Moving object detection from a moving camera

A computing system comprising:

• one or more processors;
• and
• a memory system comprising one or more non-transitory computer-readable media storing instructions that, when executed by at least one of the one or more processors, cause the computing system to perform operations, the operations comprising:
• identifying a first image that is captured at a first time, wherein the first image is captured by a camera on an aircraft in flight;
• identifying a second image that is captured at a second time, wherein the second image is captured by the camera on the aircraft in flight;
• segmenting the first image into a plurality of first image portions, wherein a variation of the aircraft, the camera, or both between the first image and the second image causes image distortion when comparing the first image as a whole and the second image as a whole, wherein the variation is in a distance between the aircraft and a ground, an attitude of the aircraft, a rotation of the aircraft, an angle of the camera, or a combination thereof, and wherein the image distortion causes an object in the first and second images to appear to move when the objects are actually stationary;
• segmenting the second image into a plurality of second image portions;
• determining a phase correlation between one of the plurality of first image portions and a corresponding one of the plurality of second image portions, wherein the phase correlation is due to the image distortion;
• comparing the one of the plurality of first image portions and the corresponding one of the plurality of second image portions based at least partially upon the phase correlation;
• determining an affine translation between the one of the plurality of first image portions and the corresponding one of the plurality of second image portions based at least partially upon the comparison;
• transforming the one of the plurality of first image portions into a transformed first image portion based at least partially upon the affine translation;
• identifying a pixel in the transformed first image portion that differs from a corresponding pixel in the corresponding one of the plurality of second image portions by more than a predetermined amount;
• determining that the pixel in the transformed first image and the corresponding pixel in the corresponding one of the plurality of second image portions represent a moving object;
• and
• causing the aircraft to navigate based at least partially upon the moving object.

A computing system comprising:

• one or more processors;
• and
• a memory system comprising one or more non-transitory computer-readable media storing instructions that, when executed by at least one of the one or more processors, cause the computing system to perform operations, the operations comprising:
• identifying a first image that is captured at a first time, wherein the first image is captured by a camera on a first aircraft in flight, and wherein the camera is pointed substantially straight down at a ground region below the first aircraft;
• identifying a second image that is captured at a second time, wherein the second image is captured by the camera on the first aircraft in flight, wherein the second time is after the first time, and wherein a time difference between the first and second times is less than 10 seconds, wherein a variation of the first aircraft, the camera, or both between the first image and the second image causes image distortion when comparing the first image as a whole and the second image as a whole, wherein the variation is in a distance between the first aircraft and a ground, an attitude of the first aircraft, a rotation of the first aircraft, an angle of the camera, or a combination thereof, and wherein the image distortion causes an object in the first and second images to appear to move when the objects are actually stationary;
• segmenting the first image into a plurality of first image portions;
• segmenting the second image into a plurality of second image portions;
• determining a phase correlation between one of the plurality of first image portions and a corresponding one of the plurality of second image portions, wherein the phase correlation is due to the image distortion;
• comparing the one of the plurality of first image portions and the corresponding one of the plurality of second image portions based at least partially upon the phase correlation;
• determining an affine translation between the one of the plurality of first image portions and the corresponding one of the plurality of second image portions based at least partially upon the comparison;
• transforming the first image into a transformed first image based at least partially upon the affine translation, wherein the transformed first image comprises a prediction of what the second image will look like based at least partially upon the first image and the affine translation;
• combining the transformed first image and the second image to produce a combined image;
• identifying an area in the combined image that represents a difference between the transformed first image and the second image, wherein the difference is greater than a predetermined amount;
• identifying the area in the first image, the second image, or both;
• and
• causing the aircraft to navigate based at least partially upon the area.