System and method for detecting crop loss via imaging processing

Abstract

The invention relates to a system and method for detecting crop loss via imaging processing. A detection and control system for an agricultural harvester includes a controller having at least one memory and at least one processor. The controller is configured to receive a sensor signal indicative of an image of harvested crop material within a feed chamber of the agricultural harvester; analyzing the image to detect at least one husked corn ear of harvested crop material within the feed chamber; analyzing the image to identify one or more parameters of the at least one husked corn ear; and determining appropriate adjustments to one or more components of the row unit based on the one or more parameters of the at least one husked corn ear.

Description

Background Art

[0001] This section is intended to introduce the reader to various technical aspects that may be related to the various aspects of the present disclosure described below. It is believed that this discussion helps to provide background information to the reader to facilitate a better understanding of the various aspects of the present disclosure. Therefore, it should be understood that these statements should be read from this perspective, rather than admitting that they are prior art.

[0002] Harvesters can be used for harvesting crops, such as barley, beans, beets, carrots, corn, cotton, flax, oats, potatoes, rye, soybeans, wheat or other plant crops. Harvesters can include or be coupled to headers, which can be designed to efficiently harvest certain types of crops. For example, corn headers can be designed to efficiently harvest corn. In particular, corn headers can include row units, which are included in the parts that operate to separate corn ears from stalks when the harvester passes through the field. A conveyor (e.g., an auger) transports corn ears to the processing machinery and / or storage compartment of the harvester, and the stalk is put back in the field. Summary of the invention

[0003] Certain embodiments comparable to the scope of the originally claimed subject matter are summarized below. These embodiments are not intended to limit the scope of the claimed subject matter, but rather, these embodiments are intended only to provide a brief overview of possible forms of the present disclosure. In fact, the present disclosure may encompass a variety of forms that may be similar or different from the embodiments set forth below.

[0004] In certain embodiments, a detection and control system for an agricultural harvester includes a controller having at least one memory and at least one processor. The controller is configured to receive a sensor signal indicative of an image of harvested crop material in a feed chamber of the agricultural harvester; analyze the image to detect at least one shelled ear of corn with the harvested crop material in the feed chamber; analyze the image to identify one or more parameters of the at least one shelled ear of corn; and determine appropriate adjustments to one or more components of a row unit based on the one or more parameters of the at least one shelled ear of corn.

[0005] In certain embodiments, a method of operating a detection and control system for an agricultural harvester includes receiving, at a controller, a sensor signal indicative of an image of harvested crop material within a feed chamber of the agricultural harvester; analyzing, using the controller, the image to detect at least one shelled ear of corn of the harvested crop material within the feed chamber; analyzing, using the controller, the image to identify one or more parameters of the at least one shelled ear of corn; and determining, using the controller, appropriate adjustments to one or more components of a row unit based on the one or more parameters of the at least one shelled ear of corn.

[0006] In certain embodiments, an agricultural harvester includes a header having a plurality of row units distributed across a width of the header, and a controller including at least one memory and at least one processor. The controller is configured to receive a sensor signal indicative of an image of harvested crop material; and determine, based on the image via machine learning, an undesirable level of grain loss in the harvested crop material due to one or more parameters of one or more components of the plurality of row units. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] These and other features, aspects, and advantages of the present disclosure will become more readily understood when the following detailed description is read with reference to the accompanying drawings, in which like characters represent like parts throughout the drawings, and in which:

[0008] Figure 1 is a side view of an agricultural harvester according to aspects of the present disclosure;

[0009] Figure 2 According to aspects of the present disclosure, it can be used for Figure 1 A perspective view of a header of an agricultural harvester;

[0010] Figure 3 According to aspects of the present disclosure Figure 2 A schematic top view of a portion of a header;

[0011] Figure 4 According to aspects of the present disclosure Figure 2 A perspective view of a row unit of a header;

[0012] Figure 5 According to aspects of the present disclosure, the harvested crop material and Figure 2 Schematic diagram of the contact between the deck plate of the header;

[0013] Figure 6 According to aspects of the present disclosure Figure 2 An example of an image of harvested crop material within a feed chamber of a header of FIG.

[0014] Figure 7 is a flow chart of a method for detecting and responding to dehulled harvested crop material according to aspects of the present disclosure. DETAILED DESCRIPTION

[0015] One or more specific embodiments of the present disclosure are described below. In order to provide a concise description of these embodiments, all features of an actual implementation may not be described in the specification. It should be recognized that in the development of any such actual implementation, as in any engineering or design project, many implementation-specific decisions must be made to achieve the developer's specific goals, such as complying with system-related and business-related constraints, which may vary from implementation to implementation. Moreover, it should be recognized that such development work may be complex and time-consuming, but it is still a routine task of design, fabrication, and manufacturing for ordinary technicians who benefit from this disclosure.

[0016] When introducing elements of various embodiments of the present disclosure, the articles "a," "an," "the," and "said" are intended to indicate that there are one or more elements. The terms "comprising," "including," and "having" are intended to be inclusive and indicate that there may be additional elements in addition to the listed elements. Any examples of operating parameters and / or environmental conditions do not exclude other parameters / conditions of the disclosed embodiments.

