Method for identification of contamination upon a lens of a stereoscopic camera

1. A method for identifying contamination upon a lens of a stereoscopic camera, wherein said stereoscopic camera

is arranged such that a capturing area of said stereoscopic camera is predefined such that images from said stereoscopic camera have the same capturing area over time,

is provided with a first camera adapted to cover said capturing area by providing first images of said capturing area, and

is provided with a second camera adapted to cover said capturing area by providing second images of said capturing area, wherein

said first images are divided into at least one evaluation area and said second images are divided into at least one evaluation area, wherein the respective evaluation area of said first and said second images are correspondently located in respective image, wherein said method comprises the steps of:

forming historical image data for said evaluation areas, wherein said historical image data comprises an image parameter representing the respective evaluation area from a predetermined number of previously captured first and second images,

comparing said historical image data for the evaluation area of said first image with historical image data for the evaluation area of said second image, and

indicating that at least one lens of said stereoscopic camera is contaminated, if a deviation larger than a threshold value between the compared historical image data is identified.

2. The method according to claim 1 , wherein the method further comprises the steps of:

identifying a minimum value (Bmin1, Bmin2) and a maximum value (Bmax1, Bmax2) of said image parameter from said historical image data for each evaluation area,

calculating a first difference value between said minimum and said maximum value (Bmin1, Bmin2; Bmax1, Bmax2) for each evaluation area,

comparing said first difference value from the evaluation area of said first images with said first difference value from the evaluation area of said second images, and

identifying the evaluation area of said first or second images associated with the lowest first difference value as obstructed by contamination on the lens.

3. The method according to claim 2 , wherein when a new first and second image has been captured by said stereoscopic camera, the method further comprises the steps of:

calculating an average value of said parameter of said historical image data for the respective evaluation areas of said first and second images,

calculating a second difference value between said average values, and

adding said second difference value to the parameter value from the evaluation area of a newly taken image in which the evaluation area is identified as obstructed.

4. The method according to claim 1 , wherein said historical image data is represented by an average value of said parameter.

5. The method according to claim 1 , wherein said historical image data is represented by a histogram of said parameter.

6. The method according to claim 5 , wherein the histogram has a separate class for each possible value of said parameter, or the histogram has classes for bundles of values of said parameter.

7. The method according to claim 5 , wherein a normal value (n0, nc) is defined as the most frequent image parameter value for the evaluation area of respective first and second images, wherein the method further comprises the step of:

adjusting the histogram of the evaluation area identified as obstructed such, that its shape and position corresponds to the histogram of its corresponding evaluation area.

8. The method according to claim 1 , wherein said parameter is selected among the following parameters; brightness, colour channel, contrast or any other image parameter.

9. The method according to claim 1 , wherein said first and second images are divided into a plurality of correspondent evaluation areas.

10. The method according to claim 9 , wherein each evaluation area is defined as an individual pixel in the respective first and second image.

11. The method according to claim 1 , wherein the historical image data is collected from a predetermined number of previously captured images.

12. The method according to claim 1 , wherein the historical image data is continuously updated.

13. The method according to claim 12 , wherein the historical image data is updated at predetermined time intervals by replacing the image parameter value from the oldest previously captured image by the corresponding image parameter value from a newly captured image.

14. The method according to claim 1 , wherein the method further comprises the step of generating a warning message when at least one lens of said stereoscopic camera is identified to have reached a predetermined level of contamination.

15. A traffic surveillance facility, comprising:

a stereoscopic camera, wherein said stereoscopic camera

is arranged such that a capturing area of said stereoscopic camera is predefined such that images from said stereoscopic camera have the same capturing area over time,

is provided with a first camera adapted to cover said capturing area by providing first images of said capturing area, and

is provided with a second camera adapted to cover said capturing area by providing second images of said capturing area, wherein

said first images are divided into at least one evaluation area and said second images are divided into at least one evaluation area, wherein the respective evaluation area of said first and said second images are correspondently located in respective image; and

an electronic control unit provided to:

control said stereoscopic camera; and

perform an identification of contamination upon a lens of the stereoscopic camera, the electronic control unit configured to:

form historical image data for said evaluation areas, wherein said historical image data comprises an image parameter representing the respective evaluation area from a predetermined number of previously captured first and second images,

compare said historical image data for the evaluation area of said first image with historical image data for the evaluation area of said second image, and

indicate that at least one lens of said stereoscopic camera is contaminated, if a deviation larger than a threshold value between the compared historical image data is identified.