Method for controlling the operation of a machine for harvesting root crops

Abstract

The invention relates to a method for controlling the operation of a machine (2) for harvesting root crops (4) and / or for separating root crops (4) from other harvested material comprising admixtures, in which method at least one test image (8) of at least one harvested material portion moved relative to a machine frame (12) by means of at least one conveying element (10, 10A, 10B, 10C, 10D, 10E, 10F, 10K, 10T) is recorded by at least one optical image acquisition unit (6), wherein an evaluation device generates a separating device adjusting signal on the basis of a test data set generated with the aid of the test image (8) or formed by same for adjusting at least one operating parameter of a separating device of the machine (2). The invention also relates to the aforementioned machine.

Description

[0001] Method for regulating the operation of a machine for harvesting root crops

[0002] The invention relates to a method for controlling the operation of a machine for harvesting root crops and / or separating root crops from other harvested material, as well as to the machine itself. In the method, at least one optical image acquisition unit captures at least one test image of at least one part of the harvested material being moved relative to a machine frame by means of at least one conveying element.

[0003] The test image shows crop that was previously picked up or fed into the machine for harvesting root crops. The conveying element, as part of the machine, serves to move the crop within the machine, with at least some of the crop being in direct contact with the conveying element.

[0004] Patent application US 2018 / 0047177 A1 discloses a method in which a recorded test image is used to calculate the speed of the conveying element. Based on the calculated speed, the actual speed of the conveying element is then adjusted.

[0005] A disadvantage of the known, generic methods is that, depending on harvesting conditions, significant damage to the root crops or a large amount of impurities may occur among the root crops unloaded by the machine. Therefore, US 2018 / 0047177 A1 additionally proposes generally modifying the harvesting rate or one or more machine configurations based on a server-based evaluation of three-dimensional data of the harvested crop acquired by the machine's sensors.

[0006] The object of the present invention is to provide a method in which the preservation of root crops is improved, whereby as large a proportion as possible of impurities is separated from the root crops.

[0007] According to the invention, the problem is solved by a generic method in which the evaluation device generates a separation device setting signal for setting at least one operating parameter of a separation device of the machine on the basis of a test data set generated from or formed by the test image.

[0008] The machine is a self-propelled or towed vehicle for harvesting root crops, in particular potatoes, beets, carrots, or chicory. During operation, the machine moves, in particular, along rows, especially planting ridges, of the root crops to be harvested, and picks them up from the ground as part of the harvested material in a continuous process. After picking up the harvested material, at least a portion of it, in particular root crops and / or admixtures, is at least partially conveyed by at least one conveying element relative to the machine frame. In particular, the conveying element is circumferential and designed as a conveyor belt, preferably as a sieve belt, or as a rotating sieve star.

[0009] Alternatively, the machine could also be a machine for separating root crops from impurities in the harvested crop, e.g. clods, stones or soil.

[0010] The separating device, with optionally individually adjustable separating elements, is part of the machine and preferably works in conjunction with one or more conveying elements. Alternatively, the separating device is part of the conveying element, is at least partially formed by it (e.g., in sieve belts equipped with vibrating devices), or forms one or more of the conveying elements (e.g., in roller threshers). During operation, a movement of the harvested crop relative to the separating device exerts a momentum on at least one component of the harvested crop, in particular on the root crops or impurities. The separating device is, for example, in the form of a roller thresher, especially equipped with rotating separating elements in the form of rollers, whereby different components of the harvested crop are at least not moved in the same direction by the separating device.

[0011] The optical image acquisition unit is arranged in a fixed position on the machine, particularly above the conveying element, and is directed towards the conveying element and thus, during operation, towards a flow of harvested material or a component thereof, particularly root vegetables or impurities, located between the image acquisition unit and the conveying element. The method according to the invention is carried out particularly during harvesting or separation with the machine and is preferably repeated.

[0012] The test image is, in particular, a multidimensional, preferably two-dimensional, image depicting at least a portion of the harvested crop, including root vegetables, admixtures, and / or the conveying element. Based on the test image captured by the image acquisition unit, the test data set is generated either by the image acquisition unit itself or by the evaluation device. Alternatively, the test data set can be formed by the test image itself. This applies particularly to image acquisition units whose test images already possess a format suitable for subsequent analysis in the evaluation device. The test data set is, in particular, a data set created by processing, for example, filtering and / or other rendering, and which is at least temporarily present in the system. Its information, e.g., color information in the form of one or more color values, is evaluated in the evaluation device. It can, for example,The test image, or the test data set already created in the image acquisition unit, is transferred from the image acquisition unit to the evaluation device. The optical image acquisition unit is designed, in particular, as a digital photo or video camera for two-dimensional recording of the test image or as a line scan camera. Where the test image is subsequently referred to in connection with the processing of the image information in the evaluation device, this may refer to the test data set. The evaluation device serves to evaluate the test data set. The evaluation device comprises at least one processor and is designed either as a central processing unit or as a decentralized system comprising at least one processor and at least one memory with different positions on machine components.It is therefore a local system to carry out any evaluations directly on site and to make the results available immediately.

[0013] The operating parameter is a variable that relates to the geometry of the separating device or one of its separating elements, its position or orientation relative to the machine frame or the conveying element, the speed of the separating device, and / or the engine power. The operating parameter determines how and to what extent the separating device interacts with the harvested crop or at least a component thereof. In particular, varying the operating parameter allows for adjusting the amount of impurities that remain with the root crops downstream of the separating device, relative to the conveying distance over which the root crops are transported within the machine. The operating parameter is specifically independent of the conveying speed of the conveying element, which transports at least the root crops while they rest on the conveying element and move in the same direction as the conveying element.The operating parameter defines how aggressively the separator operates when separating root crops and impurities. If the aggressiveness is too low, too much of the impurity will not be separated from the root crops. If the aggressiveness is too high, not only impurities but also root crops will be damaged or separated, thus reducing the yield. By generating the separator setting signal and, in particular, by transmitting it to a separator control unit, the operating parameter is preferably set according to the crop or part of the crop depicted in the test image. Based on the separator setting signal, the separator control unit increases or decreases the operating parameter. For this purpose, the separator control unit sends an electrical signal, preferably to an actuator.The control unit for the separating device, in particular through its electrical signal, adjusts hydraulic pressure, pneumatic pressure, current, voltage, force and / or torque to drive or adjust the separating device or its operating parameters. The control unit is, in particular, part of the same processing unit as the evaluation device.

[0014] This process allows for continuous optimization of the separator's operation. In particular, the aggressiveness of the separating element or separator can be continuously optimized, thus ensuring both gentle handling of the root crops and effective separation of impurities from the harvested material. Specifically, the evaluation unit for providing the separator's setting signal analyzes the test data sets locally on the machine or on a tractor directly connected to it. This enables near-instantaneous adjustment upon detection of an undesired condition on the separator, thereby preventing blockages or damage.

