Image processing method for autonomous driving and apparatus thereof

1. A processor implemented image processing method, comprising:

recognizing a road in a first frame of an input image;

setting a region on the road as a recognition region;

scanning the recognition region instead of scanning an entire region of the input image;

recognizing a target object in the recognition region in the first frame, wherein the target object is a light of an approaching vehicle in an opposite lane during nighttime;

adjusting an exposure of a second frame of the input image containing the approaching vehicle; and

generating a synthesized image by synthesizing the first frame and the second frame, thereby allowing safe autonomous driving using the synthesized image with respect to the approaching vehicle in the opposite lane during the nighttime,

wherein the second frame is a next frame following the first frame.

2. The method of claim 1 , wherein the adjusting the exposure of the second frame of the input image is based on a difference between a brightness of the target object in the first frame and a second brightness of a remaining region not including the target object in the first frame.

3. The method of claim 2 , wherein the adjusting the exposure of the second frame of the input image comprises comparing the difference with a predetermined threshold.

4. The method of claim 1 , wherein the adjusting of the exposure of the second frame comprises:

decreasing the exposure of the second frame in response to a brightness of the target object in the first frame being greater than a second brightness of a remaining region not including the target object in the first frame; and

increasing the exposure of the second frame in response to the brightness of the target object being less than the brightness of the remaining region.

5. The method of claim 1 , further comprises continuously monitoring presence of the target object in the input image.

6. An image processing apparatus, comprising:

a processor configured to:

recognize a road in a first frame of an input image,

set a region on the road as a recognition region,

scan the recognition region instead of scanning an entire region of the input image,

recognize a target object in the recognition region in the first frame, wherein the target object is a light of an approaching vehicle in an opposite lane during nighttime,

adjust an exposure of a second frame of the input image containing the approaching vehicle, and

generate a synthesized image by synthesizing the first frame and the second frame, thereby allowing safe autonomous driving using the synthesized image with respect to the approaching vehicle in the opposite lane during the nighttime,

wherein the second frame is a next frame following the first frame.

7. The image processing apparatus of claim 6 , wherein the processor is configured to adjust the exposure of the second frame of the input image based on a brightness of the target object.

8. The image processing apparatus of claim 7 , wherein the processor is configured to adjust the exposure of the second frame of the input image based on a difference between the brightness of the target object in the first frame and a second brightness of a remaining region not including the target object in the first frame.

9. The image processing apparatus of claim 8 , wherein the processor is configured to compare the difference with a predetermined threshold.

10. The image processing apparatus of claim 7 , wherein the processor is configured to:

decrease the exposure of the second frame in response to the brightness of the target object being greater than a second brightness of a remaining region not including the target object in the first frame, and

increase the exposure of the second frame in response to the brightness of the target object being less than the brightness of the remaining region.

11. The image processing apparatus of claim 6 , wherein the processor is configured to continuously monitoring presence of the target object in the input image.

12. The method of claim 1 , wherein the scanning the recognition region instead of scanning the entire region of the input image further comprises avoiding scanning a portion of the input image other than the recognition region.

13. The image processing apparatus of claim 6 , wherein the processor is further configured to scan the recognition region instead of scanning the entire region of the input image by avoiding scanning a portion of the input image other than the recognition region.

14. The method of claim 1 , wherein the generating the synthesized image comprises synthesizing: i) a remaining region not including the target object in the first frame of the input image and ii) a region corresponding to the target object in the second frame of the input image.

15. A processor implemented image processing method, comprising:

recognizing a road in a first frame of an input image;

setting a region on the road as a recognition region;

scanning the recognition region instead of scanning an entire region of the input image;

recognizing a target object in a first region of the recognition region in the input image;

adjusting an exposure of a second frame of the input image based on a brightness difference between the target object and a remaining region not including the target object in the first frame; and

generating a synthesized image by synthesizing: i) the remaining region not including the target object in the first frame and ii) the first region corresponding to the target object in the second frame.

16. The method of claim 15 , wherein the adjusting the exposure of the second frame of the input image comprises comparing the brightness difference with a predetermined threshold.

17. The method of claim 15 , wherein the recognizing the target object comprises recognizing the target object based on position information of a vehicle.

18. The method of claim 15 , wherein the target object comprise any one or any combination of two or more of a tunnel entrance, a tunnel exit, and a light of a vehicle.