Device, system and method for a sample analysis

Abstract

For carrying out an analysis of a sample (19) which has eukaryotic material (19) and / or a soil sample a device (10) has a detection apparatus (30). The detection apparatus (30) is configured for detecting at least one optical measurement value on the sample (19), wherein the optical measurement value can be evaluated in order to analyse the sample. The device (10) has a state checking apparatus (40) for checking a state of the detection apparatus (30). The state checking apparatus (40) is configured to check the state of at least one component (27, 31) of the detection apparatus (30) on the basis of images.

Description

[0001] DEVICE, SYSTEM AND METHOD FOR SAMPLE ANALYSIS

[0002] TECHNICAL FIELD

[0003] The invention relates to devices, systems, and methods that can be used in connection with the examination of eukaryotic material and / or a soil sample. The invention particularly relates to such devices, systems, and methods that can be used in agricultural engineering, for example, to examine a plant or plant-like material in order to detect a supply of nutrients and / or water and / or pathological conditions. The invention particularly relates to such devices, systems, and methods that can be used in crop plants.

[0004] BACKGROUND

[0005] The analysis of plants, other plant-like eukaryotes, and / or soil samples is of great importance. It serves to ensure adequate nutrient supply and / or to detect possible diseases.

[0006] Conventional methods for analyzing plant health have traditionally relied on human expert assessment. Techniques that use measuring devices to objectively and quantitatively analyze plant health are becoming increasingly important. For example, such techniques can be used to detect over- or under-supply of certain nutrients.

[0007] WO 2019 / 169434 A1, US 7 804 588 B2, US 11 320 307 B2 and DE 10 2018 103 509 B3 and EP 3 695 209 B1 disclose exemplary techniques for this purpose.

[0008] Devices configured for use in plant cultivation areas may experience damage and / or contamination during field use, which could negatively impact the validity of sample analysis results. DE 10 2018 103 509 B3 therefore proposes monitoring a window for damage or contamination based on an internal or external referencing of an optical spectrometer. Detecting damage or contamination based on a referencing of an optical spectrometer is complex. Furthermore, it cannot detect damage or contamination outside the field of view of the optical spectrometer.

[0009] Thus, there is still a need for devices, systems, and methods that reliably further reduce the risk of sample analysis falsification. SUMMARY

[0010] The invention is based on the object of providing improved devices, systems, and methods that can be used for sample analysis of a sample containing eukaryotic material (e.g., a plant or a plant component) and / or a soil sample. In particular, the invention is based on the object of providing devices, systems, and methods that offer improvements with regard to the detection of potential sources of error.

[0011] According to the invention, a device, a system, and a method are provided as defined in the independent claims. The dependent claims define preferred and advantageous embodiments.

[0012] According to one aspect, the invention relates to a device or a system for sample analysis of a sample comprising eukaryotic material and / or a soil sample. The device or the system comprises a detection device configured to detect at least one optical measurement variable on the sample, wherein the optical measurement variable can be evaluated for sample analysis. The device or the system comprises a status control device configured to monitor a status of the detection device, wherein the status control device comprises an image recording device for image-based status control of at least one component of the detection device.

[0013] The device provides various technical effects and advantages. The device enables image-based condition monitoring using an image recording device (e.g., a camera). This means that the detection of anomalies such as damage, contamination, and / or other deviations from the target condition is not limited to the field of view of the detection device used for sample analysis. The design of the device or system for performing sample analysis based on an optical measurement parameter enables the in-situ acquisition of the optical measurement parameter required for sample analysis using a mobile device.By combining a sample analysis based on an optical measurement variable with an image-based condition detection, with which deviations from the target state of the detection device can be detected, the quality of sample analysis results is improved in a reliable and efficient manner.

[0014] The image capture device may have a first field of view. The detection device may have a second field of view that is completely contained within the first field of view.

[0015] As a result, the condition control device can detect deviations from the target state within the second field of view (i.e. within the field of view of the detection device).

[0016] The first field of view can be larger than the second field of view. This allows the condition monitoring device to detect deviations from the target state even outside the second field of view (i.e., outside the field of view of the detection device).

