Apparatus and method for size-by-size recording of storage locations of harvested tubers

Abstract

The invention relates to a device and a method for recording storage positions of harvested tubers by size. The present invention provides an apparatus and method for sizing storage locations of harvested tubers stacked by a boxing stacker. The method may include capturing one or more images of harvested tubers; segmenting individual tuber depictions visible in the image; determining one or more shape characteristics depicted by the plurality of individual tubers; identifying one or more unoccluded tuber depictions by determining that at least one of the single tuber depictions is an unoccluded tuber depiction based on shape characteristics of the single tuber depictions; measuring a size of a tuber corresponding to at least one of the unoccluded tuber depictions; assigning an average size to at least a tuber corresponding to the occluded tuber depiction based on the measured size; determining a storage location of the harvested tubers based at least in part on the boxing stacker location data; and recording the storage location and the average size in a memory.

Description

Technical Field

[0001] The present document relates to an apparatus and a method for recording the storage location by size of harvested tubers piled by a bin piler. Background Art

[0002] A tuber is an enlarged structure that serves as a storage organ for nutrients in some plants. Tuber crops are one of the most important crops in the world and form a large part of the world's food supply. Tubers include root tubers (e.g., potato, yam) and stem tubers (e.g., sweet potato, cassava).

[0003] In agricultural applications, tubers can often be graded based on their physical attributes, such as size. The intended application and pricing of the tubers can be based on the grading results. For example, large, elongated potatoes that can be used to produce French fries and potato chips may be priced higher. Smaller, irregularly shaped potatoes that can be used for other processed products, such as mashed potatoes or potato chips, may be priced lower.

[0004] Timely grading of tubers after harvesting allows for proper pricing and logistics operations (storage, processing and distribution). This can improve operational efficiency and reduce post-harvest waste. Summary of the invention

[0005] The following summary is provided to introduce the reader to the more detailed discussion below. The summary is not intended to limit or define any claimed or unclaimed invention. One or more inventions may reside in any combination or subcombination of elements or process steps disclosed in any part of this document, including its claims and drawings.

[0006] According to some aspects, a method for recording storage locations of harvested tubers stacked by a case stacker by size is provided. The method may include: using a camera to capture one or more images of harvested tubers moving on a conveyor toward the case stacker; segmenting, by a processor, individual tuber depictions visible in the one or more images; determining, by the processor, one or more shape characteristics of a plurality of individual tuber depictions; identifying, by the processor, one or more unobstructed tuber depictions by determining, based on the one or more shape characteristics of the individual tuber depictions, that at least one of the individual tuber depictions is an unobstructed tuber depiction, the tuber depictions including an occluded tuber depiction; measuring, by the processor, the size of tubers corresponding to at least one of the unobstructed tuber depictions; assigning, by the processor, an average size to at least the tubers corresponding to the occluded tuber depictions based on the size of the tubers corresponding to at least one of the unobstructed tuber depictions; determining, by the processor, a storage location of the harvested tubers based at least in part on the case stacker position data; and recording the storage location and the average size of the harvested tubers in a memory.

[0007] According to some aspects, a device for recording storage locations of harvested tubers by size is provided. The device may include a packer, a conveyor, a camera, a positioning system, and a control system. The packer may be configured to stack the harvested tubers at a storage location. The conveyor may be configured to move the harvested tubers toward the packer. The camera may be configured to capture one or more images of the harvested tubers moving on the conveyor. The positioning system is configured to generate packer position data indicating the location of the packer. The control system may have at least one processor and a memory. At least one processor can be configured to collectively: receive one or more images from a camera; segment individual tuber depictions visible in one or more images; determine one or more shape characteristics of multiple individual tuber depictions; identify one or more unobstructed tuber depictions by determining that at least one of the individual tuber depictions is an unobstructed tuber depiction based on the one or more shape characteristics of the individual tuber depictions, the tuber depictions including occluded tuber depictions; measure the size of tubers corresponding to at least one of the unobstructed tuber depictions; assign an average size to at least the tubers corresponding to the occluded tuber depictions based on the size of the tubers corresponding to at least one of the unobstructed tuber depictions; determine a storage location for harvested tubers based at least in part on case stacker position data; and record the storage location and average size of the harvested tubers in a memory.

[0008] According to some aspects, a non-transitory computer-readable medium including instructions executable by a processor is provided. The instructions, when executed, may configure the processor to: receive one or more images of harvested tubers moving on a conveyor toward a case stacker from a camera; segment individual tuber depictions visible in the one or more images; determine one or more shape characteristics of a plurality of individual tuber depictions; identify one or more unobstructed tuber depictions by determining that at least one of the individual tuber depictions is an unobstructed tuber depiction based on the one or more shape characteristics of the individual tuber depictions, the tuber depictions including an occluded tuber depiction; measure the size of a tuber corresponding to at least one of the unobstructed tuber depictions; assign an average size to at least the tubers corresponding to the occluded tuber depiction based on the size of the tuber corresponding to at least one of the unobstructed tuber depictions; determine a storage location of the harvested tubers based at least in part on the case stacker position data; and record the storage location and the average size of the harvested tubers in a memory. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings included herein are for illustrating various examples of the articles, methods and apparatus of the present specification and are not intended to limit the scope of the teachings in any way. In the drawings:

[0010] Figure 1is a schematic diagram of an apparatus for recording storage locations of harvested tubers by size according to an embodiment;

[0011] Figure 2 yes Figure 1 A block diagram of the control system of the device;

[0012] Figure 3 is a flow chart illustrating an example method for recording storage locations by size of harvested tubers stacked by a case stacker;

[0013] Figure 4A is an example captured image of a plurality of potatoes moving on a conveyor according to an embodiment;

[0014] Figure 4B It is for Figure 4A The segmented image generated by the example captured image;

[0015] Figure 4C Shown in Figure 4B Binary masks of the five unoccluded potato depictions identified in the single potato depiction shown;

[0016] Figure 5A shows a bounding box fitted around an example binary mask depicted on a single tuber in accordance with an embodiment;

[0017] Figure 5B shows an angle-adjusted bounding box fitted around an example binary mask depicted by a single tuber in accordance with an embodiment;

[0018] Figure 5C shows the fit to Figure 5B An example binary mask of the outer contour of the ellipse; and

[0019] Figure 5D Shown are deviations of an example binary mask of a single tuber depiction from its convex hull, in accordance with an embodiment. DETAILED DESCRIPTION

[0020] Many embodiments are described in this application, and these embodiments are presented for illustrative purposes only. The described embodiments are not intended to be limiting in any sense. The present invention is widely applicable to many embodiments, as is apparent from the disclosure herein. Those skilled in the art will recognize that the present invention can be practiced by modification and change without departing from the teachings disclosed herein. Although the specific features of the present invention may be described with reference to one or more specific embodiments or the accompanying drawings, it should be understood that such features are not limited to the use in one or more specific embodiments or the accompanying drawings to which they are described with reference.

[0021] The terms "embodiment," "one or more embodiments," "some embodiments," and "one embodiment" mean "one or more (but not all) embodiments of the invention," unless expressly stated otherwise.

