Pest infestation detection for horticultural grow operations

12106,560

17/162673

January 29, 2021

Disclosed are techniques for detecting and mitigating pest infestations within a grow operation. In some embodiments, such techniques comprise receiving, from a visual observer device, image data associated with a location within a grow operation. The image data is then used to determine at least one pest classification and a count associated with the pest classification. Distribution data is generated based on the at least one pest classification, count, and location. A level of risk can then be determined based on the distribution data. In some embodiments, if the level of risk is greater than a threshold level of risk, a recommendation may be generated based at least in part on the distribution data.

iUNU, Inc. (Seattle, WA, US)

iUNU, INC. (Seattle, WA, US)

1. A computer-implemented method, comprising: receiving, from a camera device, image data, the image data being associated with a location within a grow operation; determining, from the image data, at least one pest classification and a count associated with the at least one pest classification; generating a distribution data based on the at least one pest classification, count, and location; determining a level of risk associated with the distribution data; based on the at least one pest classification, determining a threshold level of risk; comparing the level of risk associated with the distribution data to the threshold level of risk that is based on the at least one pest classification; based on comparing the level of risk associated with the distribution data to the threshold level of risk that is based on the at least one pest classification, determining that the level of risk associated with the distribution data is greater than a threshold level of risk that is based on the at least one pest classification; identifying a pest infestation based at least in part on the distribution data; based on determining that the level of risk associated with the distribution data is greater than a threshold level of risk that is based on the at least one pest classification, generating a recommendation that includes a set of potential solutions that address the pest infestation; based on the at least one pest classification, determining an effectiveness rating for each potential solution of the set of potential solutions that address the pest infestation; determining preferences associated with the grow operation; based on the preferences associated with the grow operation, determining a weight for each potential solution of the set of potential solutions; ranking the set of potential solutions based on the weight of each potential solution; comparing each effectiveness rating for each potential solution of the set of potential solutions to an effectiveness rating threshold; based on the ranking of the potential solutions and based on comparing each effectiveness rating for each potential solution of the set of potential solutions to an effectiveness rating threshold, selecting a subset of the set of potential solutions; providing, for output to a robotic device, the subset of the set of potential solutions and an instruction to implement the subset of the set of potential solutions, wherein the robotic device implements the subset of the set of potential solutions in response to receiving the instruction to implement the subset of the set of potential solutions; determining that the robotic device has implemented the subset of the set of potential solutions; receiving, from the camera device, additional image data, the additional image data being associated with the location within the grow operation; determining, from the image data, an updated count associated with the at least one pest classification; and based on the updated count associated with the at least one pest classification, updating each effectiveness rating for each potential solution of the subset of the set of potential solutions.

2. The computer-implemented method of claim 1 , wherein determining the level of risk associated with the distribution data comprises retrieving a destructiveness profile associated with the at least one pest classification.

3. The computer-implemented method of claim 2 , wherein the destructiveness profile associated with the at least one pest classification is specific to a type of plant located at the location within the grow operation.

4. The computer-implemented method of claim 1 , wherein the image data comprises an image of an insect trap.

5. The computer-implemented method of claim 1 , wherein the image data comprises an image of a portion of a plant or group of plants.

6. The computer-implemented method of claim 1 , wherein the distribution data comprises information about a population of the at least one pest classification throughout an area.

7. The computer-implemented method of claim 1 , wherein the distribution data comprises information about at least one of a pest size or a pest maturity.

8. The computer-implemented method of claim 1 , wherein the threshold level of risk comprises a percentage probability that a crop failure will occur.

9. A computing device comprising: a processor; and a memory including instructions that, when executed with the processor, cause the computing device to, at least: receive, from a visual observer device, image data, the image data being associated with a location within a grow operation; determine, from the image data, at least one pest classification and a count associated with the at least one pest classification; generate a distribution data based on the at least one pest classification, count, and location; determine a level of risk associated with the distribution data; based on the at least one pest classification, determine a threshold level of risk; compare the level of risk associated with the distribution data to the threshold level of risk that is based on the at least one pest classification; based on comparing the level of risk associated with the distribution data to the threshold level of risk that is based on the at least one pest classification, determine that the level of risk associated with the distribution data is greater than a threshold level of risk that is based on the at least one pest classification; identify a pest infestation associated with the pest classification based on the generated distribution data; based on determining that the level of risk associated with the distribution data is greater than a threshold level of risk that is based on the at least one pest classification, generating a recommendation that includes a set of potential solutions to address the pest infestation; based on the at least one pest classification, determine an effectiveness rating for each potential solution of the set of potential solutions that address the pest infestation; determining preferences associated with the grow operation; based on the preferences associated with the grow operation, determining a weight for each potential solution of the set of potential solutions; ranking the set of potential solutions based on the weight of each potential solution; compare each effectiveness rating for each potential solution of the set of potential solutions to an effectiveness rating threshold; based on the ranking of the potential solutions and based on comparing each effectiveness rating for each potential solution of the set of potential solutions to an effectiveness rating threshold, selecting a subset of the set of potential solutions: provide, for output to a robotic device, the subset of the set of potential solutions and an instruction to implement the subset of the set of potential solutions, wherein the robotic device implements the subset of the set of potential solutions in response to receiving the instruction to implement the subset of the set of potential solutions; determine that the robotic device has implemented the subset of the set of potential solutions; receive, from the visual observer device, additional image data, the additional image data being associated with the location within the grow operation; determine, from the image data, an updated count associated with the at least one pest classification; and based on the updated count associated with the at least one pest classification, update each effectiveness rating for each potential solution of the subset of the set of potential solutions.

