System and method for filtering data captured by a 3D camera
View Patent ↗In an exemplary embodiment, a system includes a three-dimensional camera and a processor. The processor is operable to access a first portion of visual data captured by the three-dimensional camera wherein the visual data comprises an image of a dairy livestock and access a second portion of the visual data, wherein the first portion and the second portion are aligned in a first dimension. The processor is further operable to determine a first coordinate, wherein the first coordinate comprises a location of the first portion in a second dimension, the second dimension orthogonal to the first dimension and determine a second coordinate, wherein the second coordinate comprises a location of the second portion in the second dimension. The processor is further operable to determine a first distance exceeds a distance threshold, wherein the first distance is the distance between the first coordinate and the second coordinate in the second dimension.
1. A system comprising:
a three-dimensional camera; and
a processor communicatively coupled to the three-dimensional camera, the processor operable to:
access a first portion of visual data captured by the three-dimensional camera wherein the visual data comprises an image of a dairy livestock;
access a second portion of the visual data;
determine a first coordinate, wherein the first coordinate comprises a depth location of the first portion of the visual data;
determine a second coordinate, wherein the second coordinate comprises a depth location of the second portion of the visual data;
determine that a first distance exceeds a distance threshold, wherein the first distance is the distance between the depth locations of the first coordinate and the second coordinate; and
discard the second portion of the visual data in response to the first distance exceeding the distance threshold.
2. The system of claim 1 , wherein filtering the second portion of the visual data comprises:
determining a third coordinate, wherein the third coordinate comprises a depth location of a third portion of the visual data;
determining a second distance is within the distance threshold, wherein the second distance is the distance between the depth locations of the first coordinate and the third coordinate; and
determining a third distance exceeds the distance threshold, wherein the third distance is the distance between the depth locations of the second coordinate and the third coordinate.
3. The system of claim 1 , wherein the second coordinate is further away from the three-dimensional camera than the first coordinate.
4. The system of claim 1 , wherein the second coordinate is closer to the three-dimensional camera than the first coordinate.
5. The system of claim 1 , wherein the image of the dairy livestock comprises an image of a portion of the hind legs.
6. The system of claim 1 , wherein the visual data comprises depth information for the first and second portions of visual data.
7. A method comprising:
accessing a first portion of visual data captured by a three-dimensional camera wherein the visual data comprises an image of a dairy livestock;
accessing a second portion of the visual data;
determining a first coordinate, wherein the first coordinate comprises a depth location of the first portion of the visual data;
determining a second coordinate, wherein the second coordinate comprises a depth location of the second portion of the visual data;
determining that a first distance exceeds a distance threshold, wherein the first distance is the distance between the depth locations of the first coordinate and the second coordinate; and
discarding the second portion of the visual data in response to the first distance exceeding the distance threshold.
8. The method of claim 7 , wherein filtering the second portion comprises:
determining a third coordinate, wherein the third coordinate comprises a depth location of a third portion of the visual data;
determining a second distance is within the distance threshold, wherein the second distance is the distance between the depth locations of the first coordinate and the third coordinate; and
determining a third distance exceeds the distance threshold, wherein the third distance is the distance between the depth locations of the second coordinate and the third coordinate.
9. The method of claim 7 , wherein the second coordinate is further away from the three-dimensional camera than the first coordinate.
10. The method of claim 7 , wherein the second coordinate is closer to the three-dimensional camera than the first coordinate.
11. The method of claim 7 , wherein the image of the dairy livestock comprises an image of a portion of the hind legs.
12. The method of claim 7 , wherein the visual data comprises depth information for the first and second portions of visual data.
13. A system comprising:
a three-dimensional camera; and
a processor communicatively coupled to the three-dimensional camera, the processor operable to:
analyze a plurality of neighboring pixels captured by the three-dimensional camera;
determine first depth information for a first pixel of the neighboring pixels;
determine second depth information for a second pixel of the neighboring pixels;
determine that the distance between the first depth information and the second depth information exceeds a distance threshold; and
discard the second pixel.
14. The system of claim 13 , wherein the second pixel is further away from the three-dimensional camera than the remaining neighboring pixels.
15. The system of claim 13 , wherein the second pixel is closer to the three-dimensional camera than the remaining neighboring pixels.
16. The system of claim 13 , wherein the plurality of neighboring pixels comprise an image of a dairy livestock.