Method for monitoring the operation of a machine for conveying root crop

Abstract

The invention relates to a method for monitoring the operation of a machine (1) for conveying root crop (16) and possibly other harvested material (17, 18), comprising: - obtaining images (15) from an image capturing unit (11), of root crop (16) and possibly other harvested material (17, 18), conveyed in the machine (1), wherein the obtained images (15) are taken with a specific field of view (42); - determining for the obtained images (15) a detection sub-area (20, 21, 22, 24) within the specific field of view (42); - detecting at least part of the root crop (16) and / or other harvested material (17, 18) in the detection sub-area (20, 21, 22, 24); and - determining one or more root crop parameters (29) and / or other harvest parameters (30, 31, 32) associated therewith. The invention further relates to a machine (1), provided with a monitoring system provided for carrying out such method.

Description

[0001] METHOD FOR MONITORING THE OPERATION OF A MACHINE FOR CONVEYING ROOT CROP

[0002] The invention relates to a method for monitoring the operation of a machine for conveying root crop and possibly other harvested material, comprising steps:

[0003] Step a) obtaining images from an image capturing unit, of root crop and possibly other harvested material, conveyed in the machine, wherein the obtained images are taken with a specific field of view of the image capturing unit;

[0004] Step b) detecting at least part of the root crop and / or other harvested material in a detection sub-area of the images;

[0005] Step c) determining one or more root crop parameters and / or other harvest parameters associated with the at least part of the root crop and / or other harvested material detected in step b.

[0006] This application concerns machines for conveying root crop and possibly other harvested material, such as planters, cleaners, graders and harvesters. Root crops in this respect not only include harvested crops such as e.g. potatoes, onions, red beets, carrots, . . ., but also seedlings thereof, e.g. potato planters.

[0007] Such harvesters and planters may vary from machines which are pulled by a tractor or which are carried on a tractor to self-propelled vehicles.

[0008] Such a harvester comprises a crop-digger section, which e.g. comprises one or more harvesting shares for the harvesting of root crops. Conveying means, comprising e.g. sieving units, a hedgehog, elevators, further conveyors... are provided in order to transport the root crops from the crop-digger section towards a bunker, which may be part of the harvester, or may be separate from the harvester, such as e.g. part of a truck or an external trailer, e.g. towed by a tractor or alternatively towards a big bag or a crate or a bag. Still more alternatively, the root crops may be deposited on the ground. At harvesting of root crops, besides the root crops typically also by-products, such as haulm, soil, stones, ... are included in the harvested material. During transport from the crop-digger section towards the bunker, as much by-products as possible are removed, to minimize the by-products in the bunker. It is important to clean the root crops as well as possible, limiting damage to the root crops as much as possible. Given changing harvesting conditions, such as a change in soil type or soil humidity, or changes in harvest requirements for specific markets, or different crop types and crop varieties, amount and size of stones in the ground, inclination of the ground, . . . this is not a sinecure. To this end, in a harvester, one or more operating parameters are typically adjustable by a harvester operator by taking operation actions in order to ensure that the harvester remains efficient in various operating circumstances. For example, the speed of a conveyor belt can be set and / or agitators and / or an interval between cleaning units and / or a slope of a hedgehog, . . .

[0009] Such a planter comprises furrow openers for forming furrows in the ground. Seedlings are fed from bunker to one or more planting element. The seedlings are singulated and the planting element distributes the singulated seedlings at a desired planting distance in the furrow. Such planting element can be a cup elevator or a belt bed, etc. The furrow is thereafter filled with soil and seedlings are covered. It however happens that misses occur, with no seedlings being planted at certain times, where it was intended to be. This may mean that the bunker is empty or that there is a blockage before or behind the feed opening. When this is noticed, the feed opening can typically be adjusted, to reduce it. Sometimes two or more seedlings, called doubles, are planted at the same time, where only one, singulated, seedling should be planted. This typically indicates that too many seedlings pass the feed opening at a time. When this is noticed, the feed opening can typically be adjusted, to enlarge it.

[0010] Said cleaners and graders can be part of a harvester, or can be fixed machines. Similar settings can be included therein.

[0011] Known such machines typically include image capturing units for imaging one or more of the conveying means in the machine, with the root crop and possibly other harvested material fed thereon. Images from these capturing units are visualised on a screen e.g. in the cabin, so that the operator is able to see how efficiently the different conveying means are operating and how the root crop and possibly harvested material moves through the machine. Based on what the operator observes on these images, the operator can typically adjust one or more operating parameters, such as e.g. a conveyor inclination, a conveying speed, ... For example, when the operator observes that there is a lot of soil on a sieving web of a potato harvester, the operator can increase the sieving web speed so that more soil falls through the sieving web. By allowing the operator to adjust the operating parameters based on what he sees on the images, the machine can be set more efficient in various operating circumstances.

[0012] Different camera-based vision systems are already described for detecting at least part of the root crop and possibly other harvested material in images obtained with a camera. Such root crop and possibly other harvested material can therein be detected e.g. using one or more neural network algorithms or more classic machine vision systems.

[0013] In order to aid the operator, it is e.g. in research projects and e.g. in WO2021160607A1, CA3118799A1, US2018 / 271015A1 and US2023 / 0012175A1 proposed to use evaluation devices for automatically analysing images obtained by such image capturing units for determining related harvest parameters and generating operating parameters of the root crop harvester based on this analysis. Also in e.g. EP3557972B1 it is described to automatically derive one or more harvest parameters with a processing unit analysing such images.

[0014] In theory, the operator will be able to make more informed decisions using such harvest parameters and possibly operating parameters generated in function thereof, to improve the efficiency even further. Such harvest parameters can e.g. be the number of root crops, the size of root crops, the amount of haulm and / or clods within the harvested material, quality of the root crops, ... In practice, however, the computer vision systems which are proposed in this respect have a myriad of issues which make them not practically usable for monitoring the operation of a root crop harvester in real-life conditions.

[0015] One of such issues is that in agricultural machinery and more particularly root crop harvesters equipped with camera-based vision systems, images are taken with a specific field of view of a camera, i.e. a specific shot. Such field of view typically however comprises zones which are of less interest for therein detecting at least part of the harvested material. It is not always possible to mount a camera in such a way that it can be pointed directly at an area that is most interesting to therein detect at least part of the harvested material. Or, even if this would be possible, several zones of interest to be analysed can be captured with a same camera, so that irrelevant zones are also captured, or a camera can be mounted in such a way that e.g. an operator is offered visual information going beyond one or more zones of interest to be analysed for detecting at least part of the harvested material therein. The field of view can broadly cover entire modules or machine sections, whereas vision algorithms which are applied for detecting at least part of the harvested material, such as for identifying crop distribution, soil contamination, or cleaning performance, require image analysis only in a specific and limited region within that field of view. Within such field of view, it is then possible, as e.g. described in CA3118799A1, to predefine a detection sub-area as a region of interest, for therein detecting at least part of the harvested material. By detecting at least part of the harvested material in a said detection subarea within such field of view, the quality of the analysis can thus be improved.

[0016] During use of a harvester, a said field of view can however e.g. be altered by turning such image capturing unit or by moving it otherwise with respect to a target, or when such target moves with respect to the image capturing unit or by zooming, or when such image capturing unit is replaced or maintained, etc. Inconsistent results can also be obtained when e.g. lighting changes, for example when lights on the harvester are switched on. With a changing field of view, the detection of at least part of the harvested material within the predefined detection sub-area will adversely be affected.

[0017] In EP4298882A1 a control system for controlling the operation of a planter based on such camera-based vision system is described. Also this system is open to improvement just like known systems for cleaning and grading. It is an objective of the invention to ensure that a desired quality of the detection of at least part of the harvested material can be maintained even when such changes occur.

[0018] This objective is achieved by providing a method for monitoring the operation of a machine for conveying root crop and possibly other harvested material, comprising: step a) obtaining images from an image capturing unit, of root crop and possibly other harvested material, conveyed in the machine, wherein the obtained images are taken with a specific field of view of the image capturing unit; step b) detecting at least part of the root crop and / or other harvested material in a detection sub-area of the images; step c) determining one or more root crop parameters and / or other harvest parameters associated with the detected at least part of the root crop and / or other harvested material; step d) determining for the obtained images a detection sub-area within the specific field of view for performing step b with this detection sub-area.

[0019] The object of the invention is further achieved by providing a machine for conveying root crop and possibly other harvested material, comprising a monitoring system for monitoring the operation of machine by carrying out an abovementioned method according to the invention, the monitoring system comprising: an image capturing unit for taking images of root crop and possibly other harvested material conveyed in the machine, with a specific field of view; one or more processing units coupled to the image capturing unit or part thereof, comprising computer-executable instructions:

[0020] - for controlling the image capturing unit for taking images;

[0021] - for detecting at least part of the root crop and / or other harvested material in a detection sub-area of the images;

[0022] - for determining one or more root crop parameters and / or other harvest parameters associated with the at least part of the harvested material; and - for determining for the obtained images the detection sub-area within the specific field of view for detecting the at least part of the harvested material with this detection sub-area.

[0023] Instead of predefining said detection sub-area as with the prior art, the detection subarea is according to the invention now actively determined, such that a desired quality of the detection of at least part of the root crop and / or other harvested material can be maintained even when mentioned changes occur. Unnecessary computational load from processing irrelevant areas can thus further be reduced and reduced accuracy due to background or peripheral visual noise can further be limited.

[0024] An operator e.g. wants to see more than just conveyor means in the cabin, such as for example when the operator wants to have a field of view wherein a conveyor is observed from an angle even though this angle is less suitable for detecting at least part of the root crop and / or other harvested material in the obtained images, or when different parts are visualised in one image, whereas only one or more smaller parts are to be analysed for the desired detection.

