Method for processing wide angle images with barrel distortion and a surveillance system

• US 9,413,927 B2
• Filed: 08/12/2013
• Issued: 08/09/2016
• Est. Priority Date: 08/13/2012
• Status: Active Grant

• Current Assignee: Politechnika Poznanska
• Original Assignee: Politechnika Poznanska
• Inventors: Luczak, Adam, Mackowiak, Slawomir
• Primary Examiner(s): Teitelbaum, Michael
• Application Number:
  US13/964,114
• Publication Number:
  US 20140043477A1
• Time in Patent Office: 1,093 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10004 Still image; Photographic image
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/20104 Interactive definition of region of interest [ROI]
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 3/12 Panospheric to cylindrical image transformations
• CPC Class Codes: G06T 5/80 Geometric correction
• 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: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

A computer-implemented method for processing a distorted wide angle image with barrel distortion to obtain a plurality of reconstructed images, characterized in that it comprises the steps of: determining, on the basis of a mapping function representing the distortion of the distorted image, a reconstruction function representing the position of points (r₀) of reconstructed image as a function of position of points (r₁) on the distorted image, determining a plurality of viewing angles (α₁, α₂, α₃), for each viewing angle (α₁, α₂, α₃), generating a reconstructed image comprising points of the distorted image comprised within the particular viewing angle (α₁, α₂, α₃) and reconstructed using the reconstruction function, such that each reconstructed image has its central point corresponding to the same point of the distorted image.

Claims

1. A computer-implemented method for processing a distorted wide angle image with a barrel distortion, the method comprising the steps of:

determining, on a basis of a mapping function representing a distortion of said distorted image, a reconstruction function representing a position of points of a reconstructed image as a function of a position of points of said distorted image and a scale coefficient,

determining a plurality of different viewing angles,

for each viewing angle of the plurality of different viewing angles generating a reconstructed image comprising the points of said distorted image comprised within, said viewing angle and reconstructed using said reconstruction function with a different scale coefficient for each of said reconstructed images;

wherein each of said reconstructed images has its central point corresponding to the same point of said distorted image;

identifying coordinates of an object on one of said reconstructed images as a function of coordinates of that object on another one of said reconstructed images and a quotient of scale coefficients of the reconstruction functions used for generating that reconstructed images.

2. The method according to claim 1, wherein each of said reconstructed images has dimensions which are not greater than dimensions of said distorted image.

3. The method according to claim 1, wherein each of said reconstructed images has the same dimensions.

4. The method according to claim 1, further comprising the step of generating a display screen comprising at least two of said reconstructed images.

5. The method according to claim 1, further comprising the step of determining at least one viewing angle in response to an input received via a user interface.

6. The method according to claim 1, further comprising the step of identifying a graphical object on one of said reconstructed images and highlighting points corresponding to said graphical object on at least two of said reconstructed images.

7. A non-transitory computer readable medium comprising program code means for performing, when said program is run on a computer, processing a distorted wide angle image with a barrel distortion by the following steps:

determining, on a basis of a mapping function representing a distortion of said distorted image, a reconstruction function representing a position of points of a reconstructed image as a function of a position of points of said distorted image and a scale coefficient,

determining a plurality of different viewing angles,

for each viewing angle of the plurality of different viewing angles, generating a reconstructed image comprising the points of said distorted image comprised within said viewing angle and reconstructed using said reconstruction function with a different scale coefficient for each of said reconstructed images;

wherein each of said reconstructed images has its central point corresponding to the same point of said distorted image;

identifying coordinates of an object on one of said reconstructed images as a function of coordinates of that object on another one of said reconstructed images and a quotient of scale coefficients of the reconstruction functions used for generating that reconstructed images.

8. A surveillance system comprising a camera with a wide angle lens coupled with an image sensor providing a distorted wide angle image with a barrel distortion, the system further comprising an image processor configured to:

determine, on a basis of a mapping function representing a distortion of said distorted image, a reconstruction function representing a position of points of a reconstructed image as a function of a position of points of said distorted image and a scale coefficient,

determine a plurality of different viewing angles,

for each viewing angle of the plurality of different viewing angles generate a reconstructed image comprising the points of said distorted image comprised within said viewing angle and reconstructed using said reconstruction function with a different scale coefficient for each of said reconstructed images;

wherein each of said reconstructed images has its central point corresponding to the same point of said distorted image;

identify coordinates of an object on one of said reconstructed images as a function of coordinates of that object on another one of said reconstructed images and a quotient of scale coefficients of the reconstruction functions used for generating that reconstructed images.

9. The system according w claim 8, wherein each, of said reconstructed images has dimensions which are not greater than the dimensions of said distorted image.

10. The system according to claim 8, wherein each of said reconstructed images has the same dimensions.

11. The system according to claim 8, further comprising a display interface configured to generate a display screen comprising at least two of said reconstructed images.

12. The system according to claim 8, wherein said image processor is further configured to determine at least one viewing angle in response to an input received via a user interface.

13. The system according to claim 8, wherein said image processor is further configured to identify a graphical object on one of said reconstructed images and to highlight points corresponding to said graphical object on at least two of said reconstructed images.