System and method for managing water
A device includes an image data receiving component, a vegetation index generation component, a GLC matrix component, a plurality of classifying components and a voting component. The image data receiving component receives multiband image data of a geographic region. The vegetation index generation component generates a normalized difference vegetation index based on the received multiband image data. The GLC matrix component generates a grey level co-occurrence matrix image band based on the received multiband image data. The classifying components generate land cover classifications based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix image band. The voting component generates a final land cover classification based a majority vote of the land cover classifications.
1. A device comprising:
an image data receiving component operable to receive multiband image data of a geographic region;
a vegetation index generation component operable to generate a normalized difference vegetation index based on the received multiband image data;
a grey level co-occurrence matrix generation component operable to generate a grey level co-occurrence matrix image band based on the received multiband image data;
a first classification component operable generate a first land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix image band;
a second classification component operable generate a second land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix image band;
a third classification component operable generate a third land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix image band; and
a voting component operable to generate a final land cover classification based a majority vote of the first land cover classification, the second land cover classification and the third land cover classification.
2. The device of claim 1 , wherein said image data receiving component is operable to receive the multiband image data of a geographic region as an RGB and near infra-red image data of the geographic region.
3. The device of claim 2 ,
wherein the multiband image data corresponds to an array of pixels, and
wherein said first classification component is operable to generate the first land cover classification by classifying each of the pixels of the multiband image data as one of the group consisting of grass, a tree, a shrub, a man-made surface, a man-made pool, a natural water body and artificial turf.
4. The device of claim 3 , wherein said first classification component comprises one of the group consisting of a simple classification and regression tree classifier, a naïve Bayes classifier, a random forecasts classifier, a GMO max entropy classifier, an MCP classifier, a Pegasos classifier, an ICPamir classifier, a voting SVM classifier, a margin SVM classifier and a Winnow classifier.
5. The device of claim 4 , further comprising a parcel data receiving component operable to receive parcel data.
6. The device of claim 1 ,
wherein the multiband image data corresponds to an array of pixels, and
wherein said first classification component operable generate the first land cover classification by classifying each of pixels of the multiband image data as one of the group consisting of grass, a tree, a shrub, a man-made surface, a man-made pool, a natural water body and artificial turf.
7. The device of claim 1 , wherein said first classification component comprises one of the group consisting of a simple classification and regression tree classifier, a naïve Bayes classifier, a random forecasts classifier, a GMO max entropy classifier, a MCP classifier a, a Pegasos classifier, an ICPamir classifier, a voting SVM classifier, a margin SVM classifier and a Winnow classifier.
8. The device of claim 1 , further comprising a regression component operable to extrapolate a predicted water usage based on a history of water usage.
9. A method comprising:
receiving, via an image data receiving component, multiband image data of a geographic region;
generating, via a vegetation index generation component, a normalized difference vegetation index based on the received multiband image data;
generating, via a grey level co-occurrence matrix generation component, a grey level co-occurrence matrix based on the received multiband image data;
generating, via a first classification component, a first land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix;
generating, via a second classification component, a second land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix;
generating, via a third classification component, a third land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix; and
generating, via a voting component, a final land cover classification based a majority vote of the first land cover classification, the second land cover classification and the third land cover classification.
10. The method of claim 9 , wherein said receiving multiband image data of a geographic region comprises receiving the multiband image data of a geographic region as an RGB and near infra-red image data of the geographic region.
11. The method of claim 10 ,
wherein the multiband image data corresponds to an array of pixels, and
wherein said generating a first land cover classification comprises generating the first land cover classification by classifying each of pixels of the multiband image data as one of the group consisting of grass, a tree, a shrub, a man-made surface, a man-made pool, a natural water body and artificial turf.
12. The method of claim 11 , wherein said generating a first land cover classification comprises generating comprises generating via one of the group consisting of a simple classification and regression tree classifier, a naïve Bayes classifier, a random forecasts classifier, a GMO max entropy classifier, an MCP classifier, a Pegasos classifier, an ICPamir classifier, a voting SVM classifier, a margin SVM classifier and a Winnow classifier.
13. The method of claim 12 , further comprising receiving, via a parcel data receiving component, parcel data.
14. The method of claim 9 ,
wherein the multiband image data corresponds to an array of pixels, and
wherein said generating a first land cover classification comprises generating the first land cover classification by classifying each of pixels of the multiband image data as one of the group consisting of grass, a tree, a shrub, a man-made surface, a man-made pool, a natural water body and artificial turf.
15. The method of claim 9 , wherein said generating a first land cover classification comprises generating comprises generating via one of the group consisting of a simple classification and regression tree classifier, a naïve Bayes classifier, a random forecasts classifier, a GMO max entropy classifier, an MCP classifier, a Pegasos classifier, an ICPamir classifier, a voting SVM classifier, a margin SVM classifier and a Winnow classifier.
16. The method of claim 9 , further comprising extrapolating, via a regression component, a predicted water usage based on a history of water usage.
17. 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 the method comprising:
receiving, via an image data receiving component, multiband image data of a geographic region;
generating, via a vegetation index generation component, a normalized difference vegetation index based on the received multiband image data;
generating, via a grey level co-occurrence matrix generation component, a grey level co-occurrence matrix based on the received multiband image data;
generating, via a first classification component, a first land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix;
generating, via a second classification component, a second land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix;
generating, via a third classification component, a third land cover classification based on the received multiband image data, the normalized difference vegetation index and the grey level co-occurrence matrix; and
generating, via a voting component, a final land cover classification based a majority vote of the first land cover classification, the second land cover classification and the third land cover classification.
18. The non-transitory, tangible computer-readable media of claim 17 , wherein the computer-readable instructions are capable of instructing the computer to perform the method such that said receiving multiband image data of a geographic region comprises receiving the multiband image data of a geographic region as an RGB and near infra-red image data of the geographic region.
19. The non-transitory, tangible, computer-readable media of claim 18 ,
wherein the multiband image data corresponds to an array of pixels, and
wherein said generating a first land cover classification comprises generating the first land cover classification by classifying each of pixels of the multiband image data as one of the group consisting of grass, a tree, a shrub, a man-made surface, a man-made pool, a natural water body and artificial turf.
20. The non-transitory, tangible, computer-readable media of claim 19 , wherein the computer-readable instructions are capable of instructing the computer to perform the method such that said generating a first land cover classification comprises generating comprises generating via one of the group consisting of a simple classification and regression tree classifier, a naïve Bayes classifier, a random forecasts classifier, a GMO max entropy classifier, an MCP classifier, a Pegasos classifier, an ICPamir classifier, a voting SVM classifier, a margin SVM classifier and a Winnow classifier.