Auxilliary Navigation System For Use In Urban Areas

• US 20070286459A1
• Filed: 10/16/2005
• Published: 12/13/2007
• Est. Priority Date: 10/18/2004
• Status: Active Grant

• Current Assignee: Mistral Detection Ltd.
• Original Assignee: O.D.F. Optronics Ltd.
• Inventors: Gal, Ehud, Greenhut, Yaron
• Granted Patent: US 7,860,269 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle

Abstract

The invention is a method for the extraction, identification and comparison of skylines from imagery data specifically, for the purpose of accurately locating the position of a mobile platform for navigation purposes. The invention is carried out by processing images obtained using a compact mobile electro-optic device that is designed to be mounted on any mobile platform, such that a clear circumferential view of the surroundings is available. The method allows the position of the mobile platform to be determined in three dimensional space, relative to any arbitrary coordinate system; the exact map coordinates of the platform to be determined; or allows the platform to follow a previously traversed track.

Claims

1. A method for the extraction, identification and comparison of skylines from imagery data for the purpose of accurately locating the position of a mobile platform for navigation purposes, said method comprised of:

a. providing an image sensor capable of acquiring an image of a circumferential field of view;

b. providing a processing means capable of processing the image acquired by the image sensor;

c. obtaining one or more images of said skyline with said image sensor and transferring said images to said processing means;

d. providing an algorithm implemented by the processing means, said algorithm including the following steps;

i. identification and extraction of a first set of one or more skylines from a first set of one or more circumferential images obtained by said image sensor;

ii. modeling and representation of said skylines;

iii. assigning coordinates to the skylines in said first set, wherein the coordinates represent the location from which each image was taken;

iv. storing said first set of skylines together with the coordinates assigned to it in a database;

v. identification and extraction of a second skyline from a second circumferential image obtained by said image sensor;

vi. modeling and representation of said second skyline;

vii. comparing said second modeled skyline to the skylines of said first set of modeled skylines stored in said database to find a match between said second skyline and at least one of said skylines from said first set of skylines stored in said database; and

viii. assigning said coordinates assigned to said first skyline to said second skyline, after a match is found.

2. A method according to claim 1, wherein step i comprises performing Canny edge detection, wherein the gradient image produced during the operator operation is enhanced using the gradient image histogram to increase the dynamic range of the gradient image.

3. A method according to claim 2, wherein step i includes performing an opening operator on the Canny edge detection operator'"'"'s edge image.

4. A method according to claim 1, wherein the first skylines and the second skyline are found by using a tracking operation along the sky edges found in the image.

5. A method according to claim 1, wherein modeling of the first skylines and the second skyline is performed by building a Curvature State Spaces graph and location and registration of prominent points in said graph.

6. A method according to claim 1, wherein the first skylines stored in the database and the second skyline, are obtained by the same image sensor on two different occasions.

7. A method according to claim 1, wherein the first skylines stored in the database and the second skyline are obtained by two different circumferential image sensors, of essentially the same kind.

8. A method according to claim 6, wherein the location of the image sensor when it obtained the image of the second skyline is found based on the known location of the image sensor when it obtained the matching image of the first skyline.

9. A method according to claim 7, wherein the location of the image sensor which obtained the image of the second skyline is found based on the known location of the image sensor which obtained the matching image of the first skyline.

10. A method according to claim 7, wherein a second mobile platform having an image sensor which obtains an image of the second skyline follows a first mobile platform having an image sensor which obtains an image of the first skyline, the first mobile platform transmits skyline data to the second mobile platform, said second mobile platform performs image processing to find a match between said first skyline and said second skyline, and said second mobile platform adjusts its travel route in order to obtain a match between said first skyline and said second skyline.

11. A method according to claim 1, wherein the relative azimuth of the locations from which two different images were taken is calculated from at least two prominent points that recur in both of said skylines and are extracted from both of said images.

12. A method according to claim 1, wherein the image sensor comprises a 180°

“

Fish Eye”

lens.

13. A method according to claim 1, wherein the image sensor comprises a reflective lens covering a circumferential field of view.

14. A method according to claim 1, wherein the image sensor comprises a refractive lens covering a circumferential field of view.

15. A method according to claim 1, comprising the extraction and identification of infrastructure and natural terrain outlines and any other special features in the image for the purpose of orientation and navigation.

16. A method according to claim 1, comprising the extraction and identification of celestial bodies in the image together with information of date and time for the purpose of orientation and navigation.

17. A method according to claim 1, comprising the extraction and identification of road boundary lines if such exists in the image in order to calculate the vehicle position and orientation relative to said road.