Automated geospatial image mosaic generation with automatic source selection
View Patent ↗Automatic generation of a mosaic comprising a plurality of geospatial images. An embodiment of the automatic mosaic generation may include automated source image selection that includes comparison of source images to base layer image to determine radiometric similar source images. Additionally, an embodiment of an automatic cutline generator may be provided to automatically determine a cutline when merging two images such that radiometric differences between the images along the cutline are reduced. In this regard, less perceivable outlines may be provided. Further still, an embodiment of a radiometric normalization module may be provided that may determine radiometric adjustments to source images to match certain properties of the base layer image. In some embodiments, when processing source images, the source images may be downsampled during a portion of the processing to reduce computational overhead. Additionally, some highly parallel computations may be performed by a GPU to further enhance performance.
1. A method for automatic source image selection for generation of a composite orthomosaic image from a plurality of geospatial source images, comprising:
accessing a plurality of geospatial source images corresponding to an area of interest;
comparing the plurality of geospatial source images to a base layer image, wherein each of the plurality of geospatial source images are compared to a corresponding portion of the base layer image that is geographically concurrent with a respective one of the plurality of geospatial source images; and
selecting at least one of the plurality of geospatial source images for use in the orthomosaic image based on the comparing.
2. The method of claim 1 , wherein the area of interest corresponds to the geographical extent of the orthomosaic image.
3. The method of claim 2 , wherein the area of interest is divided into a plurality of tiles, wherein for each tile, the comparing includes comparing a source image chip of each source image with coverage relative to the tile to a base layer chip from the base layer image corresponding to the tile.
4. The method of claim 3 , wherein the base layer image comprises a lower spatial resolution image than the geospatial images.
5. The method of claim 4 , wherein the base layer image is a color balanced global mosaic.
6. The method of claim 3 , wherein the selecting comprises calculating a cost function based on the selection of each geospatial image source for use in the mosaic.
7. The method of claim 6 , wherein the cost function includes calculating for each tile portion a merit value for each source image chip having coverage for the tile.
8. The method of claim 7 , wherein the merit value includes a similarity metric and a blackfill metric, wherein the similarity metric is a quantifiable value that is indicative of radiometric similarities between the source image chip and the base layer chip, and wherein the blackfill metric is a quantifiable value that corresponds to the amount of coverage of the source image chip relative to the tile.
9. The method of claim 8 , wherein the similarity metric is calculated using at least one of a spatial correlation metric, a block based spatial correlation metric, a Wang-Bovik quality index metric, a histogram intersection metric, a Gabor textures metric, and an imaging differencing metric.
10. The method of claim 9 , wherein the similarity metric is calculated at least utilizing a Wang-Bovik quality index metric.
11. The method of claim 10 , wherein the similarity metric comprises a weighted value based on a plurality of metrics.
12. The method of claim 9 , wherein the base layer image and the plurality of geospatial source image include image data comprising a different number of spectral bands.
13. The method of claim 8 , wherein the cost function includes calculating a coverage metric corresponding to the degree to which a given geospatial source image provides unique coverage in the overall mosaic, wherein the greater the coverage the geospatial source image provides relative to the area of interest, the greater the coverage metric for the source image, and wherein the lesser the coverage the geospatial source images provides relative to the area of interest, the greater the source image is penalized with respect to the coverage metric for the geospatial image.
14. The method of claim 13 , further comprising:
establishing a global score for the entire orthomosaic at least partially based on the merit values for each selected source image and the coverage metric for each selected source image;
wherein the selecting includes determining the geospatial source images for each tile that maximizes the global score of the mosaic.
15. The method of claim 1 , wherein the plurality of geospatial source images comprise a subset of a set of geospatial source images, wherein the subset is determined by filtering the set of geospatial images based on metadata for each of the set of geospatial source images.
16. The method of claim 15 , wherein the metadata comprises at least one of a date of acquisition, a solar elevation angle, and a satellite off-nadir angle.
17. The method of claim 1 , further comprising:
downsampling the plurality of geospatial images prior to the comparing.
18. The method of claim 17 , wherein the downsampling includes reducing the resolution of the plurality of geospatial source images to a resolution matching the base layer image.
19. The method of claim 1 , wherein the base layer image is a prior orthomosaic generated utilizing the method of claim 1 .
20. The method of claim 1 , wherein at least one of the comparing and the selecting comprise algorithms executed by a graphics processing unit.