[0017] The farming process usually begins with sowing in a field. Over time, the seeds grow and eventually become harvestable crops. Typically, only a portion of each crop has commercial value, so each crop is harvested to separate the usable material from the rest of the crop. For example, a harvester may include or be coupled to a header to harvest crops in a field. The header may be a corn header, which is configured to efficiently harvest corn in a field. The corn header may include a plurality of row units arranged on the width of the corn header, and each row unit may include a layer plate, a stalk roller, and / or (one or more) other components that operate to separate corn ears from stalks and materials other than grain (e.g., MOG; crop residues) when the harvester passes through the field. For example, a corn header may be able to quickly chop stalks and collect corn ears, while leaving stalks and any other unwanted and biodegradable byproducts (materials other than grain). A conveyor (e.g., an auger) transports the corn ears to a harvester (e.g., a chassis of a harvester), such as to a processing machine of a harvester.

[0018] The row unit of the corn harvester includes a stalk roller that engages with the stalk and drives the stalk downward and backward. When the stalk is driven downward, the corn ear engages with the layer plate located above the stalk roller, thereby separating the corn ear from the stalk. The corn ear moves backward to the conveyor and the stalk is placed on the field. Using the embodiments disclosed herein, the operator can actively adjust the stalk roller, the gap between the layer plates, and other features of the corn harvester to effectively and efficiently harvest the corn ear as crop conditions change as the harvester traverses the field. This process may be demanding on the operator and prone to error.

[0019] It is currently recognized that when an ear of corn impacts a corn header or a component thereof (e.g., a deck), particles (e.g., individual kernels) on the ear of corn may fall off the ear of corn (e.g., fall off the corn cob; due to impact with the corn header), causing the particles to be lost before being directed to the processing machinery and / or storage compartment of the harvester, thereby reducing the yield of the harvesting operation. As discussed herein, "butt-shelling" can refer to the process of separating individual kernels of the ear of corn from the ear of corn via engagement of the ear of corn with one or more components of the header (e.g., a deck, stalk roller). Therefore, it may be desirable to employ one or more sensors (e.g., imaging sensors) to detect an exposed portion of the ear of corn at a location upstream of the processing machinery and / or storage compartment of the harvester (e.g., a feed chamber of the corn header) so that one or more operating parameters of the corn header can be adjusted to reduce the amount of product loss (e.g., reduce the amount of bottom shelling, reduce the bottom shelling rate) due to impact with components of the corn header, thereby increasing the yield of the harvesting operation.

[0020] Therefore, the present embodiment relates to an imaging system comprising at least one imaging sensor (e.g., a camera). At least one imaging sensor may be located in a feed chamber of a corn header, and at least one imaging sensor may be configured to generate an image (e.g., a video) of an ear of corn as the ear of corn passes through the feed chamber of the corn header. The imaging system may be configured to analyze the image to detect or identify exposed portions of the ear of corn (e.g., due to missing kernels) as the ear of corn passes through the feed chamber of the corn header. For example, the imaging system may be configured to analyze the image to determine (e.g., estimate) the corresponding surface area of ​​the exposed portion of the ear of corn, the number of exposed portions, the number of missing kernels, or any combination thereof. Additionally or alternatively, the imaging system may be configured to analyze the image to determine (e.g., classify) the corresponding position (e.g., the end portion or the center portion) of the exposed portion of the ear of corn.

[0021] As an example, a high number of ears of corn passing through a feed chamber of a corn head having bare portions at the end portions of the ears of corn (e.g., counts or percentages over time; exceeding a threshold) indicates excessive product loss due to impacts at row units (e.g., decks of row units). Therefore, in this case, it may be desirable to increase the gaps between decks of the row units and / or make other changes to the row units to reduce or prevent product loss at the row units. However, a low number of ears of corn passing through a feed chamber of a corn head having bare portions at the end portions of the ears of corn (e.g., counts or percentages over time; below a threshold) indicates that there is no or acceptable product loss due to impacts at row units (e.g., decks of row units). In this case, it may be desirable to maintain the gaps between decks of the row units and / or other settings of the row units to continue to manage or avoid product loss at the row units.

[0022] In certain embodiments, the imaging system may implement machine learning (e.g., artificial intelligence; computer vision) to analyze the image and determine the amount of product loss due to impact at the row unit based on the image. As used herein, machine learning refers to algorithms and statistical models that can be used to perform specific tasks without using explicit instructions but relying on patterns and reasoning. In particular, machine learning generates mathematical models based on data (e.g., samples or training data, which may include historical data) to make predictions or decisions without explicit programming to perform specific tasks. Therefore, when performing a harvesting operation, the pattern of grain loss can be updated and better reflect whether the product loss is excessive or acceptable, and whether it is appropriate or beneficial to adjust the components at the row unit and / or some other location of the corn header (e.g., to reduce grain loss). Therefore, the imaging system can reference and analyze the pattern of grain loss (e.g., the current grain loss pattern of the current harvesting operation can be analyzed via machine learning) to accurately determine the condition of the corn header (e.g., the amount of product loss; excessive and acceptable product loss; due to the row unit), and the appropriate response (e.g., a control signal for adjusting the components of the corn header; a notification for visualization by the operator on the user interface).

