Method and system for harvesting a crop

Abstract

Crop management systems are provided to enable high-throughput harvesting and phenotyping of a field crop, such as sweet corn. Methods of high-throughput harvesting and phenotyping are also provided that enable market-need trait-driven genomic selection of plants grown on a plot. Plants selected by such systems and methods can be advanced into a breeding pipeline and / or moved forward for commercial use. The systems and methods enable the early advancement of plants displaying a high desired trait phenotype value, such as for a marketability trait.

Description

[0001] METHOD AND SYSTEM FOR HARVESTING A CROP

[0002] Field of the invention

[0003] The presently disclosed subject matter relates generally to systems, apparatus, and methods for phenotyping and harvesting a field crop, such as a com crop. In particular embodiments, the systems, apparatus, and methods are applicable to the phenotyping and harvesting of crops grown on research plots or trialing plots.

[0004] Background of the invention

[0005] The invention relates generally to a system, apparatus, and methods of use thereof, that enable high-throughput harvesting and phenotyping of a crop, particularly sweet com, to enable market-need trait-driven genomic selection of plants. The selected plants may be moved onwards into a breeding pipeline and / or moved forward for commercial use.

[0006] Data collected for phenotype analysis and selection of sweet corn plants following trialing, in a research plot for example, is typically done when the ears are mature (typically 21 days after pollination). At this time, the sweet corn plant is still actively developing, and harvest data assessments need to be made in as short a time as possible to prevent the harvested date from becoming a confounding factor. Ideally, when possible, the entire location (e.g., the entire plot) needs to be harvested in a day or two.

[0007] To perform the analysis, a human subject goes into the plot and selects, at random, a number of ears that is expected to be representative of the entire plot. In one example, the subject may select 5 “typical” ears from a plot and then, based on an analysis of these 5 ears (e.g., using estimated averages), make an assessment for all plants of the plot. The analyses typically include a mix of objective measurements and subjective assessments.

[0008] Applicants have recognized that there may be potential issues with such an approach. As a first example, larger sample sizes (e.g., 10 or more ears per plot) may provide a more accurate representation of the true plot average. However, acquiring larger sample sizes is more labor and time intensive. As a second example, reliance of some of the analyses on subjective assessments may skew the results due to biases in sample selection and estimation.

[0009] Accordingly, there is a need to provide higher quality trialing plot data, for at least sweet com crop analysis, where sample size can be easily scaled up, where a balance between accuracy and labor intensity can be achieved, and where trait-significant data can be gathered with reduced subjectivity and biases.

[0010] Summary of the invention

[0011] According to certain aspects of the present disclosure, systems, apparatus, and methods for harvesting a crop, such as a sweet corn crop, are disclosed. In particular aspects, the systems, apparatus, and methods enable phenotyping and harvesting of a sweet corn crop grown on a research or trialing plot. Plants selected using the system, apparatus, and / or methods disclosed herein can be moved into a breeding pipeline for commercial purposes. In one example embodiment, modifications to existing harvest machinery are disclosed that allow for precise optical, mass, and positional information to be gathered during harvesting, while enabling the equipment to be transportable and easily deployed in a research trialing network. By relying on high throughput, quantitative measurements, research plot data collection is improved, thereby providing crop breeders with data that can be better relied on for making selection, breeding and advancement decisions. In example embodiments, sweet corn crop data collection and analysis can be used to assess for marketability trait and advance sweet corn plants with high marketability earlier in a breeding pipeline. In particular embodiments, the system, apparatus, and methods disclosed herein allow for automated sample collection and data tracking, such as by using precision GPS and plot barcode information.

[0012] Aspects of the invention include harvesters configured for harvesting and phenotyping a crop grown in a plot. Example embodiments of the invention relate to a harvester, comprising a harvester head for engaging a corn plant and removing one or more corn ears from the engaged com plant; a collection platform; a collection container operatively coupled to the collection platform for receiving the one or more removed corn ears; a first sensor configured to sense receipt of the one or more removed corn ears at the collection platform or the collection container; a second geo-positioning sensor; and a controller comprising one or more processors communicatively coupled to the first and second sensors and configurable to execute instructions stored in computer readable storage media. In particular embodiments, the instructions comprise: continuously collecting output from the first sensor indicative of receipt of one or more removed corn ears from the com plant at the collection platform or the collection container; and optionally associating the output of the first sensor with an output of the second sensor indicative of a position of the corn plant within the plot. In embodiments, the harvester is further configurable to execute instructions comprising: in response to an operator input, discarding the removed one or more ears from the collection platform and discarding the output of the first and second sensor. Herein, the harvester can be operated in a first mode to collect the ears upon harvesting ears from a com plant and retain the output from the first and second sensor associated with the collected ear(s); and operating in a second mode to discard the ears upon harvesting ears from a com plant and discard any output from the first and second sensor associated with the collected ear(s).

[0013] In embodiments, the harvester continuously collecting output from the first sensor includes continuously collecting first sensor output and extracting individual corn ear weight from the output.

[0014] Aspects of the invention further include methods of operating a harvester configured for harvesting and phenotyping a crop grown in a plot. Example embodiments of such a method include a method for collecting and associating com information, the method comprising: harvesting an unhusked com ear from a com plant of a corn plot; weighing the corn ear and collecting the weight; collecting a field position of the corn plant; associating the weight of the com ear with the field position of the corn plant; uniquely identifying the com ear, the corn plant, and / or the corn plot; imaging the unhusked corn ear with an imaging device capable of providing an internal image of the corn ear; and extracting from the image, trait phenotype data for the com ear. In particular embodiments, the corn plant is a sweet com plant. In particular embodiments, the collecting and associating is performed continuously while a harvester is operated through the corn plot. In particular embodiments, the collecting and associating is performed substantially concurrently.

[0015] In embodiments, the method further comprises associating the trait phenotype data of the com ear with the weight of the com ear and field position of the corn plant. In particular embodiments, uniquely identifying the corn ear, the corn plant, and / or the com plot includes assigning a unique identifier to the com ear, the corn plant, and / or the corn plot, and the method further comprises associating the trait phenotype data of the com ear with the unique identifier. Aspects of the invention include imaging systems for phenotyping a crop. Embodiments of such a crop imaging system comprise: an image capturing device; and a processor communicatively coupled to the imaging device, the processor configured to execute instructions to: capture an image of a crop sample; and extract, from the captured image, data associated with a trait of the crop sample; and based on the data, assign an individual value to each of one or more attributes associated with the trait of the crop sample. In embodiments, the processor is further configured to capture an image for a plurality of crop samples harvested from a common plot, and estimate, from the individual value of each of the more or more attributes of each crop sample for the common plot, an average value of each of attribute associated with the trait of the crop harvested from the common plot. In particular embodiments, the crop is sweet corn. In particular embodiments, the image capturing device is an imaging device capable of providing an internal image of the corn ear, for example, where the image capturing device is an X-ray image capturing device. In other particular embodiments, the image capturing device is any one of a UV, visible, NIR, CT, or MR image capturing device, or combinations thereof and / or combined with the X-ray imaging capability. In particular embodiments, the trait is a marketability phenotype. In particular embodiments, the one or more attributes associated with the marketability phenotype includes shank length, shank diameter, husk length, ear length, and ear width.

