Systems and methods for adaptive bi-directional people counting

• US 8,325,976 B1
• Filed: 03/14/2008
• Issued: 12/04/2012
• Est. Priority Date: 03/14/2008
• Status: Active Grant

• Current Assignee: Cognyte Technologies Israel Ltd. (Cognyte Software Ltd.)
• Original Assignee: Verint Systems Limited (Verint Systems Incorporated)
• Inventors: Golan, Oren, Steinmetz, Ofer
• Primary Examiner(s): Park, Chan S
• Assistant Examiner(s): YENTRAPATI, AVINASH
• Application Number:
  US12/048,996
• Time in Patent Office: 1,726 Days
• Field of Search: 348169-172
• US Class Current: 382/103
• CPC Class Codes: G06V 20/53 Recognition of crowd images, e.g. recognition of crowd congestion
• CPC Class Codes: G07C 9/00 Individual registration on entry or exit

Abstract

People are counted in a segment of video with a video processing system that is configured with a first set of parameters. This produces a first output. Based on this first output, a second set of parameters is chosen. People are then counted in the segment of video using the second set of parameters. This produces a second output. People are counted with a video played forward. People are counted with a video played backwards. The results of these two counts are reconciled to produce a more accurate people count.

Claims

1. A method of counting people in a video using multiple passes in multiple directions, comprising:

processing a segment of a video played in a first direction with a video processing system to produce a first set of trajectories corresponding to people in a scene;

processing the segment of the video played in a second direction with the video processing system to produce a second set of trajectories corresponding to the people in the scene;

processing the first set of trajectories and the second set of trajectories using equivalence criteria to identify groups of equivalent trajectories from the first set and the second set of trajectories;

selecting a third set of trajectories comprising only one trajectory from each of the groups of equivalent trajectories; and

,

determining a count of the people in the scene based on the third set of trajectories.

2. The method of claim 1, wherein the third set of trajectories further comprises at least one unique trajectory unique across the first and the second set of trajectories.

3. The method of claim 1, wherein the equivalence criteria considers at least two trajectories equivalent if a correlation between the at least two trajectories meets a threshold value.

4. The method of claim 1, wherein the segment of the video played in the first direction is processed using a first set of parameters and the segment of the video played in a second direction is processed using a second set of parameters.

5. The method of claim 4, wherein the first set of parameters and the second set of parameters include at least one parameter that determines an algorithm used to count people.

6. A multi-directional video people counting system, comprising:

a video processing system configured to process a segment of a video played in a first direction to produce a first set of trajectories corresponding to people in a scene;

and wherein the video processing system also processes the segment of the video played in a second direction to produce a second set of trajectories corresponding to the people in the scene;

and wherein the video processing system processes the first set of trajectories and the second set of trajectories using equivalence criteria to identify groups of equivalent trajectories from the first set and the second set of trajectories;

and wherein the video processing system selects a third set of trajectories comprising only one trajectory from each of the groups of equivalent trajectories; and

,

wherein the video processing system determines a count of the people in the scene based on the third set of trajectories.

7. The multi-directional video people counting system of claim 6, wherein the third set of trajectories further comprises at least one unique trajectory unique across the first and second set of trajectories.

8. The multi-directional video people counting system of claim 6, wherein the equivalence criteria considers at least two trajectories equivalent if a correlation between the at least two trajectories meets a threshold value.

9. The method of claim 6 wherein the segment of the video played in the first direction is processed using a first set of parameters and the segment of the video played in a second direction is processed using a second set of parameters.

10. The method of claim 9 wherein the first set of parameters and the second set of parameters include at least one parameter that determines an algorithm used to count people.

11. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for counting people in a video using multiple passes in multiple directions, the method comprising:

processing a segment of a video played in a first direction with a video processing system to produce a first set of trajectories corresponding to people in a scene;

processing the segment of the video played in a second direction with the video processing system to produce a second set of trajectories corresponding to the people in the scene;

processing the first set of trajectories and the second set of trajectories using equivalence criteria to identify groups of equivalent trajectories from the first set and the second set of trajectories;

selecting a third set of trajectories comprising only one trajectory from each of the groups of equivalent trajectories; and

,

determining a count of the people in the scene based on the third set of trajectories.

12. The program storage device of claim 11, wherein the third set of trajectories further comprises at least one unique trajectory unique across the first and the second set of trajectories.

13. The program storage device of claim 11, wherein the equivalence criteria considers at least two trajectories equivalent if a correlation between the at least two trajectories meets a threshold value.

14. The program storage device of claim 11, wherein the segment of the video played in the first direction is processed using a first set of parameters and the segment of the video played in a second direction is processed using a second set of parameters.

15. The program storage device of claim 14, wherein the first set of parameters and the second set of parameters include at least one parameter that determines an algorithm used to count people.