Generation of high resolution population density data sets through exploitation of high resolution overhead imagery data and low resolution population density data sets

1. A method for use in generating population density data sets of a geographic region from overhead imagery data, comprising:

identifying areas of built-up structures in at least one input overhead image of a geographic region;

obtaining population density estimates of a plurality of first portions of the geographic region; and

with a processor, using the areas of built-up structures in the at least one input overhead image within the first portions to allocate the population density estimates across a plurality of second portions within the first portions, wherein the using comprises:

obtaining a plurality of multi-scale morphological image descriptors for each of a respective plurality of pixels of the areas of built-up structures in the at least one input overhead image of the geographic region;

training, with the processor, a linear model with a) the plurality of population density estimates of the plurality of first portions and b) the multi-scale morphological image descriptors within each of the plurality of first portions to obtain a plurality of respective weighting factors; and

using the weighting factors to allocate the population density estimate of each first portion across the plurality of second portions of the first portions.

2. The method of claim 1 , wherein the training includes:

decomposing the multi-scale morphological image descriptors within each of the first portions into a respective plurality of first vectors that each include a plurality of entries, wherein each entry includes a value that represents a frequency of one of a plurality of fixed values of the multi-scale morphological image descriptors within a respective one of the first portions; and

combining the plurality of first vectors into a matrix.

3. The method of claim 2 , wherein the training includes:

generating a second vector that includes a plurality of entries, wherein each entry includes a value that represents one of the population density estimates of a respective one of the plurality of first portions of the geographic region.

4. The method of claim 3 , wherein the linear model comprises a difference between a) the second vector and b) a product of the matrix and a third vector that includes a plurality of entries, and wherein the training includes:

determining a particular plurality of values of the respective plurality of entries of the third vector that minimizes the difference, wherein the particular plurality of values is the plurality of weighting factors.

5. The method of claim 4 , wherein the plurality of weighting factors are arranged in a weighting vector, and wherein the using the weighting factors to allocate the population density estimate of the first portions across the plurality of second portions includes, for each second portion:

decomposing the multi-scale morphological image descriptors within the second portion into a fourth vector that includes a plurality of entries, wherein each entry includes a value that represents a frequency of one of the plurality of fixed values of the multi-scale morphological image descriptors within the second portion;

obtaining an inner product of the fourth vector and the weighting vector; and

multiplying the inner product by a total quantity of pixels in the second portion of the at least one input overhead image to obtain the population density estimate for the second portion.

6. The method of claim 1 , further comprising:

mapping the plurality of population density estimates of the respective plurality of second portions into a resultant image of the geographic region.

7. The method of claim 1 , further comprising for each first portion of the geographic region:

using the population density estimate of the first portion to constrain the population density estimates of the second portions of the first portion.

8. The method of claim 7 , wherein the using the population density estimate of the first portion to constrain the population density estimates of the second portions of the first portion includes:

normalizing the population density estimates of the second portions into a plurality of normalized population density estimates; and

multiplying the population density estimate of the first portion by the normalized population density estimates to obtain corrected population density estimates for the second portions.

9. The method of claim 1 , wherein the at least one input overhead image comprises a spatial resolution of no greater than 10 meters.

10. The method of claim 9 , wherein the at least one input overhead image comprises a spatial resolution of no greater than 2 meters.

11. The method of claim 1 , wherein the first and second portions are squares.

12. A method for use in generating population density data sets of a geographic region from overhead imagery data, comprising:

identifying areas of interest in at least one input overhead image of a geographic region;

generating, with a processor for each first portion of a plurality of first portions of the geographic region, a first vector of entries that respectively indicate the frequency of each of a plurality of different multi-scale image descriptors in the areas of interest in the first portion;

combining the first vectors of the plurality of first portions into a matrix;

generating, with the processor for each first portion, a second vector of entries that respective indicate population density estimates for the plurality of first portions of the geographic region;

determining, with the processor, a third vector of weighting values that respectively minimize a difference between a) the second vector and b) a product of the matrix and the vector of weighting values; and

with the processor, using the third vector of weighting values to allocate the population density estimates of the plurality of first portions across a plurality of second portions within each of the first portions.

13. The method of claim 12 , wherein the using comprises:

generating, with the processor for each second portion of the plurality of second portions, a fourth vector of entries that respectively indicate the frequency of each of the plurality of different multi-scale image descriptors in the areas of interest in the second portion; and

manipulating the fourth vector of entries with the third vector of weighting values to obtain population density estimates for each second portion of the plurality of second portions.

14. The method of claim 13 , wherein the manipulating includes:

obtaining an inner product of the fourth vector and the third vector of weighting values; and

multiplying the inner product by a total quantity of pixels in the second portion of the at least one input overhead image to obtain the population density estimate for the second portion.

15. The method of claim 12 , further comprising:

mapping the plurality of population density estimates of the respective plurality of second portions into a resultant image of the geographic region.

16. The method of claim 12 , further comprising for each first portion of the geographic region:

using the population density estimate of the first portion to constrain the population density estimates of the second portions of the first portion.

17. The method of claim 16 , wherein the using the population density estimate of the first portion to constrain the population density estimates of the second portions of the first portion includes:

normalizing the population density estimates of the second portions into a plurality of normalized population density estimates; and

multiplying the population density estimate of the first portion by the normalized population density estimates to obtain corrected population density estimates for the second portions.

18. The method of claim 12 , wherein the areas of interest are built-up structures.