Systems and methods for dewarping images

• US 9,609,197 B1
• Filed: 08/19/2016
• Issued: 03/28/2017
• Est. Priority Date: 08/19/2016
• Status: Active Grant

• Current Assignee: Intelligent Security Systems (ISS)
• Original Assignee: Intelligent Security Systems (ISS)
• Inventors: Stepanenko, Oleg Vladimirovich
• Primary Examiner(s): Liew, Alex
• Application Number:
  US15/241,572
• Time in Patent Office: 221 Days
• US Class Current: 1/1
• CPC Class Codes: B60R 2300/30 characterised by the type of image processing
• CPC Class Codes: B60R 2300/80 characterised by the intended use of the viewing arrangement
• CPC Class Codes: G06T 1/0007 Image acquisition
• CPC Class Codes: G06T 3/4007 based on interpolation, e.g. bilinear interpolation image demosaicing G06T3/4015; edge-driven or edge-based scaling G06T3/403
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: H04N 23/661 Transmitting camera control signals through networks, e.g. control via the Internet
• CPC Class Codes: H04N 23/698 for achieving an enlarged field of view, e.g. panoramic image capture
• CPC Class Codes: H04N 23/81 for suppressing or minimising disturbance in the image signal generation

Abstract

A computer-implemented method comprises: extracting a setting from a description file of a virtual pan-tilt-zoom (PTZ) camera used to capture an original image through a wide-angle lens; determining a first set of coordinates of a pixel of the original image for each cell of a sparse conversion map represented as a first look-up table, wherein the sparse conversion map corresponds to a sparse grid of pixels of an output image; determining, via interpolating the first set of coordinates, a second set of coordinates of a pixel of the original image for each cell of a full conversion map, wherein the second set of coordinates is represented as a second look-up table, wherein the full conversion map corresponds to a full grid of pixels of the output image; instructing a display to present the output image, wherein the original image is less rectilinear than the output image.

Claims

1. A method of image conversion, the method comprising:

extracting, by a processor, a setting from a description file of a virtual pan-tilt-zoom (PTZ) camera used to capture an original image through a wide-angle lens;

determining, by the processor, based on the extracting, a first set of coordinates of a pixel of the original image for each cell of a sparse conversion map represented as a first look-up table, wherein the sparse conversion map corresponds to a sparse grid of pixels of an output image;

determining, by the processor, via interpolating the first set of coordinates, a second set of coordinates of a pixel of the original image for each cell of a full conversion map, wherein the second set of coordinates is represented as a second look-up table, wherein the full conversion map corresponds to a full grid of pixels of the output image;

instructing, by the processor, based on the interpolating, a display to present the output image, wherein the original image is less rectilinear than the output image.

2. The method of claim 1, wherein the description file of the virtual PTZ camera includes a position identifier of the virtual PTZ camera informative of a virtual PTZ camera position to be at least one of a ceiling position, a wall position, or a floor position, wherein the setting includes the position identifier.

3. The method of claim 1, wherein the description file of the virtual PTZ camera includes a minimum lens angle and a maximum lens angle, wherein the setting includes the minimum lens angle and the maximum lens angle.

4. The method of claim 1, wherein the description file of the virtual PTZ camera includes a resolution of the output image in pixels, wherein the setting includes the resolution.

5. The method of claim 1, further comprising:

determining, by the processor, a distance to an image plane based on a zoom value and a normal vector of the image plane based on a pan value and a tilt value, wherein the description file of the virtual PTZ camera contains the pan value and the tilt value, wherein the setting includes at least one of the pan value or the tilt value, wherein a description of the image plane in the description file is based on the determining the distance and the normal vector.

6. The method of claim 1, wherein the determining comprises:

converting, by the processor, a coordinate associated with a cell of the sparse conversion map into a mathematical representation on an image plane;

converting, by the processor, the mathematical representation into a set of 3d coordinates and then into a set of 2d coordinates, wherein the set of 3d coordinates and the set of 2d coordinates are from a coordinate system associated with the original image.

7. The method of claim 6, wherein the determining comprises:

converting, by the processor, the set of 2d coordinates associated with the original image into a set of pixel coordinates of the original image;

recording, by the processor, the set of pixel coordinates of the original image into the first look-up table.

8. The method of claim 1, wherein the determining and the interpolating is local to the virtual PTZ camera.

9. The method of claim 1, wherein the determining and the interpolating is remote to the virtual PTZ camera.

10. The method of claim 1, wherein the display includes a video wall.

11. The method of claim 1, wherein the interpolating comprises at least one of a neighbor interpolation, a bilinear interpolation, or a bi-cubic interpolation.

12. A system comprising:

a server storing a set of instructions instructive for a processor to perform an image conversion based at least in part on;

extracting, by the processor, a plurality of settings from a description file of a virtual pan-tilt-zoom (PTZ) camera used to capture an original image through a wide-angle lens;

determining, by the processor, based on the extracting, a first set of coordinates of a pixel of the original image for each cell of a sparse conversion map represented as a first look-up table, wherein the sparse conversion map corresponds to a sparse grid of pixels of an output image;

determining, by the processor, via interpolating the first set of coordinates, a second set of coordinates of a pixel of the original image for each cell of a full conversion map, wherein the second set of coordinates is represented as a second look-up table, wherein the full conversion map corresponds to a full grid of pixels of the output image;

instructing, by the processor, based on the interpolating, a display to present the output image, wherein the original image is less rectilinear than the output image.

13. The system of claim 12, wherein the description file of the virtual PTZ camera includes a position identifier of the virtual PTZ camera informative of a virtual PTZ camera position to be at least one of a ceiling position, a wall position, or a floor position, wherein the setting includes the position identifier.

14. The system of claim 12, wherein the description file of the virtual PTZ camera includes a minimum lens angle and a maximum lens angle, wherein the setting includes the minimum lens angle and the maximum lens angle.

15. The system of claim 12, wherein the description file of the virtual PTZ camera includes a resolution of the output image in pixels, wherein the setting includes the resolution.

16. The system of claim 12, wherein the image conversion is further based at least in part on:

determining, by the processor, a distance to an image plane based on a zoom value and a normal vector of the image plane based on a pan value and a tilt value, wherein the description file of the virtual PTZ camera contains the pan value and the tilt value, wherein the setting includes at least one of the pan value or the tilt value, wherein a description of the image plane in the description file is based on the determining the distance and the normal vector.

17. The system of claim 12, wherein the determining comprises:

converting, by the processor, a coordinate associated with a cell of the sparse conversion map into a mathematical representation on an image plane;

converting, by the processor, the mathematical representation into a set of 3d coordinates and then into a set of 2d coordinates, wherein the set of 3d coordinates and the set of 2d coordinates are from a coordinate system associated with the original image.

18. The system of claim 17, wherein the determining comprises:

converting, by the processor, the set of 2d coordinates associated with the original image into a set of pixel coordinates of the original image;

recording, by the processor, the set of pixel coordinates of the original image into the first look-up table.

19. The system of claim 12, wherein the determining and the interpolating is local to the virtual PTZ camera.

20. The system of claim 12, wherein the determining and the interpolating is remote to the virtual PTZ camera.