[0016] Aspects of the invention also relate to a crop management system capable of harvesting and phenotyping a crop. An example embodiment of a crop management system comprises: a harvesting sub-system comprising a harvester head for engaging a com plant and removing one or more unhusked corn ears from the engaged corn plant; an imaging sub-system comprising an image capturing device; one or more sensors including a first sensor for sensing collection of the com ears at the harvesting sub-system and a second sensor for sensing a position of the harvesting sub-system; and a processor with code for executing instructions to sense collection of the removed one or more com ears in the harvesting subsystem. In embodiments, the crop management system further comprises a transfer sub-system for transferring the corn ears from the harvesting sub-system to the imaging sub-system. In example embodiments, the transfer sub-system comprises a conveyor belt.

[0017] Aspects of the invention also relate to method of operating a crop management system, including operating a harvesting sub-system and / or an imaging sub-system of the crop management system, to harvest and phenotype a crop. An example embodiment of a method of collecting a plant phenotype comprises: obtaining an image of an unhusked corn ear comprising a com cob; and extracting from the image, trait significant phenotype data for the com cob. In embodiments, the method further comprises assessing a phenotype of the com cob based on the extracted image. In embodiments, the method further comprises, in response to the assessed phenotype matching a desired phenotype, advancing seeds from the corn ear in a breeding pipeline. In embodiments, the method further comprises, in response to the assessed phenotype matching a desired phenotype, assigning a phenotype value to the corn ear, the corn cob, and / or the corn plant. In particular embodiments, the corn is sweet com. In particular embodiments, the phenotype is a marketability trait and phenotype data includes ear length, ear width, or shank length. In embodiments of the method, the image of the unhusked corn ear is obtained via an imaging sub-system comprising a camera, optical sensor, or other known imaging device; and the phenotype of the com is assessed via a controller coupled to the imaging sub-system configured with code for extracting and / or processing the trait significant phenotype data from the image and assessing the phenotype based on the data.

[0018] Aspects of the invention further include a method of collecting a phenotype for a sweet com plant. An example embodiment of such a method comprises: obtaining an image of an unhusked com ear comprising a corn cob from a sweet com plant grown on a plot; extracting from the image, data associated with a marketability phenotype for the corn cob; and assigning a marketability value for the sweet corn cob, plant and / or plot based on data extracted from the image.

[0019] Aspects of the invention further include a breeding method comprising: harvesting com ears from corn plants of a plot; obtaining an image of each com ear; extracting from the image of each com ear, trait significant phenotype data for the com cob; estimating a statistical traitsignificant phenotype value for com plants of the plot; and based on the statistical value, advancing one or more corn plants of the plot into a breeding pipeline. In particular embodiments, the com ears are harvested from corn plants of a trialing plot, and their traitsignificant phenotype is assessed for determining whether to advance them into a breeding pipeline for commercial purposes. In particular embodiments, the trait significant phenotype is a marketability phenotype. In embodiments, com plants having a statistical trait-significant phenotype value that is higher than a threshold are selected for advancement into a breeding pipeline. In example embodiments, the threshold is selected based on a desired trait significant phenotype, such as a marketability phenotype for com cob. Brief Description of the Drawings

[0020] FIG. l is a high-level block diagram of a crop management system, herein also referred to as the “PhenoHarvester”, for high-throughput harvesting and phenotyping of a crop grown on a plot, such as sweet corn, to enable market-need trait-driven genomic selection of plants, in accordance with the present disclosure.

[0021] FIG. 2 shows an image of a precision harvester in accordance with the present disclosure, herein also referred to as a “PhenoPuller”, for harvesting sweet com from a research plot.

[0022] FIG. 3 shows an example embodiment of the PhenoPuller of the present disclosure comprising modifications to an existing harvester (such as the OxBo CP 100) that enable acquisition of higher quality trialing plot data.

[0023] FIG. 4 shows an example embodiment of an adjustable RAM mount of the PhenoPuller.

[0024] FIGS. 5-6 depict an X-ray imaging system, herein also referred to as a “PhenoProcessor”, that is couplable to the precision harvester for gathering trait-significant data with high accuracy from a harvested crop. FIG. 5 shows an example housing for the X-ray imaging equipment and associated support equipment, while FIG. 6 shows an example X-ray imaging equipment.

[0025] FIG. 7 is an example X-ray image of a harvester sweet com ear. The image is analyzed for extracting trait-significant ear data such as “marketability” phenotype.

[0026] FIG. 8 is an example X-ray image of a harvester sweet com ear being analyzed for attributes associated with a “marketability” phenotype.

[0027] FIG. 9 is a high-level embodiment of a portable crop imaging system that may be used for high accuracy phenotyping of harvested com ears on a field.

[0028] FIG. 10 is an example embodiment of a portable version of the Phenoprocessor that may be used for portable high speed x-ray phenotyping of harvested corn ears.

[0029] FIG. 11 is an example method of operating a crop management system in accordance with the present disclosure.

[0030] Detailed Description

[0031] Systems are disclosed that enable high-throughput harvesting of a crop from a research plot and phenotyping of the harvested crop for trait-significant data collection and analysis. The system enables trait-driven genomic selection of plants that can be advanced into a breeding pipeline at an early stage. Particular embodiments of the system enable high throughput harvesting of ears of a sweet com crop from a research plot and data collection related to sweet com significant traits such as a marketability phenotype. In particular embodiments, the system comprises one or more sub-systems, one or more apparatuses, or combinations thereof.

[0032] FIG. 1 shows a system 10 for high-throughput harvesting of a crop from a research plot and phenotyping of the harvested crop for trait-significant data collection and analysis. In one embodiment, the system 10 of FIG. 1 is referred to herein as the PhenoHarvester.

[0033] The PhenoHarvester comprises a first sub-system 100, herein also referred to as a “PhenoPuller”, for precision harvesting of the crop. As detailed below, the PhenoPuller comprises one or more apparatuses for precision harvesting of ears of a sweet corn crop from a research plot. In particular embodiments, the PhenoPuller comprises a sweet com harvester conversion modified to operate as a high throughout research plot harvester, for example, through the addition of sensors and a sample collection system.

[0034] The PhenoPuller for precision harvesting is couplable to a second sub-system 200, herein also referred to as a “PhenoProcessor”, for high accuracy image capture of trait-significant data from the harvested crop. In one particular embodiment, as described herein, the PhenoProcessor comprises one or more apparatuses for high accuracy X-ray imaging of harvested sweet corn ears. In other embodiments, the high accuracy imaging comprises high accuracy red-green-blue or RGB imaging, greyscale imaging, ultraviolet or UV imaging, Near Infra-Red or NIR imaging, any other imaging technology known in the art, and / or combinations thereof.

[0035] The PhenoProcessor 200 may further comprise one or more processors configurable to execute instructions stored in non-transitory computer readable storage media for analyzing the images of the harvester crop and extracting trait-significant data extraction from the images. In alternate embodiments, the PhenoProcessor may be couplable to a third sub-system comprising one or more processors for image analysis and trait-significant data extraction from the images of the harvested crop. As used herein, the software module 300 used for image analysis is referred to as “SweetTEA” (Sweetcorn Trait Extraction Algorithm). In one particular embodiment, SweetTEA comprises computer-readable code that is implemented via the software module 300 to analyze X-ray images of harvested sweet corn ears for data related to a marketability phenotype. As used herein, the marketability phenotype refers to a sweet corn ear trait pertaining to ear size that enables an industry-set number of ears to be packaged in an industry standard shipping box. The marketability trait comprises multiple ear size aspects including, but not limited to, shank length, shank diameter, husk length, ear length, ear width, etc.

