Systems and methods for image regularization based on a curve derived from the image data

1. A method comprising:

receiving an image comprising a plurality of original image data that is non-uniformly distributed across a space represented by the image;

obtaining a tree-based representation of the plurality of original image data, wherein the tree-based representation comprises a plurality of leaf nodes with each leaf node of the plurality of leaf nodes being associated with different original image data from the plurality of original image data, and a plurality of parent nodes with each parent node of the plurality of parent nodes being associated with decimated image data representing a different set of the plurality of original image data associated with a different set of the plurality of leaf nodes under that parent node with different amounts of decimation;

generating a regularization curve based on a distribution of the plurality of original image data;

applying the regularization curve to the tree-based representation; and

generating a regularized image from rendering the original image data associated with a subset of the plurality of leaf nodes and the decimated image data associated with a subset of the plurality of parent nodes that intersect or are within the regularization curve.

2. The method of claim 1 , wherein the regularized image comprises a more uniform distribution of image data across the space than the plurality of original image data from the image.

3. The method of claim 1 further comprising:

selecting the subset of parent nodes that intersect or are within the regularization curve from the tree-based representation; and

rendering the decimated image data associated with the subset of parent nodes as a substitute for the original image data associated with each leaf node under the subset of parent nodes in the tree-based representation.

4. The method of claim 1 further comprising:

determining the distribution of the plurality of original image data based on a density with which the plurality of original image data is distributed across the space.

5. The method of claim 4 , wherein generating the regularized image comprises:

decimating a first set of regions in the space comprising the original image data with a first density; and

preserving the original image data in a second set of regions in the space comprising the original image data with a second density that is less than the first density.

6. The method of claim 5 ,

wherein the subset of parent nodes comprises the decimated image data for the first set of regions, and

wherein the subset of leaf nodes comprises the original image data in the second set of regions.

7. The method of claim 1 further comprising:

determining the distribution of the plurality of original image data based on positional elements of the plurality of original image data, wherein the positional elements define positions for the plurality of original image data in the space.

8. The method of claim 1 further comprising:

determining the distribution of the plurality of original image data based on non-positional elements of the plurality of original image data, wherein the non-positional elements define visual characteristics for the plurality of original image data in the space.

9. The method of claim 1 , further comprising:

determining the distribution of the plurality of original image data based on a return intensity with which each of the original image data from the plurality of original image data is measured by a scanner.

10. The method of claim 9 , wherein generating the regularized image comprises:

filtering noise from the regularized image by removing the original image data associated with leaf nodes that do not intersect or are outside of the regularization curve.

11. The method of claim 1 , wherein the regularization curve is a model of the distribution of the plurality of original image data in the plurality of leaf nodes.

12. The method of claim 1 , wherein the regularization curve comprises:

one or more peaks corresponding to a first set of the plurality of original image data distributed in the space where the distribution has a maximum value, and

one or more troughs corresponding to a second set of the plurality of original image data distributed in the space where the distribution has a minimum value.

13. The method of claim 1 , wherein generating the regularized image comprises:

excluding the original image data associated with leaf nodes under the subset of parent nodes from the regularized image.

14. The method of claim 1 further comprising:

receiving a request to generate the regularized image with a particular amount of uniformity; and

wherein applying the regularization curve comprises:

aligning the regularization curve at a first layer of the tree-based representation for a first amount of uniformity in a rendering of the regularized image; and

aligning the regularization curve at a different second layer of the tree-based representation for a second amount of uniformity in the rendering of the regularized image.

15. The method of claim 1 , wherein applying the regularization curve comprises:

preserving a first amount of the plurality of original image data in the regularized image in response to applying the regularization curve at a first height in the tree-based representation; and

preserving a second amount of the plurality of original image data in the regularized image in response to applying the regularization curve at a different second height in the tree-based representation.

16. The method of claim 1 , wherein the decimated image data associated with a particular parent node replaces the original image data associated with all leaf node under that particular parent node.

17. The method of claim 1 , wherein generating the regularization curve comprises:

generating a curve or waveform that models the distribution across an arrangement of the plurality of leaf nodes in the tree-based representation.

18. The method of claim 1 further comprising:

presenting the regularized image on a display device.

19. An imaging system comprising:

one or more hardware processors configured to:

receive an image comprising a plurality of original image data that is non-uniformly distributed across a space represented by the image;

obtain a tree-based representation of the plurality of original image data, wherein the tree-based representation comprises a plurality of leaf nodes with each leaf node of the plurality of leaf nodes being associated with different original image data from the plurality of original image data, and a plurality of parent nodes with each parent node of the plurality of parent nodes being associated with decimated image data representing a different set of the plurality of original image data associated with a different set of the plurality of leaf nodes under that parent node with different amounts of decimation;

generate a regularization curve based on a distribution of the plurality of original image data;

apply the regularization curve to the tree-based representation; and

generate a regularized image from rendering the original image data associated with a subset of the plurality of leaf nodes and the decimated image data associated with a subset of the plurality of parent nodes that intersect or are within the regularization curve.

20. A non-transitory computer-readable medium storing program instructions that, when executed by one or more hardware processors of an imaging system, cause the imaging system to perform operations comprising:

receive an image comprising a plurality of original image data that is non-uniformly distributed across a space represented by the image;

obtain a tree-based representation of the plurality of original image data, wherein the tree-based representation comprises a plurality of leaf nodes with each leaf node of the plurality of leaf nodes being associated with different original image data from the plurality of original image data, and a plurality of parent nodes with each parent node of the plurality of parent nodes being associated with decimated image data representing a different set of the plurality of original image data associated with a different set of the plurality of leaf nodes under that parent node with different amounts of decimation;

generate a regularization curve based on a distribution of the plurality of original image data;

apply the regularization curve to the tree-based representation; and

generate a regularized image from rendering the original image data associated with a subset of the plurality of leaf nodes and the decimated image data associated with a subset of the plurality of parent nodes that intersect or are within the regularization curve.