[0022] The terms "include," "comprising," and variations thereof mean "including but not limited to," unless expressly stated otherwise. A listing of items does not imply that any or all of the items are mutually exclusive, unless expressly stated otherwise. The terms "a," "an," and "the" mean "one or more," unless expressly stated otherwise.

[0023] As used herein and in the claims, two or more parts are referred to as being "coupled," "connected," "attached," "engaged," "attached," or "fastened," wherein the parts are joined or work together directly or indirectly (i.e., through one or more intermediate parts) as long as there is a link. As used herein and in the claims, two or more parts are referred to as being "directly coupled," "directly connected," "directly attached," "directly engaged," "directly attached," or "directly fastened," wherein the parts are connected in physical contact with one another. As used herein, two or more parts are referred to as being "rigidly connected," "rigidly connected," "rigidly attached," "rigidly engaged," "rigidly attached," or "rigidly fastened," wherein the parts are coupled so as to move as a unit while maintaining a constant orientation relative to one another. The terms "coupled," "connected," "attached," "engaged," "attached," and "fastened" do not distinguish the manner in which two or more parts are joined together.

[0024] In addition, although the method steps may be described in order (in the disclosure and / or claims), such methods may be configured to work in an alternating order. In other words, any order or sequence of steps that may be described does not necessarily imply a requirement to perform the steps in that order. The steps of the method described herein may be performed in any feasible order. In addition, some steps may be performed simultaneously.

[0025] As used herein and in the claims, a group of elements is referred to as 'jointly' performing an action where the action is performed by any one of the elements of the group or is performed cooperatively by two or more (or all) of the elements of the group.

[0026] As used herein and in the claims, a first element is referred to as being “received” in a second element, wherein at least a portion of the first element is received in the second element, unless specifically stated otherwise.

[0027] Some elements herein may be identified by a part number consisting of a base number followed by an alphabetic or subscript numeric suffix (e.g., 112a or 1121). Multiple elements herein may be identified by part numbers that share the same base number but with different suffixes (e.g., 1121, 1122, and 1123). All elements having a common base number may be collectively or generally referred to using the base number without the suffix (e.g., 112).

[0028] As used herein and in the claims, "upward," "downward," "above," "below," "upward," "vertical," "elevated," "up," "down," and similar terms refer to directionality that is generally aligned with (e.g., parallel to) gravity. However, none of the terms mentioned in this paragraph imply any particular alignment between elements. For example, a first element may be referred to as being "vertically above" a second element, wherein the first element is at a greater height than the second element, regardless of whether the first element is vertically aligned with the second element.

[0029] Timely grading of tubers after harvest may require a large amount of labor resources. The tubers harvested can be transported and stacked in a storage facility using a conveyor with high throughput. For example, the conveyor throughput can be greater than 200 kilograms per minute. Manually measuring the size of each tuber harvested, or even a small portion of each tuber harvested, may be expensive, time-consuming and labor-intensive, requiring a large amount of labor and time investment. The disclosed apparatus and method can reduce (or almost eliminate) cost, time and labor requirements by automating the size measurement and position tracking of the tubers harvested.

[0030] Timely and accurate grading of harvested tubers can be challenging. Conveyors used to transport harvested tubers to storage can be narrow and result in the accumulation and obstruction of tubers. For example, where a stack of tubers is formed on a conveyor, the top tubers will be located above the lower tubers, thereby obscuring (i.e., blocking visibility) at least a portion of the lower tubers. When imaging is used to assist in this process, the obscured portions may hinder accurate size assessment. Other challenges may include the presence of mechanical vibrations, ambient light, and foreign matter (e.g., dirt), which can vary depending on the plot of land in the field where the tubers are harvested.

[0031] The disclosed apparatus and method can improve the accuracy of size measurements by distinguishing fully visible (i.e., unobstructed) tubers from occluded tubers in an image. A sampling of unobstructed tubers can be selected for size measurement, and an average size can be assigned to nearby harvested tubers (including occluded tubers) based on the size measurements of the sampled unobstructed tubers. This avoids inaccuracies associated with assessing the size of occluded tubers.

[0032] Reference now Figure 1 , Figure 1 A schematic diagram of an apparatus 100 for recording storage locations of harvested tubers 10 by size is shown in FIG. In the example shown, the apparatus 100 includes a case stacker 104, a conveyor 108, a camera 112, a control system 116, and a positioning system 120.

[0033] The harvested tubers 10a can be stacked at a first end 128 of the conveyor 108. For example, the harvested tubers can be transported from the field and stacked at the first end 128 by the cart 14. The second end 132 of the conveyor 108 can be connected to the box stacker 104. The conveyor 108 can have any suitable design to move the harvested tubers 10 from the first end 128 toward the box stacker 104 located at the second end 132. For example, the conveyor 108 can be a belt conveyor or a motorized roller conveyor.

[0034] The conveyor 108 may include one or more conveyor segments. For example, the conveyor 108 may include a single straight conveyor segment extending from the first end 128 to the second end 132. In the illustrated embodiment, the conveyor 108 is shown as including a plurality of conveyor segments (e.g., conveyor segments 110a-110d extending from the grading station 22 to the storage facility 18). Each conveyor segment 110 may extend in the same or different direction as an adjacent segment and / or may be inclined the same or differently than an adjacent segment. In some examples, each conveyor segment may move at the same or different speed as an adjacent segment.

[0035] Optionally, the apparatus 100 may include one or more sensors 148 configured to measure the speed of the conveyor segments 110. The apparatus 100 may include any suitable number of sensors 148 (e.g., 1 to 100 sensors). For example, the apparatus 100 may include one speed sensor 148 for each conveyor segment 110, each conveyor segment 110 being capable of moving at a different speed than an adjacent segment 110. In the illustrated embodiment, the sensors 148a-148d are configured to measure the speed of the conveyor segments 110a-110d, respectively.

[0036] The sensor 148 may be of any design suitable for measuring the speed of the conveyor segment 110. For example, the sensor 148 may be a reed 59060 magnetic shaft sensor that may provide a non-contact measurement of the conveyor speed. The reed 59060 magnetic shaft sensor may measure the revolutions per second of any one of the shafts about which the conveyor rotates. The speed of the conveyor segment may be determined based on the measured revolutions per second and the radius of the shaft according to the following equation 1:

[0037] Conveyor speed (m / s) = measured revolutions (1 / s) x 2π x shaft radius (m) (Equation 1).

[0038] In other examples, different sensors may be used. Sensor 148 may be wired or wireless (e.g., using LoRa, or Bluetooth TM ) provides measured conveyor speed data to the control system 116. Measuring the speed of the conveyor segment 110 allows the control system 116 to more accurately predict when tubers passing through the grading station 22 will be discharged from the box stacker 104 into the tuber pile 10c in the storage facility 18. This allows the storage location 30 of the tubers within the storage facility 18 to be more accurately determined and recorded. For example, the speed of the conveyor segment 110 can vary over time, depending on factors such as the density of the tubers carried by the conveyor segment 110. Therefore, the transit time between the grading station 22 and the discharge end of the box stacker 104 may also vary over time. Such variations can be accounted for by continuously measuring the speed of the conveyor segment 10.

