Reduced-bias multi-resolution coherent change detection
View Patent ↗Described herein are various technologies relating to generating a reduced bias multi-resolution (RBMR) coherent change detection (CCD) image product. A low bias, low resolution (LBLR) CCD image product is generated based upon a first SAR image of a scene and a second SAR image of the scene, where each pixel in the LBLR CCD image has a respective value that is computed using a large box-filter. The LBLR CCD image is segmented into a plurality of segments, and a box-filter size is determined for each segment based upon values of pixels in the segment. A RBMR CCD image product is then computed using box-filters of the determined sizes.
1. A computing device comprising:
a processor; and
memory that comprises instructions that, when executed by the processor, cause the processor to perform acts comprising:
receiving a first synthetic aperture radar (SAR) image and a second SAR image for a scene, the first SAR image registered with the second SAR image, the first SAR image corresponds to a first time, the second SAR image corresponds to a second time;
generating a coherence change detection (CCD) image, wherein generating the CCD image comprises computing a sample correlation of a product of the first SAR image with a complex-conjugate of the second SAR image, wherein each pixel of the CCD image has a value that is computed using a box filter of a first size;
responsive to generating the CCD image, segmenting the CCD image into a plurality of segments, wherein the CCD image is segmented into the plurality of segments based upon the values of the pixels of the CCD image,
identifying a plurality of box filter sizes for the plurality of segments of the CCD image such that each pixel of the CCD image is assigned a respective box filter size, wherein the plurality of box filter sizes are identified based upon the values of the pixels of the CCD image, and
generating, a reduced bias multi-resolution (RBMR) CCD image, wherein generating the RBMR CCD image comprises computing the sample correlation of the product of the first SAR image with the complex-conjugate of the second SAR image, wherein each pixel of the RBMR CCD image has a value that is computed using the respective box titter size assigned to the pixel.
2. The computing device of claim 1 , wherein identifying a plurality of box filter sizes for the plurality of segments of the CCD image comprises:
for a first segment in the plurality of segments:
computing a property for the first segment; and
determining a first box filter size for the first segment based upon the property for the first segment, wherein each pixel in the first segment is assigned the first box filter size;
for a second segment in the plurality of segments:
computing a second property for the second segment; and
determining a second box filter size for the second segment based upon the property for the second segment, wherein each pixel in the second segment is assigned the second box filter size.
3. The computing device of claim 2 , the property for the first segment and the property for the second segment being estimated coherence for the first segment and the second segment in the CCD image.
4. The computing device of claim 1 , wherein bias of the RBMR CCD image is beneath a predefined threshold amount of bias.
5. The computing device of claim 1 , wherein the box filter of the first size is a 25×25 box filter.
6. A method executed by a processor of a computing device, the method comprising:
receiving a first synthetic aperture radar (SAR) image and a second SAR image for a scene, the first SAR image registered with the second SAR image, the first SAR image corresponds to a first time, the second SAR image corresponds to a second time;
generating a coherence change detection (CCD) image, wherein generating the CCD image comprises computing a sample correlation of a product of the first SAR image with a complex-conjugate of the second SAR image, wherein each pixel of the CCD image has a value that is computed using a box filter of a first size;
responsive to generating the CCD image, segmenting the CCD image into a plurality of segments, wherein the CCD image is segmented into the plurality of segments based upon the values of the pixels of the CCD image;
identifying a plurality of box filter sizes for the plurality of segments of the CCD image such that each pixel of the CCD image is assigned a respective box filter size, wherein the plurality of box filter sizes are identified based upon the values of the pixels of the CCD image; and
generating a reduced bias multi-resolution (RBMR) CCD image, wherein generating the RBMR CCD image comprises computing the sample correlation of the product of the first SAR image with the complex-conjugate of the second SAR image, wherein each pixel of the RBMR CCD image has a value that is computed using the respective box filter size assigned to the pixel.
7. The method of claim 6 , wherein identifying a plurality of box filter sizes for the plurality of segments of the CCD image comprises:
for a first segment in the plurality of segments:
computing a property for the first segment; and
determining a first box filter size for the first segment based upon the property for the first segment, wherein each pixel in the first segment is assigned the first box filter size;
for a second segment in the plurality of segments:
computing a second property for the second segment; and
determining a second box filter size for the second segment based upon the property for the second segment, wherein each pixel in the second segment is assigned the second box filter size.
8. The method of claim 7 , the property for the first segment and the property for the second segment being estimated coherence for the first segment and the second segment in the CCD image.
9. The method of claim 6 , wherein bias of the RBMR CCD image is beneath a predefined threshold amount of bias.
10. The method of claim 6 , wherein the box filter of the first size is a 25×25 box filter.