10. The computing device of claim 9 , wherein the distribution data is generated by interpolating count data across an area based on the determined count.

11. The computing device of claim 9 , wherein the instructions further cause the computing device to determine, based on the distribution data, a location of the pest infestation associated with the pest classification.

12. The computing device of claim 11 , wherein determining a location of a pest infestation comprises identifying a variance between count data for the location in the distribution data and an expected count data for the location.

13. The computing device of claim 9 , wherein the recommendation comprises a recommendation for a good or service available to the grow operation.

14. The computing device of claim 9 , wherein the distribution data is generated using a machine learning model.

15. A non-transitory computer-readable media collectively storing computer-executable instructions that upon execution cause one or more computing devices to collectively perform acts comprising: receiving, from a visual observer device, image data, the image data being associated with a location within a grow operation; determining, from the image data, at least one pest classification and a count associated with the at least one pest classification; generating a distribution data based on the at least one pest classification, count, and location; determining a level of risk associated with the distribution data; based on the at least one pest classification, determining a threshold level of risk; comparing the level of risk associated with the distribution data to the threshold level of risk that is based on the at least one pest classification; based on comparing the level of risk associated with the distribution data to the threshold level of risk that is based on the at least one pest classification, determining that the level of risk associated with the distribution data is greater than a threshold level of risk that is based on the at least one pest classification; identifying a pest infestation associated with the pest classification based on the distribution data; based on determining that the level of risk associated with the distribution data is greater than a threshold level of risk that is based on the at least one pest classification, generating a recommendation that includes a set of potential solutions that address the pest infestation; based on the at least one pest classification, determining an effectiveness rating for each potential solution of the set of potential solutions that address the pest infestation; determining preferences associated with the grow operation; based on the preferences associated with the grow operation, determining a weight for each potential solution of the set of potential solutions; ranking the set of potential solutions based on the weight of each potential solution; comparing each effectiveness rating for each potential solution of the set of potential solutions to an effectiveness rating threshold; based on the ranking of the potential solutions and based on comparing each effectiveness rating for each potential solution of the set of potential solutions to an effectiveness rating threshold, selecting a subset of the set of potential solutions; providing, for output to a robotic device, the subset of the set of potential solutions and an instruction to implement the subset of the set of potential solutions, wherein the robotic device implements the subset of the set of potential solutions in response to receiving the instruction to implement the subset of the set of potential solutions; determining that the robotic device has implemented the subset of the set of potential solutions; receiving, from the visual observer device, additional image data, the additional image data being associated with the location within the grow operation; determining, from the image data, an updated count associated with the at least one pest classification; and based on the updated count associated with the at least one pest classification, updating each effectiveness rating for each potential solution of the subset of the set of potential solutions.

16. The non-transitory computer-readable media of claim 15 , wherein the distribution data is generated using one or more relationships between different locations within the grow operation.

17. The non-transitory computer-readable media of claim 15 , wherein the recommendation is provided to a horticultural management device operated by a user associated with the grow operation.

18. The non-transitory computer-readable media of claim 17 , wherein the recommendation comprises instructions to perform the subset of the set of potential solutions at a location associated with the pest infestation.

11631475 April 2023 Fuxman
20140279600 September 2014 Chait
20160078375 March 2016 Ethington et al.
20170273290 September 2017 Jay
20170287160 October 2017 Freudenberg
20190034736 January 2019 Bisberg
20190223431 July 2019 Rustia
20200253127 August 2020 McCall et al.
20210027056 January 2021 Koch
20210279923 September 2021 Singh
20210375390 December 2021 Fuxman
20220155054 May 2022 Humpston
20240016137 January 2024 Siepe
2021015468 January 2021

PCT Patent Application No. PCT/US2022/012450, International Search Report mailed May 16, 2022, 4 pages. cited by applicant
PCT Patent Application No. PCT/US2022/012450, Written Opinion mailed May 16, 2022, 5 pages. cited by applicant

edspgr.12106560