[0017] The condition monitoring device can be configured to monitor the condition of the detection device purely based on images. The condition monitoring device can be configured to determine the condition of the detection device based on images and without using measurements acquired with a spectrometer or without using (internal or external) standards.

[0018] This allows condition monitoring to be carried out in a simple, efficient and robust manner.

[0019] The image recording device may comprise a camera configured to record an image of the at least one component.

[0020] This allows the image recording required for condition monitoring to be carried out in order to detect deviations of the recording device from the target state in an efficient and reliable manner.

[0021] The camera may have a near-field camera.

[0022] This allows the image recording required for condition monitoring to be carried out in order to detect deviations of the recording device from the target state in an efficient and reliable manner.

[0023] The detection device may comprise a housing and a measuring window arranged on the housing. The measuring window may have an outer surface for contact with the sample during detection of the at least one optical measurement variable, and an inner surface.

[0024] This allows the optical measurement parameter to be captured in a well-defined state, with the sample in contact with the measurement window. Capturing the optical measurement parameter under such consistent conditions improves the quality of sample analysis.

[0025] The image recording device can have a depth of field that covers at least the inner surface of the measuring window.

[0026] This allows the condition control device to detect deviations from the target state of the measuring window, at least on its inner surface, using images.

[0027] The image recording device can have a depth of field that covers at least the outer surface of the measuring window.

[0028] This allows the condition control device to detect deviations from the target state of the measuring window, at least on its outer surface, using images.

[0029] The image recording device may have a depth of field that covers at least a depth of field range from the inner surface to the outer surface of the measuring window.

[0030] This enables the condition control device to detect deviations from the target condition of the measuring window across its entire thickness using an image-based method. The condition control device is then able to detect foreign matter or foreign bodies on the outer or inner surface of the

[0031] Due to the arrangement of the measuring window and / or its direct contact with the sample during use, there is a particular risk that the actual state of the measuring window deviates from the target state. The condition control device can detect this and thus reduce or eliminate a particularly relevant source of uncertainty for sample analysis.

[0032] The image recording device may have a depth of field that extends beyond the outer surface of the measuring window, in particular by at least 1 mm or at least 2 mm.

[0033] This allows the condition monitoring system to detect deviations from the target state even beyond the measurement window. These ranges particularly take into account the sample-specific characteristics of eukaryotic samples, such as plant leaves.

[0034] The detection device may comprise a sample holding device configured to position the sample at the measurement window. The sample holding device may be configured as a sample clamping device for clamping the sample against the measurement window.

[0035] This allows the sample to be positioned securely, increasing the quality of sample analysis.

[0036] The image recording device may have a depth of field that covers at least a surface of the sample holding device that faces the measuring window.

[0037] This allows the condition control device to detect deviations from the target condition of the surface of the sample holding device facing the measuring window.

[0038] The image recording device can have a depth of field which has at least a depth of field range from the inner surface of the measuring window to the surface of the sample holding device facing the measuring window.

[0039] This allows the condition monitoring device to detect deviations from the actual state to the target state for both the sample-contacting surface of the sample holder and the measurement window. These areas of the acquisition device are particularly susceptible to deviations from the target state, allowing the condition monitoring device to check those components of the acquisition device where the risk of negatively impacting the quality of the sample analysis is particularly high.

[0040] The surface of the sample holder facing the measurement window can have an optical functional surface, for example, a reflection surface, to enable the detection of the optical measurement variable, for example, in a reflection or transreflection arrangement. The optical functional surface can have an absorbent coating (for example, a black coating) to enable the detection of the optical measurement variable in a reflection arrangement. The image recording device can have a depth of field that extends at least to the optical functional surface.

[0041] This allows the condition monitoring device to detect deviations from the actual state to the target state for the optical functional surface of the sample holding device. This area of ​​the detection device is particularly susceptible to deviations from the target state, so the risk of negatively affecting the quality of the sample analysis can be particularly effectively reduced.

[0042] The image-based condition control can include image-based detection of a deviation of the measurement window from a target state.

[0043] This allows the condition control system to detect deviations from the actual to the target state for the measurement window that is most susceptible to such deviations. The risk of negatively influencing the quality of the sample analysis can be particularly effectively reduced.

