Automatic asset detection, location measurement and recognition

• US 20070171431A1
• Filed: 01/19/2007
• Published: 07/26/2007
• Est. Priority Date: 01/20/2006
• Status: Active Grant

• Current Assignee: Geo-3D, Inc. (Trimble Inc.)
• Original Assignee: Geo-3D, Inc. (Trimble Inc.)
• Inventors: Laflamme, Claude
• Granted Patent: US 7,623,248 B2
• Time in Patent Office: Days
• US Class Current: 356/602
• CPC Class Codes: G01C 11/02 Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures
• CPC Class Codes: G01C 15/00 Surveying instruments or accessories not provided for in groups G01C1/00 - G01C13/00
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

From a survey vehicle, two types of data are obtained: images from a camera and reflections from a laser. The laser data is filtered in order to detect assets. The detected assets are then processed to locate and measure the assets. Once this information has been obtained, it is combined with the images captured by the camera in order to recognize the assets.

Claims

1. A method for asset inventory using a survey vehicle, the method comprising:

concurrently capturing images on a camera as said vehicle advances and assigning geographical locations to said images, and emitting a beam of light from a laser mounted on said vehicle and capturing laser data reflected off of various objects;

detecting an object of interest from said various objects by processing said laser data using at least one parameter unique to said object;

locating said object of interest geographically by aggregating proximal laser scanned points into a single object and identifying an (X,Y,Z) position thereof;

measuring said object of interest by approximating object size of said single object; and

recognizing said object of interest by locating said single object in one of said images captured by said camera.

2. A method as claimed in claim 1, wherein said recognizing said object comprises classifying said object of interest according to a specific object type.

3. A method as claimed in claim 2, wherein said recognizing said object comprises classifying said object of interest according to a specific object model of a given object type.

4. A method as claimed in claim 3, wherein said classifying comprises using a template-matching algorithm with a database of predefined object models.

5. A method as claimed in claim 1, wherein said processing said laser data comprises using at least one of retro-reflective properties, morphology, size, and height to detect said object of interest.

6. A method as claimed in claim 1, wherein said detecting comprises using at least two parameters unique to said object.

7. A method as claimed in claim 1, capturing images comprises obtaining data regarding a quality and accuracy of said geographical locations assigned to said images.

8. A method as claimed in claim 1, wherein said locating said object of interest geographically comprises determining a centroid thereof.

9. A method as claimed in claim 1, wherein said recognizing comprises using known orientation and field of view of the camera to identify an image comprising said object of interest, and translating said (X,Y,Z) coordinates into two-dimensional coordinates available on said image.

10. A survey vehicle comprising:

at least one two-dimensional laser scanning device;

reception means to capture laser data reflected off various objects;

at least one camera for capturing images as said vehicle advances;

a positioning system to attach geographical locations to said images; and

a data capture computer adapted to receive said laser data and said images.

11. A survey vehicle as claimed in claim 10, wherein said data capture computer is adapted to

detect an object of interest from said various objects by processing said laser data using at least one parameter unique to said object;

locate said object of interest geographically by aggregating proximal laser scanned points into a single object and identifying an (X,Y,Z) position thereof;

measure said object of interest by approximating object size of said single object; and

recognizing said object of interest by locating said single object in one of said images captured by said camera.

12. A survey vehicle as claimed in claim 10, wherein said data capture computer classifies said object of interest according to a specific object type.

13. A survey vehicle as claimed in claim 12, wherein said data capture computer classifies said object of interest according to a specific object model of a given object type.

14. A survey vehicle as claimed in claim 13, wherein said data capture computer uses a template-matching algorithm with a database of predefined object models to classify said object.

15. A survey vehicle as claimed in claim 10, wherein said laser scanning device is a laser having a near infrared laser beam which uses time-of-flight laser range measurement principles and fast line scanning means.

16. A survey vehicle as claimed in claim 10, wherein said camera is a digital signal-triggered camera.

17. A survey vehicle as claimed in claim 10, wherein said laser scanning device has at least one of intensity and color measurement features.

18. A survey vehicle as claimed in claim 10, wherein said positioning system has an orientation system.

19. A survey vehicle as claimed in claim 10, wherein said laser scanning device is positioned to scan a road-side laterally.

20. A survey vehicle as claimed in claim 10, wherein said laser data includes a measurement of a quality of a retro-reflection.