[0015] In an advantageous embodiment of the method according to the invention, the evaluation device calculates at least a first portion of the test image, formed by at least one image area. This at least one image area at least partially depicts a defined component of the harvested crop or the machine. Based on this first portion, a cleaning characteristic value is calculated, in particular.

[0016] Before calculating the first component, the component statistically represented by it is predefined. The test image and / or the test data set is divided into a plurality of preferably equally sized image areas. The image areas that at least partially show the component together constitute the first component. The component is specifically a proportion of these image areas, which at least partially show the component, to the total image areas, whereby the first component is determined based on a ratio of the number of image areas or on their common areas.

[0017] The first component is a measure of the extent of image areas that depict the component and thus a measure of the component's density in the field of view of the image acquisition unit or of the portion of the test image being considered. The component is, in particular, at least partially a root crop component, meaning that the first component indicates, at least approximately, the concentration of root crops. An image area is considered to depict the component and is assigned to the first component if at least 50% to 100% of its area shows the component. Alternatively, the component can be calculated by summing the individual pixels depicting the component. In particular, the at least one image area can also be assigned to the first component only partially or, preferably, partially assigned in different proportions.This is particularly advantageous when, within the framework of the preferably model-based classification method, a clear assignment of the image area to a corresponding component is not possible. In this case, probabilities for the assignment to different proportions are preferably determined. The image areas are particularly preferably assigned proportionally or partially to different proportions according to these probabilities. This results in an even more precise representation of the relationships between the components.

[0018] By calculating the first fraction, a characteristic value is determined that specifically characterizes the composition of the harvested material. Based on this, the operating parameter can be controlled particularly advantageously, since the cleaning performance of the conveying element or the separating device comprising the conveying element is highly dependent on the composition of the harvested material. If the first fraction indicates a concentration of impurities, the operating parameter can preferably be varied with increasing first fraction to generate a greater separation effect or separation performance. The cleaning characteristic value is preferably calculated at least based on the first fraction or is equal to the first fraction.

[0019] Preferably, the at least one image area forming the first part is identified, in particular based on a test subset generated from the image area, as showing the defined component. Specifically, the image area is identified based on a test value contained in the test image and / or the test subset, preferably color information. The color information includes, in particular, black and white, gray, and / or color channel values ​​of a color space.

[0020] Preferably, the test subset, the test value, or the color information is classified using a statistical classification method, particularly a model-based one. Accordingly, an image area is assigned to the first component if the result of the classification method is associated with the defined component of the harvested crop or the machine. The classification method utilizes, in particular, a neural network, a random forest, a Bayesian classifier, a support vector machine, and / or a decision tree. By applying the classification method, the result of the calculation of the first component, especially different components, is particularly robust and informative regarding the composition of the harvested crop.

[0021] Preferably, the test value or color information is compared with one or more reference values ​​or reference ranges, and based on this, an image area is either assigned to the first component or not. The reference image is preferably captured by the optical image acquisition unit in the same way as the test image, whereby a user must, in particular, mark different parts of the reference image as indicating different components. This form of differentiation enables a particularly reliable identification of a relevant component on the test image. Preferably, at least one of the test values ​​of the test sub-dataset, which in particular includes the color information, is compared with at least one reference value, and an image area is assigned to the first component, in particular if at least one test value of the test sub-dataset lies within an assigned reference value range.This reference value range is limited in particular by a maximum value and a minimum value, whereby, preferably, different test values ​​must lie in their respective assigned reference value ranges for the allocation of the image area to the first part.

[0022] In an advantageous embodiment of the invention, the evaluation device automatically develops or further develops a model underlying the classification method upon input of exemplary image areas of the reference image attributable to the first component. Alternatively or additionally, the evaluation device automatically calculates or modifies the at least one reference value range upon input of exemplary image areas of a reference image attributable to the first component. In particular, the reference values, reference value ranges, or the model or model parameters thereof do not need to be completely predefined manually by the user. Instead, it suffices to input at least one exemplary image area showing the component to start up the evaluation device. Based on the image area, the evaluation device determines the at least one reference value, the at least one reference value range, or the model parameters thereof.The model or model parameters are automatically determined. Thus, the evaluation device adapts itself largely independently to different applications. The greater the number of image areas entered, the more accurately the reference values, reference value ranges, or the model or model parameters can be determined.

[0023] The method is particularly robust when the input image areas show the component under different brightness and / or ground conditions. This ensures reliable application even under varying conditions. Preferably, the evaluation device adjusts at least one reference value or reference value ranges during repeated execution of the method, optionally with the operator identifying relevant components as examples, from which training data for the algorithm can be derived.

[0024] In particular, by using additional sensors such as ambient light sensors to measure brightness, which the evaluation device assigns to essentially simultaneously recorded test data sets, the evaluation device automatically expands the scope of the reference data. Alternatively or additionally, the user of the method, i.e., in particular the driver or operator of the machine or a machine coupled to it, has the option of manually marking at least one component on visualized test images in order to expand the scope of the evaluation device's reference data. Thus, based on information entered by the user or on data stored in the evaluation device, it can differentiate between, for example, potatoes, haulm, stones, loose soil, and clods and calculate their respective proportions.

[0025] Preferably, the method according to the invention is executed automatically after its start, with the exception of the input of any existing training data in the form of component marking. This makes it easier for the driver or operator to control the machine.

[0026] Preferably, the image areas forming the first part are additionally identified based on image sub-datasets generated by or formed from adjacent image areas. In particular, color information contained in the test sub-datasets, especially comprehensive black-and-white and / or grayscale values, is used for this purpose. The evaluation of the image areas is therefore not based solely on the data assigned to them, but also on additional data assigned to the surrounding image areas. This allows brightness and / or color gradients to be determined, thus enabling identification based on a broader data foundation.

[0027] The different image areas are preferably weighted differently when calculating the first component. The contribution of the image areas forming the first component is therefore varied. This makes it possible to calculate the first component not solely based on the perspective representation of the test image, but rather to weight image areas that show a component of the harvested crop located further away from the image capture unit more heavily than image areas that show a component closer to the image capture unit. This allows for the creation of a perspective-corrected first component and thus achieves a particularly realistic image of the harvested crop composition on the conveying element.

[0028] Preferably, the entire test image or a contiguous part of the test image is divided into sub-image areas. These sub-image areas each comprise, in particular, the same number of pixels of the test image, preferably exactly one pixel. The test image part is a portion or section of the test image that comprises a plurality of sub-image areas. To calculate the first portion, only the image areas that represent the portion and belong to the test image part are considered. For this purpose, the test image part is defined, in particular, such that it depicts sensitive zones within the machine that need to be monitored. The image area forming the first portion thus comprises, in particular, several sub-image areas of a test image part.