[0025] A said detection sub-area of the field of view can then be selected so that it is easier for therein detecting the at least part of the root crop and / or other harvested material. By choosing such a detection sub-area as a region of interest, it is possible to purposefully neglect a zone of the field of view in which the root crop and / or other harvested material to be detected would normally not be present and would thus with high certainty not be detected. It is furthermore e.g. possible to purposefully neglect the detection of root crop and / or other harvested material at places where such root crop and / or other harvested material can get stuck, such as for example besides elevators, so that the obtained detection of root crop and / or other harvested material in the images is not adversely affected by this ‘stuck’ root crop and / or other harvested material.

[0026] The detection sub-area can in some embodiments be defined manually by an operator. In a particular embodiment of a machine according to the invention, the monitoring system therefore comprises one or more user interfaces coupled to the one or more processing units, for determining the detection sub-area based on user input entered via one or more of such user interfaces. Such user interfaces can e.g. comprise a monitor and / or a touch-screen and / or a keyboard and / or a dial and / or voice activation, etc. Such user interfaces can be local devices, e.g. in machines with a cabin, such user interfaces can be provided in the cabin. But such user interfaces can also comprise mobile devices and / or remote devices. In specific embodiments, such user interface is a graphical user interface and is then preferably provided for reproducing the obtained images for presenting them to the operator, this preferably so that the operator can thereon select the desired detection sub-area.

[0027] Alternatively and / or additionally, the detection sub-area can in some embodiments be automatically determined e.g. tracked by a tracking algorithm, more specifically based on the obtained images such as more specifically the recognition of reference features or forms and / or based on further parameters detected in the machine or its surroundings.

[0028] In a method according to the invention, the at least part of the root crop and / or other harvested material can be detected in any possible way, such as any known way e.g. using one or more neural network algorithm or more classic machine vision systems. Associated with the detected at least part of the root crop and / or other harvested material, one or more root crop parameters and / or other harvest parameters are determined. In specific embodiments the machine can be adjustable with at least one adjustable operating parameter. For example, the speed of a conveyor belt can be set and / or agitators and / or a distance between cleaning units and / or a slope of a hedgehog and / or settings of said image capturing unit... The method can then comprise generating at least one operating parameter signal based on the one or more root crop parameters and / or other harvest parameters for adjusting the at least one operating parameter of the machine.

[0029] In this respect: for detecting the at least part of the root crop and / or other harvested material; and for possibly determining one or more root crop parameters and / or other harvest parameters; and for possibly generating at least one operating parameter; and for possibly setting such generated at least one operating parameter; the contents of patent applications PCT / IB2024 / 061722, EP24214856.7 and EP4298882A1 of patentee, for which priority is claimed for the present application, is hereby more specifically incorporated by reference.

[0030] The detection sub-area can be determined by defining such detection sub-area from scratch and / or pre-defined shapes for such detection sub-area may be implemented as factory settings, whereby a shape can be selected from these predefined shapes and / or it can be provided that a determined shape for a detection sub-area may be saved for later reuse and / or it can be provided that a determined detection sub-area is rescalable for determining a new detection sub-area, e.g. when zooming in or out, etc. In case a detection sub-area is selected from predetermined options, the monitoring system can more specifically comprise a storage unit wherein predetermined options are stored and the one or more processing units can comprise a selection unit for selecting one of the predetermined options.

[0031] A detection sub-area may consist of a contiguous area or several separate areas, this e.g. in case desired root crop parameters and / or other harvest parameters, e.g. for setting an operating parameter are related to root crop and / or other harvested material detected in all these separate areas.

[0032] The detection sub-area can be any shape, but is preferably defined by one or more lines. Such line can be a baseline or can define a free shape or a polygon and more preferentially a quadrilateral such as a rectangle or a trapezium. By choosing a trapezoidal shape, for example a perspective transformation can e.g. be applied to this trapezoidal shape.

[0033] A line can define one detection sub-area at a specified side thereof, or can define two detection sub-areas at both sides thereof. Multiple such lines can be set for determining multiple corresponding sub-areas. Detection sub-areas can be discrete, but may also overlap. The method in an embodiment more specifically comprises determining for the obtained images multiple detection sub-areas within the specific field of view. Then each of the multiple detection sub-areas can be used for a specific purpose, such as for example: different detection sub-areas for detecting root crop and / or other harvested material at different locations within a same field of view such as different locations on a same conveyor, e.g. for comparing characteristics thereof with respect to each other, or on different conveyors within a same field of view; a first detection sub-area being provided for a specific neural network algorithm, e.g. for detecting stuck haulm, while a second detection sub-area within the same field of view is provided for another neural network algorithm, e.g. for detecting the amount of soil, or for example to give visual information to the operator.

[0034] The detection sub-area can newly be defined for such a machine, but can also be changed over time, which will typically be desired when there are changed circumstances for which a modification of the detection sub-area would be appropriate, this by changing the detection sub-area in response to such changed circumstances.

[0035] The method then preferably comprises determining said detection sub-area more than once, this by repeating step d.

[0036] As explained further on, the image capturing unit is preferably adjustable by panning and / or tilting and / or zooming and / or setting of a shutter time or night vision (infrared), black and white versus colour, resolution, a focus parameter, etc. The objects seen in the field of view of the obtained images can vary depending on how the image capturing unit is adjusted, so that a new detection sub-area is desired.

[0037] In addition, various adjustments to the operating parameters, for example the inclination of a conveyor of the conveying means, also change the field of view of the obtained images widely, so that a previously determined detection sub-area is no longer optimal.

[0038] Also e.g. deviating lighting conditions in a zone in the detection sub-area with respect to other zones in the detection sub-area can occur, or a blockage can be detected, which is not further to be taken into account, or dust, dirt or foliage can be detected on or in front of the camera, for which it is desired temporarily to alter the detection sub-area. By allowing to determine the region of interest more than once, these various adjustments can be taken into account, minimizing their effect on the robustness and the performance of the detection of the at least part of the root crop and / or other harvested material, e.g. using a neural network algorithm or a more classic machine vision system.

[0039] More specifically, the method can thus comprise repeating steps a and b, wherein in a repeated step a, images are obtained from the image capturing unit with a same or a new specific field of view of the image capturing unit and wherein in step d a new detection sub-area is determined before repeating step b with the new detection subarea. Inconsistent results due to e.g. altering fields of view or changing lighting conditions can thus be limited.

[0040] The detection sub-area used in the previously performed step b can then be replaced with the in step d determined new detection sub-area for performing the repeated step b with this new detection sub-area.

[0041] Step d can in this respect e.g. be performed after a predetermined amount of time since a previous execution of step d, this depending on the circumstances, e.g. after 2 seconds or up to after 15 seconds, or even after a minute or even langer for some applications.

[0042] Alternatively and / or additionally, in a more specific embodiment the method can therefore comprise: step e) detecting changed circumstances; wherein step d is performed if a changed circumstance is detected in step e.

[0043] Such changed circumstances could e.g. be determined automatically, based on which a determination of a new detection sub-area is triggered automatically, and / or an operator could decide to reset such detection sub-area based on changed circumstances. Changed circumstances can e.g. be detected by the operator when he moved the image capturing unit or a part of the machine on which the image capturing unit is directed, or zoomed in or out, or changed a setting of the image capturing unit, etc.

[0044] Changed circumstances can e.g. be detected automatically e.g. based on changes viewed in the images itself or based on a detection of an adjusted setting of the image capturing unit itself or a setting of a mutual position of the image capturing unit and a part of the machine on which the image capturing unit is directed, or a detection with a detector already present in the machine or provided in the machine for this purpose, etc.

[0045] In a specific method wherein a new detection sub-area is determined in step d, this new detection sub-area can be determined based on the changed circumstance detected in step e.

[0046] Step e can in an embodiment more specifically comprise determining the specific field of view of the image capturing unit, at which the images are taken in the repeated step a, for detecting changed circumstances so that a detected new specific field of view is a said changed circumstance for performing step d.

[0047] An adjustment of the specific field of view causing a new specific field of view being detected can be: accidental, such as e.g. when fastening means, such as a bolt securing the image capturing unit comes loose, causing the field of view of the images taken to vary; or deliberate, such as with an option to set a mutual position between the image capturing unit and a part of the machine on which the image capturing unit is directed.

[0048] A new specific field of view can for accidental changes e.g. be detected by an operator and / or automatically, by detecting an unexpected movement of part of the machine within the images.

[0049] A new specific field of view due to deliberate changes can e.g. be detected: - by user input, e.g. when the operator manually adjusted or replaced parts of the machine and thereafter initiates the determination of a new detection sub-area or when the operator initiates an adjustment of an operator parameter to be adjusted by the machine; and / or

[0050] - by detectors, such as e.g. an inclination sensor, e.g. when a detected inclination remains for a predetermined amount of time within a predetermined range; and / or

[0051] - by an operating parameter signal, e.g. when such signal exceeds a predetermined value; and / or

[0052] - by detecting a movement of part of the machine within the images; etc. Examples of how a detection sub-area can be determined upon such changes are given further on.

[0053] In more specific embodiments wherein the specific field of view of the image capturing unit is detected, a first said detection sub-area can be determined at a first field of view of the image capturing unit, a second said detection sub-area can be determined at a second field of view of the image capturing unit and in step d the new detection sub-area can be based on:

[0054] - the specific field of view determined in step e; said first detection sub-area determined at the first field of view; and said second detection sub-area determined at the second field of view.

[0055] The first detection sub-area and the second detection sub-area determined at these two different fields of view can e.g. be determined in an initialisation step. It is also possible alternatively or additionally, during operation of the machine, to save different determined detection sub-areas determined at different fields of view in order to take those into account when determining further detection sub-areas at further deviating fields of view.

[0056] When the determined specific field of view as detected in step e deviates from the specific field of view in a previous step a, this is considered as a detection of an adjustment of the field of view. The new detection sub-area which is then to be determined can then more specifically be determined in relation to such first determined detection sub-area at its corresponding first field of view and such second determined detection sub-area at its corresponding second field of view, different from said first field of view and different from the deviating field of view determined in step e, taking the degree of deviation or correspondence of the specific field of view determined in step e with respect to said first field of view and said second field of view into account.

[0057] More specifically it is thus possible in step e to track changes of the field of view over time. Tracked changes can then be used as a measure of the specific field of view in which images are taken at a given moment.

