Automatic configuration of cameras in building information modeling

Patent metadata

Field	Value
Publication number	US-9942521-B2
Application number	US-201213350293-A
Country	US
Kind code	B2
Filing date	Jan 13, 2012
Priority date	Jan 13, 2012
Publication date	Apr 10, 2018
Grant date	Apr 10, 2018

Abstract

There is disclosed herein a system for automatic configuration of cameras in a Building Information Model (BIM) comprising a programmed BIM processing system. A plurality of input ports are distributed at select locations in the building. A database stores a network location map identifying locations of the input ports. The BIM processing system is operated to detect a camera connected to one of the input ports and reading camera image data; determine possible building areas in the camera's field of view based on location of the one of the input ports and extracting features from the stored building models for the determined possible building areas, and establishing mapping between the camera image data and the extracted features to determine actual location of the camera in the building and store camera location data in the database.

First claim

The invention claimed is:

1. A method for automatic configuration of cameras in a Building Information Model (BIM), comprising:
providing a programmed BIM processing system including a processor and associated memory for managing building data;
connecting a plurality of input ports to the BIM processing system, the input ports being distributed at select locations in the building;
storing a database including building models and storing a network location map identifying locations of the input ports;
detecting a camera connected to one of the input ports and reading camera image data from the camera;
determining possible building areas in the cameras field of view based on location of the one of the input ports and extracting stored building model data for the determined possible building areas; and
establishing mapping between the camera image data and the stored building models by comparing features in the camera image data to features from the extracted stored building model data for the determined possible building areas to determine actual location of the camera in the building and store camera location data in the database.

2. The method of claim 1 wherein detecting a camera comprises detecting a video camera and reading camera image date from the video camera comprises selecting an image frame.

3. The method of claim 2 wherein selecting the image frame comprises selecting an image frame when no person is in the field of view.

4. The method of claim 1 wherein each input port includes an associated unique identifier stored in the network location map and the possible building areas are determined based on the unique identifier of the one of the input ports.

5. The method of claim 1 wherein establishing mapping between the camera image data and the stored building models comprises extracting features selected from edges, corners, rectangles and the like from the camera image data.

6. The method of claim 5 wherein establishing mapping between the camera image data and the stored building models comprises extracting features selected from edges, corners, rectangles and the like from the stored building models for the determined possible building areas.

7. The method of claim 6 wherein the mapping step comprises for each possible building area where the camera could be mounted establishing correspondence between features extracted from the camera image data and features extracted from the stored building models.

8. The method of claim 7 further comprising computing camera parameters based on correspondence between features extracted from the camera image data and features extracted from the stored building models.

9. The method of claim 7 further comprising determining image distance between projected points and image points for the corresponding features extracted from the camera image data and features extracted from the stored building models.

10. The method of claim 1 wherein similar possible building areas include unique elements and the unique elements are used to distinguish between the similar possible building areas.

11. A system for automatic configuration of cameras in a Building Information Model (BIM), comprising:
a programmed BIM processing system including a processor and associated memory for managing building data;
a plurality of input ports operatively connected to the BIM processing system, the input ports being distributed at select locations in the building;
a database operatively associated with the BIM processing system storing building models and storing a network location map identifying locations of the input ports; and
operating the BIM processing system to detect a camera connected to one of the input ports and reading camera image data from the camera; determine possible building areas in the camera's field of view based on location of the one of the input ports and extracting features from the stored building models for the determined possible building areas, and establishing mapping between the camera image data and the extracted features from the stored building models by comparing features in the camera image data to the extracted features from stored building models for the determined possible building areas to determine actual location of the camera in the building and store camera location data in the database.

12. The system of claim 11 wherein detecting a camera comprises detecting a video camera and reading camera image date from the video camera comprises selecting an image frame.

13. The system of claim 12 wherein selecting the image frame comprises selecting an image frame when no person is in the field of view.

14. The system of claim 11 wherein each input port includes an associated unique identifier stored in the network location map and the possible building areas are determined based on the unique identifier of the one of the input ports.

15. The system of claim 11 wherein establishing mapping between the camera image data and the stored building models comprises extracting features selected from edges, corners, rectangles and the like from the camera image data.

16. The system of claim 15 wherein establishing mapping between the camera image data and the stored building models comprises extracting features selected from edges, corners, rectangles and the like from the stored building models for the determined possible building areas.

17. The system of claim 16 wherein the mapping step comprises for each possible building area where the camera could be mounted establishing correspondence between features extracted from the camera image data and features extracted from the stored building models.

18. The system of claim 17 further comprising computing camera parameters based on the correspondence between features extracted from the camera image data and features extracted from the stored building models.

19. The system of claim 17 further comprising determining image distance between projected points and image points for the corresponding features extracted from the camera image data and features extracted from the stored building models.

20. The system of claim 11 wherein similar possible building areas include unique elements and the unique elements are used to distinguish between the similar possible building areas.

Assignees

• Lee Mi Suen
• Popowski Paul
• Honeywell Int Inc

Inventors

• Lee Mi Suen
• Popowski Paul

Classifications

• G08B13/19645

  Multiple cameras, each having view on one of a plurality of scenes, e.g. multiple cameras for multi-room surveillance or for tracking an object by view hand-over · CPC title

• H04N7/181 Primary

  for receiving images from a plurality of remote sources · CPC title

• G06T2207/10004

  Still image; Photographic image · CPC title

• G06T7/75

  involving models · CPC title

• G06T2207/30244

  Camera pose · CPC title