Method of and arrangement for blurring an image

• US 8,571,354 B2
• Filed: 09/25/2008
• Issued: 10/29/2013
• Est. Priority Date: 09/25/2008
• Status: Active Grant

• Current Assignee: Tomtom Global Content B.V (TomTom NV)
• Original Assignee: Tomtom Global Content B.V (TomTom NV)
• Inventors: Miksa, Krzysztof, Kmiecik, Marcin Michal
• Primary Examiner(s): Patel, Jayesh A
• Application Number:
  US12/736,915
• Publication Number:
  US 20110123068A1
• Time in Patent Office: 1,860 Days
• Field of Search: 382103-107, 382/118, 382/154, 382260-264, 382283-285, 382/291, 382/270, 382/299
• US Class Current: 382/299
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing

Abstract

A method of reducing the spatial resolution of images is disclosed. At least one embodiment of the method includes: —acquiring an input image including image parts having a spatial resolution larger than SR pixels/meter; —acquiring a depth map associated with the input image; —determining for each pixel p(x,y) a spatial resolution value by means of the depth map; —processing a region of pixels of the input image for which holds that the spatial resolution value is larger than a predefined threshold corresponding to SR pixels/meter to obtain a corresponding region of pixels having a spatial resolution smaller then or equal to SR pixels/meter in an output image. The method enables to removes privacy information from images by reducing the spatial resolution to a level that the privacy information cannot be recognized in the image anymore.

Claims

1. A method of reducing the spatial resolution of images comprising:

acquiring an input image comprising image parts having a spatial resolution larger than a predefined threshold SR pixels/meter, where the spatial resolution is associated with a clarity of an image;

acquiring a depth map associated with said input image;

determining for each pixel p(x,y) a spatial resolution value by means of the depth map; and

processing a region of pixels of said input image for which holds that the spatial resolution value is larger than the predefined threshold SR pixels/meter corresponding to an object in the input image to be privacy protected to obtain a corresponding region of pixels having a spatial resolution value less than 100 pixels/meter to protect privacy information related to human faces and less than 50 pixels/meter to protect privacy information related to license plates in an output image.

2. The method according to claim 1, wherein the processing a region of pixels for which holds that the spatial resolution value is larger than the predefined threshold SR pixels/meter comprises:

determining a transformation function having a transformation window for pixel p(x,y) from the spatial resolution values associated with the region of pixels; and

performing the transformation function for pixel p(x,y) to obtain a color value for pixel p(x,y) in the output image.

3. The method according to claim 2, wherein the transformation function obtains the color value for pixel p(x,y) by calculating an average color value c from pixels of the input image associated with the transformation window.

4. The method according to claim 2, wherein the centre pixel of the transformation window coincides the pixel p(x,y) in the input image to determine the color value for pixel p(x,y) in the output image.

5. The method according to claim 1, wherein the processing a region of pixels for which holds that the spatial resolution value is larger than the predefined threshold SR pixels/meter comprises:

determining first regions of pixels having a spatial resolution value falling in a first predetermined range; and

performing an image processing operation on said first regions of pixels to obtain pixels values of the output image, wherein the image processing operation corresponds to subsequently down-sampling and up-sampling the first regions of pixels having said spatial resolution with a factor associated with said first predetermined range.

6. The method according to claim 1, wherein the method further comprises determining regions of interest which do not comprise privacy information, and wherein the processing action is adapted to exclude the pixels associated with said regions of interest from processing.

7. The method according to claim 1, wherein the method further comprises:

determining a height of a group of pixels above the earth surface; and

wherein the processing a group of pixels is arranged to reduce the spatial resolution of the group of pixels in dependence of the determined height.

8. A computer implemented system for reducing the spatial resolution of images, the system comprising:

a processor; and

memory connected to the processor, the memory comprising a computer program comprising data and instructions arranged, when executed, to allow said processor to;

acquire an input image comprising image parts having a spatial resolution larger than a predefined threshold SR pixels/meter, wherein the spatial resolution is associated with a clarity of an image;

acquire a depth map associated with said input image;

determine for each pixel p(x,y) a spatial resolution value by means of the depth map; and

process a region of pixels of said input image for which holds that the spatial resolution value is larger than the predefined threshold SR pixels/meter corresponding to an object in the input image to be privacy protected to obtain a corresponding region of pixels having a spatial resolution value less than 100 pixels/meter to protect privacy information related to human faces and less than 50 pixels/meter to protect privacy information related to license plates in an output image.

9. A computer program product for use in a data processing system, for reducing the spatial resolution of images, the computer program product comprising a non-transitory computer readable medium storing instructions, which when executed, allows a processor to:

acquire an input image comprising image parts having a spatial resolution larger than a predefined threshold SR pixels/meter, wherein the spatial resolution is associated with a clarity of an image;

acquire a depth map associated with said input image;

determine for each pixel p(x,y) a spatial resolution value by means of the depth map; and

process a region of pixels of said input image for which holds that the spatial resolution value is larger than the predefined threshold SR pixels/meter corresponding to an object in the input image to be privacy protected to obtain a corresponding region of pixels having a spatial resolution value less than 100 pixels/meter to protect privacy information related to human faces and less than 50 pixels/meter to protect privacy information related to license plates in an output image.