[0023] Consider the above situation, Figure 11 is a side view of an embodiment of an agricultural harvester 100 having a corn header 200 (e.g., an agricultural header). The agricultural harvester 100 includes a chassis 102 configured to support the corn header 200 and an agricultural crop handling system 104. The corn header 200 is configured to separate corn ears from stalks (e.g., multiple row units separate corn ears from stalks, thereby contacting the exposed stalks with the soil). The corn header 200 is configured to convey the corn ears to an inlet 106 of the agricultural crop handling system 104 (via a feed chamber 107) for further processing of the corn ears. The agricultural crop handling system 104 receives the corn ears from the corn header 200 (via the feed chamber 107) and separates the desired crop material from the crop residue. For example, the agricultural crop handling system 104 may include a thresher 108 having a cylindrical threshing rotor that conveys the corn ears in a spiral flow path through the agricultural harvester 100. In addition to conveying ears of corn, the thresher 108 can also separate certain desired crop materials (e.g., corn kernels) from crop residues (such as corn husks and corn cobs) and allow the desired crop materials to flow into a cleaning system 110 located below the thresher 108. The cleaning system 110 can remove debris from the desired crop materials and convey the desired crop materials to a storage compartment 112 within the agricultural harvester 100. The crop residues can be conveyed from the thresher 108 to a crop residue disposal system 114, which can remove the crop residues from the agricultural harvester 100 via a crop residue spreading system 116 located at the rear end of the agricultural harvester 100. For ease of discussion, the agricultural harvester 100 and / or its components (e.g., the corn head 200) can be described with reference to a lateral axis or direction 10, a longitudinal axis or direction 12, and a vertical axis or direction 14. The agricultural harvester 100 and / or its components (e.g., the corn head 200) can also be described with reference to a direction of travel 16.

[0024] The agricultural harvester 100 includes a row unit detection and control system 118. The row unit detection and control system 118 includes a control system or controller 120 (e.g., an automated controller, an electronic controller, a programmable controller, a cloud computing system, a control circuit system) having a processor 122 (e.g., a processing circuit system) and a memory 124. The processor 122 can be used to execute software code or instructions stored on the memory 124, such as for processing signals and controlling the operation of the agricultural harvester 100 and / or the corn head 200. As used herein, the term "code" or "software code" refers to any instructions or instruction sets that control the operation of the row unit detection and control system 118. The code or software code can exist in a computer executable form (such as machine code that is a set of instructions and data directly executed by the processor 122 of the controller 120), in a human understandable form (such as source code that can be compiled for execution by the processor 122 of the controller 120), or in an intermediate form (such as object code generated by a compiler). In some embodiments, the row unit detection and control system 118 may include multiple controllers.

[0025] As an example, the memory 124 may store processor executable software codes or instructions (e.g., firmware or software) that are tangibly stored on a non-transitory computer readable medium. Additionally or alternatively, the memory 124 may store data. As an example, the memory 124 may include volatile memory (such as random access memory (RAM)), and / or non-volatile memory (such as read-only memory (ROM)), flash memory, a hard drive, or any other suitable optical, magnetic or solid-state storage medium, or a combination thereof. In addition, the processor 122 may include a plurality of microprocessors, one or more "general" microprocessors, one or more special-purpose microprocessors, and / or one or more application-specific integrated circuits (ASICs), or some combination thereof. For example, the processor 122 may include one or more reduced instruction set (RISC) or complex instruction set (CISC) processors. The processor 122 may include a plurality of processors, and / or the memory 124 may include a plurality of memory devices. The processor 122 and / or the memory 124 may be located in any suitable portion of the agricultural harvester 100 (e.g., the cab of the agricultural harvester 100). In addition, the controller 120 can be a distributed controller, where multiple processors 122 and / or multiple memories 124 are located in separate housings or locations (e.g., in the corn head 200, in a remote location, in the cloud). The controller 120 can be communicatively coupled to a user interface or display 125. In some embodiments, the user interface or display 125 can be located in the cab of the agricultural harvester 100.

[0026] The row unit detection and control system 118 includes an internal imaging system 126 that is communicatively coupled to the controller 120. The internal imaging system 126 includes one or more cameras 128 that can be located on the floor 130 of the feed room 107. The one or more cameras 128 capture images (e.g., movies) of harvested crop material within the feed room 107 (before threshing) and transmit signals representing the captured images to the controller 120. These images within the feed room 107 can be captured as the agricultural harvester 100 passes through a crop field.

[0027] The controller 120 is configured to receive a signal representing an image acquired from the internal imaging system 126. The controller 120 is also configured to analyze the image by identifying specific components of the harvested crop material (e.g., corn ears) and measuring at least one metric of the harvested crop material within the feed chamber 107 (e.g., exposed portions of the corn ears, such as the surface area and / or location of each exposed portion). In some embodiments, the controller 120 is also configured to analyze the image to identify trends over time (e.g., a number of corn ears over time; the number of shelled corn ears or the number of corn ears with exposed portions over a period of time). In some embodiments, the controller 120 is configured to utilize machine learning (e.g., artificial intelligence; computer vision) to analyze the image and measure metrics within the image.