[0029] The test image or test image part is rasterized into a plurality of sub-image areas, each preferably rectangular. By defining the sub-image areas as exactly one pixel, a particularly large database is created for evaluating the condition of the harvested crop with regard to its individual components, thus enabling particularly sensitive control of the operating parameter. At the same time, the data volumes supplied by conventional 2D digital cameras, which typically have a maximum of a few million pixels, can be processed quickly and easily by an evaluation device equipped with one or more modern processors.

[0030] Preferably, the test image comprises several test image sections, for which the evaluation device calculates a first component, in particular several components, of image areas, wherein the test image sections preferably depict harvested material from different conveying elements transported by a separating device. The test image sections thus show, in particular, different sections of the same conveying element, one of which is arranged in the conveying direction upstream of a separating device or a separating element thereof, and another downstream of the separating device or a separating element thereof. Alternatively, the test image sections show different conveying elements, which represent alternative conveying paths for different components of the harvested material (for example, a conveying element for cleaned root crops, a conveying element for sorted impurities). By calculating the first component for these different test image sections, the cleaning or...The separation performance of the associated separating device can be evaluated particularly comprehensively. Specifically, the initial portion of a crop flow into the separating device can be compared with the initial portion of a mixed root crop flow out of the separating device, thus determining the separating device's effectiveness. Alternatively, the composition of the crop in the conveying elements connected to and thus conveyed by a separating device is preferably determined, once for separated crop and once for crop to be conveyed further. Depending on the separating device's effectiveness, the operating parameter is adjusted. Furthermore, the test image sections depicted or available in the respective test data sets can show a portion of a conveying element upstream of a separating or deflecting element of the separating device and a portion of the conveying element downstream of the separating or deflecting element.If the image analysis reveals that excessively large proportions of, for example, root crops appear behind a deflection element in an undesired area, this deflection element can be positioned differently, e.g., lower above the conveying element, which increases the separation efficiency.

[0031] In an advantageous embodiment of the invention, the test image parts preferably show different conveying elements downstream of a separating device, in particular a conveying element for discharging a root crop mixture and a conveying element for discharging impurities downstream of the same separating device. For both test image parts, a first proportion of a component, such as root crops, is preferably determined. Alternatively, different proportions are calculated for the different test image parts. This allows, for example, a comparison of the proportion of impurities in the root crop mixture discharge stream with a comparison of the proportion of root crops in a stream of sorted impurities, and based on this, adjustment of a separating element encompassed by the separating device with respect to its position relative to the conveying element or with respect to its speed.

[0032] Preferably, the image areas forming the first component show root crops or parts thereof, and image areas forming a second component show admixtures or parts thereof. Thus, the evaluation device calculates at least two different components. Particularly preferably, the evaluation device calculates at least four components: a first component for root crops, a second component for herbaceous components, a third component for soil, and a fourth component for damaged areas. The sum of the components is particularly less than 1. Alternatively, the first component can also be an admixture component, the second component a root crop component, etc. Additionally or alternatively, components for clods and / or stones are preferably calculated.

[0033] Alternatively or additionally, in a further embodiment of the invention, at least two image acquisition units and at least two conveying elements are provided, wherein the first image acquisition unit captures a first test image of a crop portion conveyed by a separating device via the first conveying element, the second image acquisition unit captures a further test image of a crop portion conveyed by the separating device via the second conveying element, and the separating device setting signal is generated based on at least one of the test data sets formed by, preferably both of, the two test images or generated on the basis of these. The test data sets are evaluated as described above or below, in particular with regard to the respective proportions.

[0034] By using multiple components in the evaluation device's calculations, a more accurate picture of the crop composition or the occupancy of the conveying element can be obtained. As an alternative to identifying image areas based on limit values, all image areas of the test image or a test image part are necessarily assigned to a component. Preferably, the degree of agreement between test sub-datasets calculated based on the image areas and reference sub-datasets is evaluated, and each image area is assigned to the component for which the agreement is greatest.

[0035] In an advantageous embodiment of the invention, the cleaning characteristic value is determined by means of a deviation of the first component from a threshold value, calculated by the evaluation device. In particular, the threshold value characterizes an optimal utilization of the conveying element, whereby a deviation of a defined amount from this value triggers a change in the operating parameter. The cleaning characteristic value is based in particular on a plurality of components and preferably on further data, especially sensor data.

[0036] In an advantageous embodiment of the invention, the separation device setting signal is calculated based on a plurality of cleaning parameters, particularly those calculated sequentially over time, or at least one previously calculated cleaning parameter is incorporated into the calculation of the cleaning parameter. In particular, a moving average of the cleaning parameter is calculated and forms the basis for the separation device setting signal, or the cleaning parameter curve is smoothed, particularly using a low-pass filter. These measures make the method according to the invention particularly resistant to interference and thus especially robust in its application.

[0037] In an advantageous embodiment of the invention, at least one further sensor transmits sensor data to the evaluation device, which is then used to calculate the separator setting signal. The sensor is, in particular, a sensor, preferably a touch sensor or an ultrasonic sensor, for measuring the layer thickness of the harvested crop on the conveying element; a sensor for measuring drive power, for example, in the form of a pressure sensor for measuring hydraulic oil pressure; and / or a speed sensor, in particular for measuring the rotational speed of a conveying element drive. Specifically, the speed sensor is used to determine slippage of the conveying element, which is transmitted to the evaluation device in the form of sensor data. A humidity sensor can be used to incorporate further information into the calculation of the separator setting signal.

[0038] Based on this additional information contained in the sensor data, which goes beyond that provided on the basis of the test image, the evaluation device has a significantly more accurate picture of the cleaning situation in the area of ​​the conveying element, which in turn allows the operating parameters to be better adjusted accordingly.

[0039] Preferably, the evaluation device triggers either an increase or a decrease in the operating parameter by means of different separation device setting signals. In particular, the evaluation device or the separation device control unit comprises a fuzzy controller, a PID controller, or a three-point controller, which alternatively triggers either an increase, a decrease, or a maintenance of the current operating parameter. An increase is triggered, in particular, when the cleaning characteristic exceeds a predefined first threshold value; a decrease is triggered accordingly when the cleaning characteristic falls below a predefined second threshold value.

