Iterative relabeling using spectral neighborhoods
A first image is received. An initial label is assigned to at least some pixels in the first image, including by assigning a first label to a first pixel. A determination is made, based at least in part on labels of spectral neighbors of the first pixel, that the first pixel's label should be replaced with a different label. The first pixel's label is updated with the different label.
1. A system, comprising:
a processor configured to:
receive a first image;
assign an initial label to at least some pixels in the first image, including by assigning a first label to a first pixel;
determine a set of spectral neighbors of the first pixel, wherein the spectral neighbors included in the set of spectral neighbors comprise pixels that are within a threshold Euclidian distance in an n-dimensional space to the first pixel;
determine, based at least in part on labels of the spectral neighbors included in the set of spectral neighbors of the first pixel that the first pixel's label should be replaced with a different label; and
update the first pixel's label with the different label; and
a memory coupled to the processor and configured to provide the processor with instructions.
2. The system of claim 1 wherein the determination is based at least in part on a majority vote of the spectral neighbors.
3. The system of claim 1 wherein the spectral neighbors comprise nearest neighbors.
4. The system of claim 1 wherein the initial label is assigned based at least in part by a classifier trained using a training set.
5. The system of claim 1 wherein the initial label is assigned based at least in part by a classifier that takes as input the received first image and a boundary map.
6. The system of claim 1 wherein the initial label is assigned using a historical classification.
7. The system of claim 6 wherein the historical classification is determined from crop data.
8. The system of claim 1 wherein the processor is further configured to iteratively refine the first pixel's label until convergence.
9. A method, comprising:
receiving a first image;
assigning an initial label to at least some pixels in the first image, including by assigning a first label to a first pixel;
determining a set of spectral neighbors of the first pixel, wherein the spectral neighbors included in the set of spectral neighbors comprise pixels that are within a threshold Euclidian distance in an n-dimensional space to the first pixel;
determining, based at least in part on labels of the spectral neighbors included in the set of spectral neighbors of the first pixel that the first pixel's label should be replaced with a different label; and
updating the first pixel's label with the different label.
10. The method of claim 9 wherein determining that the first pixel's label should be replaced with a different label is based at least in part on a majority vote of the spectral neighbors.
11. The method of claim 9 wherein the spectral neighbors comprise nearest neighbors.
12. The method of claim 9 wherein the initial label is assigned based at least in part by a classifier trained using a training set.
13. The method of claim 9 wherein the initial label is assigned based at least in part by a classifier that takes as input the received first image and a boundary map.
14. The method of claim 9 wherein the initial label is assigned using a historical classification.
15. The method of claim 9 further comprising iteratively refining the first pixel's label until convergence.
16. A computer program product embodied in a tangible computer readable storage medium and comprising computer instructions for:
receiving a first image;
assigning an initial label to at least some pixels in the first image, including by assigning a first label to a first pixel;
determining a set of spectral neighbors of the first pixel, wherein the spectral neighbors included in the set of spectral neighbors comprise pixels that are within a threshold Euclidian distance in an n-dimensional space to the first pixel;
determining, based at least in part on labels of the spectral neighbors included in the set of spectral neighbors of the first pixel that the first pixel's label should be replaced with a different label; and
updating the first pixel's label with the different label.
17. The method of claim 14 wherein the historical classification is determined from crop data.
18. The computer program product of claim 16 wherein the determination is based at least in part on a majority vote of the spectral neighbors.
19. The computer program product of claim 16 wherein the spectral neighbors comprise nearest neighbors.
20. The computer program product of claim 16 wherein the initial label is assigned based at least in part by a classifier trained using a training set.
21. The computer program product of claim 16 wherein the initial label is assigned based at least in part by a classifier that takes as input the received first image and a boundary map.
22. The computer program product of claim 16 wherein the initial label is assigned using a historical classification.
23. The computer program product of claim 22 wherein the historical classification is determined from crop data.
24. The computer program product of claim 16 further comprising computer instructions for iteratively refining the first pixel's label until convergence.