[0028] The controller 120 is also configured to determine the initial size of the gap between the plates of the row units of the corn head 200 based on the initial settings. In some embodiments, the controller 120 is even further configured to provide a user-perceivable indication of the initial size of the gap to the operator of the agricultural harvester 100 (e.g., on the user interface or display 125) so that the operator can know the initial gap and / or adjust the initial gap. In some embodiments, the controller 120 is also configured to determine other parameters of the corn head 200, such as, for example, the angle of the row units of the corn head 200 relative to the agricultural harvester 100 and / or the rotation rate of the feed rollers of the row units of the corn head.

[0029] In some embodiments, the controller 120 is configured to provide a user-perceivable indication of the presence of at least one shelled ear of corn to an operator of the agricultural harvester 100 (e.g., on a user interface or display 125) when the presence of at least one shelled ear of corn is detected. In some embodiments, the controller 120 is also configured to provide a suggestion to the operator (e.g., on a user interface or display 125) to adjust the gap between the tiers of the row units and / or adjust other parameters (e.g., the angle of the row units and / or the rotation rate of the stalk rollers) when the presence of at least one shelled ear of corn is detected. In some embodiments, the controller 120 is also configured to automatically adjust the size of the gap between the tiers of the row units and / or adjust other parameters of the corn head 200 (e.g., the angle of the row units and / or the rotation rate of the stalk rollers) when the presence of at least one shelled ear of corn is detected. The controller 120 is configured to receive and analyze images and provide user-perceivable indications (e.g., the initial size of the gap and / or the presence of at least one shelled ear of corn) and / or automatically adjust the gap between the decks of the row units and / or adjust other parameters of the corn header 200 (e.g., the angle of the row units and / or the rotation rate of the stalk drums) as the agricultural harvester 100 traverses a crop field so that the row units can be changed to account for changes in crop conditions and limit product losses.

[0030] The row unit detection and control system 118 includes an actuator 132 coupled to the decks and configured to change position (e.g., move the decks of each row unit toward and away from each other) to change the gap between the decks of the row units. The row unit detection and control system 118 may also include an actuator 134 that operates to adjust the angle of the corn head 200 relative to the agricultural harvester 100. The row unit detection and control system 118 may also include an actuator 136 that operates to adjust the rotation rate of the stalk rollers of the row units. The controller 120 is communicatively coupled to the actuators 132, 134, 136 and may provide control signals to the actuators 132, 134, 136 to adjust corresponding parameters of the corn head 200.

[0031] Figure 2 Yes, you can Figure 1100 . In the illustrated embodiment, the corn head 200 includes a plurality of dividers 202 configured to separate rows of crops (e.g., corn). The dividers 202 can be distributed over a width 204 of the corn head 200 (e.g., along the transverse axis 10). As the corn head 200 moves along the path, the dividers 202 can guide the crops from each row to a row unit 206. The row unit 206 is configured to receive each crop (e.g., stalk). A portion of the crop is directed to one of a pair of conveyors 208 (e.g., an auger) configured to convey the portion of the crop laterally inward to a center crop conveyor located at the center of the corn head 200, and the center crop conveyor guides the portion of the crop to the entrance 106 of the agricultural crop handling system. As shown, the conveyor 208 extends along a substantial portion of the width 204 of the corn head 200 (e.g., along the transverse axis 10). The conveyor 208 may be driven by a drive mechanism (eg, an electric motor, a hydraulic motor).

[0032] The corn head 200 can separate desired crop material and crop residue from each other to facilitate directing the desired crop material into the agricultural crop handling system via the inlet 106 and prevent the crop residue from entering the agricultural crop handling system via the inlet 106. For example, operation of the row unit 206 can direct the desired crop material to the conveyor 208 and discharge the crop residue from the conveyor 208, such as discharging the crop residue directly onto the field. As the agricultural harvester 100 cuts crops, separates the crop residue from the desired crop material, and discharges the crop residue (e.g., onto the field), the controller 120 of the row unit detection and control system 118 can provide control signals to the actuators 132, 134, 136 to adjust the positioning of components of the row unit 206. In particular, the control signal provided to actuator 132 can control the gap between the layers of the row unit 206, the control signal provided to actuator 134 can control the angle of the corn header 200 relative to the agricultural harvester 100, and the control signal provided to actuator 136 can control the rotation rate of the stalk roller of the row unit 206.

[0033] Figure 31 is a top schematic diagram of an embodiment of a corn head 200 having one or more cameras 128 that can be located on the floor 130 of a feed chamber 107 of the corn head 200. The one or more cameras 128 can generate images of the ears of corn as they travel from row units 206 distributed across the width of the corn head 200 to the entrance of an agricultural harvester. In particular, the one or more cameras 128 can generate images of the ears of corn as they pass through the feed chamber 107 between the row units 206 and the agricultural harvester. As described herein, the images can indicate the amount of bottom husking.

[0034] As shown, the feed chamber 107 can include a lower feed chamber plate 300 (e.g., a first feed chamber plate) and an upper feed chamber plate 302 (e.g., a second feed chamber plate) that define a passage 304. To facilitate discussion of the feed chamber 107, the lower feed chamber plate 300 and the upper feed chamber plate 302 are shown as being separated from one another, but it should be appreciated that the lower feed chamber plate 300 and the upper feed chamber plate 302 can be stacked on top of one another along the vertical axis 14 to form the passage 304. In operation, a drive member 306 (e.g., a chain; a conveyor; a roller) can drive or guide the ears of corn through the passage 304.