[0036] In this way, aspects of the invention relate to a crop management system capable of harvesting and phenotyping a crop, such as the PhenoHarvester of the invention. An example embodiment of a crop management system comprises: a harvesting sub-system comprising a harvester head for engaging a corn plant and removing one or more unhusked corn ears from the engaged com plant; an imaging sub-system comprising an image capturing device; one or more sensors including a first sensor for sensing collection of the com ears at the harvesting subsystem and a second sensor for sensing a position of the harvesting sub-system; and a processor with code for executing instructions to sense collection of the removed one or more com ears in the harvesting subsystem. In embodiments, the crop management system further comprises a transfer sub-system for transferring the corn ears from the harvesting sub-system to the imaging sub-system.

[0037] Aspects of the invention also relate to method of operating a crop management system, including operating a harvesting sub-system and / or an imaging sub-system of the crop management system, to harvest and phenotype a crop, such as a crop planted in a plot. An example embodiment of a method of collecting a plant phenotype comprises obtaining an image of an unhusked com ear comprising a com cob; and extracting from the image, trait significant phenotype data for the com cob. In embodiments, the method further comprises assessing a phenotype of the corn cob based on the extracted image. In embodiments, the method further comprises, in response to the assessed phenotype matching a desired phenotype, advancing seeds from the com ear in a breeding pipeline. In particular embodiments, advancing seeds in the breeding pipeline comprises selecting com seeds having a desired marketability phenotype, growing com plants from the selected seeds, and crossing the com plant comprising the desired marketability phenotype with another corn plant not comprising the desired marketability phenotype but comprising an alternate desirable agronomic performance trait (such as yield). In embodiments, the method further comprises, in response to the assessed phenotype matching a desired phenotype, assigning a phenotype value to the corn ear, the corn cob, and / or the corn plant. In particular embodiments, the com is sweet corn. In particular embodiments, the phenotype is a marketability trait and the phenotype data used to assess the marketability trait and assign a marketability trait phenotype value to the corn ear or cob and / or plant includes ear length, ear width, or shank length. In specific embodiments, the phenotype data of a statistically significant number of corn ears harvested from a com plant is used to assess the marketability trait of the corn plant and assign a marketability trait phenotype value to the corn plant. In other specific embodiments, a marketability trait phenotype value of a given com plant may be determined based on a weighted or unweighted average, mean, median, or mode value of ear length, ear width, and / or shank length data of a statistically significant number (e.g., threshold number) of com ears harvested from a com plant.

[0038] PhenoPuller

[0039] Turning first to the Precision Harvesting sub-system 100, or PhenoPuller 100, it is responsible for harvesting com and collecting key yield and yield-related data. In one example embodiment, as shown at FIGS. 2-4, the PhenoPuller is created by conversion of an existing commercially available sweet corn (SWC) harvesting equipment to a research plot harvester by modifying the machine and adding sensors and a sample collection system. In the example embodiment of PhenoPuller 100 depicted at FIGS. 2-4, an existing harvester, the OxBo CP100 in the depicted example, has been modified to enable acquisition of higher quality trialing plot data. The modified harvester 350 (herein also referred to as Phenopuller 100) can be coupled, via a coupling device 354, such as a hitch, to a vehicle 352, such as a tractor, that moves the harvester 350 (or Phenopuller 100) through a field. As the tractor-driven harvester is pulled through a plot 330 comprising corn plants, ears are harvested from the plants and transferred onto a conveyor belt 356. As such, there are two basic sweet corn (SWC) markets: Fresh and Processor. Commercial fresh market ears are typically harvested by hand by crews and the ears packed into boxes and shipped to stores or sold at market stands. Processor SWC is typically harvested by a machine that uses a conventional stripper-plate header which is fast and effective for the short trip to the processing facility but would damage the tender fresh market ears. Commercial mechanical fresh market sweet com harvesters are available in the marker. One example of such as a commercially available fresh market SWC harvester manufacturer is the OxBo CP- 100 PixAll. In the embodiment of Phenopuller 100 shown in FIGS. 2-4, an OxBo CP- 100 SWC harvester was modified for use as a research trial harvester. In general, commercial SWC harvesters comprise a grabbing and cutting system 357 designed to grab a com stalk above the ear using a pair of moving gripper belts 358, cut it below the ear with a spinning disk, and then pull the stalk between a pair of counter rotating rollers 360 which remove the ears and drop it onto a conveyor belt 356. The empty stalk is discarded next to the machine, and a fan removes any loose leaves or tassel parts that may have broken off. The ears then fall onto a conveyor belt and are piled on a trailer 320 that is pulled behind the harvester. Based on the model, the modified harvester 350 (or Phenopuller 100) may have 1 to 8 rows, either pulled behind or self- propelled.

[0040] Modifications made herein to a commercially available harvester to create the Phenopuller 100 and enable research plot harvesting are described as follows. Autosteering system 390 and capabilities Conventional harvesters are able to remove sweet corn ears from the plants with little damage and minimal foreign material (leaves / stalks). This aspect of the equipment was left unmodified as well as the tractor 352 that is used to operate it. In particular embodiments, the harvester 350 is modified to include an automated steering system 390 configured with real-time kinematic geopositioning capabilities that provides highly accurate positioning data. In particular embodiments, the harvester 350 comprises an AgLeader real-time kinematic (RTK) - Geopositioning (GPS) Autosteer system 390 coupled to the tractor 352 to assist the driver in focusing on the machine operation while keeping the harvester on course. The autosteer system 390 may include positional information (e.g., via GPS data) and enables the harvester to be automatically steered in accordance with a selected route through the plot. Use of the RTK autosteer 390 feature improves accuracy by accounting for errors in satellite signals, such as atmospheric interference and clock inaccuracies. If the trials are planted using autosteer, then the harvester can easily be set to follow the same straight line with inch accuracy. This improves the accuracy and efficiency of harvesting ears through the plot. In addition, since the autosteer system include positional information, the position of each harvested plant is known with reference to a position of the plant in the plot. This can also make it easier to determine the identity of any plants that may be eventually selected for advancing in a breeding pipe. Upper Guide Belt Height Adjuster 392

[0041] As part of the harvester settings that can be made in the field, there is a manual handwheel adjustment for setting the height of the upper guide belts that grip and pull the stalks through the machine after stem cutting. Normally a grower only has one size of corn (and typically one hybrid) in their field to harvest, so only a few adjustments would be needed. This is not so with research plots where potentially hundreds of different hybrids with potentially as many different plant heights are encountered. For this reason, the manual handwheel has been replaced in particular embodiments with a hydraulic motor that can be adjusted from the driver’s seat in the cab. Based on input received from a user in the tractor cab, the operating height of cutting system 357, at which stalks are engaged and cut, can be adjusted (e.g., raised or lowered). Camera system 394

[0042] The vertical position of the front of the harvester (both guide belt and cutting system together) is adjustable using a factory hydraulic gauge wheel system. This allows the machine to follow ground contours and for the driver to lower the cutting wheel to an optimal position just below the primary ear. In doing so, the operating height of cutting system 357 can be adjusted. Typically, the com in a farmer’s field is about the same height and all the same hybrid so little adjustment needs to be made. However, to perform well in a research plot field, it needs to be adjusted at nearly every plot. In particular embodiments, the harvester system include a closed- circuit camera system 394 that captures a harvest window and aids the driver in making both the cutter height and upper guide belt adjustments in an ergonomically improved way. From the driver’s seat in the tractor 352 it may be difficult to judge the position of these parts due to the visual parallax from the seat in the cab. By moving the camera system 394 to a more orthogonal position and putting the camera system display in a more neutral forward position for the driver, a better ergonomic position can be provided for the driver and more accurate and repeatable settings can be obtained for the harvester. The camera system also aids in the driver observing the blind spots around the machine and also the rear operator station (which is obstructed from driver view by the machine).