[0039] In other embodiments, the speed(s) of the conveyor segment(s) 110 are constant or the transit time from the grading station 22 to the discharge end of the case stacker 104 is constant, so that the time when the tubers will be discharged from the case stacker 104 after passing through the grading station can be accurately predicted without measuring the speed(s) of the conveyor segment(s) 110 (i.e., without using the sensor(s) 148).

[0040] A grading station 22 may be mounted along the conveyor 108 for inspecting and grading the harvested tubers. A camera 112 may be positioned at the grading station 22. The camera 112 may be any suitable camera for capturing images of the harvested tubers 10b moving along the conveyor 108. In some embodiments, the camera 112 may be an RGB camera that captures images of the harvested tubers in the visible range of the electromagnetic spectrum. In other embodiments, the camera 112 may be any other camera, such as a hyperspectral camera that can capture images in the visible and infrared (IR) ranges of the electromagnetic spectrum. Hyperspectral cameras may provide additional information compared to RGB cameras, such as information about internal defects of the imaged tubers. RGB cameras may achieve lower cost and system complexity compared to hyperspectral cameras.

[0041] The camera 112 may provide captured images (e.g., individually captured photos or selected video frames) to the control system 116 directly (e.g., via wires or wirelessly) or indirectly via the network 26. The network 26 may include a communications network, such as the Internet, a wide area network (WAN), a local area network (LAN), or another type of network. The camera 112 may communicate with the network 26 using a communications module 144. In some examples, the communications module 144 may be integrated into the camera 112. In other examples, the communications module 144 may be a separate module (e.g., a Wi-Fi or network dongle) that is communicatively coupled to the camera 112.

[0042] In some embodiments, the control system 116 may control the image capture rate of the camera 112 based on the positioning and image capture parameters of the camera 112 and the measured speed of the conveyor 108. For example, the control system 116 may control the image capture rate of the camera 112 based on conveyor speed data received from the sensor 148 associated with the conveyor segment 110a in the field of view of the camera 112. The image capture rate of the camera 112 may be controlled to avoid capturing multiple images including the same tuber.

[0043] For example, the field of view of the camera 112 may include a 1.3 meter length of the conveyor 108 based on the positioning and image capture parameters of the camera 112. Thus, this length of the conveyor 108 may be included in each image captured by the camera 112. For a conveyor speed of 1 meter per second, the control system may set the image capture rate of the camera 112 to 0.5 images per second (i.e., 1 image is captured every 2 seconds) to avoid capturing the same tuber in multiple images. For a conveyor speed of 0.5 meters per second, the control system may set the image capture rate of the camera 112 to 0.25 images per second (i.e., 1 image is captured every 4 seconds) to avoid capturing the same tuber in multiple images. Equation 2 below provides the relationship between the image capture rate (f) in images per second and the conveyor speed (v) in meters per second for the example case of capturing a 1.3 meter length of the conveyor in each image:

[0044] f = 0.5v (Equation 2).

[0045] The case stacker 104 can have any design suitable for stacking the harvested tubers into the storage facility 18. For example, the case stacker 104 can include an elevator 136 and a boom 140. The elevator 136 can control the vertical height from which the harvested tubers are discharged from the case stacker 104. The boom 140 can control the lateral (e.g., horizontal XY) position of the harvested tubers discharged from the case stacker 104. The combination of the elevator 136 and the boom 140 can control the storage position 30 of the harvested tubers 10c. Other case stacker designs can have alternative and / or additional movable elements that are used to control the three-dimensional position at which the harvested tubers are discharged into the storage facility 18.

[0046] The positioning system 120 may use any suitable technique to track the movement of the case stacker 104 and generate the case stacker location data. For example, the positioning system 120 may use dead reckoning or tracking to nearby anchor nodes (i.e., nodes with known fixed locations, such as Access point, beacons or ultra wideband beacons) to track the case palletizer 104.

[0047] In the illustrated schematic, the positioning system 120 includes a tag 120a positioned at the discharge end of the case palletizer 104 and a plurality of anchor tags 120b ( Figure 1 Three anchor tags are shown in FIG, but four or more tags are preferred to derive 3D coordinates). Each tag 120 may include an impulse radio ultra-wideband (IR-UWB) transmitter-receiver (e.g., with Qorvo TM The position of the discharge end of the case palletizer 104 can be tracked in real time based on two-way time-of-flight (TOF) measurements between the tag 120a and multiple anchor tags 120b. The high bandwidth provided by the IR-UWB transmitter-receiver can enable the case palletizer movement to be tracked in three dimensions with high accuracy.

[0048] The positioning system 120 may provide the generated case stacker position data to the control system 116 using any suitable communication technology. For example, the positioning system 120 (e.g., tag 120a) may use LoRa to transmit the case stacker position data to the communication module 144 to enable low power operation of the positioning system 120. Alternatively or additionally, the communication module 144 may use the network 26 to provide the received case stacker position data to the control system 116. In other examples, the positioning system 120 (e.g., tag 120a) may provide the generated case stacker position data directly to the control system 116. Alternatively or in addition to using LoRa, the positioning system 120 may use other technologies (e.g., or via a wired connection) to transmit the position data of the case stacker.

[0049] Alternatively or additionally, the positioning system 120 may include sensors that detect the tilt of the lift 136 and / or the extension of the boom 140 and / or the extension of one or more of the conveyor segments 110. For example, one or more of the conveyor segments 110 and / or the case stacker boom 140 may be extendable in length to accommodate the stacking of tubers at the target storage location. Such a change in length may affect the accuracy of the predicted delivery time from the grading station 22 to the discharge end of the case stacker 104, where the delivery time is predicted based on the measured speed of (one or more) of the (one or more) conveyor segments 110. In some embodiments, there may be a sensor associated with each of one or more (or all) of the extendable conveyor segments 110 and / or the case stacker boom 140 to determine the length of these elements, whereby the predicted delivery time (measured as the length divided by the travel speed across each element) may be determined based on the accurate length.

[0050] Any suitable sensor may be used to determine the current length of the extendable conveyor segment(s) 110 and / or the case handler boom 140. The illustrated example shows some of the conveyor segments 110 including sensors 121, which may be hardware similar to the sensors 120a or 120b described above. For example, data from the sensors 121 may be sent to the control system 116, which may determine the 2D (two-dimensional) or 3D (three-dimensional) position of each sensor 121 and derive the length of the associated segment 110 therefrom. Similarly, the length of the case handler boom 140 may be derived based on the position data from the sensor 120a and / or additional sensors.

[0051] Reference now Figure 1 and Figure 2 . Figure 2is a block diagram of an example control system 116. In some embodiments, the control system 116 can be implemented as a server device (e.g., a remote server). In other embodiments, the control system 116 can be implemented using any other suitable hardware components. For example, the control system 116 can be implemented as a device physically integrated with the grading station 22.

[0052] for Figure 2 In the illustrated example embodiment, the control system 116 includes a memory 152, an application 156, an output device 160, a display device 164, an auxiliary storage device 168, a processor 172, an input device 176, and a communication device 180. One or more (or all) of the memory 152, the application 156, the output device 160, the display device 164, the auxiliary storage device 168, the processor 172, the input device 176, and the communication device 180 may be communicatively coupled by wire and / or wirelessly.