[0058] In a more specific embodiment the image capturing unit is provided for obtaining images of root crop and / or other harvested material conveyed on a conveyor of the machine, wherein the image capturing unit and the conveyor are arranged in an adjustable mutual position. This can e.g. be a sieving conveyor, or a hedgehog, or a set of rollers, or a bottom of a bunker, etc. whether height adjustable and / or for which the inclination can be set and / or with respect to which the image capturing unit is adjustable.

[0059] The first detection sub-area can in such embodiment be determined at a first mutual position and the second detection sub-area can be determined at a second mutual position and in step e the mutual position of the image capturing unit and the conveyor can be determined in order to determine the specific field of view.

[0060] More specifically, by tracking changes in the mutual position over time in step e, these tracked changes can be used as a measure of the current mutual position.

[0061] In embodiments with a said step e, for detecting changed circumstances, step e can alternatively and / or additionally comprise detecting deviations within the detection sub-area of the obtained images so that a detected deviation is a said changed circumstance for performing step d. For determining the detection sub-area in step d, the method can more specifically comprise: a step f) determining a basic region of interest, which basic region of interest is subdivided in multiple partial regions of interest; a step g) detecting for each partial region of interest a circumstance parameter; and in step d the circumstance parameters detected in step g can be compared and based on this comparison one or more of the corresponding partial regions of interest can be selected to compose the detection sub-area.

[0062] Similarly as described above for a detection sub-area, said basic region of interest can be defined manually by an operator or automatically determined. The basic region of interest can similarly be determined by defining such basic region of interest from scratch and / or by selecting such basic region of interest from predetermined options and / or by adjusting a predefined basic region of taking user input and / or circumstances into account. The basic region of interest may similarly consist of a contiguous area or several separate areas. The basic region of interest can similarly be any shape. Multiple basic regions of interest can similarly be set, which can be discrete, but may also overlap. The basic region of interest can newly be defined for such a machine. Step f can be executed once after which said basic region of interest is reused each time step d is executed, but can also be executed more than once such that the basic region of interest is changed over time. Step f can then more specifically e.g. be executed each time step d is executed.

[0063] Step g is preferably executed each time step d is executed and is even more preferably executed in a continuous way.

[0064] The basic region of interest can manually be subdivided into multiple partial regions of interest or can automatically be subdivided, e.g. in a predetermined way or by selection from predetermined options or by adjusting a set division taking user input and / or circumstances into account. As a basic region of interest may consist of several separate areas also partial regions of interest do not necessarily have to be adjacent. Embodiments wherein a basic region of interest is determined can even more specifically comprise: a step h) detecting at least part of the root crop and / or other harvested material in each of the partial regions of interest; a step i) for each partial region of interest determining a partial root crop parameter and / or another partial harvest parameter associated with the corresponding at least part of the root crop and / or other harvested material detected in step h); and the partial root crop parameters and / or the other partial harvest parameters determined in step i can then be used as circumstance parameters in step g and d.

[0065] In this way a said deviation can more specially be detected within the detection subarea by evaluating these partial root crop parameters and / or other partial harvest parameters.

[0066] Such a deviation can in this respect e.g. comprise one or more of the following: a reduced amount of potatoes in a partial region of interest with respect to other partial regions of interest in the detection sub-area for which it is e.g. desired to take this into account for determining operating parameters, such as:

[0067] - in a machine in which root crop and possibly other harvested material is conveyed in multiple rows, in case one or more of such rows are not used; or

[0068] - in case the machine is used on a sloped field; or

[0069] - in case there is an imbalance on a conveyor that influences a setting of an operating parameter in an undesirable way, such as e.g. an imbalance between start and end of a conveyor belt; a detected blockage which is not further to be taken into account.

[0070] Step h and step i are preferably executed each time step d is executed and are even more preferably executed in a continuous way. The at least part of the root crop and / or other harvested material which are detected in step h and correspond with the one or more partial regions of interest which are selected in step d to compose the detection sub-area can more specifically also be used in step b for detecting the at least part of the root crop and / or harvested material in the detection sub-area.

[0071] The partial root crop parameters and / or the other partial harvest parameters which are determined in step i and which correspond with the one or more partial regions of interest selected in step d to compose the detection sub-area can more specifically also be used in step c to determine the corresponding root crop parameter and / or the corresponding other harvest parameter.

[0072] Depending on the root crop parameter and / or the other harvest parameter this can e.g. be done by averaging or by adding up the corresponding partial root crop parameters and / or the corresponding partial harvest parameters. In some embodiments it is possible to accord to different sub-groups of the selected partial regions of interest different weighing factors for differently weighing these partial regions of interest when determining a partial root crop parameter and / or another partial harvest parameter associated with the corresponding at least part of the root crop and / or other harvested material detected therein for determining the corresponding root crop parameter and / or the corresponding other harvest parameter in step c.

[0073] Alternatively and / or additionally to the abovementioned circumstance parameters, a circumstance parameter can be one of: a lighting parameter, which is a measure of the lighting conditions in said detection sub-area; an obstruction parameter, which is a measure of obstructions on or in front of the image capturing unit, e.g. causing a blurred area in the detection sub-area, such as e.g. dust or dirt or foliage.

[0074] Such lighting parameter and / or obstruction parameter can e.g. similarly be determined as a said harvest parameter using similarly trained neural networks algorithms or a more classic machine vision system. In case step g is repeated continuously, for step e, for each partial region of interest the detected circumstance parameter in step g can more specifically be compared with an initial circumstance parameter and if the detected circumstance parameter during a predetermined amount of time deviates more than a threshold value from the initial circumstance parameter, this is considered as a said deviation and therefore considered as a detection of a said changed circumstance and the last detected circumstance parameter is set as a new initial circumstance parameter.

[0075] The predetermined amount of time and the threshold value can more specifically be provided settable.

[0076] In a particular embodiment, the machine comprises one or more reference elements for determining a detection sub-area. The method can then more specifically comprise a step j detecting the one or more reference elements in the obtained images and step d can then comprise determining the new detection sub-area based on the one or more reference elements detected in step j.

[0077] Such reference elements can be detected by the operator or can be detected automatically e.g. similarly as a said root crop or other harvested material, using similarly trained neural networks algorithms or a more classic machine vision system. Detection can e.g. be based on contour or colour detection or e.g. detection of a moving part such as a conveyor belt, constituting a reference element.

[0078] Said reference elements can be parts of the machine itself or the machine can additionally be provided with reference elements. Reference elements can e.g. be edges, bolts, structural outlines, a geometric contour of a conveyor, such as a cleaning module, or a sticker with a fixed pattern, a QR code, . . .

[0079] The relation between detected reference elements and the detection sub-area can be defined in many different ways. An edge as a reference element can e.g. define a line for defining a sub-area or reference elements can constitute the corners of a polygon for defining a sub-area or a sub-area can e.g. be defined as the smallest overlapping polygon for a reference element etc. In a specific embodiment, the machine comprises several rollers, such as axial rollers and / or haulm rollers, which rollers are positioned in the machine such that they are moveable relative to each other.

[0080] In step d the detection sub-area can then more specifically be determined so as to encompass at least part of these rollers.

[0081] In this respect, said rollers can more specifically be detected in step j as said reference elements.

[0082] The detection sub-area can then e.g. be determined as the smallest overlapping polygon for these rollers.

[0083] Similarly in step f the basic sub-area can more specifically be determined so as to encompass these rollers, e.g. as the smallest overlapping polygon for these rollers.

[0084] Step j is preferably executed each time step d is executed and is even more preferably executed in a continuous way.

[0085] For step e it can more specifically be determined whether and to what extent a said roller at least partially extends outside the detection sub-area and if said roller during a predetermined amount of time extends for more than a predetermined portion outside the detection sub-area, this is considered as a said detected deviation and therefore as a said detection of a said changed circumstance.

[0086] Similarly for an above described basic region of interest it can be determined whether and to what extent a said roller at least partially extends outside the basic region of interest and if said roller during a predetermined amount of time extends for more than a predetermined portion outside the basic region of interest to repeat step f. Also similarly it can be determined whether and to what extent a said roller at least partially extends outside the field of view for detecting changes in the field of view.

[0087] Similarly it can e.g. be determined whether and at what distance one of these rollers is located from an outer edge of the detection sub-area (or similarly the basic region of interest or similarly the image for the field of view) and when it is located at a distance greater than a predetermined distance from this outer edge for a certain period of time, this is considered as a said detected deviation and therefore as a said detection of a said changed circumstance.

[0088] Said predetermined time(s) and / or predetermined portion(s) and / or predetermined distance(s) can more specifically be provided settable.

[0089] In a specific embodiment, the image capturing unit can be provided for obtaining images of harvested material conveyed on a hedgehog of the machine, and in step d, the detection sub-area can then be situated at the top of the hedgehog. This detection sub-area is then preferentially determined by a said baseline. Instead of providing a baseline, the detection zone can e.g. be determined in the image as a freeform shape, or a polygon such as e.g. a rectangle or as a top edge of the hedgehog.

[0090] In a further embodiment of a method according to the inventions, said root crops are seedlings, which are fed by one or more conveying elements of the machine from a bunker of the machine towards a planting element of the machine. The image capturing unit can then be provided for obtaining images of an area around a connection of the one or more conveying elements to the planting element and in step d the detection sub-area can be situated on the planting element above the connection.

[0091] In step j the planting element can then more specifically be detected as a said reference element. For step e, a piling height to which seedlings are piling up on the one or more conveying elements at the connection, can then e.g. be detected and compared with an initial piling height and if the detected piling height during a predetermined amount of time deviates more than a threshold value from the initial piling height, this is considered as a detection of a said changed circumstance and the last detected piling height is set as a new initial piling height.

[0092] The predetermined amount of time and the threshold value can more specifically be provided settable.

[0093] The machine can comprise one or more said image capturing units. The image capturing unit can comprise a regular camera or a video camera or any other type of image capturing device.

[0094] Such image capturing unit can more specifically comprise one or more cameras for taking the images. Such camera is then preferably a digital camera.

