Stereo door sensor

• US 7,400,744 B2
• Filed: 03/14/2003
• Issued: 07/15/2008
• Est. Priority Date: 09/05/2002
• Status: Expired due to Fees

• Current Assignee: Cognex Technology & Investment Corporation (Cognex Corporation)
• Original Assignee: Cognex Technology & Investment Corporation (Cognex Corporation)
• Inventors: Schatz, Dave, Silver, William, Nichani, Sanjay
• Primary Examiner(s): Mehta; Bhavesh M.
• Assistant Examiner(s): Strege; John B
• Application Number:
  US10/388,925
• Publication Number:
  US 20040045339A1
• Time in Patent Office: 1,950 Days
• Field of Search: 382/103, 382/154, 348/153
• US Class Current: 382/103
• CPC Class Codes: E05F 15/73 responsive to movement or presence of persons or objects
• CPC Class Codes: E05F 2015/767 using cameras
• CPC Class Codes: E05Y 2900/132 Doors
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69

Abstract

A stereo imaging based vision system to monitor the area on the two sides of either a door and control door motion according to the motion of 3D objects in the viewing area. The system calibrated to provide heights above the ground plane for any point in the field of view. Therefore, when any object enters the field of view, it generates interest points called “features”, the heights of which are measured relative to the ground plane. These points are then clustered in 3D space to provide “objects”. These objects are then tracked in multiple frames to provide “trajectories”. Such a system could then control the door motion (open, close, stall) based on the various pieces of information generated about the object.

Claims

1. A method of controlling a door comprising the steps of:

acquiring a stereo image of an incoming area;

computing a set of 3D features from said stereo image;

filtering from said set of 3D features to generate a set of filtered 3D features;

computing a trajectory of said set of filtered 3D features relative to said door; and

generating door control signals in response to said trajectory;

wherein said step of computing a set of 3D features includes the steps of;

edge-processing said stereo image to generate a plurality of connected features;

identifying connected features having a selected characteristic greater than a predetermined threshold as selected features;

matching said selected features to generate disparities generated from different images in said image set; and

computing 3D locations of feature points according to at least said disparities.

2. The method according to claim 1 wherein said filtering step eliminates ground plane features from said set of 3D features.

3. The method according to claim 1 wherein said filtering step eliminates shadows from said set of 3D features.

4. The method according to claim 1 wherein said filtering step eliminates background patterns from said set of 3D features.

5. The method according to claim 1 wherein said filtering step eliminates ambient lighting effects.

6. The method according to claim 1 wherein said filtering step eliminates features outside of preselected 3D zones.

7. The method according to claim 1 wherein said stereo image is acquired by combining images acquired by a stereo vision camera.

8. The method according to claim 1 wherein said stereo image is acquired by combining a plurality of images acquired by a monocular camera.

9. The method according to claim 8 wherein at least one of said plurality of images are acquired by said monocular camera via reflective optical components.

10. The method according to claim 8 wherein at least one of said plurality of images are acquired by said monocular camera via refractive optical components.

17. The method according to claim 1, wherein said filtering step comprises:

selecting 3D spaces of interest; and

eliminating features outside said selected 3D spaces of interest.

18. The method according to claim 17 further comprising performing different run-time algorithms on each of a plurality of said 3D spaces of interest.

11. A stereo vision apparatus for controlling a door, said apparatus comprising:

a stereo image acquisition device;

a 3D processor receiving stereo images from said stereo image acquisition device;

a trajectory processor receiving frames comprising 3D objects from said 3D processor;

and

a door actuator receiving door control signals from said trajectory processor in response to an object'"'"'s trajectory relative to said door;

wherein said 3D processor generates said frames comprising 3D objects by;

determining heights above a ground plane for points in a field of view of said stereo image acquisition devices; and

clustering said points in 3D space to generate objects;

wherein said clustering includes;

generating initial clusters according to chain organization of edgelets;

breaking chains of features into contiguous segments based on changes in a z coordinate between successive points; and

merging two closest clusters based on a minimum distance criteria.

12. The apparatus according to claim 11 wherein said stereo image acquisition device comprises a plurality of electronic cameras.

13. The apparatus according to claim 11 wherein objects having a close proximity to said ground plane relative to a predetermined threshold are filtered out as ground plane noise.

14. The apparatus according to claim 11 wherein said trajectory processor determines an object'"'"'s trajectory by tracking said objects in multiple frames.

15. The apparatus according to claim 11 wherein said stereo image acquisition device comprises a monocular camera configured for acquiring a plurality of images.

16. The apparatus according to claim 15 wherein at least one of said plurality of images are acquired via a mirror.