[0044] The image recording device can be configured to capture at least one image of an area that is offset relative to the measuring window along an optical axis of the capture device.

[0045] This allows the condition monitoring device to detect deviations from the actual state to the target state for components of the detection device that are offset relative to the measurement window. Alternatively or additionally, this enables image-based readout of a sensor located on or in the detection device to detect deviations from the actual state to the target state.

[0046] The image-based condition control may comprise one, several, or all of the following condition controls based on the at least one image: detection of foreign substances in or on the detection device; detection of a shift between actual and target positions of the components of the detection device; checking at least one parameter based on an image-based readout of at least one sensor of the detection device configured to detect the at least one parameter; condition control of a sample holding device for holding the sample; detection of spatial and / or temporal inhomogeneities of an illumination source of the detection device.

[0047] This allows the condition monitoring system to detect deviations from the target condition of the detection system. Control of the detection system and / or evaluation of the optical measurement value for sample analysis can be performed independently of the image-based condition monitoring system.

[0048] The image recording device can be mounted in the housing. This allows the functions of capturing the optical measurement variable and capturing at least one image for condition monitoring to be performed in a particularly simple and fail-safe manner, since the image recording device of the condition monitoring device is mounted in the same housing as the detection device.

[0049] The condition monitoring device can have at least one evaluation circuit for evaluating images captured by the image recording device and at least one memory. The at least one evaluation circuit can be configured to perform an evaluation of the images captured by the image recording device for image-based condition monitoring.

[0050] This allows deviations of the recording device from the target state to be detected by evaluating the recorded images.

[0051] Parameters influencing the evaluation for the image-based condition control can be stored in the memory and retrieved by the at least one evaluation circuit and used for the evaluation.

[0052] This allows the evaluation to be carried out automatically by the condition control device in order to detect deviations from the target state of the recording device in the image-based condition control.

[0053] The parameters can include data-driven parameters, for example model parameters of an artificial intelligence model (Kl).

[0054] This allows image-based condition monitoring to be carried out using evaluation techniques based on objective criteria, for example using data-driven, trained evaluation techniques.

[0055] The evaluation may comprise a comparison of an image acquired on a sample when used with at least one comparison image acquired with the image recording device.

[0056] This allows the evaluation to be carried out automatically by the condition control device in order to detect deviations from the target state of the recording device in the image-based condition control.

[0057] The at least one comparison image may comprise a reference image acquired on a reference object and stored in the memory and / or a sample image acquired on another sample when used.

[0058] This allows the evaluation to be carried out automatically by the condition control device using image processing techniques to detect changes in the detection device.

[0059] The condition monitoring device can have one or more controllable light sources that can be selectively activated for image acquisition by the image recording device. This allows for image-based condition monitoring using controlled illumination. The detection of any deviation of the detection device from the target state is improved. This increases the quality of sample analysis.

[0060] The condition control device may comprise a plurality of controllable light sources that can be selectively activated for image acquisition by the image acquisition device in order to carry out the image-based condition control on the basis of images acquired with the image acquisition device under different illumination conditions (e.g., different illumination angles).

[0061] This can further improve the detection of deviations of the detection device from the target state. In particular, deviations from the target state can be detected more reliably by varying the lighting conditions during image acquisition for image-based condition monitoring.

[0062] A plurality of controllable light sources can be arranged distributed around an optical axis of the detection device. Alternatively or additionally, the state control device can be configured for time-sequential activation of the plurality of controllable light sources.

[0063] This can further improve the detection of deviations of the detection device from the target state. In particular, deviations from the target state can be detected more reliably by varying the lighting conditions during image acquisition for image-based condition monitoring.

[0064] The condition control device can be configured to carry out the image-based condition control for a plurality of components of the detection device in a sequence in which an image-based condition control for a component positioned closer to the image recording device takes place before the image-based condition control for a component positioned further away from the image recording device.

