Crop health monitoring using predictive modeling

1. A method comprising:

receiving, using at least one processor, first sensor data over time, the first sensor data pertaining to plant-related parameters of each of multiple first plants;

storing the first sensor data in at least one memory;

identifying, using the at least one processor, an issue affecting at least one of the first plants;

analyzing, using the at least one processor, at least some of the stored first sensor data to generate a predictive model associated with the issue;

receiving, using the at least one processor, second sensor data pertaining to plant-related parameters of each of multiple second plants; and

identifying, using the at least one processor, at least one of the second plants to receive one or more interventions by applying the predictive model to the second sensor data.

2. The method of claim 1 , wherein:

the issue affecting the at least one first plant comprises the at least one first plant having a performance that is poor; and

identifying the at least one second plant comprises identifying any of the second plants having at least one of the plant-related parameters that is indicative of possible poor future performance.

3. The method of claim 2 , wherein the one or more interventions are selected to increase a predicted future performance of the at least one second plant above the poor performance of the at least one first plant.

4. The method of claim 1 , wherein:

the issue affecting the at least one first plant comprises the at least one first plant having a performance that is good; and

identifying the at least one second plant comprises identifying any of the second plants having at least one of the plant-related parameters that deviates from at least one of the plant-related parameters of the at least one first plant.

5. The method of claim 4 , wherein the one or more interventions are selected to increase a predicted future performance of the at least one second plant towards the good performance of the at least one first plant.

6. The method of claim 1 , further comprising at least one of:

causing the one or more interventions involving the at least one identified second plant to be performed; and

outputting, to a user, a notification suggesting that the one or more interventions involving the at least one identified second plant be performed.

7. The method of claim 1 , wherein analyzing at least some of the stored first sensor data to generate the predictive model comprises:

performing feature selection to identify attributes in the first sensor data that enhance an accuracy of the predictive model;

excluding any of the first sensor data that does not contribute to the accuracy of the predictive model or that decreases the accuracy of the predictive model; and

generating the predictive model based on the identified attributes and not based on the excluded first sensor data.

8. The method of claim 1 , wherein:

receiving the first sensor data comprises receiving (i) initial data during a pre-screening phase of at least some of the first plants and (ii) subsequent data during a secondary screening phase of one or more of the first plants that are suffering from at least one problem; and

the one or more first plants are selected for secondary screening based on at least some of the initial data.

9. The method of claim 1 , wherein identifying the issue affecting at least one of the first plants comprises:

assessing the first plants for one or more desired attributes; and

classifying the first plants into different classifications based on the assessment.

10. An apparatus comprising:

at least one interface configured to receive first sensor data over time, the first sensor data pertaining to plant-related parameters of each of multiple first plants, the at least one interface also configured to receive second sensor data pertaining to plant-related parameters of each of multiple second plants;

at least one memory configured to store the first sensor data; and

at least one processor configured to:

identify an issue affecting at least one of the first plants;

analyze at least some of the stored first sensor data to generate a predictive model associated with the issue; and

identify at least one of the second plants to receive one or more interventions by applying the predictive model to the second sensor data.

11. The apparatus of claim 10 , wherein:

the issue affecting the at least one first plant comprises the at least one first plant having a performance that is poor; and

to identify the at least one second plant, the at least one processor is configured to identify any of the second plants having at least one of the plant-related parameters that is indicative of possible poor future performance.

12. The apparatus of claim 11 , wherein the at least one processor is configured to select the one or more interventions to increase a predicted future performance of the at least one second plant above the poor performance of the at least one first plant.

13. The apparatus of claim 10 , wherein:

the issue affecting the at least one first plant comprises the at least one first plant having a performance that is good; and

to identify the at least one second plant, the at least one processor is configured to identify any of the second plants having at least one of the plant-related parameters that deviates from at least one of the plant-related parameters of the at least one first plant.

14. The apparatus of claim 13 , wherein the at least one processor is configured to select the one or more interventions to increase a predicted future performance of the at least one second plant towards the good performance of the at least one first plant.

15. The apparatus of claim 10 , wherein the at least one processor is further configured to at least one of:

cause the one or more interventions involving the at least one identified second plant to be performed; and

output, to a user, a notification suggesting that the one or more interventions involving the at least one identified second plant be performed.