[0053] In some embodiments, the control system 116 includes any one or more of the memory 152, the application 156, the output device 160, the display device 164, the secondary storage device 168, the processor 172, the input device 176, and the communication device 180. In some embodiments, the control system 116 does not include one or more of the application 156, the secondary storage device 168, the network connection, the input device 176, the output device 160, the display device 164, and the communication device 180.

[0054] The memory 152 may include one or more of a random access memory (RAM) and a read-only memory (ROM). In some embodiments, the memory 152 stores one or more applications 156 executed by the processor 172. The application 156 corresponds to a software module including computer executable instructions that, when executed by one or more processors, cause the one or more processors to collectively perform processing for the functions and methods described herein.

[0055] The memory 152 can store one or more machine learning models. The machine learning model can be trained to segment individual tubers visible in the captured image and / or identify unobstructed tubers in the segmented individual tubers. The machine learning model can be generated and trained by the control system 116. In some embodiments, the control system 116 can receive the generated machine learning model from an external device (e.g., via the network 26) and can train the received machine learning model. The control system 116 can use labeled training data to perform supervised training of the received model. In some embodiments, the control system 116 can receive the trained machine learning model from an external device (e.g., via the network 26) and store the trained machine learning model in the memory 152.

[0056] In some embodiments, the memory 152 may also store the determined storage location and the average size of the harvested tubers. The stored data may be provided to a user, for example, via the network 26.

[0057] The secondary storage device 168 may include any suitable non-transitory computer-readable medium that includes instructions that can be executed by a processor (e.g., processor 172). For example, the secondary storage device 168 may include a hard disk drive, a floppy disk drive, a CD drive, a DVD drive, a Blu-ray drive, a solid-state drive, flash memory, or other types of non-volatile data storage. The processor 172 may collectively execute the instructions included on the secondary storage device 168 to perform processing for the functions and methods described herein.

[0058] In some embodiments, the control system 116 stores information in a remote storage device (such as cloud storage) that is accessible via a network (e.g., network 26 or another network). In some embodiments, the control system 116 stores information distributed across multiple storage devices (such as memory 152 and auxiliary storage device 168) (i.e., each of the multiple storage devices stores a portion of the information, and the multiple storage devices collectively store all of the information). Therefore, as used herein and in the claims, storing data on a storage device means storing the data in a local storage device, storing the data in a remote storage device, or distributing the data across multiple storage devices, each of which may be local or remote.

[0059] The input device 176 may include any device for inputting information into the control system 116. For example, the input device 176 may be a keyboard, a keypad, a cursor device, a touch screen, a camera, or a microphone. The input device 176 may also include input ports and wireless radios (e.g., or 802.11x).

[0060] Display device 164 may include any type of device for presenting visual information. For example, display device 164 may be a computer monitor, a flat screen display, a projector, or a display panel.

[0061] Output device 160 may include any type of device for presenting a hard copy of information, such as, for example, a printer. Output device 160 may also include other types of output devices, such as, for example, a speaker. In at least one embodiment, output device 160 includes an output port and a wireless radio (e.g., or 802.11x).

[0062] The communication device 180 may be of any design suitable for receiving analog and / or digital inputs therefrom and providing analog and / or digital outputs. In some embodiments, the communication device 180 may include separate modules for analog and digital signals.

[0063] Processor 172 may be any device capable of collectively executing an application, computer-readable instruction, or program. For example, the application, computer-readable instruction, or program may be stored in memory 152 or secondary storage device 168, or may be received from a remote storage device accessible via network 26. Processor 172 may be a high-performance general-purpose processor, a standard processor (e.g., Processor or processor), dedicated hardware (e.g., a GPU), or multiple processing devices that collectively perform the functions provided by processor 172.

[0064] Figure 2 An example of a control system 116 is shown. In alternative embodiments, the control system 116 includes fewer, additional, or different components. In addition, although various aspects of the implementation of the control system 116 are described as being stored in a memory, those skilled in the art will understand that these aspects may also be stored on or read from other types of computer program products or computer-readable media (such as auxiliary storage devices, including hard disks, floppy disks, CDs, or DVDs), carrier waves from the Internet or other networks, or other forms of RAM or ROM. For example, the control system 116 may include a non-transitory computer-readable medium storing computer-readable instructions that, when executed by the processor 172, configure the processor 172 to collectively perform the methods described herein.

[0065] Reference now Figure 3 , Figure 3 Shown is a flow chart illustrating an example method 200 of a storage location of tubers of the harvest piled up by a packing stacker by size record. The tubers of the harvest can be, for example, all tubers of the harvest from a field. In certain embodiments, the tubers of the harvest can be a part of the tubers of the harvest from a field. In some examples, the part can be limited by duration (for example, the tubers of the harvest moving on a conveyor during a prescribed (for example, one minute or one hour) duration or by quantity (for example, the quantity of the cart load of tubers)).

[0066] Method 200 may, for example, use device 100 ( Figure 1 ) to execute, and also refer to Figure 1 and Figure 2The method 200 may be automatically performed in response to detecting movement of the conveyor 108. In some embodiments, the method 200 may be performed in response to user input or based on a stored schedule.

[0067] In some cases, the harvested tuber can be a potato. In other cases, the harvested tuber can be any other tuber (e.g., sweet potato, yam, cassava, etc.).

[0068] At 204, one or more images of the harvested tubers 10 moving on the conveyor toward the case stacker may be captured. For example, the images may be captured by the camera 112. The camera 112 may provide the captured images to a control system (e.g., the control system 116).

[0069] In some embodiments, a portion of the harvested tubers may be imaged at 204. Imaging on a portion of the harvested tubers may provide a reduction in computing resource requirements. In other embodiments, all of the harvested tubers may be imaged at 204. This may provide a greater amount of data to determine the size of the harvested tubers, which may allow for a more accurate size determination.

[0070] At 208, the individual tuber depictions visible in the captured image may be segmented. For example, the processor 172 of the control system 116 may segment the individual tuber depictions visible in the captured image. The individual tuber depictions may be segmented by generating a binary mask indicating a detected area of ​​each individual tuber depiction and generating a bounding box surrounding the detected area of ​​the individual tuber depiction.

[0071] In some embodiments, a suitable machine learning model may be used to segment a single tuber depiction. For example, processor 172 may use a suitable machine learning model stored in memory 152 to segment a single tuber depiction.

[0072] An appropriate machine learning model may be selected based on factors including accuracy, speed, and computational resource requirements. In some embodiments, a masked region-based convolutional neural network (R-CNN) model may be used to segment individual tuber depictions visible in a captured image. In other embodiments, a different model (e.g., a U-net) may be used to segment individual tuber depictions visible in a captured image.

[0073] Reference now Figure 4A and Figure 4B . Figure 4A An example captured image 304 of a plurality of potatoes moving on a conveyor is shown. The moving potatoes may cluster and some potatoes may be overlaid on other potatoes. As can be seen in captured image 304, clustering may result in occlusion of some of the imaged potatoes. Figure 4B An example segmented image 308 generated using a Mask R-CNN model for, for example, captured image 304 is shown. The trained Mask R-CNN model may receive captured image 304 as input and provide segmented image 308 as output. Segmented image 308 includes individual potato depictions segmented from each other (instance segmentation) and from a conveyor (semantic segmentation).