[0095] Said images obtained from such image capturing unit can be images taken with a single such camera, or images can be composed from images from different cameras.

[0096] Preferentially, this image capturing unit is adjustable by the operator and can pan and / or tilt and / or zoom in accordance with the preferences of the operator. Such settings can also be automatically settable.

[0097] Furthermore, the shutter time can be settable by setting of a shutter time or night vision (infrared), black and white versus colour, resolution, a focus parameter, etc. Since it can be very dark within e.g. a root crop harvester or planter, it is necessary for image capturing units with shutter times to provide such shutter times sufficiently long so that the images are not too dark. However, the shutter times can of course not be too long, since otherwise the images will become very blurry, which would be detrimental for the performance of automatic detection systems, such as one or more neural network algorithms or a more classic machine vision system. In specific embodiments, a root crop harvester or planter comprises supplemental internal lights so that the shutter times of the images can become smaller.

[0098] In specific embodiments, the image capturing unit comprises a depth camera (Stereo or ToF (Time of Flight)) or a LiDAR camera system, so that the captured images are 3D images, which provide a lot more information than 2D images. These 3D images additionally provide depth info which allows an accurate volume measurement to be taken as an additional root crop parameter and / or other harvest parameter so that an automatic detection system, such as one or more neural network algorithms or a more classic machine vision system is able to more efficiently detect the root crop and / or possibly other harvested material. For most applications such depth information will however not be required. In other specific embodiments, the image capturing unit comprises a hyperspectral camera, so that e.g. the product quality of the root crops can be determined as a root crop parameter. In further specific embodiments, the image capturing unit comprises a thermal camera. In addition, the object of the invention is achieved by a computer program, comprising computer-executable instructions which when loaded on the one or more processing units of an above described machine according to the invention, cause the machine to carry out the method steps of an above described method according to the invention.

[0099] Furthermore, the object of the invention is achieved by a computer-readable data carrier having stored there on such computer program according to the invention.

[0100] The present invention will now be explained in more detail by means of the following detailed description of methods for monitoring the operation of a machine for conveying root crop and possibly other harvested material, corresponding monitoring systems, computer programs, computer-readable data and machines according to the present invention. The sole aim of this description is to give explanatory examples and to indicate further advantages and particulars of the present invention, and can thus by no means be interpreted as a limitation of the area of application of the invention or of the patent rights defined in the claims.

[0101] In this detailed description, reference numerals are used to refer to the attached drawings, in which :

[0102] - Fig. 1 schematically illustrates a possible embodiment of a monitoring system according to the invention, implemented in a root crop harvester ;

[0103] - Fig. 2 illustrates the possible field of view of one or more cameras positioned above a sieving conveyor, with possible set detection sub-areas and with possible detected harvested material and possible detected conveying means using neural network algorithms of a monitoring system according to the invention;

[0104] - Fig. 3 illustrates the possible field of view of one or more cameras positioned above a hedgehog, with possible set detection sub-areas and with possible detected harvested material and possible detected conveying means using neural network algorithms of a monitoring system according to the invention; - Fig. 4 schematically illustrates a normal production mode of a possible monitoring system according to the invention;

[0105] - Fig. 5 illustrates the possible field of view of Fig. 2 with a possible set basic region of interest, subdivided in multiple partial regions of interest;

[0106] - Fig. 6 illustrates the possible field of view of Fig. 3 with a possible set basic region of interest, subdivided in multiple partial regions of interest;

[0107] - Fig. 7 schematically illustrates a field of view of one or more cameras positioned above a conveyor, with a determined basic region of interest, subdivided in multiple partial regions of interest and some possible determined detection sub-areas, wherein for each of the partial regions of interest a circumstance parameter is detected, showing a difference in circumstance parameters between start and end of the conveyor;

[0108] - Fig. 8 schematically illustrates the field of view, with the determined basic region of interest, and partial regions of interest of figure 7, with a more pronounced difference in circumstance parameters between start and end of the conveyor;

[0109] - Fig. 9 schematically illustrates the field of view, with the determined basic region of interest, and partial regions of interest of figure 7, showing a difference in circumstance parameters between the left side and the right side of the conveyor;

[0110] - Fig. 10 schematically illustrates the field of view, with the determined basic region of interest, and partial regions of interest of figure 7, showing a a more pronounced difference in circumstance parameter between the left side and the right side of the conveyor than in figure 9;

[0111] - Fig. 11 schematically illustrates the field of view, with the determined basic region of interest of figure 7, with different possible determined sub-areas, showing a gradual difference in circumstance parameters between several areas of the conveyor;

[0112] - Fig. 12 schematically illustrates the field of view, with the determined basic region of interest of figure 11, showing a difference in circumstance parameters in a limited of the conveyor with respect to the rest of the conveyor; - Fig. 14 schematically illustrates a detection of a changed circumstance in a possible monitoring system according to the invention;

[0113] - Fig. 15 schematically shows the determining of a detection sub-area in a possible monitoring system according to the invention with respect to a normal production mode;

[0114] - Fig. 16 shows an embodiment of a potato planter according to the invention in perspective view;

[0115] - Fig. 17 schematically illustrates with part of the potato planter of figure 16 with the bunker, conveying means and a planting element, how this planter can be monitored with a method according to the invention.

[0116] Figure 1 schematically illustrates a possible embodiment of a monitoring system according to the invention, implemented in a root crop harvester (1). Such a root crop harvester (1) comprises a crop-digger section with e.g. one or more harvesting shares, which are not illustrated. A sieving unit comprises sieving conveyors (3), haulm hooks (10), a feeding conveyor (4), a hedgehog (5) and axial rollers (6). This sieving unit is positioned downstream of the harvesting shares fortransporting root crops towards the ring elevator (7) and in the meantime, to sieve dirt from the harvested root crops. With the elevator (7), the root crops are then lifted towards a discharge conveyor (8) onto a transfer elevator (9). The feeding conveyor (4) preferably does not have a sieve function when passing through the ring elevator (7).

[0117] Other configurations are of course conceivable wherein e.g. the root crop harvester (1) is provided with a bunker and / or without axial rollers (6) and / or with haulm rollers and / or without hedgehog (5) and / or with the ring elevator (7) being positioned fully at the back of the root crop harvester (1) and / or without such ring elevator (7), and / or with additional or less conveyors (3, 4, 8, 9) and / or other transfer means etc.

[0118] The parts of the root crop harvester (1) can be of any known design.

[0119] In the embodiment as illustrated, the harvester (1) comprises two rows. Analogously, harvesters with one or more rows can be worked out. The illustrated harvester (1) is a self-propelled harvester (1). Alternatively, a harvester (1) according to this invention could be worked out as a harvester which is pulled by a tractor.

[0120] The in figure 1 illustrated harvester (1) is a potato harvester. The invention is however equally applicable to other type of root crop harvesters, such as e.g. harvesters (1) for harvesting onions or red beets or carrots, etc. In a carrot harvester e.g. similar monitoring with respect to the hedgehog can be implemented. Additionally and / or alternatively it is e.g. possible to implement a monitoring system for setting the harvesting depth depending on e.g. the height of haulm to be cut off and / or a monitoring system for adjusting the cleaning intensity depending on the number of roots that are excessively cut off by the cleaning brushes.

[0121] Figures 16-17 illustrate a possible embodiment of a monitoring system according to the invention, implemented in a planter (1). Other planters with analogue singulation and planting of seedlings, such as planters for tulip bulbs or onions or garlic can be worked out analogously.

[0122] The illustrated planter (1) comprises a drawbar for connecting this planter (1) to a tractor. Alternatively, a planter (1) according to this invention could be worked out as a self-propelled planter or could be lifted by a tractor.

[0123] Each planter (1) comprises a bunker (80) for keeping potatoes to be planted. In order to plant potatoes, this bunker (80) is filled with potatoes to be planted.

[0124] Conveyors (4) are provided as conveying means (4) in the planter (1) to feed the potatoes to be planted in a feed direction from the bunker (80) to one or more planting elements (81) of the planter (1). Alternatively or additionally such conveying means could e.g. comprise an inclined plate and / or a moving floor, etc.

[0125] The general structure and working of such planters (1) is known. In the embodiment as illustrated, the planter (1) is a cup planter and comprises four cup elevators as said planting elements (81). Analogously, cup planters with more or less cup elevators can be worked out.

[0126] Further alternative embodiments can be belt planters comprising one or more string beds as said planting elements (81).

[0127] The potatoes are fed by a conveyor (4) through a feed opening (82) allowing a quantity of potatoes to pass from the bunker (80) to the planting element (81). In the illustrated embodiments, the feed opening (82) is delimited at the bottom by this conveyor (4) and at the top by a delimiter (83) which is provided as a height-adjustable and plateshaped slide (83).

[0128] In alternative embodiments, the top side of the feed opening (82) could be fixed and the height of the conveyor (4) could be adjustable, such that the conveyor (4) is provided as an adjustable delimiter of the feed opening (82). In further alternative embodiments, adjustable delimiters could be provided at the side of the feed opening (82). A combination of such delimiters can be provided. Several such delimiters may be positioned side by side, possibly with space between them.

[0129] In further alternative embodiments, machines according to the invention are mutatis mutandis e.g. cleaners or graders with similar possible settings and similar corresponding possible monitoring systems according to the invention.

[0130] In figure 1, the harvester (1) is provided with different cameras (11) as image capturing units:

[0131] 2 cameras (11) are positioned above the first sieving conveyor (3);

[0132] 2 cameras (11) are positioned above the third sieving conveyor (3);

[0133] 2 cameras (11) are positioned above the feeding conveyor (4);

[0134] 1 camera (11) is positioned above the hedgehog (5).

