Digital image color space channel blending

1. A method, comprising:

capturing digital images that encompass a three-hundred and sixty degree (360°) field of view;

combining the digital images along a seam between the digital images to form a blended image, the combining comprising:

determining average values of channel gains in a color space for regions-of-interest within an overlap region for each of the digital images, a region-of-interest encompassing a number of pixels wide by a set of pixels on each side of the seam that is between the digital images;

determining mismatched color as between the average values for each of the respective regions-of-interest of each of the digital images along the seam within the overlap region that overlaps two of the digital images along the seam; and

applying the channel gains in the color space to blend the digital images, the channel gains applied to enhance pixels of one of the digital images starting within the overlap region along the seam and blending into the one digital image.

2. The method as recited in claim 1 , wherein said determining the mismatched color as between the digital images along the seam comprises:

determining average values of the channel gains in the color space for pixel rows within the overlap region for each of the digital images; and

determining the mismatched color as between the average values for each of the respective pixel rows of each of the digital images.

3. The method as recited in claim 2 , further comprising filtering the average values of the channel gains with a low-pass filter effective to minimize introducing artifacts into the blended image.

4. The method as recited in claim 2 , wherein the average values of the channel gains are determined in YCbCr color space of the digital images.

5. The method as recited in claim 1 , wherein said applying the channel gains in the color space to blend the digital images comprises tapering the blending from a difference value starting along the seam and diminishing the difference value as the blending progresses into the one digital image.

6. The method as recited in claim 5 , wherein said tapering the blending into the one digital image is adaptive based on a change in image content as the blending progresses into the one digital image.

7. The method as recited in claim 1 , wherein the overlap region that overlaps two of the digital images along the seam corresponds to overlapping image content that is captured in both of the digital images.

8. The method as recited in claim 1 , wherein the overlap region is a same number of pixels wide on each side of the seam that is between the digital images.

9. The method as recited in claim 1 , wherein the regions-of-interest horizontally subdivide the overlap region.

10. A mobile device, comprising:

a camera device to capture digital images that encompass a three-hundred and sixty degree (360°) field of view;

an image blending module implemented to combine the digital images along a seam between the digital images to form a blended image, the image blending module configured to:

determine average values of channel gains in a color space for regions-of-interest within an overlap region for each of the digital images, a region-of-interest encompassing a number of pixels wide by a set of pixels on each side of the seam that is between the digital images;

determine mismatched color as between the average values for each of the respective regions-of-interest of each of the digital images along the seam within the overlap region that overlaps two of the digital images along the seam; and

blend the digital images by the channel gains in the color space applied to enhance pixels of one of the digital images starting within the overlap region along the seam and blending into the one digital image.

11. The mobile device as recited in claim 10 , wherein to determine the mismatched color as between the digital images along the seam, the image blending module is configured to:

determine average values of the channel gains in the color space for pixel rows within the overlap region for each of the digital images; and

determine the mismatched color as between the average values for each of the respective pixel rows of each of the digital images.

12. The mobile device as recited in claim 11 , wherein the image blending module is configured to low-pass filter the average values of the channel gains effective to minimize introducing artifacts into the blended image.

13. The mobile device as recited in claim 11 , wherein the image blending module is configured to determine the average values of the channel gains in YCbCr color space of the digital images.

14. The mobile device as recited in claim 10 , wherein to blend the digital images, the image blending module is configured to taper the blending from a difference value starting along the seam and diminish the difference value as the blending progresses into the one digital image.

15. The mobile device as recited in claim 14 , wherein the image blending module is configured to adaptively taper the blending into the one digital image based on a change in image content as the blending progresses into the one digital image.

16. The mobile device as recited in claim 10 , wherein the overlap region that overlaps two of the digital images along the seam corresponds to overlapping image content that is captured in both of the digital images.

17. The mobile device as recited in claim 10 , wherein the overlap region is a same number of pixels wide on each side of the seam that is between the digital images.

18. An image color blending system, comprising:

memory to maintain digital images that are captured to encompass a three-hundred and sixty degree (360°) field of view;

a processor of a computing device to implement an image blending module that combines the digital images along a seam between the digital images to form a blended image, the image blending module configured to:

determine average values of channel gains in a color space for pixel rows within an overlap region for each of the digital images, the pixel rows in regions-of-interest encompassing a number of pixels wide by a set of pixels in the overlap region overlapping the digital images along the seam between the digital images;

determine mismatched color based on as between the average values of the channel gains in the color space for each of the respective pixel rows in the respective regions-of-interest of each of the digital images along the seam within the overlap region that overlaps the digital images;

filter the average values of the channel gains effective to minimize introducing artifacts into the blended image; and

blend the digital images by the filtered average values of the channel gains applied to enhance pixels of one of the digital images.

19. The image color blending system as recited in claim 18 , wherein to enhance the pixels of the one digital image, the image blending module is configured to apply the filtered average values of the channel gains starting within the overlap region along the seam and blend into the one digital image.

20. The image color blending system as recited in claim 18 , wherein the computing device is a mobile phone that includes the memory to maintain the digital images and includes the processor to implement the image blending module, and the mobile phone including a 360-degree field of view camera implemented to capture the digital images.