Flicker mitigation via image signal processing

1. An image processing method comprising:

receiving from an image sensor a sequence of input frames representing a scene;

identifying a light source represented within the input frames, and tracking the identified light source between consecutive input frames;

identifying flickering of the light source by detecting frame to frame color characteristic variation of a region representing the light source; and

generating a sequence of output frames corresponding to the input frames, with the flickering removed by selectively modifying image data of pixels within the region.

2. The image processing method of claim 1 , wherein said detecting frame to frame color characteristic variation comprises identifying a frame to frame change in at least one color space volume location of the light source, wherein the change satisfies at least one predefined criterion.

3. The image processing method of claim 2 , wherein the at least one color space volume location comprises a volume location in Hue Saturation Value (HSV) color space.

4. The image processing method of claim 3 , wherein the HSV color space is defined with H, S and V channels, and the flickering of the light source is identified by identifying a change in density of a histogram of the value (V) channel that satisfies the at least one predefined criterion.

5. The image processing method of claim 1 , wherein said identifying a light source comprises:

detecting an object within each of the input frames; and

identifying the object as a light source by determining that pixels representing the object are brighter than other image pixels of the input frames through use of adaptive thresholding.

6. The image processing method of claim 1 , wherein said identifying a light source and tracking the identified light source comprises:

defining a bounding box surrounding a flickering light source suspect (FSS); and

tracking the light source between consecutive frames by tracking the bounding box between consecutive frames.

7. The image processing method of claim 6 , wherein the bounding box is defined as a rectangular box having a length longer than a longest distance between perimeter points of the FSS, aligned with a major axis of the FSS, by at least 10%, and having a width longer by at least 10% than a longest distance between perimeter points of the FSS in a direction orthogonal to the major axis.

8. The image processing method of claim 6 , wherein the bounding box is defined as a rectangular box having a length and a width, the length being longer than a longest distance between perimeter points of the FSS in an x direction of a pixel grid defining the input frames by at least 10%, and the width being longer than a longest distance between perimeter points of the FSS in a y direction of the pixel grid by at least 10%.

9. The image processing method of claim 6 , further comprising segmenting components of the bounding box.

10. The image processing method of claim 9 , further co p sing:

tracking the segmented components of the bounding box between consecutive input frames.

11. An apparatus comprising:

a camera including:

an image sensor configured to capture images of a scene and provide input frames representing the same;

an image signal processor configured to execute instructions to:

receive from the image sensor a sequence of the input frames;

identify a light source represented within the input frames, and track the identified light source between consecutive input frames;

identify flickering of the light source by detecting frame to frame color characteristic variation of a region representing the light source; and

generate a sequence of output frames corresponding to the input frames, with the flickering removed by selectively modifying image data of pixels within the region.

12. The apparatus of claim 11 , further comprising a display and a response system for generating a response to the sequence of output frames.

13. The apparatus of claim 12 , wherein the apparatus is a vehicle.

14. The apparatus of claim 11 , wherein said identifying a light source and tracking the identified light source comprises:

defining a bounding box surrounding a flickering light source suspect (FSS); and

tracking the light source between consecutive frames by tracking the bounding box between consecutive frames.

15. The apparatus of claim 14 , wherein the bounding box is defined as a rectangular box having a length and a width, the length being longer than a longest distance between perimeter points of the FSS in an x direction of a pixel grid defining the input frames by at least 10%, and the width being longer than a longest distance between perimeter points of the FSS in a y direction of the pixel grid by at least 10%.