Method and apparatus for detecting non-people objects in revolving doors

• US 20070047837A1
• Filed: 08/29/2005
• Published: 03/01/2007
• Est. Priority Date: 08/29/2005
• Status: Abandoned Application

• Current Assignee: Cognex Technology & Investment Corporation (Cognex Corporation)
• Original Assignee: Cognex Technology & Investment Corporation (Cognex Corporation)
• Inventors: Nichani, Sanjay, Schwab, John
• Application Number:
  US11/215,307
• Publication Number:
  US 20070047837A1
• Time in Patent Office: Days
• US Class Current: 382/285
• CPC Class Codes: G06T 7/0008 checking presence/absence
• CPC Class Codes: G06T 7/593 from stereo images
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G07C 9/15 with arrangements to prevent the passage of more than one individual at a time

Abstract

A stereo imaging based vision system and method provides enhanced portal security through stereoscopy. In particular, a system detects non-people objects within the chamber of a revolving door by acquiring two-dimensional (2D) images from different vantage points, and computing a filtered set of three-dimensional (3D) features of the door compartment by using both the acquired 2D images and model 2D images. Applying image processing techniques to the filtered 3D feature set, non-people objects can be detected.

Claims

1. A method of detecting objects comprising:

acquiring a plurality of 2D images of a space in a revolving door;

computing a filtered set of 3D features using the plurality of acquired 2D images and a plurality of model 2D images; and

identifying non-people objects within the revolving door space.

2. A method of claim 1 wherein the filtered set of 3D features is a disparity map.

3. A method of claim 1 wherein computing a filtered set of 3D features comprises:

computing a set of acquired 3D features from the plurality of acquired 2D images;

computing a set of model 3D features from the plurality of model 2D images; and

filtering the set of model 3D features from the set of acquired 3D features.

4. A method of claim 1 wherein computing a filtered set of 3D features comprises:

filtering the plurality of model 2D images from the plurality of acquired 2D images to create a plurality of filtered 2D images; and

computing the filtered set of 3D features from the plurality of filtered 2D images.

5. A method of claim 4 further comprising:

processing the filtered set of 3D features to minimize shadow and noise.

6. A method of claim 4 wherein identifying non-people objects comprises a blob analysis.

7. A method of claim 4 wherein identifying non-people objects comprises pattern recognition.

8. A method of claim 1 further comprising:

eliminating transient non-people objects from detection by tracking identified non-people objects.

9. A method of claim 1 wherein acquiring a plurality of 2D images occurs in response to a triggering event.

10. A method of claim 9 wherein the triggering event is the detection of a particular door position.

11. A method of claim 1 wherein the model images are an average of previously acquired images taken over a period of time.

12. A method of claim 1 wherein the model images are an average of filtered previously acquired images taken over a period of time.

13. A method of claim 1 wherein the model images are an average of cleared images taken over a period of time.

14. A method of claim 13 wherein recently cleared images are weighed more heavily.

15. A method of claim 1 further comprising transmitting an alert in response to the identification of a non-people object.

16. A method of claim 1 further comprising stopping the revolving door in response to the identification of a non-people object.

17. A secured portal comprising:

a revolving door separating a first area from a second area;

a plurality of image sensors positioned to acquire a plurality of 2D images in a space in the revolving door; and

a processor for detecting non-people objects by;

(i) computing a filtered set of 3D features using the plurality of acquired 2D images and a plurality of model 2D images, and

(ii) identifying non-people objects in the revolving door space using the filtered set of 3D features.

18. A secured portal of claim 17 wherein the filtered set of 3D features is a disparity map.

19. A secured portal of claim 17 wherein computing a filtered set of 3D features comprises:

computing a set of acquired 3D features from the plurality of acquired 2D images;

computing a set of model 3D features from the plurality of model 2D images; and

filtering the set of model 3D features from the set of acquired 3D features.

20. A secured portal of claim 17 wherein computing a filtered set of 3D features comprises:

filtering the plurality of model 2D images from the plurality of acquired 2D images to create a plurality of filtered 2D images; and

computing the filtered set of 3D features from the plurality of filtered 2D images.

21. A secured portal of claim 20 further comprising:

processing the filtered set of 3D features to minimize shadow and noise.

22. A secured portal of claim 17 wherein identifying non-people objects comprises a blob analysis.

23. A secured portal of claim 17 wherein identifying non-people objects comprises pattern recognition.

24. A secured portal of claim 17 wherein the processor further eliminates transient non-people objects from detection by tracking identified non-people objects.

25. A secured portal of claim 17 wherein the plurality of image sensors acquire the plurality of 2D images in response to a triggering event.

26. A secured portal of claim 25 wherein the triggering event is the detection of a particular door position.

27. A secured portal of claim 17 wherein the model images are an average of previously acquired images taken over a period of time.

28. A secured portal of claim 17 wherein the model images are an average of filtered previously acquired images taken over a period of time.

29. A secured portal of claim 17 wherein the model 2D images are an average of cleared images taken over a period of time.

30. A secured portal of claim 29 wherein recently cleared images are weighed more heavily.

31. A secured portal of claim 17 wherein the processor is further capable of transmitting an alert in response to the identification of a non-people object.

32. A secured portal of claim 31 further comprising:

a control system for stopping movement of the revolving door upon receipt of an alert.

33. A computer readable medium having computer readable program codes embodied therein for causing a computer to function as an analysis unit that selectively designates prohibited communications connections between an origin and one or more destinations in a communications network, the computer readable medium program codes performing functions comprising:

acquiring a plurality of 2D images of a space in a revolving door;

computing a filtered set of 3D features using the plurality of acquired 2D images and a plurality of model 2D images; and

identifying non-people objects within the revolving door space.

34. A security method comprising:

acquiring a plurality of 2D images of a space in a revolving door;

identifying a non-people object within the revolving door space; and

transmitting an alert upon detection of the non-people object.

35. A method of claim 34 further comprising:

stopping the revolving door in response to the alert.