[0135] A camera (11) can also be positioned above the reading table (not illustrated). The illustrated planters (1) illustrated in figures 16-17 also comprise several cameras (11) for recording feed parameter (85, 86) of the feed of potatoes: all cameras (11) are provided for measuring a volume parameter (85) of the volume of potatoes fed from the bunker (80) towards the planting element (81): o a first camera (11) forms an image of potatoes just before the slide (83) in the bunker (80) and based thereon forms an estimate of the volume of potatoes present; o seen in the feed direction, a second camera (11) forms an image of potatoes behind the slide (83), in an intermediate buffer of potatoes where the cups (87) of the cup elevator of planting element (81) take out potatoes and based thereon forms an estimate of the volume of potatoes present in this area around the connection of the conveyor (4) to the planting element (81);

[0136] - the second camera (11) also forms an image of potatoes in the ascending part of the cup elevator and is also provided for based thereon determining a quantity parameter (86) of the amount of potatoes per cup (87) this seen from the front of the cup (18).

[0137] Some of these cameras (11) constitute an image capturing unit as such, whereas others together can be part of an image capturing unit.

[0138] Preferably at least 1 camera (11) is provided per said module (3, 4, 5). 2 cameras (11) can be placed above a said module (3, 4, 5) e.g. to provide an operator with a more pleasant view. If desired, even more cameras (11) could be provided per module (3, 4, 5). Alternatively, it is also possible to provide 1 camera (11) for several conveyors (3, 4, 5) next to each other. Such conveyors (3, 4, 5) are then preferably driven at a same speed. It is however also possible to split analysis of the obtained images (15) for conveyors (3, 4, 5) which are set up next to each other but which are running at different speeds. E.g. conveyors which fill an elevator evenly can be controlled separately.

[0139] With fewer cameras (11) related costs will be reduced. With multiple cameras (11) complexity of analysis of corresponding images (15) increases. In fact, the machine (1) can be provided with as many cameras (11) as desired. In practice, for a harvester (1) or planter (1) preferably e.g. 8 to 18 cameras (11) can be implemented standardly and a customer could request more cameras (11) as an option. Not all cameras (11) will then necessarily be used for monitoring, some might only be present to provide an operator with a desired view, e.g. for driving, such as a reverse camera or a haulm topper camera, etc.

[0140] Where a camera (11) is provided, the location thereof is preferably as central as possible on (part of) the respective module (3, 4, 5) above which it is placed.

[0141] The machine (1) is now according to the invention provided with a monitoring system (34), which is provided for monitoring the operation of the root crop harvester (1), this e.g. using a classic machine vision system and / or e.g. applying artificial intelligence, this e.g. with one or more neural network algorithms (28) provided for applying segmentation.

[0142] The monitoring system (34) is preferably provided so that it can be set how many cameras (11) are present so that the one or more neural network algorithms (28) or more classic machine vision system will process images (15) taking this setting into account. The monitoring system (34) can be provided automatically to detect which cameras (11) are present and / or an operator could set which cameras (11) are to be taken into account using e.g. a user interface (13). It can then preferably be set which camera (11) is present at which position in the machine (1), this preferably per module (3, 4, 5) and possibly also within said module (3, 4, 5).

[0143] The one or more cameras (11) could be supplemented with other image capturing units (11). In addition to one or more image capturing units (11) the harvester (1) is preferably provided with additional sensors (26), such as one or more pressure sensors and / or one or more inclination sensors and / or one or more speed sensors and / or one or more valve setting sensors, . . . Thus, preferably a pressure sensor is provided on the first sieving conveyor (3). The filling of the ring elevator (7) could also be monitored using an image capturing unit or using a pressure sensor. Speed sensors for monitoring the speed of the conveyors could further be provided. It is also e.g. possible to provide a moisture sensor. A planter (1) can furthermore be provided with one or more additional sensors as described in EP7298882A1.

[0144] Furthermore as shown in figures 1 and 17, the monitoring system (34) within the machine (1) is further preferably provided with a user interface (13) within the cabin (2) which allows the operator to make settings and / or on which obtained images (15) and possibly other sensor output data (27) can be shown to the operator and / or on which a generated operating parameter signal (33) can be presented to the operator, etc. Preferably, the control system (12) within the harvester (1) comprises one or more such user interfaces (13). Such user interfaces (13) can e.g. comprise a monitor and / or a touch-screen and / or a keyboard and / or a dial and / or voice activation, etc. Such user interfaces (13) can be provided in the cabin (2) of the harvester (1), but could also comprise mobile devices and / or remote devices.

[0145] The monitoring system (34) can further be provided with a network connection (14) which allows access to services, data storage and / or applications e.g. for obtaining settings (37) via the internet. In certain embodiments of monitoring systems (34) according to the invention, there could also be no network connection (14) or multiple network connections (14) present within the monitoring system (34).

[0146] The monitoring system (34) is further provided with a storage unit with one or more data banks which can store the data within the monitoring system (34) such as the sensor output data (15, 27, 37) and / or the current operation actions and / or settings (38) with a user interface (13), ... These data banks can comprise any type of means for storing data, e.g. flash drives, SD cards, hard drives, . . .

[0147] The one or more user interfaces (13) and / or the one or more network connections (14) and / or the one or more data banks are preferably adapted to be connectable to a data connection within monitoring system (34) such as a CAN-bus connection and / or an Ethernet connection. The monitoring systems (34) shown in figures 1 and 17 further comprise a local control unit (12) for executing the one or more neural network algorithms (28) or more classic machine vision system. Preferably, the monitoring system (34) comprises one or more such local control units (12), also known as a so-called edge device. The one or more local control units (12) are preferably adapted to be connectable to a data connection within the monitoring system (34) such as a CAN-bus connection and / or an Ethernet connection. Each local control unit (12) is preferably specialised in high performance for neural network applications, and preferably comprises a processing unit which contains a computer circuit that includes good CPU performance, good GPU performance and high-speed memory connections. Such a processing unit preferentially also includes one or more hardware and / or software GPU accelerators. In specific embodiments, such as e.g. described in EP7298882A1 for a planter (1) a processing unit can be integrated within an image capturing unit (11). Preferentially such a processing unit is a local processing unit which is provided within the machine (1).

[0148] For monitoring of the operation of the machine (1), images (15) are obtained with one or more of said possible cameras (11) in the machine (1). More specifically, e.g. 2 images (15) per second (this is preferably settable) can be taken by each camera (11) so as not to overload the local control unit (12) which is provided for processing said images (15).

[0149] As illustrated in figure 4, the images (15) can more specifically be loaded in one or more neural network algorithms (28) on one or more local control units (12). Each neural network algorithm (28) can be executed for detecting at least part of the root crop and / or possibly other harvested material (16, 17, 18) in the images (15) and possibly for detecting at least part of the conveying means (19).

[0150] Such a control unit (12) can be integrated within the respective camera (11) but is preferably provided in the cabin (2). More specifically a NVIDIA Jetson can be chosen as such control unit (12). In a normal production mode (70) the images (15) taken by the cameras (11) can in a step a be captured and pre-processed and in a step b be analysed with a material detection unit (39) for detecting at least part of the root crop (16), haulm (17), clods (18), conveying means (19) and other parts of the harvester (1) and in a step c possibly root crop parameters (29, 85, 86) and / or haulm parameters (30), clod parameters (31) and by-product parameters (32) can be determined, using a parameter determining unit (35) and in a further step possibly at least one operating parameter signal (33) can be generated, using an operating parameter signal determining unit (36), and at least one corresponding operating parameter (29, 30, 31, 32) can be set e.g. as described in patent applications PCT / IB2024 / 061722, EP24214856.7 and EP4298882A1 of patentee, which are hereby more specifically incorporated by reference.

[0151] Possible root crop parameter (29) can e.g. be an amount of detected root crops (86), a size, such as e.g. an average size of the detected root crops (16), an estimated volume (85) of the root crops (16), a colour of root crops (16), a speed or acceleration of the root crops (16), the presence of root crops (16) on a specific area of the conveying means, ...

[0152] Possible haulm parameters (30) can e.g. be an amount of haulm (17), an estimated volume, the presence of haulm (17) on a specific area of the conveying means, . . .

[0153] Possible soil parameters (31) can e.g. be an amount of soil (31), an estimated volume, the presence of soil (31) on a specific area of the conveying means, ...

[0154] Possible further by-product parameters (32) can e.g. be an amount of by-products (32,), an estimated volume, the presence of by-products (32) on a specific area of the conveyor, ...

[0155] A possible operating parameter signal (33) can e.g. be used to set:

[0156] - the speed of conveying means (3, 4, 6, 7, 8, 9); agitators; an interval between sieving conveyors (3); a slope of a hedgehog (5); adjusting of the spindles; drop height; adjust differential bypasses of e.g. a so-called flexyclean; intensity, angle, position, pressure and / or speed of a haulm roller and in case of a blockage change direction of rotation thereof; adjustment of axial rollers; opening adjustment, angle, rotation direction and / or position of a cleaning unit; fan blower control and intensity; haulm topper intensity, speed and positioning; height of crop digging section,: delimiter (83).

[0157] One or more neural network algorithms (28, 28’) which can be used in this respect, can also be trained as described in patent applications PCT / IB2024 / 061722 and EP24214856.7 of patentee, for which the contents thereof is hereby incorporated by reference.

[0158] Images (15) taken by such cameras (11) are taken with a specific field of view (42). In figures 2 and 5, a possible image (15) composed of images captured with the two cameras (11) above a sieving conveyor (3) illustrated in figure 1 is schematically illustrated. In figure 3 and 6 a possible image captured with the single camera (11) above the hedgehog (5) illustrated in figure 1 is schematically illustrated. As illustrated in figures 2, 3, 5 and 6, such field of view (42) typically comprises parts of the machine (1) which are not relevant for the desired analysis, e.g. when the operator wishes to have a view on images (15) taken by such cameras which is not optimal for the required analysis.

[0159] In order to simplify the analysis, the monitoring system (34) is provided for determining a new detection sub-area (20’, 21’, 22’, 24’) within the specific field of view (42) in a step d, this for performing step b with this detection sub-area (20, 21, 22, 24). At least part of the root crop (16) and / or possible other harvested material (17, 18) - which are of relevance for generating root crop parameters (29, 85, 86) and / or haulm parameters (30), clod parameters (31) and / or by-product parameters (32) and possibly for generating at least one operating parameter signal (33) - are then detected in this detection sub-area (20, 21, 22, 24). As illustrated in figure 15, the computer-executable code of the monitoring system (34) is in this respect provided with a detection sub-area determining unit (50), which can e.g. be executed at initialisation of the machine (1) for performing step d.