[0040] Preferably, the operating parameter is a distance between two conveying elements, whereby the conveying elements can jointly function as a separating device, or a distance between a separating element of the separating device or the separating unit and a conveying element. In particular, the operating parameter is a distance between two conveying rollers of a roller table rotating during operation, between which soil is sieved. Alternatively, the operating parameter is a distance between a conveying element designed as a sieve belt and a separating element designed as a deflecting roller, wherein the separating element extends transversely across the conveying element and causes the root crops to be deflected laterally from the conveying element. The deflecting roller rotates during operation about an axis of rotation which, in a top view of the conveying element, is angled at less than 90° to the conveying direction of the conveying element.Alternatively, the separating element is designed as a finger belt that circulates during operation, located above the conveying element, and whose outwardly projecting fingers comb through the crop arranged on the conveying element. Alternatively, the separating element is designed as a non-circulating stripping device that is positioned above a coarse weed belt interacting with a sieve belt, and strips root crops from the weeds deposited on the coarse weed belt. The distance between the two devices is adjustable, particularly by a hydraulic or mechanical adjusting device, which allows for easy modification of the separating element's aggressiveness in conjunction with the conveying element, or the separation efficiency of the conveying elements.

[0041] Alternatively, the operating parameter is the penetration depth of at least one of the machine's digging shares into the soil. This allows for a simple way to influence the amount of impurities in the harvested crop.

[0042] Alternatively or additionally to the above, the operating parameter is a separation speed, in particular a circulation or rotation speed, of the separation device or a separation element of the separation device. Specifically, the separation speed is a circulation speed of the aforementioned finger belt or a rotation speed of the aforementioned deflection roller. Alternatively, the separation speed is a circulation speed of an angled separation device that conveys impurities upwards during operation, e.g., in the form of a fine weed belt, which is operated in such a way that impurities are conveyed upwards as much as possible and scum rolls downwards against the direction of movement of the section of the separation device facing them.

[0043] Preferably, the operating parameter is alternatively defined as the angle of attack of the conveying element of the separating device, i.e., at least one separating element of the separating device. In particular, the operating parameter is the angle of attack of the separating device, also referred to as a fine weed elevator. The angle of attack changes the inclination of the conveying plane of a fine weed belt of the separating device relative to a horizontal plane, thus adjusting the aggressiveness of the separating device. In an alternative advantageous embodiment of the invention, the operating parameter causes a change in airflow velocity or air mass flow rate per unit time, with the corresponding separating device performing the separation based on airflow. Here, motor power, e.g., represented by a motor speed, can be the corresponding operating parameter.The air is used to separate root crops and impurities, in particular blowing haulm out of the harvested crop stream and thus removing it. The operating parameter for such an air separation device, especially one suitable for stationary use, is preferably the rotational speed of an associated blower or the angle of attack of an associated unit in the form of an air deflector, which, for example, divides the airflow into a main airflow and a cross-flow.

[0044] In an advantageous embodiment of the method according to the invention, several of the aforementioned operating parameters are set by the same or different isolation device setting signals. A control set can be stored in the evaluation device for this purpose, which generates corresponding signals for the respective adjustable parameters for the desired increase or decrease in isolation performance of the respective isolation device.

[0045] Preferably, after triggering an operating parameter change for a defined period or conveying distance of the conveying element, no further operating parameter changes are triggered. This applies in particular only to the same operating parameter and / or at least one operating parameter of at least one separation device located downstream in the system. This ensures that the separation element is not over-regulated and that every operating parameter change is based on sound data that already takes into account any previous operating parameter changes.

[0046] Preferably, the isolator setting signal is transmitted to the isolator control unit via wired connection, in particular via CAN bus or Ethernet, or wirelessly, with the isolator setting preferably being authorized in advance by an operator via an interface. This allows existing or at least established communication systems for setting the isolator element to be used, and the reliability of the method is increased, in particular because an operator is shown the resulting or to-be-made setting of the isolator, instead of its automatic setting, especially in the driver's cab, and must authorize it via a corresponding input at an interface.

[0047] The object of the invention is further achieved by a machine for harvesting root crops and / or separating root crops from other crop components. The machine comprises a machine frame, a conveying element, an image acquisition unit, a separating device, and an evaluation device. The machine is designed to carry out the method described above or below. Preferably, the evaluation device includes a graphics processing unit, in particular a GPU (Graphical Processing Unit) or GPGPU (General Purpose Graphical Processing Unit) and / or an FPGA (Field Programmable Gate Array)-based processor unit. This configuration of the evaluation device allows the test data set to be evaluated in a particularly resource-efficient manner and, in particular, locally. It is understood that the evaluation device, designed as a computer device, includes other conventional means, e.g., for power supply, interfaces, and working memory.

[0048] In an advantageous embodiment of the invention, the machine has at least one sensor coupled to the evaluation device, in particular a touch or ultrasonic sensor for measuring the layer thickness of the harvested crop on the conveying element, a sensor for measuring drive power, for example a pressure sensor for measuring hydraulic oil pressure, and / or a speed sensor arranged on a conveying element. This sensor allows the conveying speed signal to be calculated based on measured physical quantities, thereby significantly increasing the reliability of the quantities calculated by the evaluation device and reducing their susceptibility to errors. A humidity sensor can also provide additional information that contributes to the adjustment of one or more of the separation devices within the framework of the evaluation device's analysis.

[0049] Preferably, the machine has several image acquisition units, each capturing at least one image of the same conveying element during operation. Alternatively, the machine preferably has several image acquisition units, each capturing at least one image of different conveying elements, particularly those conveying from the same separating device. Alternatively, one of the two image acquisition units can be directed at a discharge area, for example, of a separating device that uses airflow. The multiple image acquisition units allow the composition of the harvested material, in particular the path of the first fraction along a conveying path and especially on different sections of the conveying path after a separating device of the machine, to be tracked. In particular, the conveying speeds of different conveying elements can thus be adjusted based on different first fractions.Accordingly, an analysis of the conveyor section areas captured by the respective test images is performed in at least one evaluation device, using the procedure described above or below. While preferably only one central evaluation device is provided for evaluating the data from the image acquisition units, the respective image acquisition units can also be assigned their own evaluation devices. These can then control the respective assigned separating devices, particularly in coordination with other evaluation devices. Alternatively or additionally, a central evaluation device is responsible for generating the separating device setting signals and forwarding them to a machine control system.

[0050] Preferably, the image acquisition unit is arranged such that the test image shows at least two alternative conveying paths of the entire conveying line for different components of the harvested crop, in particular for corn cobs and impurities. This allows two conveying elements to be monitored using one image acquisition unit, with each test image portion depicting a section of the different conveying elements or of the harvested crop. In particular, one of the conveying elements is designed for conveying sorted impurities and another for conveying cleaned corn cobs.

[0051] This allows for a particularly comprehensive picture of the cleaning performance and thus of the utilization of the conveying element and / or the separating device encompassing the conveying element.

