Plant provenance and data products from computer object recognition driven tracking
This disclosure sets forth systems and techniques that trace plant provenance and determine a likelihood of plant spoilage within a plant growth operation. Particularly, a plant health monitoring system may monitor and record plant growth data of one or more individual plants during a plant life cycle from germination through to harvest. The plant growth data may include data relating to environmental conditions that surround the individual plants, and plant-specific data drawn from time-lapse images captured during the plant life cycle. A plant grower may use the time-lapse image data to identify changes in plant structure that indicate the degree of health of an individual plant, or the encompassing batch of plants. Further, a plant grower may use the plant growth data to generate one or more financial instruments that may be distributed to one or more financial institutions as support for securing capital to fund the plant growth operation.
1. A system, comprising:
one or more processors; and
memory coupled to the one or more processors, the memory including one or more modules that are executable by the one or more processors to:
receive plant growth data that corresponds to one or more individual plants in a batch of plants of a plant growth operation;
receive a first image of at least one plant of the one or more individual plants;
receive a second image of the at least one plant of the one or more individual plants, the second image being captured from a different angle than the first image;
based on the first image and the second image, generate, using stereoscopic analysis or structured light analysis, a three-dimensional model of the at least one plant of the one or more individual plants;
based on the plant growth data and the three-dimensional model of the at least one plant of the one or more individual plants, quantify a plant health metric that describes a likelihood of spoilage during a life cycle of the at least one plant of the one or more individual plants prior to harvest of the batch of plants;
predict a harvest date for the batch of plants based on the plant health metric;
establish a proposed price for a harvest of the batch of plants at the predicted harvest date;
generate a financial risk assessment of a current plant growth operation that includes growth of the batch of plants, the financial risk assessment quantifying the plant health metric, including risk data that is suitable for generating a financial instrument;
generate a financial instrument based at least in part on the financial risk assessment; and
output the financial instrument to a financial institution.
2. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
determine a similarity between the first image and a third image of an additional plant of the one or more individual plants in a data-store of previously stored plant images; and
generate the plant health metric for the batch of plants, the plant health metric being based at least in part on the similarity.
3. The system of claim 2 , wherein the one or more modules are further executable by the one or more processors to:
identify a first branch of the at least one plant of the one or more individual plants,
wherein the similarity between the first image and the third image is based at least in part on a comparison of the first branch and a second branch of the additional plant of the one or more individual plants.
4. The system of claim 2 , wherein the one or more modules are further executable by the one or more processors to:
identify a first leaf of the at least one plant of the one or more individual plants,
wherein the similarity between the first image and the third image is based at least in part on a comparison of the first leaf and a second leaf of the additional plant of the one or more individual plants.
5. The system of claim 1 , wherein the plant health metric numerically quantifies a defect in a leaf structure of the plant.
6. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
determine that the at least one plant has a defect; and
trace the provenance of the at least one plant to the batch of plants;
wherein the plant health metric describes a likelihood of spoilage of the batch of plants based at least in part on the defect of the at least one plant.
7. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
determine that multiple individual plants within the plant growth operation are found to show the same defect; and
trace the provenance of a particular individual plant of the multiple individual plants to a batch of plants;
wherein the plant health metric describes a likelihood of spoilage of the batch of plants that includes the particular individual plant based at least in part on the defect of the particular individual plant.
8. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
output the plant growth data.
9. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
predicate a date for planting a new crop based on the predicted harvest date; and
align a harvest date of the new crop with peak market conditions.
10. The system of claim 1 , wherein the one or more modules are further executable by the one or more processors to:
identify a change in the at least one plant; and
update the plant health metric for the at least one plant to reflect the identified change.