[0043] Load transport selector 396

[0044] Factory settings of conventional harvesters include a lever and a pin to select between operation of the harvester in a closed “transport” position and an open “operate” position. The factory default method typically requires an operator in the tractor cab to pull back on a spring- loaded pin on the drawbar and one (or several) others to push the cutting system 357 into position so the pin pops into place. This operation can have several drawbacks. Accordingly, Applicants have modified this aspect by removing the pin and hole system and replacing it with a hydraulic cylinder with a selectable lock. This allows the harvester to be switched between an “operate position” (where the cutting system is operated to harvest corn ears) and a “transport position” (where the cutting system of the harvester is not operational) from the cab of the tractor 352 with no involvement of additional crew on the ground. Furthermore, it can be locked in any position in between the operate and transport positions to enable optimal tracking of the harvester behind the tractor.

[0045] On-Board Power 398

[0046] To provide power for the electronics, the harvester 350 further comprises one or more or each of a portable generator, air compressor, and a deep cycle battery with a charger. Each of these, when included, are mounted on the harvester. These components provide for an uninterruptible source of DC (12V nom) power to run the electronics and compressed air to operate air actuators.

[0047] Remote E-Stop(s) 364

[0048] The tractor 352 is used to power the harvester via the power take-off (PTO) output (in one embodiment, the John Deere 4066R tractor) is equipped with an electro-mechanical PTO engagement system. Modifications to the harvester include a modified circuit on the tractor so that the on-off control of the PTO can be extended from the position of a rear operator 362. The updated wiring allows any station to stop the PTO shaft by pressing an Emergency Stop switch 364 and it will only restart if all the switches are in the operate mode. This is a valuable safety improvement.

[0049] Product conveyor extension

[0050] The primary product conveyor was lengthened (in one example embodiment, by 18”) to allow for more clearance for the ears to fall into a collection bin 360.

[0051] Rear Work Platform 370

[0052] Harvester 100 includes a rear work platform 370 that allows an operator 362 to load empty bins 360, scan barcodes 372 on the bins, monitor the harvest process, and remove full bins from the weighing system 374. This platform is large enough to accommodate many totes / bins and the operator station 376. It is also equipped with safety railing and operator seating. Movable Load Cell Platform 380

[0053] As the ears drop from the end of the conveyor 356, they are guided by a metal chute 378 and fall into a lightweight plastic tub or bin 360 (such as for example, a Uline S-2773B-Y plastic bin) that is placed on the platform 380 of weighing system 374. In a particular embodiment, the weighing system 370 is an industrial balance (such as for example , a Mettler-Toledo balance). The load cell platform 380 of the weighing system 374 is on a slide mounted tray. In embodiments, the slide mounted tray allows for the platform to be moved to a rear “sample collection” setting where the ears will fall in during sample collection, or moved forward to a “bypass” setting where the tray is pulled ahead and an opening in the floor of the work platform is revealed that lets the ears drop on the ground and onto the plot / field where the ears were harvested from. The bypass setting is particularly useful in areas of the border areas of the field that need to be picked but no data or samples are needed. This modification enables the harvesting system (or PhenoPuller) 100 to be operated in a first mode where ears are harvested and collected in a collection bin 360 from where they are directed to the PhenoProcessor for data collection, and a second mode where ears are harvested and discarded and from where they are not directed to the imaging system 200 (with reference to FIG. 1) or PhenoProcessor and no data is collected for these ear samples.

[0054] Continuous Weighing System 374

[0055] Typical commercial load cell-based balances wait for stability of a load before they display the object’s weight or send the data to a printer / computer. In particular embodiments, a load cell platform 380 is included in the harvesting system that in operated to be continuously transmitting the weight readings of the harvested ears collected in the bins to an on-board computer or processor for logging weight data of the collected ears. The resulting “step” change data in the signal at the weighing system can be advantageously used to count the number of ears that fell into the sample bin over the course of the plot harvest. Software is then used to analyze or process the load cell data to identify and distinguish single ear drops from multiple ear drops. In further embodiments, the load cell data received from the weighing station is additionally used to measure individual ear weights along with an average ear weight. Ear to Ear uniformity in size and weight is a key trait for commercial sweet corn as producers prefer hybrids that make consistent ears sizes (in terms of length, width, and weight). At least a threshold number of ears need to be collected individually to measure uniformity. An ear count and a total weight would only result in an average ear weight without divulging information on the variation in ear-to-ear weight. In comparison, the continuously streaming weight data collection and analysis feature enables assessment of key phenotypic traits.

[0056] Data Collection and Sample Tracking During data collection and tracking, operator input is received at the harvesting system to indicate the start and / or stop of each plot. This allows data for each plot to be individually tracked and phenotype assessments to be accurately made for plants associated with each plot. In particular embodiments, as the harvester and associated cutting machine progresses through the field, the operator changes the sample bin 360 placed on the load cell platform 380 and provides input into a computer of the harvesting system indicative of an upcoming new plot. In one embodiment, this may be done using a barcode scanner and a sample ID card that is located at the operator workstation 376. The cards can be printed in advance of the harvest and placed in the expected harvest order that the operator can access. Once driver and operator are ready and the harvester is adjusted and aligned with the next plot to be harvested, the operator scans the correct plot barcode into software that is running on a computer or processor on-board the harvester (e.g., a Windows Toughpad) located at the operator station. This will initialize the software to begin collecting a new batch of data and the driver can start moving down the plot to the next alley between plots. While the machine is operating, the load cell data is recorded by the computer along with a precision GNSS signal being transmitted by a GPS sensor (e.g., a Trimble SPS986 GPS sensor) located above the cutting head. The positional data from the GPS is used to measure the area of the plot that the sample was collected from and the ears falling from the conveyor into the sample bin enable an individual and collective weight of the harvestable ears from the plot to be measured. When the driver observes the last plant of the plot has been cut by the cutting system of the harvester, they may pause to wait for verbal confirmation from the operator that the ears have transited the machine and that the bin of ears has been removed from the load cell and a new empty bin is in place. The operator then places the barcode card that was used for the prior plot in the tub along with the ears to maintain identity of which plot these ears are from. Bins of ears can then be transported to other data collection workstations in the field or back to the research facility for further testing. Rear operator automation

[0057] In one particular embodiment, the rear operator’s role is replaced with robotic automation. It embodiments where only the ear weight and position data are needed and not the samples themselves, a self-dumping and resetting system can be installed on the load cell platform, for example a system that uses an air-actuated piston and door setup to push the ears out of a permanent sample bin 360 in place on the load cell. This would reduce the need for manual resetting of the load cell. With this type of setup, it would also be possible to collect data continuously and allow the computer to determine when to purge and re-zero the load cell based on weight. Without stopping the harvester, it would be possible to create a precision Yield Map of the harvested area wherein weight per unit area can be visualized on the display. In addition, one or more other plot features and attributes, such as population type (e.g., hybrid or inbred) or agronomic treatment (e.g. fertilizer or biological application) could be overlaid and used to select a plant (e.g., identify a higher yielding plant) based on a combination of factors and phenotypes. Data Output Expected

[0058] After harvest, the data stored in the processor on-board the harvester may be exported (e.g., to an off-board or remote processor, or to a cloud processor) and processed to generate a report that includes plot attributes such as: Plot ID, Harvested plot length (m), Total harvest weight (g), Number of ears in harvested plot area, Average ear mass (g), measure of ear to ear variation (Std Dev in g), Number of marketable ears (ears that were in specified weight parameters), Marketable yield (expressed in T / ha and in number of cases of 48 marketable ears per acre), and so on. GIS output (e.g., as a JSON file) describing plot area vectors may be overlayed onto UAV orthomosaic images or other raster or vector layers including, but not limited to, precision planter seed drop data or Plant Stand Analyzer GIS data.