[0160] Preferably this detection sub-area determining unit (50) is executed more than once. It will largely depend on the application how often this is desired. This can e.g. be each time the machine (1) is started and / or after a predetermined amount of time since a previous execution of step b and / or at initiation of an operator, using an input device (13). Alternatively and / or additionally, as illustrated in figure 15, the computerexecutable code can be provided with a changed circumstance detection unit (60) such that in normal operation mode (70) as soon as in step c a changed circumstance (65) is detected, step d is executed. Furthermore alternatively, step d can be performed each time step b is performed.

[0161] Each time a new detection sub-area (20’, 21’, 22’, 24’) is determined, it replaces a corresponding previously used detection sub-area (20’, 21’, 22’, 24’) for performing step b in normal production mode (70). This new detection sub-area (20’, 21’, 22’, 24’) is thereafter used as detection sub-area (20, 21, 22, 24) in one or more following steps b, until a further new detection sub-area (20’, 21’, 22’, 24’) is determined.

[0162] In figure 15, the monitoring system (34) comprises an evaluation unit (71) which in a step e continuously monitors whether a changed circumstance (65) is detected.

[0163] In some embodiments, changed circumstances (65) are detected as illustrated at the bottom part of figure 14. Figure 14 illustrates a more advanced way of determining changed circumstances (65) as explained later on, but the functioning of the changed circumstances detection unit (60) is similarly applicable for other possible changed circumstances (65). A circumstance parameter (62) can be determined, e.g. an inclination detected with an inclination sensor and it can then be evaluated by this changed circumstances detection unit (60) whether this circumstance parameter (62) deviates for a predetermined amount of time (41) with a predetermined threshold value (64) from an initial circumstance parameter (63). If this is the case, this can be considered as a changed circumstance (65) and the detected circumstance parameter (62) can be set as a new initial circumstance parameter (63). The predetermined amount of time (41) is in this respect monitored with a time module (40). The predetermined threshold value (64) can be a factory setting or can be provided settable by user input (38).

[0164] Alternatively and / or additionally, detected circumstances (65) can be detected e.g. based on user input (38) or on operating parameter signals (33), etc.

[0165] With the evaluation unit (71) in figure 15, in case no changed circumstance (65) is detected, normal production mode (70) continues with the already set detection subarea (20, 21, 22, 24). In case a changed circumstance (65) is detected, the evaluation unit (71) triggers the detection sub-area determining unit (50) to determine a new detection sub-area (20’, 21’, 22’, 24’) to pass on for further use in the normal production mode (70).

[0166] In a simple embodiment, the detection sub-area determining unit (50) is provided for directly generating such new detection sub-area (20’, 21’, 22’, 24’).

[0167] In this respect, the detection sub-area determining unit (50) can e.g. be provided for prompting an operator via the user interface (13) to set one or more new detection subareas (20’, 21’, 22’, 24’) of the specific field of view (42), in order to set the detection sub-areas (20, 21, 22, 24) such as e.g. illustrated in figures 2 and 3. Alternatively and / or additionally the monitoring system (34) can be provided to have such new detection sub-area (20’, 21’, 22’, 24’) automatically determined e.g. by a neural network algorithm.

[0168] In figure 2, an image (15) of the on which the harvested material (16, 17, 18) is conveyed can be shown to the operator via the user interface (13), who can use this user interface (13) to e.g. define a first detection sub-area (20), selecting an area of the sieving conveyor (3) - to be analysed by one or more corresponding neural network algorithms (28) or a more classic machine vision system - by selecting four corners (23) defining a quadrangle. Computer-executable instructions known in e.g. drawing programmes can be used in this respect. Instead of detecting such corners (23) other ways of defining any kind of forms as known in such drawing programmes can be used. Peripheral areas of the conveyor (3) which are not relevant for the further analysis are thereby excluded. A second detection sub-area (21) is analogously defined in figure 2, by selecting four corners (23) on the haulm hooks (10) above the sieving conveyor (3).

[0169] In figure 3, a first detection sub-area (22) is analogously defined, by selecting four comers (23) on the hedgehog (5). Furthermore in figure 3, a second detection sub-area (24) is defined, by setting a baseline (25) on the hedgehog (5) above which the image (15) is more specifically further to be analysed. This second detection sub-area (24) is thereby situated at the top of the hedgehog (5).

[0170] Instead of having these detection sub-areas (20, 21, 22, 24) selected by an operator, e.g. a neural network algorithm could be trained for recognising relevant detection subareas e.g. using bounding boxes or semantic segmentation, such that a new detection sub-area can be determined more frequently. E.g. a neural network algorithm can be trained to recognise the first detection sub-area (24) in figure 3, by training it to derive the hedgehog (5) from received images (15). With an automatic detection of the hedgehog (5) and an automatic setting of the inclination of a hedgehog (5) it can e.g. be desired to perform step d continuously, e.g. every 2 seconds.

[0171] Moreover, such cameras (11) will typically be adjustable by the operator and can pan and / or tilt and / or zoom in accordance with the preferences of the operator. With an automatic detection of detection sub-areas and adjustments of the camera (11) it can e.g. be desired to perform step d continuously or at least e.g. every 15 seconds, depending on the application.

[0172] In several embodiments, modules (3, 4, 5) above which such cameras (11) are placed, can be set in different positions. In such case, it is not necessary to have said detection sub-areas (20, 21, 22, 24) directly set for each possible setting of the camera (11) and / or the respective module (3, 4, 5). In such cases, as illustrated in figure 13, it is e.g. possible to choose a limited number of possible settings, corresponding with different fields of view (42i, . . ., 42n).

[0173] In a first such embodiment it is possible to derive the detection sub-area (20’, 21’, 22’, 24’) for a specific setting via interpolation of several detection sub-areas (20i-n, 21 i-n, 22i-n, 24i-n) determined for a limited number of possible settings. For a first said field of view (42i), a first said detection sub-area (20i, 211, 22i, 24i) is then set (e.g. a first polygon (22i) as in figure 3 or a first baseline (25i) as in figure 3) and for a second said field of view (422), a second said detection sub-area (202, 212, 222, 242) (e.g. a second polygon (222) as in figure 3 or a second baseline (252) as in figure 3) is set. Thereafter at each actual setting of the camera (11) and / or each actual setting of the respective module (3, 4, 5) - which constitute a changed circumstance (65) - the specific field of view (42’) is determined in relation to said first field of view (42i) and said second field of view (422). For this specific field of view (42’), the respective detection sub-area (20’, 21’, 22’, 24’) is then not set by an operator or determined by a tracking algorithm, but obtained in function of the set first detection sub-area (20i, 21i, 22i, 24i) and the set second detection sub-area (2O2, 2h, 222, 242). More specifically e.g. for the hedgehog (5), a first polygonal detection sub-area (22i) can be set with the hedgehog (5) set at a first inclination, a second polygonal detection subarea (222) can be set with the hedgehog (5) set at a second inclination and at each third inclination at which said hedgehog (5) is set, determining the specific field of view (42’) at which the images (15) are taken for analysis, the detection sub-area (22’) is further determined as a polygon by interpolation of the first polygonal detection subarea (22i) and the second polygonal detection sub-area (222), in accordance with the ratio of the third inclination to the first inclination and the second inclination. The operating parameter signal (33) for setting the inclination of the hedgehog (5) can in this respect e.g. be used as a measure for the inclination and therefore as a measure for the specific field of view (22’).

[0174] Instead of deriving the detection sub-area (22) by interpolation, it is e.g. also possible as illustrated in figure 13, to determine a limited number of possible settings (422-n-i) equally distributed between a minimal setting (42i) and a maximal setting (42n) and to subdivide all possible settings in corresponding zones (Zi-n). For these settings (42i-n) a corresponding detection sub-areas (20i-n, 21i-n, 22i-n, 24i-n) is then determined. As soon as thereafter the detected specific field of view (42’) remains for a predetermined amount of time within such zone (Zi-n), the corresponding detection sub-area (20’, 21’, 22’, 24’) is then selected as detection sub-area (20, 21, 22, 24). In embodiments where the machine (1) comprises several rollers (6), such as axial rollers (6) and / or haulm rollers, such rollers (6) can be positioned in the machine (1) such that they are moveable relative to each other. A camera (11) above such rollers (6) can be set in a fixed place with respect to a first one of these rollers (6), but in such a way that all rollers (6) remain within the field of view (2) of this camera for all possible intermediate distances between these rollers (6). In step d with the detection sub-area determining unit (50), the detection sub-area (20, 21, 22, 24) can for such embodiments be determined so as to encompass these rollers (6), this manually or automatically, detecting these rollers (6). Depending on how the rollers (6) were set with respect to each other at executing step d, after such detection sub-area (20, 21, 22, 24) is set, the last roller (6) which is positioned at the opposite side of the rollers (6) than said first one of these rollers (6), could then however move outside of such detection sub-area (20, 21, 22, 24) and / or could move within the detection sub-area (20, 21, 22, 24), such that an undesired area outside the rollers (6) is encompassed by the detection sub-area (20, 21, 22, 24). In order to adjust such detection sub-area, the images (15) can be analysed for monitoring these rollers (6). To the one hand it can be determined whether and to what extent a said roller (6) at least partially extends outside the detection sub-area (20, 21, 22, 24) and if said roller (6) during a predetermined amount of time (41) extends for more than a predetermined portion outside the detection sub-area (20, 21, 22, 24), e.g. for 5 centimetre during 5 seconds, the detection sub-area (20, 21, 22, 24) can be rescaled accordingly. To the other hand it can be determined whether and at what distance one of these rollers (6) is located from an outer edge of the detection sub-area (20, 21, 22, 24), e.g. for 5 centimetre during 5 seconds, and the detection sub-area (20, 21, 22, 24) can be rescaled accordingly.