[0065] This allows different components to be systematically checked for any deviations from the target condition. A condition check in the sequence described above avoids the risk of a component located farther from the image capture device being falsely identified as faulty or dirty, even though the deviation from the target condition actually exists in a component located closer to the image capture device. In other words, a systematic check ensures that a component located farther from the image capture device is checked if it is ensured that a component located closer to the image capture device is assessed as non-defective.

[0066] The condition control device may be configured to detect a change in the

[0067] To cause the detection device to perform the image-based condition check for the multiple components of the detection device in such a sequence. For example, the condition check device can be configured to image-based check the measuring window for any deviations from the target state while the sample holder is positioned at a greater distance from the measuring window, and subsequently to check the sample holder for any deviations from the target state when it is positioned at a closer distance from the measuring window or in contact with the measuring window.

[0068] This allows different components to be checked systematically for any deviations from the target state.

[0069] The image recording device can have a fixed (i.e. non-variably adjustable) aperture and / or a fixed (i.e. non-variably adjustable) focusing optics.

[0070] This allows the condition check to be performed with a simple configuration. Furthermore, the aperture can be selected so small that the depth of field of the image recording device (e.g., a camera) extends at least over a depth of field range from the inner surface to the outer surface of the measuring window.

[0071] The condition control device can be configured to influence the detection of at least one optical measurement variable depending on the image-based condition control.

[0072] This allows a result of the image-based condition control to be used for an automatic control process.

[0073] The condition control device can alternatively or additionally be configured to output a result of the image-based condition control via a human-machine interface depending on the image-based condition control.

[0074] This allows the human-machine interface to be controlled in such a way that information about any deviations from the target state and / or recommended actions to correct a deviation from the target state are fed back to an operator while the device or system is in field use.

[0075] The condition control device can alternatively or additionally be configured to mark the at least one optical measurement variable or a sample analysis result derived therefrom based on the result of the image-based condition control, depending on the image-based condition control.

[0076] This allows a result of the image-based condition check to be used to provide optical measurement variables and / or derived sample analysis results with information that provides an indication of possible impairments in terms of reliability.

[0077] The condition monitoring device can be configured to perform the image-based condition monitoring based on at least one trigger criterion. The trigger criterion can include an operator input, a time-based criterion, and / or another trigger criterion (e.g., based on the number of samples checked since the last condition monitoring).

[0078] This allows image-based condition monitoring to be carried out continuously, optionally automatically, during field operation of the device or system.

[0079] The device or system can comprise at least one processing circuit configured to perform the sample analysis based on the detected optical measurement variable. The processing circuit can also be provided remotely from the device comprising the detection device and configured to establish a communication connection with the device via a communication interface.

[0080] This allows the evaluation of the recorded optical measurement variable to be carried out locally in the device having the recording device or remotely from it.

[0081] The detection device may comprise a spectral analysis detection device. The spectral analysis detection device may be configured to perform the optical measurement for several different wavelengths depending on the image-based condition monitoring.

[0082] This allows spectral analysis information to be acquired based on image-based condition monitoring. This can be achieved through user guidance during the acquisition of the optical measurement value or through automatic control of the spectral analysis acquisition device. Human error sources are reduced.

[0083] The device or system may include a device for reducing ambient light influences. The device for reducing ambient light influences may include a screen provided on the housing. The screen may protrude at least partially outward around the measurement window.

[0084] This facilitates the acquisition of the optical measurement and / or image-based condition monitoring under consistent conditions.

[0085] The device or system may be configured to detect the optical measurement quantity while the measurement window is in contact with the sample.

[0086] This makes it easier to record the optical measurement under consistent conditions.

[0087] The device can be designed as a movable device.

[0088] This allows for use in the field on living plants or other samples.

[0089] The device can be designed as a manually held device.

[0090] This facilitates use in the field on living plants or other samples without necessarily requiring vehicle access to the sample. The device can be attached to a vehicle, in particular an agricultural vehicle. Accordingly, according to one embodiment, a vehicle with the device according to the invention mounted thereon is provided.

[0091] This facilitates use in the field on living plants or other samples using a vehicle.

[0092] The device can be attached to a robot. Accordingly, according to one embodiment, a robot is provided that has the device according to the invention. The robot can be movably mounted, for example, on a rail system or in another manner.

