Automated feature extraction from imagery

1. A computer-implemented method of identifying features in imagery comprising:

receiving an image;

grouping selected pixels of the image into a plurality of superpixels;

storing in memory a delineation of the superpixels;

analyzing at least two or more of the plurality of superpixels, the superpixels having an entropy, and the analyzing including determining an indication of the entropy;

storing in memory an indication of the amount of entropy for each of the analyzed superpixels;

identifying superpixels based on an amount of entropy;

identifying clusters of superpixels based on both (1) the indication of the amount of entropy for each of the analyzed superpixels and (2) superpixels that are proximate in location, to produce identified clusters;

storing data delineating the identified clusters in memory, the clusters representing only a portion of the image;

determining whether a cluster density exceeds a predetermined value as a factor in identifying the clusters;

clipping the image to only include the identified groups of superpixels having the predetermined cluster density and entropy;

analyzing statistical parameters of the clipped image;

analyzing geometric factors of the clipped image;

determining one or more settlements based on the statistical parameters and geometric factors of the superpixels; and

identifying a shape and area of the one or more settlements based on the statistical parameters and geometric factors of the clipped image.

2. The computer-implemented method of claim 1 wherein the statistical parameters of the clipped image include one or more of a contrast, mean, mode, median, standard deviation, and entropy.

3. The computer-implemented method of claim 1 wherein the geometric factors include edge detection.

4. The computer-implemented method of claim 1 further comprising:

identifying boundaries of the one or more settlements based on spatial and spectral parameters of proximate superpixels.

5. The computer implemented method of claim 1 wherein the superpixels clustering uses an enhanced segmentation process comprising:

determining a proximity of superpixels; and

determining a best match of superpixels using statistics.

6. The computer-implemented method of claim 5 wherein the statistics include entropy.

7. The computer-implemented method of claim 1 wherein the clusters of superpixels correspond to estimated areas of human settlement.

8. A non-transitory computer-readable medium comprising instructions stored thereon for identifying features in imagery, the instructions, when executed on a processor, perform the steps of:

receiving an image;

grouping selected pixels of the image into a plurality of superpixels;

storing in memory a delineation of the superpixels;

analyzing at least two or more of the plurality of superpixels, the superpixels having an entropy, and the analyzing including determining an indication of the entropy;

storing in memory an indication of the amount of entropy for each of the analyzed superpixels;

identifying superpixels based on an amount of entropy;

identifying clusters of superpixels based on both (1) the indication of the amount of entropy for each of the analyzed superpixels and (2) superpixels that are proximate in location, to produce identified clusters;

storing data delineating the identified clusters in memory, the clusters representing only a portion of the image;

determining whether a cluster density exceeds a predetermined value as a factor in identifying the clusters;

clipping the image to only include the identified groups of superpixels having the predetermined cluster density and entropy;

analyzing statistical parameters of the clipped image;

analyzing geometric factors of the clipped image;

determining one or more settlements based on the statistical parameters and geometric factors of the superpixels; and

identifying a shape and area of the one or more settlements based on the statistical parameters and geometric factors of the clipped image.

9. The non-transitory computer-readable medium of claim 8 wherein the statistical parameters of the clipped image include one or more of a contrast, mean, mode, median, standard deviation, and entropy.

10. The non-transitory computer-readable medium of claim 8 wherein the geometric factors include edge detection.

11. The non-transitory computer-readable medium of claim 8 further comprising:

identifying boundaries of the one or more settlements based on spatial and spectral parameters of proximate superpixels.

12. The non-transitory computer-readable medium of claim 8 wherein the superpixels clustering uses an enhanced segmentation process comprising:

determining a proximity of superpixels; and

determining a best match of superpixels using statistics.

13. The non-transitory computer-readable medium method of claim 12 wherein the statistics include entropy.

14. A system for identifying features in imagery, a memory:

a processor;

a non-transitory computer-readable medium comprising instructions stored thereon, the instructions, when executed on the processor, perform the steps of:

receiving an image;

grouping selected pixels of the image into a plurality of superpixels;

storing in the memory a delineation of the superpixels;

analyzing at least two or more of the plurality of superpixels, the superpixels having an entropy, and the analyzing including determining an indication of the entropy;

storing in the memory an indication of the amount of entropy for each of the analyzed superpixels;

identifying superpixels based on an amount of entropy;

identifying clusters of superpixels based on both (1) the indication of the amount of entropy for each of the analyzed superpixels and (2) superpixels that are proximate in location to produce identified clusters;

storing data delineating the identified clusters in the memory, the clusters representing only a portion of the image;

determining whether a cluster density exceeds a predetermined value as a factor in identifying the clusters;

clipping the image to only include the identified groups of superpixels having the predetermined cluster density and entropy;

analyzing statistical parameters of the clipped image;

analyzing geometric factors of the clipped image;

determining one or more settlements based on the statistical parameters and geometric factors of the superpixels; and

identifying a shape and area of the one or more settlements based on the statistical parameters and geometric factors of the clipped image.