16. The apparatus of claim 10 , wherein, to analyze at least some of the stored first sensor data to generate the predictive model, the at least one processor is configured to:

perform feature selection to identify attributes in the first sensor data that enhance an accuracy of the predictive model;

exclude any of the first sensor data that does not contribute to the accuracy of the predictive model or that decreases the accuracy of the predictive model; and

generate the predictive model based on the identified attributes and not based on the excluded first sensor data.

17. The apparatus of claim 10 , wherein:

the first sensor data comprises (i) initial data associated with a pre-screening phase of at least some of the first plants and (ii) subsequent data associated with a secondary screening phase of one or more of the first plants that are suffering from at least one problem; and

the at least one processor is configured to select the one or more first plants for secondary screening based on at least some of the initial data.

18. The apparatus of claim 10 , wherein, to identify the issue affecting at least one of the first plants, the at least one processor is configured to:

assess the first plants for one or more desired attributes; and

classify the first plants into different classifications based on the assessment.

19. A non-transitory computer readable medium containing instructions that when executed cause at least one processor to:

obtain first sensor data over time, the first sensor data pertaining to plant-related parameters of each of multiple first plants;

store the first sensor data in at least one memory;

identify an issue affecting at least one of the first plants;

analyze at least some of the stored first sensor data to generate a predictive model associated with the issue;

obtain second sensor data pertaining to plant-related parameters of each of multiple second plants; and

identify at least one of the second plants to receive one or more interventions by applying the predictive model to the second sensor data.

20. The non-transitory computer readable medium of claim 19 , wherein:

the issue affecting the at least one first plant comprises the at least one first plant having a performance that is poor; and

the instructions that when executed cause the at least one processor to identify the at least one second plant comprise:

instructions that when executed cause the at least one processor to identify any of the second plants having at least one of the plant-related parameters that is indicative of possible poor future performance.

21. The non-transitory computer readable medium of claim 20 , wherein the instructions when executed further cause the at least one processor to select the one or more interventions to increase a predicted future performance of the at least one second plant above the poor performance of the at least one first plant.

22. The non-transitory computer readable medium of claim 19 , wherein:

the issue affecting the at least one first plant comprises the at least one first plant having a performance that is good; and

the instructions that when executed cause the at least one processor to identify the at least one second plant comprise:

instructions that when executed cause the at least one processor to identify any of the second plants having at least one of the plant-related parameters that deviates from at least one of the plant-related parameters of the at least one first plant.

23. The non-transitory computer readable medium of claim 22 , wherein the instructions when executed further cause the at least one processor to select the one or more interventions to increase a predicted future performance of the at least one second plant towards the good performance of the at least one first plant.

24. The non-transitory computer readable medium of claim 19 , further containing instructions that when executed cause the at least one processor to at least one of:

cause the one or more interventions involving the at least one identified second plant to be performed; and

output, to a user, a notification suggesting that the one or more interventions involving the at least one identified second plant be performed.

25. The non-transitory computer readable medium of claim 19 , wherein the instructions that when executed cause the at least one processor to analyze at least some of the stored first sensor data to generate the predictive model comprise:

instructions that when executed cause the at least one processor to:

perform feature selection to identify attributes in the first sensor data that enhance an accuracy of the predictive model;

exclude any of the first sensor data that does not contribute to the accuracy of the predictive model or that decreases the accuracy of the predictive model; and

generate the predictive model based on the identified attributes and not based on the excluded first sensor data.

26. The non-transitory computer readable medium of claim 19 , wherein the first sensor data comprises (i) initial data associated with a pre-screening phase of at least some of the first plants and (ii) subsequent data associated with a secondary screening phase of one or more of the first plants that are suffering from at least one problem; and

further containing instructions that when executed cause the at least one processor to select the one or more first plants for secondary screening based on at least some of the initial data.

27. The non-transitory computer readable medium of claim 19 , wherein the instructions that when executed cause the at least one processor to identify the issue affecting at least one of the first plants comprise:

instructions that when executed cause the at least one processor to:

assess the first plants for one or more desired attributes; and

classify the first plants into different classifications based on the assessment.