[0093] This makes it easier to use in greenhouses, for example.

[0094] The device can be attached to a flying object. Accordingly, according to one embodiment, a flying object is provided that has the device according to the invention.

[0095] This facilitates use in field operations using the flying object.

[0096] The vehicle, robot or flying object may comprise a receiver for receiving remote control signals and at least one actuator for moving the vehicle, robot or flying object in response to the remote control signals for positioning the device.

[0097] This enables control from a control position remote from the device.

[0098] The vehicle, robot or flying object may have a controller configured to move the vehicle, robot or flying object autonomously or semi-autonomously to position the device.

[0099] This achieves a higher degree of automation.

[0100] The vehicle, robot, or flying object may have a gripper arm for gripping and positioning the sample relative to the device. The gripper arm may be configured to guide the sample to the measurement window.

[0101] This allows even parts of the plant that are difficult to access to be correctly positioned in relation to the device.

[0102] According to a further aspect or embodiment, a method for analyzing a sample comprising eukaryotic material and / or a soil sample is provided. The method comprises: detecting at least one optical measurement variable on the sample by a detection device, wherein the optical measurement variable can be evaluated for the condition analysis of the sample; and monitoring a condition of the detection device by a condition control device, which has an image recording device for image-based condition control of at least one component of the detection device.

[0103] The method provides various technical effects and advantages. The method enables image-based condition monitoring using an image recording device (e.g., a camera). This means that the detection of anomalies such as damage, contamination, and / or other deviations from the target condition is not limited to the field of view of the detection device used for sample analysis. Performing sample analysis based on an optical measurement parameter enables the acquisition of the optical measurement parameter required for sample analysis to be performed in situ. Combining sample analysis based on an optical measurement parameter with image-based condition detection, which can detect deviations from the target condition of the detection device, improves the quality of sample analysis results in a reliable and efficient manner.

[0104] The method may be performed automatically by the device or system according to one aspect or embodiment.

[0105] Optional features of the method and the technical effects achieved thereby correspond to the features and effects disclosed with reference to the device and the system.

[0106] According to a further aspect of the invention, machine-readable instruction code is provided which, when executed by a programmable circuit, carries out the method according to one aspect or embodiment of the invention.

[0107] According to a further aspect of the invention, a storage medium is provided with machine-readable instruction code stored thereon, which, when executed by a programmable circuit, carries out the method according to one aspect or embodiment of the invention.

[0108] The devices, systems, methods, and system components can be used in various fields. This includes, but is not limited to, the condition analysis of crops or plants in agricultural engineering. The devices, systems, methods, and system components according to exemplary embodiments can be used in agricultural engineering, for example, for investigations on plants or plant-like eukaryotes (such as algae) or for the examination of soil samples.

[0109] BRIEF DESCRIPTION OF THE CHARACTERS

[0110] Embodiments of the invention are described with reference to the figures. In the figures, similar or identical reference numerals designate elements with similar or identical design and / or function.

[0111] Fig. 1 is a schematic representation of a device for sample analysis.

[0112] Fig. 2 shows a schematic representation of the device.

[0113] Fig. 3 is a schematic representation of a deviation of a measurement window from the target state to explain the operation of the device. Fig. 4 is a schematic representation of a deviation of a measurement window from the target state to explain the operation of the device.

[0114] Fig. 5 is a schematic representation of a deviation of a sample holding device from the desired state to explain the operation of the device.

[0115] Fig. 6 is a schematic representation of a depth of field of the device.

[0116] Fig. 7 is a schematic representation of a depth of field of the device.

[0117] Fig. 8 is a schematic representation of fields of view of an image pickup device and a detection device of the device.

[0118] Fig. 9 is a flowchart of a method.

[0119] Fig. 10 illustrates an implementation of an image-based condition control.

[0120] Fig. 11 illustrates another implementation of image-based condition control.

[0121] Fig. 12 is a flowchart of a method.

[0122] Fig. 13 is a flowchart of a method.

[0123] Fig. 14 shows a schematic representation of the device.

[0124] Fig. 15 shows a schematic representation of a lighting arrangement of the device.