[0175] In planters (1) such as the planter (1) illustrated in figures 16-17, seedlings (16) piling up at the connection of a conveying element (4) to the planting element (81) can move into a detection sub-area which is initially set for the second camera (11). The monitoring system (34) can then be provided for detecting a piling height (H) to which these seedlings (16) are piling up within the images (15) taking with this second camera (11). This piling height can then be compared with an initial piling height and if the detected piling height during a predetermined amount of time (41) deviates more than a threshold value from the initial piling height, e.g. for 5 centimetre during 15 seconds, the detection sub-area can be rescaled accordingly.

[0176] Similarly, if a camera is positioned at an angle to a sieving conveyor (3), it may happen that when a large number of potatoes (16) lands on this sieving conveyor (3), they accumulate to such an extent that it is desirable to adjust the detection sub-area (20,

[0177] 21, 22, 24) accordingly. Furthermore, it is e.g. possible to detect whether root crops (16) are located at the edge of a detection sub-area, which detection sub-area (20, 21,

[0178] 22, 24) can then be adjusted accordingly.

[0179] In a more advanced embodiment, the detection sub-area determining unit (50) can alternatively be provided for indirectly generating such new detection sub-area (20’, 21’, 22’, 24’), such that it is more easy to respond accurately to changed circumstances (65). The detection sub-area (20, 21, 22, 24) can then be more accurately set in different circumstances. In this respect, as illustrated in figure 15 with dashed lines, the detection sub-area determining unit (50) can be provided with a basic region of interest determining unit (45), a subdividing unit (47) and a comparison unit (49).

[0180] In the basic region of interest determining unit (45) in a step f first a basic region of interest (45) is determined similarly as described above for determining a detection sub-area (20, 21, 22, 24) which is directly determined. It will not be required to detect the basic region of interest (45) each time a changed circumstance (65) is detected and new detection sub-area (20’, 21’, 22’, 24’) is to be determined. At least this basic region of interest (45) is determined in an initialisation step. Thereafter, this basic region of interest (45) can e.g. be reset for specific changed circumstances (65) such as when an operator activates resetting thereof via user input (38) or in case a to be monitored reference element moves outside the basic region of interest (45) such as e.g. described above for a roller moving outside the detection sub-area (20, 21, 22, 24). A basic region of interest (45) will typically be a whole module of the machine (1) detected in the images (15), such as a conveyor (3, 4, 8, 9), or a larger part thereof. In figure 5 a top part of a sieving conveyor (3) is thus selected as basic region of interest (45), whereas in figure 6 the hedgehog (5) is selected.

[0181] After such basic region of interest (45) is determined, this basic region of interest (45) can then be subdivided with the subdividing unit (47) into multiple partial regions of interest (48i, ..., 48n). In this respect, the subdividing unit (47) can e.g. be provided for prompting an operator via the user interface (13) to set one or more marking elements (54) for defining a subdivision for defining the partial regions of interest (481, 482), as illustrated in figures 5 and 6. Alternatively and / or additionally a number of preset layouts for subdividing the basic region of interest (45) into partial regions of interest (481, . . ., 48n) could be presented to the operator for selecting a layout in which to subdivide the basic region of interest (45) automatically into partial regions of interest (48i, ..., 48n). For a quadrangle, an operator could alternatively e.g. set the number of columns and / or rows in which to subdivide the quadrangle equally. Furthermore alternatively and / or additionally, depending on the application, there could be a factory setting for an automatic subdivision of a basic region of interest (45) into multiple partial regions of interest (481 , . . . , 48n), etc.

[0182] Depending on the application it can be desired to have larger or smaller partial regions of interest (481 , 482) in order to be able to obtain meaningful corresponding detections therein. Such partial regions of interest (48i, 482) may not become too small and for this a minimum size for such partial region of interest (481, 482) can e.g. be predefined, this e.g. as a factory setting.

[0183] Thereafter, in a step g, as illustrated in figures 14 and 15, a circumstance parameter (62i, . . . , 62n) can be detected for each partial region of interest (481, . . . , 48n). In step d, these circumstance parameters (62i, ..., 62n) can then be compared and based on this comparison one or more of the corresponding partial regions of interest (481 , . . . , 48n) can be selected to compose the detection sub-area (20, 21, 22, 24).

[0184] As illustrated in figures 7-10 and 14, in a specific embodiment for detecting such circumstance parameter (62i, . . ., 62n) for each partial region of interest (48i, . . ., 48n), it is in a step h possible to detect at least part of the root crop (16’) and / or other harvested material (17’, 18’) in each of the partial regions of interest (48i, 48n) and in a step i to determine for each partial region of interest (48 i, . . 48n) a partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) associated with the corresponding at least part of the root crop (16’) and / or other harvested material (17’, 18’) detected in step h. The evaluation of the images (15) in these partial regions of interest (48i, ..., 48n) for detecting the at least part of the root crop (16’) and / or other harvested material (17’, 18’) can be done similarly using a similar material detection unit (39’) with similar neural network algorithms (28’) as the neural network algorithms (28) described in patent application PCT / IB2024 / 061722 of patentee. Thereafter the determination of the partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) can be done similarly using similar harvest parameter determining units (35’) as the harvest parameter determining units (35) described in patent applications PCT / IB2024 / 061722 of patentee. These detected partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) can then be used as circumstance parameters (62i, ..., 62n) which are compared in step g for selecting corresponding partial regions of interest (48i, ..., 48n) to compose the detection sub-area (20, 21, 22, 24).

[0185] For sieving conveyors (3) it is e.g. possible to set an operating parameter based on an operating parameter signal (33) which is determined based on a clod parameter (31). For longer sieving conveyors (3) such clod parameter (31) can however, depending on e.g. moisture conditions, become less relevant as illustrated in figures 7 and 8. In figure 7 it is illustrated how a clod parameter (31 ’) for the partial regions of interest (481-4) at the top of such sieving conveyor (3) can be reduced in a way with respect to a clod parameter (31’) for the partial regions of interest (48 -x) at the bottom of such sieving conveyor (3), such that a clod parameter (30) based on the whole basic region of interest (46) is representative for determining such corresponding operating parameter signal (33). In such case all of the partial regions of interest (48i-s) can be selected to constitute the detection sub-area. Depending on the circumstances it is however possible that a clod parameter (31’) for the partial regions of interest (481-4) at the top of such sieving conveyor (3) can be reduced in such a way with respect to a clod parameter (31’) for the partial regions of interest (48s.x) at the bottom of such sieving conveyor (3) that a clod parameter (30) based on the whole basic region of interest (46) is no longer representative for determining such corresponding operating parameter signal (33). In such case a clod parameter (31), which is only based on the partial regions of interest (481-4) at the top of the sieving conveyor (3), is more relevant for determining such corresponding operating parameter signal (33). In such case only the partial regions of interest (481-4) at the top of the sieving conveyor (3) are selected to constitute the detection sub-area.

[0186] Similarly, with machines (1) with different rows it is furthermore possible as illustrated in figure 9 that on sloped field the partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) used for setting of an operating parameter signal (33) becomes less relevant at a left side of the sieving conveyor, with respect to the right side of the sieving conveyor. In such case only the partial regions of interest (483-4, 7-s) are selected to constitute the detection sub-area. Alternatively, it is also possible to select all partial regions of interest (481-8), but to accord to the partial regions of interest (481-2, 5-6) at the left side a different weighing factor than to the partial regions of interest (483-4, 7-s) for differently weighing these partial regions of interest (481-8) when determining a partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) associated with the corresponding at least part of the root crop (16’) and / or other harvested material (17’, 18’) detected therein for determining the corresponding root crop parameter (16) and / or the corresponding other harvest parameter (17) in step c. Furthermore for machines (1) with different rows, as illustrated in figure 10, it could be derived from partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) that one or more rows are not in use, such that only partial regions of interest (483-4, 7-s) for the rows which are in use are selected to constitute the detection sub-area. On a sloped field it is more frequently desired to perform such evaluation for selecting a corresponding detection sub-area, this every 2 seconds up to every 15 seconds. More specifically, e.g. every 5 seconds it can be determined for which partial regions of interest (480 a partial root crop parameter (29’) representing the amount of potatoes therein amounts to less than 10% of the highest corresponding partial root crop parameter (29) detected for the other partial regions of interest (48 i.n). Such partial regions of interest (480 are then excluded from the detection sub-area.

[0187] As illustrated in figure 11 it is furthermore possible to derive a lighting parameter for each of the partial regions of interest (48i-n), showing that shadow on part of the partial regions of interest (48i.n) prevents a meaningful partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) to be determined for such region for using it to set a corresponding operating parameter signal (33). In such case only the partial regions of interest (480 without shadow therein are selected to constitute the detection sub-area. It is possible to add partial regions of interest (480 that are only partially in the shaded area to this detection sub-area but then to assign to these partial regions of interest (480 that are only partially in the shaded area a weighing factor and to take this weighing factor into account when partial root crop parameter (29’) and / or another partial harvest parameter (30’, 31’, 32’) for such partial regions of interest (480 are used together with the partial root crop parameter (29’) and / or the partial harvest parameter (30’, 31’, 32’) for the other partial regions of interest (480 in this detection sub-area for determining root crop parameter (29) and / or another partial harvest parameter (30, 31, 32).

[0188] In figure 12 it is furthermore illustrated that it is possible to derive an obstruction parameter, e.g. a parameter representing that at least part of the objects in the partial region of interest (48 i_n) is not moving. In this way, it is possible to detect fixed parts of the machine (1) within the basic region of interest (45) or dust, dirt or foliage on or in front of the camera (11). A partial region of interest (480 for which more than a certain percentage of its area is covered with not moving objects can then be excluded from the detection sub-area.

