Adaptive sampling region
View Patent ↗Properties of pixels in a digital image are sampled within different subdivisions of an editing tool impression to produce different property distributions. The different subdivisions can automatically alter their size, geometry, and/or location, based on image content within one or more of the subdivisions, in order to encompass a set of pixels having a substantially uniform distribution of a pixel property. Uniformity can be defined relative to the editing operation or context of the image. The property distributions from each region are classified to identify different edit classes within the property space, which are then used to apply an edit effect to the digital image within the tool impression. The edit classes may be represented by an edit profile in two or more dimensions.
1. A method comprising:
defining a tool impression in a digital image, the tool impression including a sampling region encompassing a set of pixels;
iteratively modifying the sampling region within the tool impression to encompass other pixels within the tool impression, the iterative modifications to the sampling region being limited by a sampling criterion;
sampling pixels in the iteratively modified sampling region to determine a first distribution of a pixel property using a processing unit; and
sampling pixels in a second region within the tool impression to determine a second distribution of the pixel property using the processing unit.
2. The method of claim 1 , further comprising:
editing at least one pixel within the tool impression based on the first and second distributions.
3. The method of claim 1 , further comprising:
editing at least one pixel within the tool impression in accordance with an edit profile.
4. The method of claim 1 , wherein the iterative modifications to the sampling region are further limited by a limit of growth, further comprising:
if the sampling region satisfies the limit of growth, editing at least one pixel within the tool impression in accordance with an edit profile; and
if the sampling region fails to satisfy the sampling criterion, editing at least one pixel within the tool impression based on the first and second distributions.
5. The method of claim 1 , wherein the sampling criterion is a pixel property uniformity threshold.
6. The method of claim 1 , wherein the sampling criterion is a pixel property uniformity threshold, further comprising:
if a pixel property uniformity of the sampling region satisfies the pixel property uniformity threshold, editing at least one pixel within the tool impression in accordance with an edit profile; and
if the pixel property uniformity of the sampling region fails to satisfy the pixel property uniformity threshold, determining a first edit class and a second edit class based on the first and second distributions of the pixel property, and applying a first editing operation to pixels in the first edit class and a different editing operation to pixels in the second edit class.
7. The method of claim 1 , wherein the sampling region has an area and the iteratively modifying operation includes:
increasing the area of the sampling region by one or more pixels to generate the iteratively modified sampling region.
8. The method of claim 1 , wherein the sampling region has an area and the iteratively modifying operation includes:
increasing the area of the sampling region by one or more pixels without violating the sampling criterion to generate the iteratively modified sampling region.
9. A non-transitory tangible computer readable medium encoding a computer program for executing on a computer system a computer process, the computer process comprising:
defining a tool impression in a digital image, the tool impression including a sampling region encompassing a set of pixels;
iteratively modifying the sampling region within the tool impression to encompass other pixels within the tool impression, the iterative modifications to the sampling region being limited by a sampling criterion;
sampling pixels in the iteratively modified sampling region to determine a first distribution of a pixel property; and
sampling pixels in a second region within the tool impression to determine a second distribution of the pixel property.
10. The non-transitory tangible computer readable medium of claim 9 , wherein the computer process further comprises:
editing at least one pixel within the tool impression based on the first and second distributions.
11. The non-transitory tangible computer readable medium of claim 9 , wherein the computer process further comprises:
editing at least one pixel within the tool impression in accordance with an edit profile.
12. The non-transitory tangible computer readable medium of claim 9 , wherein the iterative modifications to the sampling region are further limited by a limit of growth and wherein the computer process further comprises:
if the sampling region satisfies the limit of growth, editing at least one pixel within the tool impression in accordance with an edit profile; and
if the sampling region fails to satisfy the sampling criterion, editing at least one pixel within the tool impression based on the first and second distributions.
13. The non-transitory tangible computer readable medium of claim 9 , wherein the sampling criterion is a pixel property uniformity threshold.
14. The non-transitory tangible computer readable medium of claim 9 , wherein the sampling criterion is a pixel property uniformity threshold and wherein the computer process further comprises:
if a pixel property uniformity of the sampling region satisfies the pixel property uniformity threshold, editing at least one pixel within the tool impression in accordance with an edit profile; and
if the pixel property uniformity of the sampling region fails to satisfy the pixel property uniformity threshold, determining a first edit class and a second edit class based on the first and second distributions of the pixel property, and applying a first editing operation to pixels in the first edit class and a different editing operation to pixels in the second edit class.
15. A system comprising:
an impression definition module in execution upon a microprocessor of the system adapted to define a tool impression in a digital image, the tool impression including a sampling region encompassing a set of pixels;
an adaptive sampling region module in execution upon the microprocessor of the system adapted to iteratively modify the sampling region within the tool impression to encompass other pixels within the tool impression, the iterative modifications to the sampling region being limited by a sampling criterion; and
a region sampler module adapted in execution upon the microprocessor of the system to sample pixels in the iteratively modified sampling region to determine a first distribution of a pixel property and sample pixels in a second region within the tool impression to determine a second distribution of the pixel property.
16. The system of claim 15 , further comprising:
an edit profile definition module in execution upon the microprocessor of the system adapted to edit at least one pixel within the tool impression based on the first and second distributions.
17. The system of claim 15 , further comprising:
an edit profile definition module in execution upon the microprocessor of the system adapted to edit at least one pixel within the tool impression in accordance with an edit profile.
18. The system of claim 15 , wherein the iterative modifications to the sampling region are further limited by a limit of growth, further comprising:
an edit profile definition module in execution upon the microprocessor of the system adapted to:
edit at least one pixel within the tool impression in accordance with an edit profile if the sampling region satisfies the limit of growth; and
edit at least one pixel within the tool impression based on the first and second distributions if the sampling region fails to satisfy the sampling criterion.
19. The system of claim 15 , wherein the sampling criterion is a pixel property uniformity threshold.
20. The system of claim 15 , wherein the sampling criterion is a pixel property uniformity threshold, further comprising:
an edit profile definition module in execution upon the microprocessor of the system adapted to:
edit at least one pixel within the tool impression in accordance with an edit profile, if a pixel property uniformity of the sampling region satisfies the pixel property uniformity threshold; and
determine a first edit class and a second edit class based on the first and second distributions of the pixel property and apply a first editing operation to pixels in the first edit class and a different editing operation to pixels in the second edit class, if the pixel property uniformity of the sampling region fails to satisfy the pixel property uniformity threshold.