessly and by video at the same time points to see if there is a match.
18. The method of claim 14 , wherein there is a threshold for matching.
19. The method of claim 14 , wherein the comparing includes using an error correction model.
20. The method of claim 14 , the video sensor and the plurality of wireless sensors being situated at a specified location deep inside the retail space.
21. The method of claim 14 , the video sensor and the plurality of wireless sensors being situated at an entryway to the retail space.
22. The method of claim 14 , further comprising computing employee retail analytics using a compiled set of employee path data comprising a plurality of video paths, including the video path matching the wireless path, from the video paths for the plurality of individuals over the predetermined period of time, the employee retail analytics outputting data that is related to the number of employees in the predetermined area during the predetermined period of time.
23. The method of claim 14 , where the wireless path of the individual is determined using one of trilateration, Markov localization, and nearest neighbor localization to determine the location of the individual at each timestamp.
24. The method of claim 14 , the comparing further comprising a trajectory-based comparison of the wireless path with each of the video paths.