[0074] The Mask R-CNN model architecture may include a backbone stage and a head stage. In some embodiments, the backbone stage may be constructed using a Feature Pyramid Network (FPN), a Residual Network (ResNet), a Region Proposal Network (RPN), and a Region of Interest (ROI) alignment layer. The backbone stage may generate predictions for regions in a captured image that may contain tuber depictions. The head stage may include fully connected layers where classification, bounding box, and mask predictions may be generated using region predictions from the backbone stage.

[0075] Any suitable platform can be used to implement the Mask R-CNN model architecture. In some embodiments, the detectron2 library can be used. In other embodiments, a different platform (e.g., MMDetection) can be used.

[0076] Return to reference Figures 1 to 3 At 212 of method 200 , one or more shape characteristics of the individual tuber delineations may be determined. For example, processor 172 may determine one or more shape characteristics of the individual tuber delineations segmented at 208 .

[0077] The shape characteristics may include, for example, one or more color-based feature parameters (S / N_1 and S / N_2), edge-based feature parameters (avg_val_bbox, ellipticity, ellipticity ratio, roundness, convexity defect), and / or seven image invariant moments (H1-H7) obtained from the HuMoments method of the OpenCV Python library. Further information about these shape characteristics is provided below. In some embodiments, step 212 is omitted because one or more shape characteristics are pre-stored (e.g., in memory 152 or storage device 168) and can be accessed by processor 172 (see Figure 2 ).

[0078] Reference now FIG. 5A to FIG. 5D . Figure 5A A bounding box 320 of an example binary mask fit around a single tuber depiction is shown. Figure 5B An angle-adjusted bounding box 332 is shown fitting around an example binary mask depicting a single tuber. Figure 5C An ellipse 336 is shown fitted to the outer contour of an example binary mask depicting a single tuber. Figure 5DDeviations 340 of an example binary mask depicting a single tuber from its convex hull are shown.

[0079] The avg_val_bbox edge-based feature parameters may be determined based on the average density of tuber mask pixels along the edge of the bounding box. For example, the processor 172 may calculate Figure 5A 324 and 328). The processor 172 may also calculate the number of tuber mask pixels along the edge of the bounding box 320 (corresponding to edge portions 324 and 328). The processor 172 may determine the avg_val_bbox edge-based feature parameter using a ratio of the tuber mask pixels along the edge of the bounding box 320 to the total number of pixels along the edge of the bounding box 320. A high value of this ratio (e.g., above a threshold) may indicate that the tuber represented by the binary mask is substantially truncated (e.g., an occluded tuber or truncated by the edge of the captured image) and should be considered an occluded tuber.

[0080] The S / N_1 color-based feature parameter can be determined based on the ratio of the number of tuber mask pixels to the total number of pixels enclosed by the bounding box. Figure 5A Based on the example bounding box 320 shown in FIG. 1 , the processor 172 may calculate a ratio of the number of tuber mask pixels (white pixels 352 ) to the total number of pixels (black pixels 356 and white pixels 352 ) enclosed by the bounding box 320 .

[0081] The S / N_2 color-based feature parameter can be determined based on the ratio of the number of tuber mask pixels to the total number of pixels enclosed by the bounding box. An angle-adjusted bounding box can be defined to capture the angular orientation of the tuber mask. Figure 5B , the processor 172 may calculate a ratio of the number of tuber mask pixels (white pixels 360) to the total number of pixels (gray pixels 364 and white pixels 360) enclosed by the angle-adjusted bounding box 332. The angle-adjusted bounding box may be smaller than the corresponding unadjusted bounding box and result in a higher S / N_2 parameter value than the corresponding S / N_1 parameter value.

[0082] The feature parameter based on the ellipticity edge can be determined based on the ratio of the area of ​​the ellipse (fitted to the outer contour of the tuber mask pixels) to the number of tuber mask pixels. For example, the processor 172 can calculate the area of ​​the ellipse that fits to the outer contour of the tuber mask pixels. Figure 5C The area of ​​the ellipse 336 of the outer contour of the tuber mask pixels shown in . The processor 172 can calculate the area of ​​the ellipse 336 using the following equation 3:

[0083] A = πab / 4 (Equation 3) where a is the length of the major axis 368 of the ellipse 336, b is the length of the minor axis 372 of the ellipse 336, and A is the computer area of ​​the ellipse 336. The processor 172 may determine the ratio of the area A to the number of tuber mask pixels to determine the ellipticity.

[0084] The feature parameter based on the ellipse ratio edge can be determined based on the ratio of the number of tuber mask pixels enclosed by the outline of the fitted ellipse to the number of tuber mask pixels. For example, the processor 172 can calculate the ratio of the number of tuber mask pixels enclosed by the outline of the fitted ellipse to the number of tuber mask pixels. Figure 5C The ratio of the number of tuber mask pixels enclosed by the outline of the ellipse 336 (the number of white pixels enclosed by the ellipse 336) to the total number of tuber mask pixels (the total number of white pixels) is shown in FIG.

[0085] The feature parameter based on the roundness edge can be determined based on how close the shape of the tuber mask is to a circle. For example, the processor 172 can use the following equation 4 to determine the roundness parameter value:

[0086]

[0087] where area is the area of ​​the tuber mask (e.g., the number of tuber mask pixels (e.g., Figure 5C ), and perimeter is the perimeter of the tuber mask (e.g., the number of edge pixels that form the perimeter of the tuber mask).

[0088] The characteristic parameters based on the convexity defect edge can be determined according to the deviation of the tuber mask from its convex hull. Figure 5D In the example shown, the processor 172 may determine a convexity defect parameter value based on a deviation 340 of the tuber mask from its convex hull 344 .

[0089] Return to reference Figures 1 to 3 At 216 of method 200, one or more unoccluded tuber depictions may be identified by determining that at least one of the individual tuber depictions is an unoccluded tuber depiction. Figure 4A As shown, the captured image 304 may depict a plurality of occluded tubers and unoccluded tubers. Based on the shape characteristics determined for the individual tuber depictions at 212, the segmented individual tuber depictions may be identified as unoccluded tuber depictions. For example, the processor 172 of the control system 116 may identify one or more unoccluded tuber depictions at 216.

[0090] Unobstructed tuber depictions can be identified as fully visible tubers and partially visible tubers based on a binary classification of segmented individual tuber depictions. In some embodiments, a threshold-based sampling method can be used to identify unobstructed tuber depictions. In other embodiments, a machine learning-based sampling method can be used to identify unobstructed tuber depictions. Compared to the threshold-based sampling method, the machine learning-based sampling method can provide higher classification accuracy.

[0091] Any suitable number of unobstructed tuber depictions may be sampled in each image. In some embodiments, up to 1 to 5 unobstructed tuber depictions may be sampled in each image. Smaller sample sizes may reduce computing resource requirements. In other embodiments, a larger number of unobstructed tuber depictions (e.g., up to 5 to 10) may be sampled in each image. Unobstructed tuber depictions less than a defined maximum number may be sampled, wherein less than a defined maximum number of unobstructed tuber depictions are present in the image. In general, higher sample sizes may provide higher accuracy in determining the average size of tubers conferred for harvest, but may require a larger amount of computing resources.