[0059] In one example embodiment, the harvest rate target for the PhenoPuller is 750 single row plots per day (in contrast to manual options which typically are at a rate of 150 plots per day, evaluating a subset of 5 ears per plot). Typically, a crew of 2 people will look at -1500 ears in a day, whereas the PhenoPuller can objectively measure >20,000 ears with same 2 crew members. In another example embodiment, the PhenoPuller is configured to harvest a whole plot in -30 seconds, and harvest -100 plots per hour.

[0060] In addition to increased speed, the ability to collect and organize the sample bins for downstream processing is a significant advantage. For example, the barcode tag that is used to associate the plot with the data being collected can be placed in the sample bin with the collected ears and the bin can be taken to other remote or field edge processing locations. If sample collection is not needed, and only the spatial ear count and weight data are desired (e.g., for a straight yield trial), the PhenoPuller can be configured to collect the data for a plot and then dump the ears on the ground in preparation for the next plot. This may be more beneficial in later stages of product development where a multi -location yield trial is desired and other non- yield traits have already been selected for. This “Collect and Dump” mode may be operated by a single operator who is both driving the tractor and monitoring the load cell system remotely from the cab. The PhenoPuller can also be configured with features to improve portability. For example, the PhenoPuller can be configured to include a drop-deck gooseneck trailer so that it can be transported similar to typical farm equipment. In embodiments, the rear platform may also include an adjustable RAM mount 359 for the on-board processor.

[0061] In this way, a commercially available harvester may be modified to create a precision harvester, or PhenoPuller, that includes technological adaptations that improve the efficiency of research plot crop data collection through automation. As a result, the overall PhenoHarvester system is able to increase the rate of genetic gain for breeding programs. Specifically, the increased speed and efficiency in obtaining ear counts and weights on single row research plots can facilitate collection of quantitative yield performance data earlier in the product development pipeline. Not only can this increase the rate of genetic gain immediately by enabling selection earlier in the hybrid testing pipeline, but also via contribution over time to standardized, robust datasets from which relationships between yield performance, genotypes, and environments can be modeled and used to implement predictive breeding throughout the product development pipeline.

[0062] Aspects of the invention include harvesters configured for harvesting and phenotyping a crop grown in a plot, such as the PhenoPuller. Example embodiments of the invention relate to a harvester, comprising a harvester head for engaging a com plant and removing one or more com ears from the engaged corn plant; a collection platform; a collection container operatively coupled to the collection platform for receiving the one or more removed com ears; a first sensor configured to sense receipt of the one or more removed com ears at the collection platform or the collection container; a second geo-positionary sensor; and a controller comprising one or more processors communicatively coupled to the first and second sensors and configurable to execute instructions stored in computer readable storage media. In particular embodiments, the instructions comprise: continuously collecting output from the first sensor indicative of receipt of one or more removed corn ears from the com plant at the collection platform or the collection container; and optionally associating the output of the first sensor with an output of the second sensor indicative of a position of the com plant within the plot. In embodiments, the harvester is further configurable to execute instructions comprising: in response to an operator input, discarding the removed one or more ears from the collection platform and discarding the output of the first and second sensor. Herein, the harvester can be operated in a first mode to collect the ears upon harvesting ears from a com plant and retain the output from the first and second sensor associated with the collected ear(s); and operating in a second mode to discard the ears upon harvesting ears from a com plant and discard any output from the first and second sensor associated with the collected ear(s).

[0063] In embodiments, the harvester continuously collecting output from the first sensor includes continuously collecting first sensor output and extracting individual corn ear weight from the output.

[0064] Aspects of the invention further include methods of operating a harvester configured for harvesting and phenotyping a crop grown in a plot. Example embodiments of such a method include a method for collecting and associating com information, the method comprising: harvesting an unhusked com ear from a com plant of a corn plot; weighing the corn ear and collecting the weight; collecting a field position of the corn plant; associating the weight of the com ear with the field position of the corn plant; uniquely identifying the com ear, the corn plant, and / or the corn plot; imaging the unhusked corn ear with an imaging device capable of providing an internal image of the corn ear; and extracting from the image, trait phenotype data for the com ear. In particular embodiments, the corn plant is a sweet com plant. In particular embodiments, the collecting and associating is performed continuously while a harvester is operated through the corn plot. In particular embodiments, the collecting and associating is performed substantially concurrently.

[0065] In embodiments, the method further comprises associating the trait phenotype data of the com ear with the weight of the com ear and field position of the corn plant. In particular embodiments, uniquely identifying the corn ear, the corn plant, and / or the com plot includes assigning a unique identifier to the com ear, the corn plant, and / or the corn plot, and the method further comprises associating the trait phenotype data of the com ear with the unique identifier.

[0066] PhenoProcessor

[0067] Turning now to the PhenoProcessor 200 of FIG. 1, it is configured as a Precision Imaging sub-system of the Harvesting system 10 or PhenoHarvester. In one example embodiment, as shown in FIGS. 5-6, the PhenoProcessor 200 is configured as a portable, field edge operating, high resolution imaging system 500 that is placed in the field near the PhenoPuller and plot ear samples are delivered to it therefrom (such as by members of a harvest team or via a conveyor belt or other known transfer mechanism). In the depicted embodiment (FIG. 6), the imaging system comprises, within housing 502 (depited herein as a trailer), an X-ray imaging system 504 (such as a Loma X-5 Space Saver industrial cabinet X-ray system). In the depicted embodiment the imaging system operates as a Field Research Platform (FiRP), or trailer 502, housing the imaging equipment 504. The FiRP is equipped, in addition to the imaging equipment, with support equipment such as a generator (e.g., to provide electrical energy for imaging), an HVAC system (e.g., to maintain environmental conditions including temperature at and in the vicinity of the imaging equipment), and a display monitor 602. One or more of the support equipment may be housed within housing 502 (e.g., HVAC system or display monitor 602) or may be positioned externally (such as generator 604). An example view 550 of the inside of housing or trailer 502 showing the position of the imaging equipment and related support equipment is shown in FIG. 6.

[0068] In particular embodiments, the x-ray imaging system 504 is a dual beam x-ray imager providing x-rays of differing intensity (e.g., soft and hard x-rays) to image different portions of the harvested ears (e.g., to image the silk of the ears differently from the cob).

