Creating a boundary map

1. A system, comprising:

a processor configured to:

receive, as input, a boundary map and a first image, wherein the received boundary map indicates a plurality of fields, and wherein the boundary map comprises at least one of: (1) a set of pixel locations previously determined to meet at least one boundary criterion, and (2) a set of pixel locations previously determined to not meet the at least one boundary criterion;

select a first field indicated by the boundary map, and use the boundary map to regularize, across a first set comprising a plurality of pixels in the first image bounded by the first field in the plurality of fields, a first set of one or more pixel brightness values;

select a second field indicated by the boundary map, and use the boundary map to regularize, across a second set comprising a plurality of pixels in the first image bounded by the second field in the plurality of fields, a second set of one or more pixel brightness values; and

generate as output a regularized image, wherein, for each pixel in the first image bounded by the first field, a set of values in the corresponding pixel in the regularized image is assigned the set of regularized pixel brightness values; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system of claim 1 wherein the processor is configured to regularize the set of one or more pixel brightness values at least in part by averaging, for a given spectral band, pixel brightness values of each of the pixels included in the plurality of pixels.

3. The system of claim 1 wherein the processor is further configured to determine a land use classification for at least some of the pixels included in the regularized image.

4. The system of claim 1 wherein the processor is further configured to determine a crop yield estimate for at least some of the pixels included in the regularized image.

5. The system of claim 1 wherein the boundary map was created by:

receiving a set of data comprising at least a first tile and a second tile;

determining, for a first pixel location included in both the first tile and the second tile, a measure of whether the first pixel location corresponds to a boundary; and

generating as output a boundary map comprising at least one of boundary pixels and region pixels.

6. The system of claim 5 wherein determining whether the first pixel location corresponds to a boundary includes determining a gradient magnitude for the first pixel location in each of the first tile and the second tile.

7. The system of claim 6 wherein a set of gradient magnitudes collectively determined for each pixel included in the first tile is used to generate a gradient magnitude image.

8. The system of claim 5 wherein determining whether the first pixel location corresponds to a boundary includes determining a quotient of an edge count and a presence count for the first pixel location.

9. The system of claim 1 wherein the plurality of fields comprise common land units.

10. A method, comprising:

receiving, as input, a boundary map and a first image, wherein the received boundary map indicates a plurality of fields, and wherein the boundary map comprises at least one of: (1) a set of pixel locations previously determined to meet at least one boundary criterion, and (2) a set of pixel locations previously determined to not meet the at least one boundary criterion;

selecting a first field indicated by the boundary map, and using the boundary map to regularize, across a first set comprising a plurality of pixels in the first image bounded by the first field in the plurality of fields, a first set of one or more pixel brightness values;

selecting a second field indicated by the boundary map, and using the boundary map to regularize, across a second set comprising a plurality of pixels in the first image bounded by the second field in the plurality of fields, a second set of one or more pixel brightness values; and

generating as output a regularized image, wherein, for each pixel in the first image bounded by the first field, a set of values in the corresponding pixel in the regularized image is assigned the set of regularized pixel brightness values.

11. The method of claim 10 wherein the set of one or more pixel brightness values is regularized at least in part by averaging, for a given spectral band, pixel brightness values of each of the pixels included in the plurality of pixels.

12. The method of claim 10 further comprising determining a land use classification for at least some of the pixels included in the regularized image.

13. The method of claim 10 further comprising determining a crop yield estimate for at least some of the pixels included in the regularized image.

14. The method of claim 10 wherein the boundary map was created by:

receiving a set of data comprising at least a first tile and a second tile;

determining, for a first pixel location included in both the first tile and the second tile, a measure of whether the first pixel location corresponds to a boundary; and

generating as output a boundary map comprising at least one of boundary pixels and region pixels.

15. The method of claim 14 wherein determining whether the first pixel location corresponds to a boundary includes determining a gradient magnitude for the first pixel location in each of the first tile and the second tile.

16. The method of claim 15 wherein a set of gradient magnitudes collectively determined for each pixel included in the first tile is used to generate a gradient magnitude image.

17. The method of claim 14 wherein determining whether the first pixel location corresponds to a boundary includes determining a quotient of an edge count and a presence count for the first pixel location.

18. The method of claim 10 wherein the plurality of fields comprise common land units.