[0125] Fig. 16 shows a schematic representation of the device.

[0126] Fig. 17 shows a schematic representation of the device.

[0127] Fig. 18 shows a schematic representation of the device.

[0128] Fig. 19 shows a schematic representation of the device.

[0129] Fig. 20 shows a schematic representation of the device.

[0130] Fig. 21 is a flowchart of a method.

[0131] Fig. 22 is a flowchart of a method.

[0132] Fig. 23 shows a schematic representation of the device.

[0133] Fig. 24 shows a schematic representation of an agricultural vehicle having the device.

[0134] Fig. 25 shows a schematic representation of a robot having the device.

[0135] Fig. 26 shows a schematic representation of an aircraft having the device.

[0136] Fig. 27 is a flowchart of a method.

[0137] DETAILED DESCRIPTION OF EMBODIMENTS

[0138] Embodiments of the invention are described with reference to the figures. In the figures, similar or identical reference numerals designate elements with similar or identical design and / or function.

[0139] While embodiments are described in connection with a sample analysis, for example, a nutrient analysis of a crop or cultivated plant, the embodiments are not limited to this. The features of the embodiments can be combined with one another, unless this is expressly excluded in the following description.

[0140] A device, a system, and a method according to embodiments of the invention are configured to perform a sample analysis based on an optical measurement variable. The sample analysis can include a nutrient analysis, for example, a quantitative nutrient concentration analysis, a detection of deficiency or oversupply states, a detection of pathological conditions, or other sample analyses performed based on the optical measurement variable. The device, the system, or the method are configured such that, using a condition monitoring device, an image-based condition monitoring of the detection device configured to detect the optical measurement variable can be carried out. Deviations of one or more components of the detection device from a target state can thereby be detected in an image-based manner, and the risk of impairing the meaningfulness of the sample analysis can be reduced.

[0141] The detection device can be configured to detect spectral analytical information. For this purpose, the detection device can be configured to detect the optical measurement variable at a same spatial measurement range of the sample for multiple (two or more than two) wavelengths. The optical measurement variable detected by the detection device can, for example, comprise reflectivity, reflected light intensity, scattered light intensity, Raman scattered light intensity, fluorescent light intensity, phosphor light intensity, or other optical measurement variable detected for the multiple wavelengths, which can also be determined, for example, in a transmission arrangement (e.g., by detecting transmitted intensity as a function of wavelength) or a transflection arrangement.

[0142] The term "detection device" as used here encompasses a device that is capable of and configured to detect an optical measurement variable detected for multiple (two or more than two) wavelengths at the same measurement range of the sample. The detection device may comprise a spectrometer. However, this is not necessarily required. For example, as an alternative or in addition to using a spectrometer, the detection device may be configured to detect the optical measurement variable for the multiple wavelengths sequentially in time, for example by actively irradiating with different wavelengths and determining the intensity of the scattered or reflected light detected in each case. In a sample analysis that is also carried out image-based, the detection device may comprise a further camera.

[0143] The condition monitoring device has an image recording device, which can be configured as a near-field camera. The term "near-field camera," as used here, particularly encompasses an image recording device with near-field optics that allows images of components of the detection device (e.g., a measurement window and / or a sample holder) to be captured.

[0144] The term "device for sample analysis," as used here, refers to a device that can be used for such a sample analysis. The device is specifically configured to perform at least some, and advantageously all, of the measurements required for the condition analysis to capture the optical measurand. It is possible, but not mandatory, for the evaluation for sample analysis to also be performed by the device itself.

[0145] The term image-based condition control includes, in particular, a purely image-based condition control that can check the condition of the detection device without the use of measurements acquired with a spectrometer and without the use of (internal or external) standards.

[0146] The device, system, or method is configured for analyzing a sample that may include eukaryotic material and / or a soil sample. The eukaryotic material may, for example, include a plant part, in particular a leaf of a plant. The eukaryotic material may also include one or more plant-like eukaryotes, for example, algae.

