Systems and methods for dewarping images

• US 9,947,079 B1
• Filed: 02/14/2017
• Issued: 04/17/2018
• 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:
  US15/432,770
• Time in Patent Office: 427 Days
• 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
• CPC Class Codes: H05K 999/99 dummy group

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 comprising:

extracting, by a processor, a setting from a file associated with a wide-angle camera;

determining, by the processor, based on the setting, a first set of coordinates of a pixel of a first image, wherein the determining of the first set of coordinates is 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 a second image, wherein the first image is captured via the wide-angle camera;

determining, by the processor, via interpolating the first set of coordinates, a second set of coordinates of a pixel of the first image such that the first image would be less rectilinear than the second image when displayed, wherein the determining of the second set of coordinates is 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 second image; and

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

2. The method of claim 1, wherein the file includes a position identifier of the wide-angle camera informative of a wide-angle 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 file 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 file 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 file 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 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; and

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 first image.

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

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

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

8. The method of claim 1, wherein the determining and the interpolating is local to the wide-angle camera.

9. The method of claim 1, wherein the determining and the interpolating is remote to the wide-angle 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 method comprising:

extracting, by a processor, a setting from a file associated with a probe camera;

determining, by the processor, based on the setting, a first set of coordinates of a pixel of a first image, wherein the determining of the first set of coordinates is 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 a second image, wherein the first image is captured via the probe camera;

determining, by the processor, via interpolating the first set of coordinates, a second set of coordinates of a pixel of the first image such that the first image would be less rectilinear than the second image when displayed, wherein the determining of the second set of coordinates is 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 second image; and

instructing, by the processor, based on the interpolating, an output device to output the second image, wherein the first image is less rectilinear than the second image.

13. A method comprising:

receiving, by a processor, a first image from a pan-tilt-zoom (PTZ) camera, wherein the first image captures a plurality of zones;

for each of the zones depicted in the first image;

extracting, by the processor, a setting from a file associated with the PTZ camera;

determining, by the processor, a first set of coordinates of a pixel of the first image, wherein the determining of the first set of coordinates is 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 a second image;

determining, by the processor, via interpolating the first set of coordinates, a second set of coordinates of a pixel of the first image such that the first image would be less rectilinear than the second image when displayed, wherein the determining of the second set of coordinates is 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 second image; and

instructing, by the processor, based on the interpolating, an output device to output the second image, wherein the first image is less rectilinear than the second image.

14. A device comprising:

a vehicle including a processor, a memory, and a camera, wherein the processor is in communication with the memory and the camera, wherein the memory stores a set of instructions executable via the processor to perform a method, wherein the method comprises;

instructing, by the processor, the camera to capture a first image;

extracting, by the processor, a setting from a file associated with the camera;

determining, by the processor, based on the setting, a first set of coordinates of a pixel of the first image, wherein the determining of the first set of coordinates is 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 a second image;

determining, by the processor, via interpolating the first set of coordinates, a second set of coordinates of a pixel of the first image such that the first image would be less rectilinear than the second image when displayed, wherein the determining is 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 second image; and

instructing, by the processor, based on the interpolating, an output device to output the second image, wherein the first image is less rectilinear than the second image.