[0052] Preferably, the image acquisition unit is arranged such that, during operation, the test image at least partially depicts two conveyor element sections separated by a separating device, in particular a separating element of a separating device. The conveyor element sections are only separated by the separating element of the separating device in the representation of the test image and are each encompassed by the conveyor element. The separating element is closer to the image acquisition unit than the conveyor element, and the latter is therefore covered by the separating element in the test image. This positioning of the image acquisition unit makes it possible to calculate at least a first component for each of the two individual test image parts and thus directly evaluate the effectiveness of the separating element and the associated separating device.In particular, the composition of a harvested crop before reaching the separating element is compared with the composition of at least a portion of the harvested crop after passing the separating element.

[0053] Preferably, the conveying element is designed as a sieve belt or hedgehog belt, which, in operation, runs in particular under at least one deflecting roller extending transversely across the conveying element and diverting the harvested material laterally. Alternatively, the conveying element is designed as a sieve star or conveying roller, wherein the conveying roller is in particular enclosed by a roller table.

[0054] The machine is designed as an alternative or supplement to the above-mentioned design for separating root crops from other harvested material. According to an advantageous further development, the machine is operated in a stationary manner, i.e., without continuous, local forward movement of the machine during operation.

[0055] According to another advantageous further development, the machine is a potato harvester or a beet harvester.

[0056] Further details and advantages of the invention can be found in the schematically illustrated embodiments described below;

[0057] It shows:

[0058] Fig. 1 shows a flowchart of a method according to the invention,

[0059] Fig. 2 shows an associated program flowchart for determining the separator setting signal,

[0060] Fig. 3 shows an associated program flowchart for the processing of the cleaning parameters, Fig. 4 shows an associated program flowchart for the separation device setting,

[0061] Fig. 5 shows a view of a test image and its partial evaluation.

[0062] Fig. 6 shows an object according to the invention,

[0063] Figs. 7 and 8 show the object according to Fig. 6 in different side views,

[0064] Fig. 9 shows a partial view of the object similar to Fig. 6 with a conveyor element.

[0065] Fig. 10 shows a detailed view of an area of ​​the device according to Fig. 6, which is shown section by section in Fig. 9.

[0066] Fig. 1 1 shows the object according to Fig. 10 from a different perspective,

[0067] Fig. 12 shows a representation of the test image of the image acquisition unit according to Fig. 10,

[0068] Fig. 13 shows a separating device of the machine similar to Fig. 6 with an image capture unit.

[0069] Fig. 14 shows a schematic test image taken from the perspective of the image acquisition unit shown in Fig. 13; Fig. 15 shows another separating device of the machine similar to Fig. 6 with an image acquisition unit.

[0070] Fig. 16 shows a schematically represented test image taken from the perspective of the image acquisition unit shown in Fig. 15.

[0071] Fig. 17 shows another detailed view of a machine similar to Fig. 6, with an additional image acquisition unit.

[0072] Fig. 18 shows a schematic representation of a test image viewed from the perspective of the image acquisition unit according to Fig. 17.

[0073] Fig. 19 shows a detailed view of another device according to the invention,

[0074] Fig. 20 shows a further detailed view of another device according to the invention.

[0075] Parts with identical or similar effects are provided with identical reference numerals, where applicable. Individual technical features of the embodiments described below can also be combined with the features of the embodiments described above to lead to further developments according to the invention, but always at least in combination with the features of one of the independent claims. The items listed in the list of figures are sometimes only partially shown in individual figures.

[0076] The method according to the invention serves to control the operation of a machine 2 for harvesting root crops 4 (see Figs. 6 to 8). In the method, at least one optical image acquisition unit 6 captures at least one test image 8, which shows harvested crops comprising root crops 4, moved relative to a machine frame 12 of the machine 2 by means of at least one conveying element, initially generally numbered 10.

[0077] The test image 8 is transmitted to an evaluation device, which, based on a test data set generated from or created by the test image, generates a separator setting signal for setting at least one operating parameter of a separator of the machine (2). The illustrations shown as test images only schematically depict the parts relevant to the invention without any borders or boundaries. Images taken by a camera, especially digital images, may contain further information not shown in the illustrations. This information can, for example, be masked or filtered by the camera itself or during the creation or processing of a test data set.

[0078] In an embodiment according to the invention, starting from a material flow 1.1, the composition of the harvested material is evaluated using the method described above before a first separating element (block 1.2) (Fig. 1). Furthermore, the composition of the harvested material flow is also calculated before the inlet and after the outlet of a further separating device (blocks 1.3 and 1.4). Finally, the composition of the harvested material flow is again determined at the inlet of a third separating device. This yields the respective proportions A1 and A2 of root crops and impurities (blocks 1.6, 1.7, 1.8 and 1.9). Depending on the desired separation efficiency at the individual separating devices, the individual proportions A1 and A2 of root crops or impurities for the respective separating devices are calculated together in the evaluation device (block 1.10).The operating parameters of the respective separation devices are then set (Block 1 .1 1) in order to optimize the performance of the respective separation device.

[0079] The determination of the separator setting signal is shown in greater detail in Fig. 2. Starting with a test image 8, the relevant parts of the test image are first extracted (Block 2.1) to position the test data set. For this purpose, a mask or region of interest (ROI) can be predefined based on the position of the image acquisition unit (Block 2.2), which distinguishes between distances of the test image that are to be considered and those that are not. Based on the relevant image section of the test image and the test data set now available for processing, the proportions of the image areas showing the individual crop components are calculated (Block 2.3). For this purpose, the color information can be evaluated in particular. These values ​​can be taken from a reference table or specified by an operator (Block 2.4). Based on a threshold value definition (Block 2.5) The deviations of the calculated proportions from the threshold value are calculated (Block 2.6). The threshold value is, for example, an ideal value for the respective proportion under consideration (e.g., root crop, component 1, component 2). A low-pass filter is then applied to smooth the determined deviations (Block 2.7). A filter time constant defined in Block 2.8 is used for this purpose. Subsequently, a cleaning characteristic RS is calculated based on the smoothed values ​​of the deviations for the individual positions along the conveying path and the respective proportions (Block 2.9). Controller parameters can be used for simplification here, according to which, for example, a cleaning characteristic RS_1 (A1) on a first monitored conveying path for a proportion A1 in the form of root crops is set to 1 if the deviation is too large and therefore too many components are present in the considered test image (section).The cleaning characteristic RS for a component A1 can be set to zero if the deviations from an ideal value are sufficiently small. Based on the cleaning characteristic RS_1 (A1) of block 2.9, the separation device setting signal can be generated, for example, using a three-point controller (block 2.10).