[0069] Plot samples (that is ears harvested by the PhenoPuller) are brought to the PhenoProcessor, for example, harvested ears are brought to the FiRP and dumped into a container (e.g., wire cart). A barcode tag associated with the plot sample is scanned before data collection is started so that an association between the plot sample and the subsequently collected data is made.

[0070] Imaging settings are selected based on the crop and the imaging modality. In one example, where the imaging modality is X-ray imaging, and the harvested samples are sweet com ears, the X-ray settings include a wavelength setting between 30 and 40 keV and a current setting between 1.5 and 4.0 mA using a belt speed of 20m / min. The pixel size used may be 0.4mm H and V.

[0071] Each harvested ear is placed on belt 606 in a selected common orientation (e.g., a butt first orientation) and allowed to pass through the imaging equipment 504. Ears may be imaged individually or in groups with individual ears of the group positioned side-by-side to allow for a comparison. For example, ears may be imaged as singles or two at a time, side-by-side, or more than two at a time, side-by-side, to speed up the process. In one example embodiment, samples can be imaged at a rate of ~2 plots per minute.

[0072] Each ear (or pair of ears or group of ears) is imaged and the corresponding image in stored in a memory storage medium coupled to the imaging equipment in a folder labeled in accordance with the associated plot information (that was scanned prior to initiating imaging of ears of that plot). In one example, each X-ray image of an individual ear is stored as a 256 gray scale BMP image in the computer memory inside a folder named after the scanned barcode data. An example of such an X-ray image 700 of a harvested corn ear is shown at FIG. 7. Taking an x-ray image allows trait significant phenotype data to be collected without requiring husking of com ears. For example, X-ray image 700 enables the corn cob region of the ear to be displayed as a shaded region 702 distinct from the husk region of the ear (unshaded region 704). Corn cob features and husk features can then be gathered without requiring prior husking of the ear. As non-limiting examples, and as elaborated at FIG. 8, the X-ray image can be assessed to segment the ear into a cob region (comprising the kernels), a shank region (comprising the shank via which the ear was connected to the corn plant) located posterior to the cob region, and a husk region (comprising the husk and therebeyond silk) located anterior to the cob region. Following segmentation, the length, width, diameter, and other parameters of each region can be estimated.

[0073] After all samples of a plot (that is, all harvested ears of a given plot) are imaged, further image acquisition is stopped. Before starting samples of the next plot, the next plot barcode is scanned, thereby opening a new folder in the computer’s memory associated with the new plot, and subsequently, all samples associated with that plot are imaged and the images are stored in the newly created folder.

[0074] After all the plots are imaged, the image file folders can be transferred to a downstream image processing system (such as via a USB drive or a wired or wireless network connection) to the downstream image processing system where the images are analyzed to extract traitsignificant data from the images.

[0075] In this way, aspects of the invention include imaging systems for phenotyping a crop, such as the Phenoprocessor. Embodiments of such a crop imaging system comprise: an image capturing device; and a processor communicatively coupled to the imaging device, the processor configured to execute instructions to: capture an image of a crop sample; and extract, from the captured image, data associated with a trait of the crop sample; and based on the data, assign an individual value to each of one or more attributes associated with the trait of the crop sample. In embodiments, the processor is further configured to capture an image for a plurality of crop samples harvested from a common plot, and estimate, from the individual value of each of the more or more attributes of each crop sample for the common plot, an average value of each of attribute associated with the trait of the crop harvested from the common plot. In particular embodiments, the crop is sweet corn. In particular embodiments, the image capturing device is an imaging device capable of providing an internal image of the corn ear, for example, where the image capturing device is an X-ray image capturing device. In other particular embodiments, the image capturing device is any one of a UV, visible, NIR, CT, or MR image capturing device, or combinations thereof and / or combined with the X-ray imaging capability. In particular embodiments, the trait is a marketability phenotype. In particular embodiments, the one or more attributes associated with the marketability phenotype includes shank length, shank diameter, husk length, ear length and ear width.

[0076] SweetTEA (Image Analysis)

[0077] Turning now to SweetTEA, it is the processing sub-system of the PhenoHarvester. In some embodiments, a controller is coupled to the PhenoHarvester (e.g., to the PhenoProcessor) and comprises one or more processors configurable to execute instructions stored in non- transitory computer readable storage media, the instructions comprising code for analyzing images of samples of a crop and extracting, from each image of each sample, trait-significant data for that crop (e.g., data associated with a marketability phenpotype of sweetcorn ears harvested from a given research plot). In some embodiments, the controller is on-board the PhenoProcessor. In other embodiments, the controller is coupled to the PhenoProcessor but located off-board or remote from the PhenoProcessor. In embodiments, the Marketability Trait Extraction Algorithm (TEA) is a machine learning (ML) based software algorithm built into a software package that uses the raw images output by the PhenoProcessor to generate a data output table that summarizes the image derived objective measurements.

[0078] In particular embodiments, the SweetTEA algorithm is trained using a training data set comprising images that have been intentionally masked for specific features to enable the algorithm to learn how to identify those features (e.g., the same feature or similar features) in unannotated images of a test data set. The ML algorithm can be improved over time to identify one or more “standard” features with higher accuracy and learn new, trait-significant features as they become available. Example trait-significant features that may be extracted from an X-ray image of a sweet corn sample and used to assess a marketability phenotype for sweet com include husk length, ear length, ear width, and shank length.

[0079] An example of processing of an X-ray image 800 of a harvested corn ear is shown at FIG. 8. Trait significant phenotype data, including husk length, ear length, ear width, and shank length, is extracted from the image 800 without requiring husking of corn ears. As described previously with reference to FIG. 7, X-ray image 800 enables the com cob region of the ear to be displayed as a shaded region 802 distinct from the husk region of the ear (unshaded region 804). Corn cob features and husk features can then be gathered without requiring prior husking of the ear. As non-limiting examples, and as elaborated at FIG. 8, the x-ray image can be assessed to segment the ear into a cob region (comprising the kernels; depicted in the figure by the central segment), a shank region (comprising the shank via which the ear was connected to the corn plant) located posterior to the cob region (depicted in the figure as the right most segment), and a husk region (comprising the husk and therebeyond silk) located anterior to the cob region (depicted in the figure as the left most segment). Following segmentation, the length, width, diameter, and other parameters of each region can be estimated. For example, following segmentation, one or more of a shank length 816, husk length 812, and ear length 814 can be determined. Likewise, one or more of a shank or core width 808, an ear width 806, and a cob width 810 can be estimated.

[0080] FIG. 9 shows a high-level example embodiment of a portable imaging system 900 or Phenoprocessor 900 that may be used for portable high speed x-ray phenotyping of harvested com ears. The portable imaging system 900 may be configured for use as a field side modular platform for use with conventional harvesters. Bags 902 of harvested corn ears 906 are unloaded onto a scale 904 of the portable imaging system 900 for weighing and a unique identifier (e g., barcode) of the bag is scanned. The identifier provides an indication regarding the corn plant, germline, and / or plot from which the ears were harvested and allows for images of the ears to be associated with the identity of the com plant, germline, and / or plot. In one example embodiment, bags of corn from a first plot are assessed for phenotype value before bags of corn from a second, different plot are processed.