11. A server, comprising:
one or more processors; and
memory coupled to the one or more processors, the memory including one or more modules that are executable by the one or more processors to:
receive an indication that at least one individual plant within a plant growth operation has a defect;
trace an origin of the at least one individual plant to a batch of plants;
access plant growth data that corresponds to the batch of plants;
receive a first image of at least one plant of the batch of plants;
receive a second image of the at least one plant of the batch of plants, the second image being captured from a different angle than the first image;
based on the first image and the second image, generate, using stereoscopic analysis or structured light analysis, a three-dimensional model of the at least one plant of the batch of plants;
analyze the plant growth data and the three-dimensional model of the at least one plant of the batch of plants;
determine whether the plant growth data and the three-dimensional model of the at least one plant of the batch of plants indicates a sign of an impending defect in another plant within the batch of plants;
in response to determining that the plant growth data indicates a sign of an impending defect in another plant within the batch of plants, determine other individual plants within the batch of plants that are subject to the same defect; and
generate a financial risk assessment of an unharvested crop of plants in the plant growth operation, the financial risk assessment quantifying in financial terms the risk of plant spoilage of the unharvested crop based at least in part on the individual plants within the batch of plants that are subject to the same defect;
generate a financial instrument based at least in part on the financial risk assessment; and
output the financial instrument to a financial institution.
12. The server of claim 11 , wherein the defect includes a discoloration.
13. The server of claim 11 , wherein the plant growth data includes data of plant spoilage.
14. The server of claim 11 , wherein the one or more modules are executable by the one or more processors to:
receive a third image of a first plant of the at least one individual plant;
identify a dataset that includes a fourth image of a health condition of a second plant of the at least one individual plant, the second plant being a same plant type as the first plant;
determine a similarity between the third image and the fourth image; and
generate a plant health metric for the first plant that quantifies the risk of plant spoilage of the unharvested crop, the plant health metric being based at least in part on the similarity;
wherein the plant growth data is based at least in part on the plant health metric.
15. The server of claim 14 , wherein the one or more modules are executable by the one or more processors to:
determine an additional similarity between the third image and a fifth image stored in an additional dataset that includes individual images of additional plants within the batch of plants; and
assign a plant identifier to the third image, based at least in part on the additional similarity, the plant identifier being associated with the fifth image.
16. The server of claim 11 , wherein the batch of plants is already harvested.
17. One or more non-transitory computer-readable media storing computer-readable instructions, that when executed on one or more processors, cause the one or more processors to:
receive plant growth data that corresponds to one or more individual plants in a batch of plants of a plant growth operation;
receive a first image of at least one plant of the one or more individual plants;
receive a second image of the at least one plant of the one or more individual plants, the second image being captured from a different angle than the first image;
based on the first image and the second image, generate, using stereoscopic analysis or structured light analysis, a three-dimensional model of the at least one plant of the one or more individual plants;
based on the plant growth data and the three-dimensional model of the at least one plant of the one or more individual plants, quantify a plant health metric that describes a likelihood of spoilage during a life cycle of the at least one plant of the one or more individual plants prior to harvest of the batch of plants;
predict a harvest date for the batch of plants based on the plant health metric;
establish a proposed price for a harvest of the batch of plants at the predicted harvest date;
generate a financial risk assessment of a current plant growth operation that includes growth of the batch of plants, the financial risk assessment quantifying the plant health metric, including risk data that is suitable for generating a financial instrument;
generate a financial instrument based at least in part on the financial risk assessment; and
output the financial instrument to a financial institution.
18. The one or more non-transitory computer-readable media of claim 17 , wherein the computer-readable instructions, when executed on the one or more processors, cause the one or more processors to:
receive a plurality of images that correspond to the at least one plant, the plurality of images providing a time-lapse sequence of images that illustrate growth of the at least one plant over a predetermined period of time;
wherein the plant health metric is based at least in part on a rate of change in shape, size, or color of the at least one plant.
19. The one or more non-transitory computer-readable media of claim 17 , wherein the computer-readable instructions, when executed on the one or more processors, cause the one or more processors to:
access a reference lookup table that correlates an estimate price of harvest with harvest dates; and
determining the predicted harvest date based at least in part on data accessed from the reference lookup table.
20. The one or more non-transitory computer-readable media of claim 17 , wherein the computer-readable instructions, when executed on the one or more processors, cause the one or more processors to:
determine that multiple individual plants in the plant growth operation are found to exhibit the same defect; and
recall the multiple individual plants based at least in part on traceability of plant provenance of at least one of the multiple individual plants to a particular plant batch.