[0080] Using the method according to the invention, the ratios between the products present in a material flow upstream of a separating device and then downstream of a separating device, for example on a product discharge conveyor element (4) and a product discharge conveyor element (5), are evaluated separately. If a single camera monitors both conveyor elements, separate regions of interest are defined for the product discharge element and for the product discharge conveyor element. Depending on whether there are too many potatoes (root crops 4) on the product discharge conveyor element or too many product discharge conveyor elements, parameters of the separating device that influence the separation threshold are adjusted accordingly. In particular, the fingers of a finger belt or...Brushes of a brush belt can be raised or lowered and / or the finger or brush belt can run slower or faster.

[0081] In an exemplary implementation of the three-point control system according to Fig. 3, the system first checks the cleaning parameters RS_1 (A2) - RS_n(A2), i.e., the cleaning parameters for the components A2 describing the impurities (Block 3.1). This example involves considering the proportions of impurities in the harvested crop in the form of stones before and after a separator, or before and after a conveying section comprising several separators. The cleaning parameters are, according to the above definition, "zero" or "one". Subsequently, Block 3.2 checks whether the sum of the cleaning parameters RSJ (A2) (with i=1 ...m) is equal to zero. If this is not the case, Block 3.3 first queries memory 3.4 for the last cleaning parameters and, if applicable, the associated crop compositions at the respective positions. If a check in block 3.5 shows that a change to the isolation device settings has been sufficiently long ago, the following will be done in block 3.6. A degressive operating parameter signal is generated for the respective separator. The queried cleaning parameters from memory can also be taken into account. The setting of the separator(s) (Block 3.7) is less aggressive, so that fewer root crop components remain in the admixture or harvested crop flow on the conveying line. The last cleaning parameter(s) RSJ (A2) are entered in memory 3.4, possibly with the corresponding A2 values ​​(Block 3.15).

[0082] If the sum of the cleaning parameters RSJ(A2) (with i=1 ...m) is zero, block 3.8 then checks for cleaning parameters RSJ(A1) relating to the components A1, e.g., root crops. It is then checked whether the sum of these cleaning parameters is again zero (block 3.9). If so, the last cleaning parameters RSJ(A1) are also written back to memory 3.4 (block 3.10). No change to the operating parameters is necessary, and a neutral operating parameter signal or no operating parameter signal is output (block 3.11). If the sum of the cleaning parameters for the components A1 is not zero, the old values ​​are retrieved from memory in block 3.12, and block 3.13 checks whether sufficient time has elapsed since the last change to the operating parameters. If so, block 3.12...14. An operating parameter for a more aggressive setting of the separation devices was issued.

[0083] To implement the specifications according to Fig. 3 in block 3.7, separator actuators must be controlled. For this purpose, separator-specific changes to the operating parameters are determined in control 4.1 according to Fig. 4, based on the specifications of blocks 3.6, 3.11, and 3.14, taking into account the current settings (4.2) of the respective separator. For this purpose, for example, speeds of the separator elements, rotational speeds, distances, or inclinations are defined. From this, control variables for the separator actuators of the respective separators are defined in block 4.3, via which the separators are adjusted (block 4.4).

[0084] Fig. 5 shows, as an example, a test image 8 in the upper part of the figure, which depicts the transition from a conveying element 10a to a conveying element 10b. Root crops 4 and impurities 5, which may include stones and foliage, are located on this conveying section. According to the classifiers defined in the algorithm's training or specified via a database, for example, a table with color information (e.g., in HSV format), individual sub-image areas 16 are checked for the presence of identical components. Thus, based on the assignment of the respective image areas to the individual components, shown as an example in the lower left of Fig. 5, a distribution of the proportions of root crops 4 and impurities 5 in the test image 8 is obtained. A1 therefore shows the proportion of root crops 4 in the test image or the corresponding test data set, A2 the proportion of foliage, and A3 the proportion of stones.Preferably, this assignment is based on the color information of the individual pixels, i.e., an image area 19 that is assigned to a proportion corresponds in particular to an area of ​​a pixel.

[0085] The cleaning characteristic value, generally designated RS, is based, by way of example, but preferably, on a deviation of the first fraction A1 from a threshold value R, which indicates an optimal distribution of root crops 4 at the observed point of the conveying path. For example, the cleaning characteristic value RS is set to 1 if the deviation is > 50% from the cleaning value and to 0 if the deviation is < 50%. These values ​​are then stored accordingly and processed in the further program sequence as shown in Figures 1 to 4.

[0086] The machine 2 according to the invention is designed as a trailed potato harvester, as shown in Fig. 6, wherein a plurality of conveying elements 10 and their associated separating devices are held by a machine frame 12, which is only partially numbered. Along the conveying path, a plurality of image capture units 6 are provided, which capture the harvested crop, comprising root crops 4, transported on the conveying elements 10. The positions for image capture units 6 indicated in Fig. 6 are a transition from a first conveying element 10A in the form of a sieve belt to a second conveying element 10B in the form of a sieve belt, which is additionally enclosed by a coarse haulm belt, and the transition from this second sieve belt 10B to a further conveying element 10C comprising a further separating device.Furthermore, at the output side of this separation device, a conveyor element 10E leading to the sorting table is monitored by a further image acquisition unit 6, while at the same time a further conveyor element 10F intended for residues of impurities 5, in particular stones, is detected.

[0087] An evaluation device can be positioned at any centrally accessible location, preferably near the selection table.

[0088] The evaluation device can, for example, transmit a vehicle speed signal or information regarding the setting of the separation devices to a towing vehicle via a cable 12.1 visible in Fig. 6. The machine shown in side view in Figs. 7 and 8 can be equipped with optical image acquisition units 6 at further positions. Additional image acquisition units can thus be arranged directly in the area of ​​a harvesting device 29 or a drop step leading to a bunker 33.

[0089] Figures 9 and 10 show the arrangement of an optical image acquisition unit 6, positioned on the frame side above a first drop step between a conveying element 10A and a conveying element 10B. The unit's field of view is directed downwards. A light source 7 illuminates the field of view to capture a sufficiently illuminated test image 8. The conveying element 10A is a sieve belt which, coming from a harvesting device 29, already sieves out a portion of impurities 5, particularly soil and / or clods, and transfers them via a drop step to another conveying element 10B, also designed as a sieve belt. This conveying element 10B additionally features a coarse haulm belt, which is designed to separate the haulm present on the potatoes or in the harvested crop. Accordingly, stripping devices 32 are arranged across the width of the conveying element 10B.

[0090] The height H of the scraper 32 above the conveying plane of the conveying element 10B can be adjusted by means of the separator setting signal. This provides a way to influence the separation performance of the separator, which is designed as a haulm belt. Furthermore, the relative speed between the screen belt and the coarse haulm belt 43 can be adjusted. For clarity, only the coarse haulm belt 43 is shown in Fig. 10, and not the actual conveying element 10B, which is designed as a screen belt (see Fig. 14).