[0035] One or more cameras 128 can be mounted along the channel 304 to capture images of the ears of corn as they move through the channel 304. For example, the one or more cameras 128 can be mounted to the lower feed chamber plate 300 (e.g., via one or more fasteners, such as threaded fasteners, welds, and / or adhesives). In some embodiments, the one or more cameras 128 can be mounted below the lower feed chamber plate 300 relative to the vertical axis 14 (e.g., in a housing mounted below the lower feed chamber plate 300) such that the one or more cameras 128 obtain images through a portion 308 (e.g., a window; at least partially transparent) defined in the lower feed chamber plate 300.

[0036] In some embodiments, the one or more cameras 128 may include one camera 128 (e.g., a single camera; only one camera) at the feed chamber 107. In this case, the image may not capture all ears of corn moving through the passage 304 of the feed chamber 107; however, the image may represent the condition of the corn head 200 collecting all ears of corn. Therefore, the image from the one camera 128 can enable the determination or estimation of the amount of bottom husking and / or enable dynamic adjustments to address bottom husking issues during harvesting operations.

[0037] While the embodiments herein include one camera 128 along the feed chamber 107, it should be appreciated that any suitable number of cameras 128 (e.g., 1, 2, 3, 4, 5, 6, or more) may be provided along the feed chamber 107 and / or other locations of the corn head 200 to facilitate the techniques disclosed herein. For example, multiple cameras 128 may be distributed (e.g., spaced apart) along the transverse axis 10 and / or the longitudinal axis 12 to capture multiple images of an ear of corn as the ear of corn moves through the passage 304 of the feed chamber 107. Multiple images may be analyzed together to obtain additional data about the condition of the ear of corn (e.g., the amount of bottom husking) (e.g., compared to images from one camera 128).

[0038] Figure 4 200 . As shown, each row unit 206 is located between adjacent dividers 202, and each row unit 206 includes various components that operate to separate a portion of a crop (e.g., an ear of corn; desired crop material) from a stalk, transport the portion of the crop (e.g., an ear of corn; desired crop material) to a conveyor 208, and return the stalk to the field. For example, each row unit 206 may include a pair of stalk rollers 210 that are configured to grasp the stalk and rotate in opposite directions to push the stalk in a direction toward the field (e.g., vertically downward along the vertical axis 14; below the corn head 200). The stalk rollers 210 may be configured to operate (e.g., rotate) at a particular speed to guide the stalk to the field. In some embodiments, the speed at which the stalk rollers 210 rotate may be adjusted (e.g., via actuators 136) based on images captured by one or more cameras, as described herein.

[0039] Each row unit 206 also includes a pair of decks 212 positioned above (e.g., on) the pair of stalk rollers 210 relative to the vertical axis 14. Each deck 212 extends from a first end to a second end along the longitudinal axis 12, and the pair of decks 212 are separated from each other along the transverse axis 10 to define a gap 214. In addition, each row unit 206 can include a pair of chains 216 (e.g., having lugs) configured to drive or push the portion of the crop (e.g., an ear of corn; desired crop material) along the pair of decks 212 toward the conveyor(s) 208. The pair of decks 212 are spaced apart so that the width 215 of the gap 214 is sized to enable stalks to be pulled through the gap 214, but prevent the portion of the crop (e.g., an ear of corn; desired crop material) from falling through the gap 214. In some embodiments, the pair of plies 212 are adjustable and can be driven toward and away from each other along the lateral axis 10 (e.g., via the actuator 132) to change the width 215 of the gap 214, as described herein. A cover 218 is positioned behind each divider 202 relative to the direction of travel 16 and between adjacent row units 206 to cover various components, such as the actuators 132, 136, linkages, etc.

[0040] In order to obtain good harvesting performance, the ear of corn should be initially contacted with the pair of decks 212 toward the front of the pair of decks 212 relative to the longitudinal axis 12 or the direction of advance 16 of the corn head 200. This can enable the ear of corn to be separated from its stalk so that the stalk can be completely discharged from the corn head 200 before reaching the rear of the pair of decks 212, thereby reducing the possibility of the stalk being fed into the conveyor 208. In addition, the width 215 of the gap 214 should be sized so that the stalk of the crop can pass through the gap 214 while preventing or inhibiting the ear of corn from passing through the gap 214. For example, if the width 215 of the gap 214 is too large, then both the stalk and the ear of corn may pass through the gap 214 and be discharged from the corn head 200 before being guided to the conveyor 208, resulting in product loss. In other cases, the width 215 of the gap 214 can be sized so that bottom shelling occurs. For example, as described herein, bottom husking may refer to individual kernels or particles of a crop falling off from a crop when the crop impacts or engages with the pair of strata 212. Bottom husking may be a result of and / or may be affected by the speed at which the crop impacts the pair of strata 212. However, in some cases, bottom husking may occur when the width 215 of the gap 214 is sized so that only a portion of the ear of corn can pass through the gap 214 without interference, while different portions of the ear of corn engage with the pair of strata 212. Therefore, various operational and / or structural factors, such as the ground speed of the corn header 200, the rotation rate of the pair of stalk rollers 210, and / or the angle of the corn header 200 relative to the agricultural harvester (and therefore relative to the ground), and / or the width 215 of the gap 214 between the pair of strata 212 (e.g., relative to the size of the ear of corn), may cause bottom husking and / or affect the rate of bottom husking.