[0092] In the threshold-based sampling method, one or more of the shape characteristics can be used as thresholding parameters. The thresholding parameters (binary classification as occluded / unoccluded) can be selected based on a predetermined correlation coefficient between the parameter values ​​of the shape characteristics and the response variable. For example, the thresholding parameters can include S / N_1, S / N_2, ellipticity, roundness, convexity defects, and two Hu image moments (H3 and H4). In other examples, different thresholding parameters can be used.

[0093] Processor 172 may identify occluded tuber depictions by comparing the determined value of the thresholding parameter to a predetermined threshold value. In some embodiments, the predetermined threshold value may be based on an average, a range, a 25th percentile value, a 50th percentile value, and a 75th percentile value for both the fully visible tuber category and the partially visible tuber category.

[0094] In the sampling method based on machine learning, any suitable machine learning model can be used. In some embodiments, a trained random forest model can be used. In other embodiments, other trained models can be used, for example, a support vector machine (SVM) or a logistic regression model. The random forest model can provide higher classification accuracy than other models for identifying unobstructed tuber depictions.

[0095] In some embodiments, the processor 172 may provide a subset of one or more shape characteristics as input to a selected machine learning model (e.g., a random forest model). The subset of shape characteristics may be predetermined based on sequential forward feature selection to identify features with the highest predictive power. For example, the processor 172 may provide parameter values ​​corresponding to convexity defect, ellipticity ratio, ellipticity, roundness, avg_val_bbox, and S / N_1 as input to the machine learning model.

[0096] Reference now Figure 4C , Figure 4C Shown in FIG. 3 is an image 312 that provides a graphical illustration of the output generated by the machine learning model. Figure 4B Five unoccluded potato depictions identified in the segmented single tuber depiction shown.

[0097] Return to reference Figures 1 to 3 At 220 of method 200, a size of a tuber corresponding to at least one of the unoccluded tuber depictions (identified at 216) may be measured. For an example embodiment in which five unoccluded tuber depictions are identified in each image, processor 172 may measure a size of a tuber corresponding to each of the five unoccluded tuber depictions.

[0098] The measured size may include at least one of a one-dimensional length, a projected area, and a volume. Figure 4C In the example image shown in , processor 172 can measure the major diameter of the ellipse fitted to each unoccluded potato tuber. In some embodiments, processor 172 can measure the minor diameter of the ellipse or measure both the major diameter and the minor diameter.

[0099] In some embodiments, processor 172 may measure the projected area of ​​the tuber corresponding to the unobstructed tuber depiction. For example, processor 172 may measure the projected area of ​​the tuber corresponding to the unobstructed tuber depiction. Figure 4C The two-dimensional region corresponding to the tuber mask depicted by each unoccluded potato is shown in .

[0100] In some embodiments, processor 172 may measure the volume of the tuber corresponding to the unoccluded tuber depiction. For example, processor 172 may measure the volume based on the volume obtained by rotating the projected area of ​​the unoccluded tuber around the major axis of the fitted ellipse.

[0101] At 224, an average size may be assigned to at least the tubers corresponding to the occluded tuber depiction based on the sizes measured at 220. For example, at 220, processor 172 may assign an average size to at least the tubers corresponding to the occluded tuber depiction based on the measured sizes.

[0102] Average size can be assigned based on the measured size of multiple unobstructed tuber depictions. For example, multiple unobstructed tuber depictions can be measured in each captured image and / or multiple images of harvested tubers can be captured, wherein each image includes one or more measured unobstructed tuber depictions. Average size can be assigned to tubers corresponding to the obstructed tuber depictions in the captured image. Average size can also be assigned to tubers corresponding to the unobstructed tuber depictions in the captured image and / or any tuber that is not imaged. For example, average size can be assigned to tubers of a certain quantity or quality, or tubers that pass within a certain period of time before, between and / or after the captured image.

[0103] At 228 , a storage location for the harvested tubers may be determined based at least in part on the case stacker position data. For example, the processor 172 may determine a storage location for the harvested tubers based in part on the case stacker position data received from the positioning system 120 .

[0104] The storage location of the harvested tubers may include any suitable data indicating the physical placement of the harvested tubers. For example, a box stacker may stack the harvested tubers into boxes located within a storage facility, and the determined storage location may include a box identifier (e.g., an identification number). As another example, the storage location may include location coordinates (e.g., two-dimensional or three-dimensional coordinates) within the storage facility or a location identifier associated with a physical location within the storage facility (e.g., a storage facility quadrant or zone number).

[0105] The storage location may include data having any suitable level of granularity. For example, the storage location may include xy coordinate data indicating a ground location within a storage facility. In some examples, the storage location may also include z coordinate data indicating a height within a pile of harvested tubers.

[0106] In some embodiments, the storage location of the harvested tubers can be determined based on the packer position data and one or more movement characteristics of the conveyor. For example, the processor 172 can determine the storage location based on the packer position data received from the positioning system 120, the conveyor speed data received from the sensor 148, and the extendable conveyor segment length data and boom length data received from the sensor 121 and the positioning system 120. The processor 172 can use the conveyor speed and length data to track the movement of the harvested tubers from the grading station 22 to the packer 104. The movement tracking can be based on the real-time speed and length data received from the sensor 148 or the predetermined transportation time associated with the movement of the tubers from the grading station 22 to the packer 104. The processor 172 can also track the movement of the harvested tubers from the packer 104 to the storage location based on the packer position data received from the positioning system 120. The processor 172 may determine what average size to assign to the tubers currently being stacked by the case stacker 104 based on the moving characteristics of the conveyor. The transit time between the grading station 22 and the case stacker 104 may allow multiple groups of tubers to pass through the grading station 22, and each group of tubers to be assigned an average size before the first of those groups of tubers arrives at the case stacker 104. In this example, the transit time between the grading station 22 and the case stacker 104 may be used by the processor 172 to determine which of a series of determined average sizes to assign to the tubers currently being stacked by the case stacker 104 (e.g., an average size determined for a group that passed through the grading station 22 one transit time ago).

[0107] In some embodiments, the height (i.e., z-coordinate, elevation, or gravity orientation) position of tubers stacked by the case stacker 104 can be determined as a specified or measured offset from the z-coordinate position of the discharge end of the case stacker 104. For example, the case stacker 104 can maintain the discharge end of the boom 140 at a programmed height above the tubers or floor below (e.g., automatically using a sensor such as an ultrasonic distance sensor, or manually by an operator), wherein the height is intended to mitigate damage to the tubers from falling (e.g., 15 cm or less). In other examples, the case stacker 104 can include a sensor 122 (e.g., an ultrasonic distance sensor) for sensing a height offset between the discharge end of the case stacker 104 and the tubers or floor below. This can allow accurate determination of the height (e.g., z-coordinate position) at which the tubers are stacked within the storage facility 18.

[0108] At 232, the storage location and average size of the harvested tubers may be recorded in memory. For example, the processor 172 may record the storage location of the harvested tubers in memory 152 in association with the average size assigned to the harvested tubers (e.g., as a database entry). The stored information may be provided to one or more users of the control system 116. This may allow tubers of a desired grade (e.g., size) to be selectively located and retrieved based on the desired product (e.g., French fries or potato chips) to be made from the tubers.