[0147] Fig. 1 shows a device 10 for sample analysis of a sample 19. The sample 19 can comprise a eukaryotic material, for example a leaf of a plant or another plant part. The device 10 has a housing 20. The device 10 has a detection device 30. The detection device 30 is configured to detect an optical measurement variable on a sample 19 positioned at a measuring window 31 of the detection device 30. The detection device 30 can have a measuring system 32 configured to detect the optical measurement variable at at least two (i.e., two or more) different wavelengths. The measuring system 32 can, for example, have an illumination source 33 and a spectrometer 34. Other configurations are possible. For example, the measuring system 32 can be configured to perform an image-based status analysis (for example, a quantitative nutrient concentration determination) of the sample 19.In this case, the device 10 can have the design and mode of operation disclosed in DE 10 2023 113 716.6 of the applicant.

[0148] The detection device 30 has a sample holding device 27, which is configured to position the sample 19 at the measuring window 31 (for example, in contact with the measuring window 31) for detecting the optical measurement variable. The sample holding device 27 can be configured to be positioned in various positions relative to the measuring window 31. The sample holding device 31 can be arranged on the housing 20 via a mounting arrangement 26 such that the device 10 enables the positioning of the sample holding device 31 in different positions relative to the measuring window 31 and optionally enables the locking in at least one, optionally multiple positions, relative to the measuring window 31.

[0149] A surface of the sample holding device 27 facing the measuring window 31 can extend along, for example, substantially parallel to, an outer surface of the measuring window 31. The surface of the sample holding device 27 facing the measuring window 31 can be provided with an optical functional surface or configured as an optical functional surface, for example, to enable the detection of the optical measurement variable in a reflection and / or transflection arrangement.

[0150] The device 10 has at least one circuit 21, which can be configured to evaluate the optical measurement variable. Alternatively or additionally, the at least one circuit 21 can be configured to control data transmission via a data interface 24 based on the optical measurement variable. This enables further evaluation of the optical measurement variable in an evaluation device separate from the device 10 and / or logging of the optical measurement variable (for example, as a function of location and / or time) and / or the sample analysis results derived therefrom. The at least one circuit 21 can be configured to perform the evaluation using evaluation instructions stored non-volatilely in a memory 22.

[0151] The device 10 advantageously has a human-machine interface 25. The at least one circuit 21 can be configured to control the human-machine interface 25 based on a sample analysis result derived from the optical measurement variable. This allows the operator, when using the device 10 in the field, to receive feedback on the sample analysis result at the respective measurement location. The human-machine interface 25 can have an optical output device, for example a display device and / or a virtual reality (VR) and / or augmented reality (AR) output device. The device 10 can be configured to display sample analysis results as an overlay on a visually perceivable representation of the sample.

[0152] The device 10 has a condition monitoring device 40. The condition monitoring device 40 has an image recording device, which may have a camera 41 or may be configured as a camera 41. The camera 41 may be a near-field camera. The camera 41 may have near-field optics. The condition monitoring device 40 is configured to perform an image-based condition monitoring of one or more components of the detection device 30. The condition monitoring device 40 is configured to evaluate an image of the component(s) of the detection device 30, captured by the camera 41, for the image-based condition monitoring. The evaluation can be performed in different ways, as will be described in more detail below.For image-based condition monitoring, the at least one circuit 21 can be configured to perform an image analysis that can detect deviations from the target state of at least one component of the detection device 30. The device 10 can, in particular, be configured to detect deviations of the measuring window 31 and / or the surface of the sample holder 27 facing the measuring window from their respective target state. The device 10 can be configured to detect damage and / or contamination of the measuring window 31 and / or the surface of the sample holder 27 facing the measuring window based on the image.

[0153] The camera 41 can be coupled to a beam path of the detection device 30 via an optical coupling arrangement 29 in order to enable the camera 41 to capture at least one image of one or more components of the detection device 30.

[0154] The device 10 is configured to use the result of the image-based method in the acquisition and / or evaluation of the optical measurement variable. The image-based condition monitoring can be used by the device 10 in various ways to influence the acquisition, evaluation, or subsequent assessment of the optical measurement variable. In this way, the device 10 can, for example, control and / or support the acquisition and / or evaluation of the optical measurement variable to enable reliable sample analysis.