[0091] A test image 8 resulting from the field of view of the optical image acquisition unit 6, shown in dashed lines in Fig. 11, is shown in detail in Fig. 12. The evaluations described above are carried out on the basis of the deviations of the proportions of the detected and classified objects from the threshold values ​​R, using a test data set created from or generated by this test image 8.

[0092] Starting from conveying element 10B, the remaining harvested material is transferred to another conveying element 10C with a conveying direction 1C. This element is equipped with a separating device in the form of several vertically stacked, rotating deflecting rollers 24. A pulse exerted by this device transports the harvested material towards conveying element 10D (Fig. 13).

[0093] A distance H between the conveying element 10C and the lower deflection roller 24 is adjustable to vary the separation performance and thus represents the adjustable operating parameter. If necessary, further distances between the individual deflection rollers 24 can be varied to adjust the intensity of the deflection or any separation function in which haulm is drawn between the deflection rollers 24. Alternatively or additionally, the separation performance or deflection can be varied by adjusting the rotational speeds of the deflection rollers 24. The height H of the lower ends of the fingers 26 of a separating device designed as a finger belt 26.1, which belongs to the conveying element 10D, is also adjustable as one of several operating parameters. The height H describes the distance of the fingers 26 from the upper edge of the conveying element designed as a hedgehog belt. Furthermore, the angle of attack of the finger belt 26 can be adjusted.1. The conveyor element must be designed to be adjustable to a perpendicular to the conveying plane. The same applies to the rotational speed of the guide belt 26.1.

[0094] The image acquisition unit 6 shown in Fig. 13 generates the test image 8 shown in Fig. 14, in which a test image part 8A relevant in the present embodiment is defined by means of filtering or masking. For monitoring the performance of a separating device, in this case the separation performance of the deflecting rollers 24, an additional test image part 8B, which is located behind the deflecting rollers 24 when viewed from a conveying direction 1C, can be selected. In particular, the area in front of the deflecting rollers 24 is monitored for the setting of the separating device. The test data set then results from the corresponding test image part 8A.

[0095] If a corresponding cleaning characteristic RS for test image section 8A indicates an insufficient separation performance of an upstream or depicted separation device, the separation device can be set to a more aggressive setting. Alternatively, if the cleaning characteristic in test image section 8B indicates an excessive separation performance, for example due to excessive amounts of impurities 5 in the form of clods behind the deflection rollers 24, which are at least partially required for potato protection on the subsequent conveying section, the distance H of the deflection rollers 24 from the conveying element 10 can be reduced, and the separation device can thus be set to a less aggressive setting.

[0096] Another optical image acquisition unit 6, arranged in the area of ​​conveyor belts 10C and 10D, is shown in Figs. 15 and 16. This image acquisition unit 6 can be used in addition to or as an alternative to the image acquisition unit 6 shown in Fig. 6. In particular, it serves to monitor the operation of the separating and diverting device formed by the deflection rollers 24. This monitoring unit also has a light source 7 for better illumination of the monitored area.

[0097] A further optical image acquisition unit 6, with an associated light source 7, is arranged above a sorting table, facing a conveying element 10E and a conveying element 10F (Fig. 17). By means of masking, the test image parts 8A and 8B depicted in the test image 8 according to Fig. 18 are selected. These parts monitor, on the one hand, the conveying element 10E with conveying direction 1E for transporting root crops 4, and on the other hand, the conveying element 10F with conveying direction 1F for transporting impurities 5 in the form of stones and / or clods. The evaluation described above is used to check whether the proportion of root crops 4 on the conveying element 10F is too high. If this is the case, the upstream [unclear text] is [unclear text] using the method according to the invention.

[0098] The separating device is adjusted more precisely. This separating device is located above the conveying element 10D, which is designed as a hedgehog belt, and is specifically designed as a finger belt with fingers 26, shown as examples and with dashed lines, although in the illustration shown they are arranged behind the cover 40 in front of it. For example, the distance between the fingers 26 and the conveying element 10D is reduced to convey more harvested material in the form of root crops 4 onto the conveying element 10E via an associated chute 41. If too many impurities 5 in the form of stones and / or clods are detected on the conveying element 10E, the rotational speed of deflection rollers 24 can be reduced, for example, since a lower impulse exerted on impurities 5 leads to better deflection of any stones towards the conveying element 10F. Impurities 5 then slide more easily onto the conveying element 10F via a chute 42.

[0099] Fig. 19 illustrates the application of the inventive method to a further separation device, in this case in the form of a fine haulm elevator. Accordingly, the test image 8 shown in Fig. 19 depicts a conveying element 10K with a conveying direction 1K, which leads via a drop step onto a separation device designed as a fine haulm elevator with a haulm belt 30. This transports any fine haulm present in the harvested crop stream in direction 31, while potatoes and heavier admixtures of the harvested crop stream, which land on the haulm belt 30, fall back into the gap between the conveying element 10K and the fine haulm elevator due to the inclined position of the haulm belt 30, which can be set via an adjustable angle a. If the proportions of root crops 4 and 8B in test image sections 8A and 8B areIf the additive quantities 5 are outside the reference values, the separating device can be made more aggressive or less aggressive, for example by adjusting the angle of attack a or by changing the rotational speed of the weed belt 30, in order to achieve the desired separation performance.

[0100] In the roller separator shown in Fig. 20, which is preferably used in storage technology, the distance H between individual conveying elements 10T is adjustable. The conveying element 10T thus directly forms the separating device. In the individual test image sections 8A and 8B of the test image 8, the proportion of the material is preferably calculated and used for adjusting the separating device's performance. As described above, perspective correction is performed based on the "fish-eye" representation of the image acquisition unit 6.

[0101] Other alternative embodiments of the invention can be designed, for example, as self-propelled beet harvesters or as a cleaning line in potato or beet storage technology.

Claims

Claims 1. Method for controlling the operation of a machine (2) for harvesting root crops (4) and / or for separating root crops (4) from further crop material, comprising admixtures, in which at least one test image (8) of at least one part of the crop material moved relative to a machine frame (12) by means of at least one conveying element (10, 10A, 10B, 10C, 10D, 10E, 10F, 10K, 10T) is recorded by at least one optical image acquisition unit (6), characterized in that an evaluation device generates a separation device setting signal for setting at least one operating parameter of a separation device of the machine (2) on the basis of a test data set generated on the basis of or formed by the test image (8).

2. Method according to claim 1, characterized in that the evaluation device evaluates the test data sets locally on the machine (2) or a directly attached towing vehicle.

3. Method according to claim 1 or 2, characterized in that the evaluation device calculates at least one first part (A1 ) of the test image (8) formed by at least one image area (19), wherein the at least one image area (19) at least partially depicts a defined component of the harvested crop or the machine (2), and a cleaning characteristic value (14) is calculated on the basis of the first part (A1 ).