[0109] Although the above description provides examples of embodiments, it should be understood that some features and / or functions of the described embodiments are susceptible to modification without departing from the spirit and principles of operation of the described embodiments. Therefore, the above description is intended to illustrate the present invention and is non-limiting, and those skilled in the art will understand that other variations and modifications may be made without departing from the scope of the present invention as defined by the appended claims. The scope of the claims should not be limited by the preferred embodiments and examples, but should be given the broadest interpretation consistent with the entire specification.

[0110] project

[0111] Item 1: A method for recording the storage location of harvested tubers stacked by a box stacker by size, the method comprising: using a camera to capture one or more images of harvested tubers moving toward the box stacker on a conveyor; segmenting, by a processor, individual tuber depictions visible in the one or more images; determining, by the processor, one or more shape characteristics of a plurality of the individual tuber depictions; identifying, by the processor, one or more unobstructed tuber depictions by determining, based on the one or more shape characteristics of the individual tuber depictions, that at least one of the individual tuber depictions is an unobstructed tuber depiction, the tuber depictions including occluded tuber depictions; measuring, by the processor, the size of tubers corresponding to at least one of the unobstructed tuber depictions; assigning, by the processor, an average size to at least the tubers corresponding to the occluded tuber depictions based on the size of the tubers corresponding to at least one of the unobstructed tuber depictions; determining, by the processor, the storage location of the harvested tubers based at least in part on the box stacker position data; and recording the storage location and average size of the harvested tubers in a memory.

[0112] Item 2: A method as in any of the preceding items, wherein the segmentation comprises deriving a binary mask for each individual tuber delineation using a masked region-based convolutional neural network (R-CNN) model.

[0113] Item 3: The method of any preceding item, wherein determining one or more shape characteristics comprises at least one of: determining color-based feature parameters and edge-based feature parameters for a binary mask.

[0114] Item 4: A method as in any preceding item, wherein identifying one or more unobstructed tuber depictions comprises using a random forest model to perform a binary classification of fully visible tuber depictions and partially visible tuber depictions.

[0115] Item 5: The method of any preceding item, wherein the size comprises at least one of volume, projected area, and one-dimensional length.

[0116] Item 6: The method of any preceding item, wherein the measuring comprises fitting an ellipse to the outer contour of at least one of the unoccluded tuber depictions and determining a major diameter and a minor diameter of the ellipse.

[0117] Item 7: The method of any preceding item, wherein determining the storage location of the harvested tubers is further performed based on one or more movement characteristics of the conveyor.

[0118] Item 8: The method of any preceding item, wherein the case palletizer position data is received from one or more sensors positioned on the case palletizer or adjacent to the storage location.

[0119] Item 9: The method of any preceding item, wherein the harvested tubers include potatoes.

[0120] Item 10: An apparatus for recording storage locations of harvested tubers by size, the apparatus comprising: a box stacker, the box stacker configured to stack harvested tubers at a storage location; a conveyor, the conveyor configured to move the harvested tubers toward the box stacker; a camera, the camera configured to capture one or more images of the harvested tubers moving on the conveyor; a positioning system, the positioning system configured to generate box stacker position data indicating the location of the box stacker; and a control system, the control system having at least one processor and memory, the at least one processor configured to collectively: receive one or more images from the camera; segment individual tuber depictions visible in the one or more images; Determine one or more shape characteristics of a plurality of individual tuber depictions; identify one or more unobstructed tuber depictions by determining that at least one of the individual tuber depictions is an unobstructed tuber depiction based on the one or more shape characteristics of the individual tuber depictions, the tuber depictions including an occluded tuber depiction; measure the size of tubers corresponding to at least one of the unobstructed tuber depictions; assign an average size to at least the tubers corresponding to the occluded tuber depiction based on the size of the tubers corresponding to at least one of the unobstructed tuber depictions; determine a storage location for harvested tubers based at least in part on case stacker position data; and record the storage location and average size of the harvested tubers in a memory.

[0121] Item 11: An apparatus as in any preceding item, wherein the segmentation comprises deriving a binary mask for each individual tuber delineation using a masked region-based convolutional neural network (R-CNN) model.

[0122] Item 12: The apparatus of any preceding item, wherein determining one or more shape characteristics comprises determining at least one of: a color-based feature parameter and an edge-based feature parameter of a binary mask.

[0123] Item 13: The apparatus of any preceding item, wherein identifying one or more unobstructed tuber depictions comprises using a random forest model to perform a binary classification of fully visible tuber depictions and partially visible tuber depictions.

[0124] Item 14: The apparatus of any preceding item, wherein size comprises at least one of volume, projected area, and one-dimensional length.

[0125] Item 15: The apparatus of any preceding item, wherein the measuring comprises fitting an ellipse to an outer contour of at least one of the unobstructed tuber depictions and determining a major diameter and a minor diameter of the ellipse.

[0126] Item 16: The apparatus of any preceding item, wherein said determining a storage location for the harvested tubers is further performed based on one or more movement characteristics of the conveyor.

[0127] Item 17: The apparatus of any preceding item, wherein the positioning system includes one or more sensors positioned on the case stack or adjacent to the storage location, the one or more sensors configured to generate on-stack position data.

[0128] Item 18: The apparatus of any preceding item, wherein the harvested tubers include potatoes.

[0129] Item 19: A non-transitory computer-readable medium comprising instructions executable by a processor, wherein the instructions, when executed, configure the processor to: receive one or more images of harvested tubers moving on a conveyor toward a box stacker from a camera; segment individual tuber depictions visible in the one or more images; determine one or more shape characteristics of a plurality of the individual tuber depictions; identify one or more unobstructed tuber depictions by determining that at least one of the individual tuber depictions is an unobstructed tuber depiction based on the one or more shape characteristics of the individual tuber depictions, the tuber depictions including occluded tuber depictions; measure the size of tubers corresponding to at least one of the unobstructed tuber depictions; assign an average size to at least the tubers corresponding to the occluded tuber depictions based on the size of the tubers corresponding to at least one of the unobstructed tuber depictions; determine a storage location of the harvested tubers based at least in part on the box stacker position data; and record the storage location and average size of the harvested tubers in a memory.

[0130] Item 20: A non-transitory computer-readable medium as in any preceding item, wherein the segmenting comprises deriving a binary mask for each individual tuber delineation using a masked region-based convolutional neural network (R-CNN) model.

[0131] Item 21: The non-transitory computer-readable medium of any preceding item, wherein determining one or more shape characteristics comprises determining at least one of: a color-based feature parameter and an edge-based feature parameter of a binary mask.

[0132] Item 22: The non-transitory computer-readable medium of any preceding item, wherein identifying one or more unobstructed tuber depictions comprises using a random forest model to perform a binary classification of fully visible tuber depictions and partially visible tuber depictions.

[0133] Item 23: The non-transitory computer readable medium of any preceding item, wherein the size comprises at least one of a volume, a projected area, and a one-dimensional length.

[0134] Item 24: The non-transitory computer readable medium of any preceding item, wherein the measuring comprises fitting an ellipse to an outer contour of at least one of the unoccluded tuber depictions and determining a major diameter and a minor diameter of the ellipse.