[0081] Corn ears 906 are then transferred to conveyor 908 after weighing for transfer to an onboard X-ray imager 910. Corn ears are imaged via the x-ray imaging as they pass through the imager 910 on the conveyor 908. After imaging, corn ears 906 are collected in bins 912 and can be further processed for determination of other traits. For example, the ears can then be transferred or moved through another imager (e.g., NIR, RGB, or UV camera) to collect other trait significant or agronomic quality data, such as color, taste, starch content, etc. Data collected from the imaging system can then be used to assign the corn ears associated with the unique identifier a trait significant phenotype value. Based on this value, a breeder may make breeding selections and decisions.

[0082] FIG. 10 shows another example embodiment of a portable imaging system 1000 or Phenoprocessor 100 that may be used for portable high speed x-ray phenotyping of harvested com ears. The portable imaging system 1000 may be configured as a rugged all-in-one device. In particular embodiments, the portable imaging system 1000 is configured to be resistant to dust, heat (e.g., at least up to 50°C), and vibration, and furthermore may be wet tolerant. The device may be sized to be ergonomic and to have a total weight that enables a small team or farm vehicle to be able to easily transport the device. In a particular embodiment, the portable imaging system 1000 weighs 80 lbs and can be transported by 2 people.

[0083] Harvested com ears 1002 are received at a first opening 1005 of the portable imaging system 1000. The first opening 1005 is positioned at a first proximal end of the device and may comprise a flap or door that can be closed once imaging is completed. Prior to insertion for imaging, bags comprising the ears may be weighed and a unique identifier (e.g., barcode) 1006 of the bags can be scanned via a scanner 1004. The identifier provides an indication regarding the corn plant, germline, and / or plot from which the ears were harvested and allows for images of the ears to be associated with the identity of the corn plant, germline, and / or plot. Ears are placed on a conveyor or moving platform inside the device wherein the ears are moved to and through an x-ray imager 1008. A shield 1010 protects the operators from the x-rays. In one example embodiment, the x-ray imager 1008 captures images at a rate of 1 ear per second and generates high resolution images, such as in the range of 0.2mm pixels. In particular embodiments, the x-ray imager 1008 is a dual beam x-ray imager providing soft and hard x-rays to image the silk of the ears differently from the cob. Optionally, the imaging system is communicatively coupled to an on-board or off-board database 1012 for transfer and storage of the images (e.g., a cloud connected database to which the images can be uploaded). Further, the imaging system is coupled to an on-board or off-board processor 1020 (such as a processor in the field or at a station) where the uploaded images can be processed and trait significant phenotype data can be extracted from the images. A breeder 1022 may make breeding selections and decisions based on the images and extracted trait significant phenotype data and values. For example, they may analyze the data and make decisions regarding advancement in a pipeline.

[0084] FIG. 11 is an example method 1100 of operating a crop management system in accordance with the present disclosure. The method allows for collecting and imaging corn ears and associating com information with the collected ears.

[0085] At 1102, the method comprises harvesting an unhusked com ear from a corn plant of a com plot. The unhusked ear may be harvested by a harvesting system such as the phenopuller. Alternatively, previously harvested unhusked ears are received. In one example embodiment, the ears are received at a bin on the rear platform of the phenopuller.

[0086] At 1104, it is determined if the harvesting system is in a collection mode, such as based on the position of a lever or switch, or based on another form of operator input. If not, then the system is operating in a bypass mode and at 1130, the collected ears are discarded and no data is collected for them. For example, the bins of ears may be placed on a movable platform that can be moved to a position wherefrom the ears can be easily discarded onto the floor of a field.

[0087] If the system is in the collection mode, at 1106, the method includes weighing the com ear and collecting the weight. In particular embodiments, the collecting and associating is performed continuously while a harvester is operated through the com plot. Further, the collecting and associating is performed substantially concurrently.

[0088] At 1108, the method includes collecting a field position of the corn plant, such as via a GPS sensor coupled to the harvester. At 1110, the method comprises associating the weight of the com ear with the field position of the corn plant. At 1112, based on the association, the method comprises uniquely identifying the com ear, the com plant, and / or the com plot.

[0089] At 1114, the method comprises imaging the unhusked corn ear with an imaging device capable of providing an internal image of the corn ear. For example the unhusked ear is imaged using x-ray imaging at a phenoprocessor. At 1116, the method includes extracting trait phenotype data for the com ear from the image. In one example, the trait phenotype data includes marketability trait phenotype data including shank length, cob length, cob width, etc. Furthermore, a trait associated phenotype value may be assigned to the corn ear based on the marketability trait phenotype data extracted from the ear image. For example, a marketability trait associated phenotype value may be assigned.

[0090] At 1118, the method includes associating the trait phenotype data of the corn ear with the weight of the corn ear and field position of the corn plant. In particular embodiments, uniquely identifying the corn ear, the corn plant, and / or the com plot includes assigning a unique identifier to the com ear, the com plant, and / or the com plot, and associating the trait phenotype data of the com ear with the unique identifier.

[0091] At 1120, it is determined if the trait associated phenotype value (individual or absolute or relative or statistically significant value) for the ear, plant, and / or plot is higher than a threshold. In one example, where the trait associated phenotype value is a marketability value, it may be determined if the com ears have a desired uniform size. If the trait associated phenotype value is higher than a threshold (wherein the threshold is selected based on the trait associated phenotype being assessed), then at 1122, the associated corn plant is selected for advancement in a breeding pipeline, or an alternate favorable breeding associated decision is made. Further, other plants from the associated plot may be deemed favorable or may be indicated as needing further assessment for other phenotypes.

[0092] If the trait associated phenotype value is lower than the threshold, then at 1124, the associated com plant is not selected for advancement in a breeding pipeline, or an alternate unfavorable breeding associated decision is made. The associated plants may be discarded. Further, other plants from the associated plot may be deemed unfavorable or may be indicated as not needing further assessment for other phenotypes. In this way, the PhenoHarvester system holds the potential to increase the rate of genetic gain for crop breeding programs, in particular sweet com breeding programs, by improving the efficiency of data collection through automation. Specifically, the increased speed and efficiency to obtain ear counts and weights on single row research plots will facilitate collection of quantitative yield performance data earlier in the product development pipeline. Not only will this increase the rate of genetic gain immediately by enabling selection earlier in the hybrid testing pipeline, but also via contribution over time to standardized, robust datasets from which relationships between yield performance, genotypes, and environments can be modeled and used to implement predictive breeding throughout the product development pipeline.

[0093] In this way, aspects of the invention include methods of collecting a phenotype for a sweet com plant. An example embodiment of such a method comprises: obtaining an image of an unhusked corn ear comprising a com cob from a sweet corn plant grown on a plot; extracting from the image, data associated with a marketability phenotype for the corn cob; and assigning a marketability value for the sweet corn cob, plant and / or plot based on data extracted from the image.

[0094] Aspects of the invention further include breeding methods. An example embodiment of such as method comprises harvesting corn ears from com plants of a plot; obtaining an image of each com ear; extracting from the image of each com ear, trait significant phenotype data for the com cob; estimating a statistical trait-significant phenotype value for corn plants of the plot; and based on the statistical value, advancing one or more com plants of the plot into a breeding pipeline. In particular embodiments, the trait significant phenotype is a marketability phenotype.

[0095] In further embodiments, the crop management system or PhenoHarvester 10 may include one or more of the following features:

[0096] 1. Stalk detection in the PhenoPuller as an additional or alternative population calculation tool. In particular embodiments, a stem detection “finger” is added just below and behind the stalk cutter on the PhenoPuller (directly beneath the Trimble GPS detector). This finger attachment sends a stalk detection trigger signal in response to engagement with the stalk, thereby providing an indication of an actual count of harvested stalks in the plot area. This can be used for understanding harvest population and spacing issues.

