Image analysis of multiband images of geographic regions

1. A computer-implemented method for image analysis by one or more computer executing executable instructions stored in one or more non-transitory, tangible, computer readable media, the method comprising:

receiving one or more multiband image of a geographic region, the one or more multiband image having pixels;

generating a surface index for the one or more multiband image containing information indicative of a surface type represented by one or more of the pixels in the one or more multiband image;

classifying the pixels of the one or more multiband image into one of a group of predefined land cover classes, based on the surface index;

receiving information indicative of features within a first multiband image of the one or more multiband image;

dividing the first multiband image into segments; and

creating a segment feature index of the features for each segment, wherein the segment feature index comprises a primary feature index, a secondary feature index, and a tertiary feature index.

2. The computer-implemented method of claim 1 , wherein generating the surface index includes generating a vegetation index for the multiband image containing information indicative of whether one or more of the pixels in the one or more multiband image represents vegetation.

3. The computer-implemented method of claim 1 , wherein classifying pixels in the one or more multiband image into one of a group of predefined land cover classes, based on the surface index, further comprises classifying pixels in the one or more multiband image into one of the group of predefined land cover classes, based on the surface index and based on a majority vote of results of a plurality of classification algorithms determining a plurality of surface cover classifications of the pixels.

4. The computer-implemented method of claim 3 , wherein the plurality of classification algorithms include three or more of a CART classification, a Nave Bayes classification, a random forests classification, a GMO Max Entropy classification, an MCP classification, a Pegasos classification, an IKPamir classification, a voting SVM classification, a margin SVM classification, and a Winnow classification.

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

dividing the pixels of a first multiband image of the one or more multiband image into two or more segments of the first multiband image; and

generating a land cover classification for each of the two or more segments of the first multiband image based on the classification of the pixels.

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

dividing the pixels of a first multiband image of the one or more multiband image into a first class image and a second class image, wherein classifying the pixels of the first multiband image comprises classifying pixels of the first class image and classifying pixels of the second class image; and

reassembling the first multiband image from the first class image and the second class image.

7. The computer-implemented method of claim 1 , wherein the primary feature index includes a number of measured features per segment;

wherein the secondary feature index includes predetermined Boolean relationships of features of each segment; and

wherein the tertiary feature index includes predetermined likelihoods of Boolean relationships of features of each segment and a predetermined threshold of the predetermined likelihoods.

8. The computer-implemented method of claim 7 , wherein the information indicative of features includes one or more of land cover attributes information, demographic information, man-made structure information, and weather information.

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

generating land cover classifications for one or more geographic area based on the classification of the pixels in a first multiband image and a second multiband image of the one or more multiband image.

10. A non-transitory, tangible, computer-readable media having computer-readable instructions stored thereon, for use with a computer and being capable of instructing the computer to perform a method comprising:

receiving one or more multiband image of a geographic region, the one or more multiband image having pixels;

generating a surface index for the one or more multiband image containing information indicative of a surface type represented by one or more of the pixels in the one or more multiband image;

classifying the pixels of the one or more multiband image into one of a group of predefined land cover classes, based on the surface index;

receiving information indicative of features within a first multiband image of the one or more multiband image;

dividing the first multiband image into segments; and

creating a segment feature index of the features for each segment, wherein the segment feature index comprises a primary feature index, a secondary feature index, and a tertiary feature index.

11. The non-transitory, tangible, computer-readable media of claim 10 , wherein generating the surface index includes generating a vegetation index for the multiband image containing information indicative of whether one or more of the pixels in the one or more multiband image represents vegetation.

12. The non-transitory, tangible, computer-readable media of claim 10 , wherein classifying pixels in the one or more multiband image into one of a group of predefined land cover classes, based on the surface index, further comprises classifying pixels in the one or more multiband image into one of the group of predefined land cover classes, based on the surface index and based on a majority vote of results of a plurality of classification algorithms determining a plurality of surface cover classifications of the pixels.

13. The non-transitory, tangible, computer-readable media of claim 12 , wherein the plurality of classification algorithms include three or more of a CART classification, a Nave Bayes classification, a random forests classification, a GMO Max Entropy classification, an MCP classification, a Pegasos classification, an IKPamir classification, a voting SVM classification, a margin SVM classification, and a Winnow classification.

14. The non-transitory, tangible, computer-readable media of claim 10 , further comprising:

dividing the pixels of a first multiband image of the one or more multiband image into two or more segments of the first multiband image; and

generating a land cover classification for each of the two or more segments of the first multiband image based on the classification of the pixels.

15. The non-transitory, tangible, computer-readable media of claim 10 , further comprising:

dividing the pixels of a first multiband image of the one or more multiband image into a first class image and a second class image, wherein classifying the pixels of the first multiband image comprises classifying pixels of the first class image and classifying pixels of the second class image; and

reassembling the first multiband image from the first class image and the second class image.

16. The non-transitory, tangible, computer-readable media of claim 10 :

wherein the primary feature index includes a number of measured features per segment;

wherein the secondary feature index includes predetermined Boolean relationships of features of each segment; and

wherein the tertiary feature index includes predetermined likelihoods of Boolean relationships of features of each segment and a predetermined threshold of the predetermined likelihoods.

17. The non-transitory, tangible, computer-readable media of claim 16 , wherein the information indicative of features includes one or more of land cover attributes information, demographic information, man-made structure information, and weather information.

18. The non-transitory, tangible, computer-readable media of claim 10 , further comprising:

generating land cover classifications for one or more geographic area based on the classification of the pixels in a first multiband image and a second multiband image of the one or more multiband image.