4. Method according to claim 3, characterized in that the at least one image area (19) forming the first part (A1 ) is preferably identified as showing the defined component of the harvested crop or machine (2) on the basis of a test sub-data set generated on the basis of the image area (19), in particular at least one of which includes color information.

5. Method according to claim 3 or 4, characterized in that the test sub-data set, in particular at least one test value included therein, preferably the color information, is classified by a statistical classification method, in particular a model-based classification method, and an image area (19) is assigned to the first part (A1) in particular when the result of the classification method is assigned to the defined component of the harvested crop or the machine (2).

6. Method according to one of claims 3 to 5, characterized in that the at least one test value of the test sub-dataset, in particular the color information, is compared with at least one reference value (R) and an image area (19) is assigned to the first part (A1) in particular if at least the at least one test value of the test sub-dataset lies within an assigned reference value range.

7. Method according to claim 5 or 6, characterized in that the evaluation device, upon input of exemplary image areas (19) attributable to the first part (A1) of a reference image, assigns a classification method to the classification method. The underlying model is automatically further developed and / or at least one reference value range is automatically calculated or changed.

8. Method according to one of claims 3 to 7, characterized in that different image areas (19) are weighted differently when calculating the first part (A1 ).

9. Method according to one of claims 3 to 8, characterized in that the entire test image (8) or a connected part of the test image (8A) is divided into partial image areas (16), which in particular each comprise the same number of pixels of the test image (8), preferably exactly one pixel.

10. Method according to one of claims 3 to 9, characterized in that the test image (8) comprises several test image parts (8A, 8B) for which the evaluation device calculates a first part (A1), in particular several parts (A1, A2, A3) of image areas (19), wherein preferably the test image parts (8A, 8B) depict harvested material from different conveying elements conveyed by a separating device. 1 1. Method according to one of claims 3 to 10, characterized in that the image areas (19) forming the first part (A1 ) show root crops (4) or parts thereof and image areas forming a second part (A2) show admixtures (5) or parts thereof.

12. Method according to one of claims 3 to 1 1 , characterized in that the cleaning characteristic value (14) is determined by means of a deviation of the first component (A1 ) from a threshold value (R) calculated by the evaluation device.

13. Method according to any of the preceding claims 3 to 12, characterized in that the separator setting signal is calculated on the basis of a plurality of cleaning characteristic values ​​(RS) calculated in particular successively over time or at least a previously calculated cleaning characteristic value (RS) is included in the calculation of the cleaning characteristic value (RS).

14. Method according to one of the preceding claims with at least two image acquisition units (6) and at least two conveying elements (10), characterized in that the first image acquisition unit (6) captures a first test image of a portion of the harvested crop conveyed by a separating device by means of the first conveying element (6), the second image acquisition unit (6) captures a further test image of a portion of the harvested crop conveyed by the separating device by means of the second conveying element (6), and the separating device setting signal is generated on the basis of at least one of the two test data sets formed by or generated on the basis of the two test images.

15. A method according to one of the preceding claims, characterized in that at least one further sensor, in particular an ultrasonic or touch sensor, for measuring a crop layer thickness on the conveying element, a sensor for measuring the drive power, a moisture sensor and / or a rotary sensor, is used. The number sensor transmits sensor data to the evaluation device, which is used in the calculation of the separator setting signal.

16. Method according to one of the preceding claims, characterized in that the evaluation device triggers either an increase or a decrease of the operating parameter by means of different separation device setting signals.

17. Method according to claim 16, characterized in that after triggering a change in operating parameters for a defined period or a defined conveying distance of the conveying element (10, 10A, 10B, 10C, 10D, 10E, 10F,10K,10T) does not trigger any further change in operating parameters.

18. Method according to one of the preceding claims, characterized in that the operating parameter is a distance (Fl) between two conveying elements (10T), a separating element (32) or the separating device to the or a further conveying element (10C).

19. Method according to one of the preceding claims, characterized in that the operating parameter is a separation speed, in particular a rotational or cyclical speed, of a separation element (24, 26, 30) or of the separation device.

20. Method according to one of the preceding claims, characterized in that the operating parameter is an angle of attack (a) of the conveying element (10) or of the separating device.

21. Method according to one of the preceding claims, characterized in that the operating parameter is an engine power and / or an angle of attack of an associated unit.

22. Method according to one of the preceding claims, characterized in that the isolating device setting signal is transmitted to an isolating device control unit by wired means, in particular by means of CAN bus or Ethernet, or wirelessly, wherein preferably the isolating device setting is to be released in advance by an operator via an input at an interface.

23. Method according to one of the preceding claims, characterized in that the separation device setting signal and / or the operating parameter is displayed for an operator and / or is automatically used to set the operating parameter.

24. Method according to one of the preceding claims, characterized in that the optical image acquisition units are only 1D or 2D Gather information.

25. Machine (2) for harvesting root crops (4) and / or for separating root crops (4) from other harvested material, comprising at least a machine frame (12), a conveying element (10), an image acquisition unit (6), a separation device and an evaluation device, and designed to carry out the method according to one of the preceding claims.

26. Machine (2) according to claim 24, characterized in that the evaluation device comprises a graphics processing unit, in particular a GPGPU, and / or an FPGA-based processing unit.

27. Machine (2) according to claim 25 or 26, characterized by at least one sensor coupled to the evaluation device, in particular an ultrasonic or touch sensor for measuring a crop layer thickness on the conveying element (10), a sensor for measuring a drive power, a speed sensor arranged on a conveying element (10) and / or a moisture sensor.

28. Machine (2) according to one of claims 25 to 27, characterized by several image acquisition units (6) which, during operation, each receive at least one test image (8) of the same conveying element (10) or of the same separating device or of different conveying elements (10) or separating devices, in particular conveying from the same separating device.

29. Machine (2) according to one of claims 25 to 28, wherein the image acquisition unit (6) is arranged such that, during operation, the test image (8) is at least two at least one separating element (24, 32) represents separate conveying element sections, each at least partially.

30. Machine (2) according to one of claims 25 to 29, characterized in that the image acquisition unit (6) is arranged such that the test image (8) shows at least two alternative conveying paths forming conveying elements (10) of the machine (2) for different crop components, in particular a conveying path for corn crops (4) and a conveying path for impurities.

31. Machine (2) according to one of claims 25 to 29, characterized in that the conveying element (10) is a sieve belt (10A, 10B, 10E) or a hedgehog belt (10C, 10D), which in operation runs in particular under at least one deflection roller (24) extending transversely over the conveying element (10) and diverting the harvested crop from it, or is designed as a sieve star (1 OP, 10Q, 10S) or as a conveying roller (10T), in particular encompassed by a roller table (10M), or as a harvesting share.

Images