Claims

CLAIMS1. A method for monitoring the operation of a machine (1) for conveying root crop (16) and possibly other harvested material (17, 18), comprising:- step a) obtaining images (15) from an image capturing unit (11), of root crop (16) and possibly other harvested material (17, 18), conveyed in the machine (1), wherein the obtained images (15) are taken with a specific field of view (42) of the image capturing unit (11);- step b) detecting at least part of the root crop (16) and / or other harvested material (17, 18) in a detection sub-area (20, 21, 22, 24) of the images (15);- step c) determining one or more root crop parameters (29) and / or other harvest parameters (30, 31, 32) associated with the at least part of the root crop (16) and / or other harvested material (16, 17, 18) detected in step b; characterised in that the method comprises a step:- step d) determining for the obtained images (15) the detection sub-area (20, 21, 22, 24) within the specific field of view (42) for performing step b with this detection sub-area (20, 21, 22, 24).

2. A method according to claim 1, characterised in that the method comprises repeating steps a and b, wherein in a repeated step a, images (15) are obtained from the image capturing unit (11) with a same or a new specific field of view (42) of the image capturing unit (11) and wherein in step d a new detection sub-area (20’, 21’, 22’, 24’) is determined before repeating step b with the new detection sub-area (20’, 21’, 22’, 24’).

3. A method according to claim 2, characterised in that step d is performed after a predetermined amount of time since a previous execution of step d.

4. A method according to any of the preceding claims, characterised in that the method comprises- a step e) detecting changed circumstances (65); and that step d is performed if a changed circumstance (65) is detected in step e.

5. A method according to claim 4 when dependent on claim 2, characterised in that in step d the new detection sub-area (20’, 21’, 22’, 24’) is determined based on the changed circumstance (65) detected in step e.

6. A method according to claim 4 or 5, characterised in that step e comprises determining the specific field of view (42) of the image capturing unit (11), at which the images (15) are taken in the repeated step a, for detecting changed circumstances (65) so that a detected new specific field of view (42’) is a said changed circumstance (65) for performing step d.

7. A method according to claim 6 when dependent on claim 5, characterised in that a first detection sub-area (20i, 211, 22i, 24i) is determined at a first field of view (42i) of the image capturing unit (11), that a second detection sub-area (202, 2h, 222, 242) is determined at a second field of view (422) of the image capturing unit (11) and that in step d the new detection subarea (20’, 21’, 22’, 24’) is determined based on:-the specific field of view (42’) determined in step e;- said first detection sub-area (20i, 211, 22i, 24i); and- said second detection sub-area (2O2, 212, 222, 242).

8. A method according to claim 7, characterised in that the image capturing unit (11) is provided for obtaining images (15) of root crop and / or other harvested material (16, 17, 18) conveyed on a conveyor (3, 4, 5, 6, 7, 8, 9) of the machine (1), that the image capturing unit (11) and the conveyor (3, 4, 5, 6, 7, 8, 9) are arranged in an adjustable mutual position, that the firstdetection sub-area (20i, 211, 22i, 24i) is determined at a first mutual position and the second detection sub-area (202, 2h, 222, 242) is determined at a second mutual position and that in step e the mutual position of the image capturing unit (11) and the conveyor (3, 4, 5, 6, 7, 8, 9) is determined in order to determine the specific field of view (42).

9. A method according to one of claims 5 to 8, characterised in that for detecting changed circumstances (65) step e comprises detecting deviations within the detection sub-area (20, 21, 22, 24) of the obtained images (15) so that a detected deviation is a said changed circumstance (65) for performing step d.

10. A method according to any of the preceding claims, characterised in that for determining the detection sub-area in step d, the method comprises: a step f) determining a basic region of interest (45), which basic region of interest (45) is subdivided in multiple partial regions of interest (48i, ..., 48n); a step g) detecting for each partial region of interest (48i, ..., 48n) a circumstance parameter (62i, . . ., 62n); and in step d the circumstance parameters (62i, . . . , 62n) detected in step g are compared and based on this comparison one or more of the corresponding partial regions of interest (48i, ..., 48n) are selected to compose the detection sub-area (20, 21, 22, 24).

11. A method according to claim 10, characterised in that the method comprises: a step h) detecting at least part of the root crop (16’) and / or other harvested material (17’, 18’) in each of the partial regions of interest (48i, ..., 48n); a step i) for each partial region of interest (481, . . . , 48n) determining a partial root crop parameter (29’) and / or another partial harvestparameter (30’, 31’, 32’) associated with the corresponding at least part of the root crop (16’) and / or other harvested material (17’, 18’) detected in step h); and that the partial root crop parameters (29’) and / or the other partial harvest parameters (30’, 31’, 32’) determined in step i are used as circumstance parameters (62i, . . ., 62n) in step g and d.

12. A method according to claim 11, characterised in that the at least part of the root crop (16’) and / or other harvested material (17’, 18’) which are detected in step h and correspond with the one or more partial regions of interest (481 , . . . , 48n) which are selected in step d to compose the detection sub-area (20, 21, 22, 24) are used in step b for detecting the at least part of the root crop (16) and / or harvested material (17, 18) in the detection subarea (20, 21, 22, 24).

13. A method according to claim 11 or claim 12, characterised in that the partial root crop parameters (29’) and / or the other partial harvest parameters (30’, 31’, 32’) which are determined in step i and which correspond with the one or more partial regions of interest (48i, ..., 48n) selected in step d to compose the detection sub-area (20, 21, 22, 24) are used in step c to determine the corresponding root crop parameter (29) and / or the corresponding other harvest parameter (30, 31, 32).

14. A method according to claims 10, characterised in that the circumstance parameter (62i, . . ., 62n) is one of: a lighting parameter; an obstruction parameter.

15. A method according to one of claims 10 to 14 when dependent on claim 9, characterised in that step g is repeated continuously and that for step e, for each partial region of interest (48i, ..., 48n) the detected circumstanceparameter (62i, . . 62n) in step g is compared with an initial circumstance parameter (631, 63n) and if the detected circumstance parameter (62i,. . 62n) during a predetermined amount of time (41) deviates more than a threshold value (64) from the initial circumstance parameter (631, . . 63n), this is considered as a detection of a said changed circumstance (65) and the last detected circumstance parameter (62i, . . . , 62n) is set as a new initial circumstance parameter (63 i, . . 63n).

16. A method according to any of the preceding claims, characterised in that the machine (1) comprises one or more reference elements for determining a detection sub-area (20, 21, 22, 24), that the method comprises: a step j detecting the one or more reference elements in the obtained images (15); and that in step d the detection sub-area (20, 21, 22, 24) is determined based on the one or more reference elements detected in step j .

17. A method according to any of the preceding claims, characterised in that the machine (1) comprises several rollers (6), such as axial rollers (6) and / or haulm rollers, which rollers (6) are positioned in the machine (1) such that they are moveable relative to each other and that in step d the detection subarea (20, 21, 22, 24) is determined so as to encompass these rollers (6).

18. A method according to claim 17 when dependent on claim 16, characterised in that said rollers (6) are detected in step j as said reference elements.

19. A method according to claim 18, when dependent on claim 9, characterised in that step j is repeated continuously and that for step e it is determined whether and to what extent a said roller (6) at least partially extends outside the detection sub-area (20, 21, 22, 24) and if said roller (6) during a predetermined amount of time (41) extends for more than a predeterminedportion outside the detection sub-area (20, 21, 22, 24), this is considered as a detection of a said changed circumstance (65).

20. A method according to one of the preceding claims, characterised in that the image capturing unit (11) is provided for obtaining images (15) of root crop and possibly other harvested material (16, 17, 18) conveyed on a hedgehog (5) of the machine (1), and that in step d the detection sub-area (22, 24) is situated at the top of the hedgehog (5).

21. A method according to one of the preceding claims, characterised in that seedlings as said root crop (16), are fed by one or more conveying elements (4) of the machine (1) from a bunker (80) of the machine (1) towards a planting element (81) of the machine (1), in that the image capturing unit (11) is provided for obtaining images (15) of an area around a connection of the one or more conveying elements (4) to the planting element (81) and in that in step d the detection sub-area is situated on the planting element (81) above the connection.

22. A method according to claim 21, when dependent on claim 4, characterised in that for step e, a piling height to which seedlings are piling up on the one or more conveying elements (4) at the connection, is detected and compared with an initial piling height and if the detected piling height during a predetermined amount of time (41) deviates more than a threshold value from the initial piling height, this is considered as a detection of a said changed circumstance (65) and the last detected piling height is set as a new initial piling height.

23. A machine (1) for conveying root crop (16) and possibly other harvested material (17, 18), comprising a monitoring system (34) for monitoring the operation of the machine (1), the monitoring system (34) comprising:- an image capturing unit (11), for taking images (15) of root crop (16) and possibly other harvested material (17, 18) conveyed in the machine (1), with a specific field of view (42);- one or more processing units coupled to the image capturing unit (11) or part thereof, comprising computer-executable instructions for controlling the image capturing unit (11) for taking images (15) and for detecting at least part of the root crop (16) and / or other harvested material (17, 18) in a detection sub-area (20, 21, 22, 24) of the images (15) and for determining one or more root crop parameters (29) and / or other harvest parameters (30, 31, 32) associated with the at least part of the root crop (16) and / or other harvested material (17, 18); characterised in that the monitoring system (34) is configured for carrying out a method according to any of claims 1 to 22 and the computerexecutable instructions are therefore provided for determining for the obtained images (15) the detection sub-area (20, 21, 22, 24) within the specific field of view (42) for detecting the at least part of the root crop (16) and / or other harvested material (17, 18) with this detection sub-area (20, 21, 22, 24).

24. A machine (1) according to claim 23, characterised in that the monitoring system (34) comprises a user interface (13) coupled to the one or more processing units, for determining the detection sub-area (20, 21, 22, 24) based on user input (38) entered via the user interface (13).

25. A machine (1) according to claim 24, characterised in that the user interface (13) comprises a graphical user interface.

26. A machine (1) according to any of claims 23 to 25, characterised in that the image capturing unit (11) comprises a digital camera for taking the images (15).

27. Computer program, comprising computer-executable instructions which when loaded on the one or more processing units of the machine (1) of any of claims 23-26, cause the machine (1) of any of claims 23-26 to carry out the method steps of any of claims 1 to 22.

28. Computer-readable data carrier having stored there on the computer program of claim 27.

Images