[0135] Item 25: The non-transitory computer readable medium of any preceding item, wherein the determining the storage location of the harvested tubers is further performed based on one or more movement characteristics of the conveyor.

[0136] Item 26: The non-transitory computer readable medium of any preceding item, wherein the case palletizer position data is received from one or more sensors positioned on the case palletizer or adjacent to the storage location.

[0137] Item 27: The non-transitory computer readable medium of any preceding item, wherein the harvested tubers include potatoes.

Claims

1. A method for recording the storage location of harvested tubers stacked by a box stacker by size, the method comprising: capturing, with a camera, one or more images of the harvested tubers moving on a conveyor toward the case stacker; segmenting, by a processor, individual tuber depictions visible in the one or more images; determining, by the processor, one or more shape characteristics depicted by a plurality of the individual tubers; identifying, by the processor, one or more unoccluded tuber depictions by determining, based on the one or more shape characteristics of the individual tuber depictions, that at least one of the individual tuber depictions is an unoccluded tuber depiction, the tuber depictions including an occluded tuber depiction; measuring, by the processor, a size of a tuber corresponding to at least one of the unoccluded tuber depictions; assigning, by the processor, an average size to tubers corresponding to at least one of the unoccluded tuber depictions based on sizes of tubers corresponding to at least one of the unoccluded tuber depictions; determining, by the processor, the storage location of the harvested tubers based at least in part on case stacker position data; as well as The storage location and the average size of the harvested tubers are recorded in a memory.

2. The method of claim 1, wherein: The segmentation involves deriving a binary mask for each individual tuber delineation using a masked region-based convolutional neural network (R-CNN) model.

3. The method of claim 2, wherein: The determining one or more shape characteristics comprises at least one of: determining color-based feature parameters and edge-based feature parameters for the binary mask.

4. The method of claim 1, wherein: The identifying one or more unobstructed tuber depictions includes performing a binary classification of fully visible tuber depictions and partially visible tuber depictions using a random forest model.

5. The method of claim 1, wherein: The size includes at least one of volume, projected area, and one-dimensional length.

6. The method of claim 1, wherein: The measuring includes fitting an ellipse to an outer contour of at least one of the unobstructed tuber depictions and determining a major diameter and a minor diameter of the ellipse.

7. The method of claim 1, wherein: The determining of the storage location of the harvested tubers is further performed based on one or more movement characteristics of the conveyor.

8. The method of claim 1, wherein: The case palletizer position data is received from one or more sensors positioned on the case palletizer or adjacent to the storage location.

9. The method of claim 1, wherein: The harvested tubers include potatoes.

10. An apparatus for recording the storage location of harvested tubers by size, the apparatus comprising: a case stacker configured to stack the harvested tubers at the storage location; a conveyor configured to move the harvested tubers toward the case stacker; a camera configured to capture one or more images of the harvested tubers moving on the conveyor; a positioning system configured to generate case palletizer position data indicative of a position of the case palletizer; as well as A control system having at least one processor and a memory, the at least one processor being configured to collectively: receiving the one or more images from the camera; segmenting individual tuber depictions visible in the one or more images; determining one or more shape characteristics depicted by a plurality of said individual tubers; identifying one or more unoccluded tuber depictions by determining that at least one of the individual tuber depictions is an unoccluded tuber depiction based on the one or more shape characteristics of the individual tuber depictions, the tuber depictions including an occluded tuber depiction; measuring a size of a tuber corresponding to at least one of the unobstructed tuber depictions; assigning an average size to tubers corresponding to at least one of the occluded tuber depictions based on the size of tubers corresponding to at least one of the unoccluded tuber depictions; determining said storage location of said harvested tubers based at least in part on said case stacker position data; as well as The storage location and the average size of the harvested tubers are recorded in the memory.

11. The device of claim 10, wherein: The segmentation involves deriving a binary mask for each individual tuber delineation using a masked region-based convolutional neural network (R-CNN) model.

12. The device of claim 11, wherein: The determining one or more shape characteristics comprises determining at least one of: a color-based feature parameter and an edge-based feature parameter of the binary mask.

13. The apparatus of claim 10, wherein: The identifying one or more unobstructed tuber depictions includes performing a binary classification of fully visible tuber depictions and partially visible tuber depictions using a random forest model.

14. The apparatus of claim 10, wherein: The size includes at least one of volume, projected area, and one-dimensional length.

15. The apparatus of claim 10, wherein: The measuring includes fitting an ellipse to an outer contour of at least one of the unobstructed tuber depictions and determining a major diameter and a minor diameter of the ellipse.

16. The apparatus of claim 10, wherein: The determining of the storage location of the harvested tubers is further performed based on one or more movement characteristics of the conveyor.

17. The apparatus of claim 10, wherein: The positioning system includes one or more sensors positioned on the case palletizer or adjacent to the storage location, the one or more sensors configured to generate the case palletizer position data.

18. The apparatus of claim 10, wherein: The harvested tubers include potatoes.

19. A non-transitory computer readable medium comprising instructions executable by a processor, wherein: The instructions, when executed, configure the processor to: receiving one or more images from a camera of harvested tubers moving on a conveyor toward a case stacker; segmenting individual tuber depictions visible in the one or more images; determining one or more shape characteristics depicted by a plurality of said individual tubers; identifying one or more unoccluded tuber depictions by determining that at least one of the individual tuber depictions is an unoccluded tuber depiction based on the one or more shape characteristics of the individual tuber depictions, the tuber depictions including an occluded tuber depiction; measuring a size of a tuber corresponding to at least one of the unobstructed tuber depictions; assigning an average size to tubers corresponding to at least one of the occluded tuber depictions based on the size of tubers corresponding to at least one of the unoccluded tuber depictions; determining a storage location for the harvested tubers based at least in part on the case stacker position data; as well as The storage location and the average size of the harvested tubers are recorded in a memory.

20. The non-transitory computer readable medium of claim 19, wherein: The segmentation involves deriving a binary mask for each individual tuber delineation using a masked region-based convolutional neural network (R-CNN) model.

21. The non-transitory computer readable medium of claim 20, wherein: The determining one or more shape characteristics comprises determining at least one of: a color-based feature parameter and an edge-based feature parameter of the binary mask.

22. The non-transitory computer readable medium of claim 19, wherein: The identifying one or more unobstructed tuber depictions includes performing a binary classification of fully visible tuber depictions and partially visible tuber depictions using a random forest model.

23. The non-transitory computer readable medium of claim 19, wherein: The size includes at least one of volume, projected area, and one-dimensional length.

24. The non-transitory computer readable medium of claim 19, wherein: The measuring includes fitting an ellipse to an outer contour of at least one of the unobstructed tuber depictions and determining a major diameter and a minor diameter of the ellipse.

25. The non-transitory computer readable medium of claim 19, wherein: The determining of the storage location of the harvested tubers is further performed based on one or more movement characteristics of the conveyor.

26. The non-transitory computer readable medium of claim 19, wherein: The instructions, when executed, further configure the processor to receive the case palletizer position data from one or more sensors positioned on the case palletizer or proximate to the storage location.

27. The non-transitory computer readable medium of claim 19, wherein: The harvested tubers include potatoes.

Images