Localization of individual plants based on high-elevation imagery
Implementations are described herein for localizing individual plants by aligning high-elevation images using invariant anchor points while disregarding variant feature points, such as deformable plants. High-elevation images that capture the plurality of plants at a resolution at which wind-triggered deformation of individual plants is perceptible between the high-elevation images may be obtained. First regions of the high-elevation images that depict the plurality of plants may be classified as variant features that are unusable as invariant anchor points. Second regions of the high-elevation images that are disjoint from the first set of regions may be classified as invariant anchor points. The high-elevation images may be aligned based on invariant anchor point(s) that are common among at least some of the high-elevation images. Based on the aligned high-elevation images, individual plant(s) may be localized within one of the high-elevation images for performance of one or more agricultural tasks.
1. A method for localizing one or more individual plants of a plurality of plants, the method implemented using one or more processors and comprising:
obtaining high-elevation images that depict the plurality of plants at a resolution at which wind-triggered deformation of individual plants is perceptible between the high-elevation images;
classifying a first set of regions of the high-elevation images that depict deformable plants as variant visual features that are unusable as invariant anchor points, wherein each region of the first set includes a respective cluster or bounded area of pixels that depict deformable plants;
classifying a second set of regions of the high-elevation images that are disjoint from the first set of regions as invariant anchor points, wherein each region of the second set includes a respective cluster or bounded area of pixels;
spatially aligning the high-elevation images based on one or more of the invariant anchor points that are common among at least some of the high-elevation images; and
based on the spatially-aligned high-elevation images, localizing one or more individual plants of the plurality of plants within one of the high-elevation images for performance of one or more agricultural tasks.
2. The method of claim 1 , wherein the localizing includes assigning position coordinates to the one or more individual plants based on position coordinates generated by an airborne vehicle that acquired the high-elevation images.
3. The method of claim 1 , wherein the localizing includes mapping the one or more individual plants to one or more rows of a plurality of rows in which the plurality of plants are arranged.
4. The method of claim 3 , further comprising processing the aligned high-elevation images to generate a map of the plurality of rows relative to the invariant anchor points.
5. The method of claim 1 , wherein each region of the first and second sets of regions are pixel-wise classified.
6. The method of claim 1 , further comprising deploying one or more agricultural robots to the one or more localized individual plants to perform one or more of the agricultural tasks.
7. The method of claim 1 , wherein the high-elevation images are acquired by an unmanned aerial vehicle (UAV).
8. The method of claim 1 , wherein classifying the first set of regions includes processing the high-elevation images based on one or more machine learning models that are trained to recognize one or more plant types.
9. The method of claim 1 , wherein classifying the second set of regions includes detecting one or more lodged plants among the plurality of plants.
10. The method of claim 1 , wherein classifying the second set of regions includes detecting agricultural equipment in spatial proximity with the plurality of plants.
11. The method of claim 1 , wherein classifying the second set of regions includes detecting one or more water features or roads in spatial proximity with the plurality of plants.
12. A system for localizing one or more individual plants of a plurality of plants, the system comprising one or more processors memory storing instructions that cause the one or more processors to:
obtain high-elevation images that depict the plurality of plants at a resolution at which wind-triggered deformation of individual plants is perceptible between the high-elevation images;
classify a first set of regions of the high-elevation images that depict deformable plants as variant visual features that are unusable as invariant anchor points, wherein each region of the first set includes a respective cluster or bounded area of pixels that depict deformable plants;
classify a second set of regions of the high-elevation images that are disjoint from the first set of regions as invariant anchor points, wherein each region of the second set includes a respective cluster or bounded area of pixels;
spatially align the high-elevation images based on one or more of the invariant anchor points that are common among at least some of the high-elevation images; and
based on the spatially-aligned high-elevation images, localize one or more individual plants of the plurality of plants within one of the high-elevation images for performance of one or more agricultural tasks.
13. The system of claim 12 , further comprising instructions to assign position coordinates to the one or more individual plants based on position coordinates generated by an airborne vehicle that acquired the high-elevation images.
14. The system of claim 12 , further comprising instructions to map the one or more individual plants to one or more rows of a plurality of rows in which the plurality of plants are arranged.
15. The system of claim 14 , further comprising instructions to process the aligned high-elevation images to generate a map of the plurality of rows relative to the invariant anchor points.
16. The system of claim 12 , wherein each region of the first and second sets of regions are pixel-wise classified.
17. The system of claim 12 , further comprising instructions to deploy one or more agricultural robots to the one or more localized individual plants to perform one or more of the agricultural tasks.
18. The system of claim 12 , wherein the high-elevation images are acquired by an unmanned aerial vehicle (UAV).
19. The system of claim 12 , further comprising instructions to process the high-elevation images based on one or more machine learning models that are trained to recognize one or more plant types.
20. A non-transitory computer-readable medium for localizing one or more individual plants of a plurality of plants, the medium comprising instructions that, in response to execution of the instructions by a processor, cause the processor to:
obtain high-elevation images that depict the plurality of plants at a resolution at which wind-triggered deformation of individual plants is perceptible between the high-elevation images;
classify a first set of regions of the high-elevation images that depict deformable plants as variant visual features that are unusable as invariant anchor points, wherein each region of the first set includes a respective cluster or bounded area of pixels that depict deformable plants;
classify a second set of regions of the high-elevation images that are disjoint from the first set of regions as invariant anchor points, wherein each region of the second set includes a respective cluster or bounded area of pixels;
spatially align the high-elevation images based on one or more of the invariant anchor points that are common among at least some of the high-elevation images; and
based on the spatially-aligned high-elevation images, localize one or more individual plants of the plurality of plants within one of the high-elevation images for performance of one or more agricultural tasks.