[0041] For example, Figure 5An example of a portion of a crop 250 (e.g., a corn cob, an ear of corn, a desired crop material) is shown, the crop 250 having a core 252 and a plurality of kernels 254 extending radially outward from the core 252 relative to a central axis 251 of the portion of the crop 250. The portion of the crop 250 may be non-uniform in shape, and thus may have a plurality of different diameters along a length 256 of the portion of the crop 250. For example, in some embodiments, the portion of the crop 250 may have a conical shape. Generally speaking, the diameter of the portion of the crop 250 may gradually increase from the apex 258 of the portion of the crop 250 to the base 260 of the portion of the crop 250. For example, the portion of the crop 250 may have a first diameter 262 located at or near the apex 258 of the portion of the crop 250 and a second diameter 264 located at or near the base 260 of the portion of the crop 250. As shown, the second diameter 264 is greater than the first diameter 262. In addition, the first diameter 262 may be smaller than the width 215 of the gap 214, so that the core 252 and the kernels 254 of the portion of the crop 250 near the apex 258 can pass through the gap 214, while the second diameter 264 of the portion of the crop 250 near the base 260 may be larger than the width 215 of the gap 214, so that the core 252 can pass through the gap 214, while the kernels 254 are restricted or prevented from passing through the gap 214 by the pair of plies 212 (e.g., engaging with the plies 212). When the stalk roller 210 guides the stalk and the crop 250 toward the pair of plies 212, the portion of the crop 250 may hit one or both of the plies 212, and due to the engagement of the kernels 254 with one and / or both of the plies 212, one or more kernels 254 may fall off (e.g., shear) from the core 252, thereby resulting in one or more exposed sections 266 on the portion of the crop 250, which may be undesirable. Indeed, when kernels 254 break away from the portion of the crop 250, the kernels 254 may fall to the ground and / or otherwise be directed to escape the corn head 200 (eg, not be captured by the corn head 200), thereby resulting in yield loss.

[0042] refer to Figure 3 and Figure 4, the width 215 of the gap 214 can be adjusted so that the stem can pass through the gap 214, the portion of the crop can be prevented from passing through the gap 214, and the amount of bottom hulling can be limited (e.g., controlled, reduced), as described herein. Another factor in good harvesting performance is the speed at which the pair of stem rollers 210 direct the stem and the portion of the crop to the pair of layer plates 212. For example, if the speed of the pair of stem rollers 210 is too high (e.g., too fast), then when the stem is directed through the gap 214, the impact force from the portion of the crop hitting the pair of layer plates 212 may be sufficient to cause the kernels to fall off the portion of the crop (e.g., bottom hulling). However, if the speed of the pair of stalk rollers 210 is too low (e.g., too slow), then reduced efficiencies associated with harvesting the field may occur, resulting in increased costs associated with harvesting (e.g., more gas, more time required for the operator to harvest the field) (e.g., as the equipment runs for a longer amount of time, fuel costs increase, and the time a farmer spends on the harvester may increase, resulting in less time to perform other harvesting operations). Therefore, it may be desirable to operate the pair of stalk rollers 210 at a target speed (e.g., a speed or speed range) that limits the amount of bottom husking, while still enabling the operator to efficiently perform harvesting operations.

[0043] Figure 6 500 is an example of an image 500 of harvested crop material in a feed chamber. As described herein, one or more cameras in the feed chamber can capture image 500 as the harvested crop material passes through the feed chamber. As depicted, identification of an ear of corn is indicated by box 502. Further, as depicted, one or more parameters or characteristics of the ear of corn can be determined based on image 500. For example, the one or more parameters can include an exposed section 504 of the ear of corn (e.g., a hulled section), missing kernels 506 of the ear of corn, or any combination thereof. Further, the one or more parameters can include the location of the exposed section 504 (e.g., at or near the top of the ear of corn).

[0044] It should be appreciated that any of the image 500 and / or various markings, including the box 502, one or more segment markings identifying (e.g., highlighting, outlining, marking) exposed segments 504, and / or one or more kernel markings identifying (e.g., highlighting, outlining, marking) missing kernels 506, can be presented to an operator, such as via a user interface display. In some cases, the image 500 and additional images can be presented as a video feed, enabling an operator to view multiple ears of corn as they pass by one or more cameras. In some such cases, the video feed may include some of all the markings superimposed on the ears of corn. Alternatively, the video feed may not include any markings or only a subset of the markings (e.g., only one or more segment markings); however, the operator may provide input to stop the video feed and present a still image (e.g., image 500) with some or all of the markings or additional markings (e.g., one or more segment markings and one or more kernel markings). In addition, the operator may be able to provide input to scroll through multiple still images with some or all of the markings or additional markings. This can enable an operator to visualize the quality or condition of an ear of corn, as well as understand why the farming system recommends and / or implements certain operational changes.

