Image shadow detection using multiple images

1. A computer program product, the computer program product being tangibly embodied on a non-transitory computer-readable storage medium and comprising instructions that, when executed by at least one computing device, are configured to cause the at least one computing device to:

obtain a first image of an object and a second image of the object, captured by a camera;

obtain a first intensity map of the first image and a second intensity map of the second image, each cell of the first intensity map generated from a corresponding cell of the first image and at least one other adjacent cell of the first image, each cell of the second intensity map generated from a corresponding cell of the second image and at least one other adjacent cell of the second image;

compare intensity variations between the first intensity map of the first image and at least one of the second intensity map of the second image and a combination intensity map obtained from at least the first intensity map and the second intensity map; and

identify a shadow within the first image, based on the intensity variations.

2. The computer program product of claim 1 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

divide the first intensity map into a first portion intensity map;

divide the second intensity map into a second portion intensity map; and

generate the combination intensity map including combining each portion of the first intensity map with each corresponding portion of the second intensity map.

3. The computer program product of claim 2 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

generate, for each portion of the first portion intensity map, a corresponding first local mean intensity;

generate, for each portion of the second portion intensity map, a corresponding second local mean intensity; and

generate the combination intensity map including generating a global mean intensity for each portion of the combination intensity map, using the corresponding first mean local intensity and the corresponding second local mean intensity.

4. The computer program product of claim 2 , wherein the instructions, when executed to compare the intensity variations, are further configured to cause the at least one computing device to:

generate a first variation map with portions corresponding to the portions of the first portion intensity map, and further corresponding to portions of the combination intensity map,

wherein each portion of the first variation map includes a relative difference between the corresponding portions of the combination intensity map and the first portion intensity map.

5. The computer program product of claim 4 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

determine, for each portion of the first variation map, whether the relative difference therein is outside of a threshold range; and

determine that each portion of the first variation map outside of the threshold range is a candidate portion for representing a corresponding shadow portion of the shadow.

6. The computer program product of claim 5 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

determine whether each candidate portion of the first variation map is adjacent to a minimum number of additional candidate portions; and

designate each minimally-adjacent candidate portion as the corresponding shadow portion of the shadow.

7. The computer program product of claim 5 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

generate a second variation map with portions corresponding to the portions of the second portion intensity map, and further corresponding to portions of the combination intensity map,

wherein each portion of the second variation map includes a relative difference between the corresponding portions of the combination intensity map and the second portion intensity map.

8. The computer program product of claim 7 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

determine, for each portion of the second variation map, whether the relative difference therein is outside of the threshold range;

determine that each portion of the second variation map outside of the threshold range is a second candidate portion for representing the corresponding shadow portion of the shadow; and

determine, based on each candidate portion and each second candidate portion, whether each corresponding portion of the first image should be designated as the corresponding shadow portion of the shadow.

9. The computer program product of claim 1 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

activate a light source of the camera, based on the identifying of the shadow.

10. The computer program product of claim 1 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

obtain the first image and the second image during a framing period that follows activation of a viewing frame of the camera, during which the camera is positioned relative to the object within the viewing frame, prior to an initiation of image capture being received at the camera.

11. A computer program product, the computer program product being tangibly embodied on a non-transitory computer-readable storage medium and comprising instructions that, when executed by at least one computing device, are configured to cause the at least one computing device to:

generate a first intensity map of a first image from local mean intensity values of the first image;

generate a second intensity map of a second image from local mean intensity values of the second image;

divide the first intensity map for the first image into a first portion intensity map;

divide the second intensity map for the second image into a second portion intensity map;

generate a combined intensity map, based on the first portion intensity map and the second portion intensity map;

generate at least one variation map based on a comparison of the combined intensity map with at least one of the first intensity map portions and the second intensity map portions; and

identify a shadow within at least one of the first image and the second image, based on the at least one variation map.

12. The computer program product of claim 11 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

divide the first intensity map into the first portion intensity map, each portion of the first portion intensity map including a cell of a first matrix of cells;

divide the second intensity map into the second portion intensity map, each portion of the second portion intensity map including a cell of a second matrix of cells that is the same size and dimension as the first matrix of cells; and

generate the combined intensity map including a third matrix of cells in which each cell of the third matrix of cells is a combination of a corresponding cell of the first matrix and the second matrix.

13. The computer program product of claim 12 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

generate the variation map as including a fourth matrix of cells, in which each cell represents a relative difference between the corresponding cells of the third matrix and at least one of the first matrix and the second matrix.

14. The computer program product of claim 11 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

obtain the first image and the second image from a camera, while a user positions the camera relative to an object.

15. The computer program product of claim 11 , wherein the instructions, when executed, are further configured to cause the at least one computing device to:

obtain the first image and the second image during a framing period that follows activation of a viewing frame of the camera by the user, during which the user positions and orients the camera relative to the object within the viewing frame, prior to an initiation of image capture being received from the user at the camera.

16. In a digital medium environment for obtaining digital images, a computer-implemented method of shadow detection comprising:

a step for comparing corresponding portions of a plurality of images obtained from a camera during a framing period that occurs after an object is framed within a view frame of the camera and before an image capture signal is received at the camera;

wherein the comparing generates an intensity map for each of the images, combines the intensity maps into a combined intensity map, generates a variation map including relative differences between the combined intensity map and one of the intensity maps; and

identifying a shadow within at least one of the images from the relative differences that are outside a threshold range.

17. The method of claim 16 , wherein the step for comparing comprises:

dividing a first intensity map of the first image into a first portion intensity map;

dividing a second intensity map of the second image into a second portion intensity map; and

generating the combined intensity map including combining each portion of the first intensity map with each corresponding portion of the second intensity map.

18. The method of claim 17 , wherein the step for comparing comprises:

generating a first variation map with portions corresponding to the portions of the first portion intensity map, and further corresponding to portions of the combined intensity map, wherein each portion of the first variation map includes a relative difference between the corresponding portions of the combined intensity map and the first portion intensity map.

19. The method of claim 18 , comprising:

determining, for each portion of the first variation map, whether the relative difference therein is outside of the threshold range; and

determining that each portion of the first variation map outside of the threshold range is a candidate portion for representing a corresponding shadow portion of the shadow.

20. The method of claim 18 , comprising:

determining, for each portion of the first variation map, whether the relative difference therein is outside of a threshold range for a border detection error; and

determining that each portion of the first variation map outside of the threshold range for the border detection error represents a border detection error in the first image.