[0097] 2. PhenoPuller GIS data layer analysis to help understand the exact area harvested (e.g., to a 5 cm accuracy) that created a given harvester sample (e.g., the position of the plant from which a given ear is harvested). This knowledge can give a very precise estimation of green weight yield (Mg / ha). In combination with PhenoProcessor data, a very accurate measure of marketable ears per Ha can be obtained.

[0098] 3. PhenoPicker as a PhenoPuller modification. Herein, instead of a harvester, a stripper plate header may be used to remove the corn ear shank for use in processor-type sweet com or commercial dent com. 4. Leveraging emergence data to discover improved germination traits. Germination and early stand development data is an important trait in sweet corn due to the endosperm mutations that limit starch development and thus starve the new seedling of energy. Using the combined precision seed drop date, Plant Stand Analyzer, or UAV emergence data, and the PhenoPuller stem detection data overlayed together, a controller may be coded to determine how well a seed germinated and ultimately developed into a functional plant.

[0099] 5. Downstream phenotyping from the PhenoProcessor including yellow camera and / or sensory testing can be used. Since the x-ray is a non-destructive sensor system, after the ears are measured, they may be transferred to another sensor or imaging system for further analysis. One example is to use a camera-based system (yellow camera) that could image a subsample of the husked ears to measure kernel color or other visual traits.

[0100] 6. Phenoprocessor X-ray analysis for starch content. In particular embodiment, the X-ray system may be used to detect or measure the relationship between sugar (sucrose) water and starch in situ as starch crystals would have a different x ray absorption than sugars in solution. As starch formation in the sweet com ear is not desirable, it may be possible to develop a high throughput, non-destructive method for starch detection. This approach allows the same x-ray data of a corn ear to be used for both marketability phenotype associated data as well as sugar content estimation.

[0101] Many modifications and other implementations of the disclosure will come to mind to one skilled in the art to which this disclosure pertains having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the disclosure is not to be limited to the specific embodiments disclosed herein and that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims

Claims1. A harvester, comprising: a harvester head for engaging a corn plant and removing one or more corn ears from the engaged corn plant; a collection platform; a collection container operatively coupled to the collection platform for receiving the one or more removed com ears; a first sensor configured to sense receipt of the one or more removed corn ears at the collection platform or the collection container; a second geo-positionary sensor; a controller comprising one or more processors communicatively coupled to the first and second sensors and configurable to execute instructions stored in computer readable storage media, the instructions comprising: continuously collecting output from the first sensor indicative of receipt of one or more removed com ears from the corn plant at the collection platform or the collection container; and optionally associating the output of the first sensor with an output of the second sensor indicative of a position of the com plant within the plot.

2. The harvester of claim 1, further configurable to execute instructions comprising: in response to an operator input, discarding the removed one or more ears from the collection platform and discarding the output of the first and second sensor.

3. The harvester of claim 1, wherein continuously collecting output from the first sensor includes continuously collecting first sensor output and extracting individual com ear weight from the output.

4. A method for collecting and associating com information, the method comprising: harvesting an unhusked corn ear from a corn plant of a com plot; weighing the corn ear and collecting the weight; collecting a field position of the com plant; associating the weight of the com ear with the field position of the corn plant; uniquely identifying the corn ear, the corn plant, and / or the corn plot; imaging the unhusked corn ear with an imaging device capable of providing an internal image of the com ear; and extracting from the image, trait phenotype data for the com ear.

5. The method of claim 4, wherein the com plant is a sweet com plant.

6. The method of claim 4, wherein the collecting and associating is performed continuously while a harvester is operated through the corn plot.

7. The method of claim 6, wherein the collecting and associating is performed substantially concurrently.

8. The method of claim 4, further comprising, associating the trait phenotype data of the com ear with the weight of the com ear and field position of the com plant.

9. The method of claim 4 or 8, wherein uniquely identifying the com ear, the com plant, and / or the com plot includes assigning a unique identifier to the com ear, the corn plant, and / or the com plot, the method further comprising, associating the trait phenotype data of the com ear with the unique identifier.

10. A crop imaging system, comprising: an image capturing device; and a processor communicatively coupled to the imaging device, the processor configured to execute instructions to: capture an image of a crop sample; and extract, from the captured image, data associated with a trait of the crop sample; and based on the data, assigning an individual value to each of one or more attributes associated with the trait of the crop sample.

11. The system of claim 10, wherein the processor is further configured to: capture an image for a plurality of crop samples harvested from a common plot, and estimate, from the individual value of each of the more or more attributes of each crop sample for the common plot, an average value of each of attribute associated with the trait of the crop harvested from the common plot.

12. The system of claim 10 or 11, wherein the crop is sweet corn.

13. The system of claim 10, wherein image capturing device is an imaging device capable of providing an internal image of the corn ear.

14. The system of claim 13, wherein the image capturing device is an X-ray image capturing device.

15. The system of claim 10, wherein image capturing device is any one of a UV, visible, NIR, CT, or MR image capturing device, and wherein the system is portable.

16. The system of claim 10, wherein the trait is a marketability phenotype.

17. The system of claim 16, wherein the one or more attributes associated with the marketability phenotype includes shank length, shank diameter, husk length, ear length and ear width.

18. A crop management system, comprising: a harvesting sub-system comprising a harvester head for engaging a corn plant and removing one or more unhusked com ears from the engaged corn plant; an imaging sub-system comprising an image capturing device; one or more sensors including a first sensor for sensing collection of the com ears at the harvesting sub-system and a second sensor for sensing a position of the harvesting subsystem; and a processor with code for executing instructions to sense collection of the removed one or more corn ears in the harvesting subsystem.

19. The system of claim 18, further comprising a transfer sub-system for transferring the corn ears from the harvesting sub-system to the imaging sub-system.

20. A method of collecting a plant phenotype, the method comprising: obtaining an image of an unhusked corn ear comprising a corn cob; and extracting from the image, trait significant phenotype data for the corn cob.

21. The method of claim 20, further comprising assessing a phenotype of the corn cob based on the extracted image.

22. The method of claim 21, further comprising in response to the assessed phenotype matching a desired phenotype, advancing seeds from the corn ear in a breeding pipeline.

23. The method of claim 21, further comprising in response to the assessed phenotype matching a desired phenotype, assigning a phenotype value to the com ear, the com cob, and / or the com plant.

24. The method of claim 21, wherein said corn is sweet com.

25. The method of claim 24, wherein the phenotype is a marketability trait.

26. The method of claim 25, wherein the marketability trait is ear length, ear width, or shank length.

27. A method of collecting a phenotype for a sweet com plant, the method comprising: obtaining an image of an unhusked com ear comprising a corn cob from a sweet com plant grown on a plot; and extracting from the image, data associated with a marketability phenotype for the corn cob; and assigning a marketability value for the sweet com cob, plant and / or plot based on data extracted from the image.

28. A breeding method, comprising: harvesting com ears from com plants of a plot; obtaining an image of each corn ear; extracting from the image of each com ear, trait significant phenotype data for the corn cob; estimating a statistical trait-significant phenotype value for com plants of the plot; and based on the statistical value, advancing one or more corn plants of the plot into a breeding pipeline.

29. The method of claim 28, wherein the trait significant phenotype is a marketability phenotype.

Images