[0045] refer to Figure 1-6 , the row unit detection and control system 118 may include one or more cameras 128 to obtain images (e.g., such as image 500) during harvesting operations. The controller 120 may be configured to receive images from the one or more cameras 128 and analyze the images to assess the condition of the ear of corn passing through the feed chamber 107 of the corn head 200. The controller 120 may implement machine learning (e.g., artificial intelligence; computer vision) to identify one or more parameters of the ear of corn based on the image, and make a determination related to the amount of bottom threshing based on the image and / or make an appropriate response to address an undesirable level of bottom threshing. For example, the controller 120 may determine an undesirable level of bottom threshing based on the image, and also determine that the undesirable level of bottom threshing is due to a component of the row unit 206. The controller 120 may then determine and implement appropriate adjustments to the components of the row unit 206 (e.g., via the actuators 132, 134, 136) to prevent or reduce bottom threshing.

[0046] For example, in response to determining that the image indicates an undesirable level of bottom threshing (e.g., excessive kernel loss on an ear of corn), the controller 120 may send a control signal to reduce the rotation rate of the pair of stalk rollers 210 and / or adjust the width 215 of the gap 214 to prevent or reduce bottom threshing. In some embodiments, a suitable output may include providing a control signal to reduce the ground speed of the corn head 200. In some embodiments, a suitable output may include a visual alert (e.g., presented via the user interface 125; a text message with an explanation and / or such as a suggested adjustment to the width 215 of the gap 214 and / or an adjustment to the rotation rate of the stalk rollers 210) and / or an audible alert (e.g., presented via a speaker).

[0047] Figure 7 It is used for detection and control Figure 1 Flow chart of an embodiment of a method 700 for configuring components of a corn head 200 in FIG. One or more steps of the method 700 may be performed by Figure 1 The steps of method 700 may be repeated as agricultural harvester 100 traverses a crop field (where crop conditions may change). One or more steps of method 700 may be performed simultaneously or in parallel. Figure 7 The order in which they are drawn is different from the order in which they are executed.

[0048] The method 700 includes receiving a sensor signal indicative of an image of harvested crop material (before threshing) within a feed chamber of an agricultural harvester (block 702). The image is an interior image provided from within the feed chamber by one or more cameras, which may be located at or along the floor of the feed chamber (e.g., to capture the image through a window in the floor of the feed chamber).

[0049] The method 700 also includes analyzing the image to detect the presence of at least one shelled ear of corn (block 704). As described herein, the controller can analyze the image via machine learning (e.g., artificial intelligence; computer vision) to detect the presence of at least one shelled ear of corn, such as based on detection of a pattern in the image. The method 700 also includes analyzing the image to identify one or more parameters of the at least one shelled ear of corn (block 706). As described herein, the controller can analyze the image via machine learning to identify one or more parameters, which can include the surface area of ​​the exposed segment, the number of missing kernels, the location of the exposed segment, or any combination thereof.

[0050] The method 700 also includes determining an appropriate output based on the image, such as based on one or more parameters of the at least one shelled ear of corn (block 708). For example, the controller can utilize machine learning to determine that the one or more parameters of the at least one shelled ear of corn indicate excessive product loss due to one or more factors and / or that can be addressed (e.g., reduced or prevented) via adjusting one or more factors (e.g., gaps between decks, angles of the corn head relative to the agricultural harvester, rotation rates of stalk rollers, or any combination thereof).

[0051] The method 700 also includes providing a control signal to adjust one or more components of the row unit, such as providing a control signal to an actuator to adjust the gap between decks, providing a control signal to an actuator to adjust the angle of the corn head relative to the agricultural harvester, providing a control signal to an actuator to adjust the rotation rate of the stalk drum, or any combination thereof (block 710). As described herein, the controller may also adjust the ground speed of the agricultural harvester, adjust the auger gap of the conveyor, and / or make other adjustments based on or to achieve an appropriate output.

[0052] The method 700 also includes providing a notification via the user interface (block 712). For example, the notification may include providing a notification to an operator of an agricultural harvester (e.g., Figure 1 A user-perceivable indication of an image (e.g., a video feed and / or a static image) is provided on a user interface or display 125 in the agricultural harvester. A user-perceivable indication of an image can be provided in real time (e.g., within milliseconds or seconds to allow processing and communication) as the agricultural harvester moves across the field. In some embodiments, the notification can include a suggested adjustment to one or more components of the row unit for operator approval and / or input (e.g., upon input approval, the controller sends a control signal to an actuator). In some embodiments, the notification includes a current setting, a command setting, or any combination thereof for one or more components of the row unit (e.g., a gap between layer plates) to enable an operator to understand the automated adjustments being made to address (e.g., limit) product loss due to bottom hulling.

[0053] By combining the systems and methods disclosed herein with an agricultural harvester, the process of harvesting crop material (e.g., corn) can be made more efficient. For example, the components of the row unit can be detected and controlled based on image processing of images of the crop material in the feed chamber, so that the components of the row unit can be placed as desired as crop conditions vary across the field.

[0054] Although only certain features have been illustrated and described herein, many modifications and changes may occur to those skilled in the art. It should be understood, therefore, that the appended claims are intended to cover all modifications and changes that fall within the true spirit of the present disclosure. It should be recognized that Figure 1-7As shown in or referenced in Figure 1-7 Any features described may be combined in any suitable way.

[0055] The technology presented and claimed herein is cited and applied to material objects and specific examples of a practical nature that significantly improve the art and are therefore not abstract, intangible, or purely theoretical. In addition, if any claim appended at the end of this specification contains one or more elements designated as "means for (performing) (function) ..." or "steps for (performing) (function) ...", it is intended that such elements should be interpreted in accordance with 35 U.S.C. § 112(f). However, for any claim containing elements designated in any other manner, it is intended that such elements not be interpreted in accordance with 35 U.S.C. § 112(f).

Claims

1. A detection and control system for an agricultural harvester, the detection and control system comprising: A controller comprising at least one memory and at least one processor, wherein the controller is configured to: receiving a sensor signal indicative of an image of harvested crop material within a feed chamber of an agricultural harvester; analyzing the image to detect at least one shelled ear of corn in the harvested crop material within the feed chamber; analyzing the image to identify one or more parameters of the at least one shelled ear of corn; as well as Appropriate adjustments to one or more components of the row unit are determined based on the one or more parameters of the at least one shelled ear of corn.

2. The detection and control system of claim 1, wherein the controller is configured to cause the image to be displayed to an operator of an agricultural harvester.

3. The detection and control system of claim 2, wherein the image comprises a video feed, a still image, or both.

4. The detection and control system of claim 1, wherein the controller is configured to provide control signals to one or more actuators to automatically effectuate appropriate adjustments to the one or more components of the row unit.

5. A detection and control system as described in claim 4, wherein the controller is configured to provide control signals to one or more actuators to adjust the corresponding gaps between the corresponding layer plates of the row unit to automatically achieve appropriate adjustment of the one or more components of the row unit.

6. The detection and control system of claim 4, wherein the controller is configured to provide control signals to one or more actuators to adjust the rotation rate of the corresponding stalk rollers of the row unit to automatically achieve appropriate adjustment of the one or more components of the row unit.

7. The detection and control system of claim 4, wherein the controller is configured to provide control signals to one or more actuators to adjust the angle of the row unit relative to the agricultural harvester to automatically achieve appropriate adjustment of the one or more components of the row unit.

8. The detection and control system of claim 1, wherein the controller is configured to provide a control signal to adjust a ground speed of the agricultural harvester based on the one or more parameters of the at least one shelled ear of corn.

9. The detection and control system of claim 1, wherein the controller is configured to utilize machine learning to detect the at least one shelled ear of corn of the harvested crop material in the feed chamber, identify the one or more parameters of the at least one shelled ear of corn, or any combination thereof.

10. The detection and control system of claim 1, wherein the controller is configured to utilize machine learning to determine appropriate adjustments to the one or more components of the row unit based on the one or more parameters of the at least one shelled ear of corn.

11. The detection and control system of claim 1 , comprising an internal imaging system comprising at least one camera configured to be located in or along a feed chamber between a row unit and a processing system of an agricultural harvester, wherein the at least one camera is configured to provide a sensor signal indicative of the image.

12. The detection and control system of claim 11, wherein the at least one camera is mounted below a lower plate defining a passage for harvested crop material to flow through the feed chamber.

13. A method for operating a detection and control system for an agricultural harvester, the method comprising: receiving, at a controller, a sensor signal indicative of an image of harvested crop material within a feed chamber of an agricultural harvester; analyzing the image using a controller to detect at least one shelled ear of corn in the harvested crop material in the feed chamber; analyzing the image using a controller to identify one or more parameters of the at least one shelled ear of corn; as well as A controller is used to determine appropriate adjustments to one or more components of the row unit based on the one or more parameters of the at least one shelled ear of corn.

14. The method of claim 13, wherein receiving sensor signals indicative of the image of harvested crop material comprises receiving sensor signals from an internal imaging system comprising at least one camera located at or along a feed chamber.

15. The method of claim 13, comprising displaying the image to an operator of the agricultural harvester via a controller.

16. The method of claim 13, comprising using a controller to provide control signals to one or more actuators to adjust respective gaps between respective plies of a row unit to automatically achieve appropriate adjustment of the one or more components of the row unit.

17. The method of claim 13, comprising using a controller to provide control signals to one or more actuators to adjust the rotation rate of corresponding stalk rollers of the row units to automatically achieve proper adjustment of the one or more components of the row units.

18. The method of claim 13, comprising using machine learning with a controller to detect the at least one shelled ear of harvested crop material in the feed chamber, identify the one or more parameters of the at least one shelled ear of corn, determine appropriate adjustments to the one or more components of the row unit based on the one or more parameters of the at least one shelled ear of corn, or any combination thereof.

19. An agricultural harvester comprising: a header, comprising a plurality of row units distributed across a width of the header; as well as A controller comprising at least one memory and at least one processor, wherein the controller is configured to: receiving a sensor signal indicative of an image of harvested crop material; as well as An undesirable level of seed loss in harvested crop material due to one or more parameters of one or more components of the plurality of row units is determined based on the images via machine learning.

20. The agricultural harvester of claim 19, further comprising an internal imaging system comprising at least one camera located within the feed chamber, wherein the internal imaging system is communicatively coupled to the controller to provide a sensor signal indicative of the image of harvested crop material within the feed chamber.

Images