[0155] The device 10 can be configured to control the detection device 30 depending on the image-based condition monitoring. For example, detection of the optical measurement variable can be prevented or made dependent on a dedicated user input if the image-based condition monitoring indicates that at least one component deviates from the target state, which may impair the detection and / or evaluation of the optical measurement variable. Alternatively or additionally, the detection device 30 can be controlled such that the optical measurement variable is detected at a measuring point at which, based on the image-based condition monitoring, such detection is expedient with regard to sample analysis. The detection device 30 can be controlled by a detection device controller of the at least one circuit 21 of the device 10.The at least one circuit 21 may alternatively or additionally be configured to control the human-machine interface 25 to enable dedicated user confirmation depending on the image-based status control.

[0156] The device 10 can be configured to control the human-machine interface 25 of the device depending on the image-based status monitoring. For example, the human-machine interface 25 can be controlled to indicate visually, acoustically, and / or tactilely whether detecting the optical measurement variable is appropriate with regard to the state of the at least one component of the detection device 30. The human-machine interface 25 can have a display device that can be controlled by the at least one circuit 21 in such a way that it is displayed (for example, in the form of overlay(s)) which measurement points are suitable for detecting the optical measurement variable.

[0157] Alternatively or additionally, the device 10 can be configured to transmit the information via the data interface 24 for performing the sample analysis. The device 10 can use the image-based status check to selectively transmit the optical measurement variable only if the image-based status check indicates that the optical measurement variable is suitable for sample analysis.

[0158] The device 10 thus provides a device with which it can be checked whether a detection and / or evaluation of spectral analytical information with regard to the state of at least one component of the device 10 (in particular with regard to the state of the measuring window 31 and / or the surface of the sample holding device 27 facing the measuring window) can be expected to yield reliable results of a sample analysis based on the optical measurement variable.

[0159] To perform the various control and processing functions, the device 10 has at least one circuit 21. The at least one circuit 21 can comprise one or more integrated circuits to control the data interface 24, the human-machine interface 25, the capture device 30, and / or the camera 41. The one or more integrated circuits can, for example, comprise any one or any combination of the following circuits or circuit components: an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a processor (e.g., a GPU ("graphic processor unit"), CPU ("central processing unit"), or TPU ("tensor processor unit")), a controller, one or more quantum gates, a circuit for quantum information processing, and other integrated circuits.

[0160] The device 10 has a memory system 22. The memory system 22 can store machine-readable instruction code which, when executed by the at least one circuit 21, causes the functions and steps disclosed here to be carried out. If the evaluation of the image captured by the camera 41 for state control is carried out by the at least one circuit 21, i.e., locally in the device 10, the memory system 22 can also store parameters that enable an evaluation of the image captured by the camera 41 to determine whether at least one component of the device 10 exhibits a deviation from the desired state. The parameters stored in the memory system 22 for evaluation can, for example, include one or more of the following parameters: filter parameters; definition of decision boundaries of a support vector model; parameters of a trained machine learning model.Further examples are described in more detail with reference to the evaluation device or the evaluation system. The sample analysis based on the optical measurement variable can be carried out locally in the device 10 or in a computer system remote from the device 10. The sample analysis, which can include, for example, the determination of nutrient concentrations based on spectral analytical information and / or the image-based determination of nutrient concentrations, can be carried out using techniques as described, for example, in WO 2019 / 169434 A1, US 7 804 588 B2 and US 11 320 307 B2. The sample analysis can be carried out using principal component techniques, partial least squares regression (PLS) techniques, and / or"Partial Least Square") or other techniques that allow a conclusion to be drawn as to whether a nutrient concentration for at least one and advantageously several nutrients lies within a target range or outside a target range. The sample analysis can alternatively or additionally also be carried out using the techniques for image-based quantitative nutrient concentration determination disclosed in DE 102023 113716.6 of the applicant.

[0161] Fig. 2 shows a schematic representation of a device 10 in which the direction of an illumination 11 is tilted relative to the direction of the camera beam path 12. The device 10 can also comprise a spectrometer (for example, for more detailed measurements) whose beam path 13 is tilted relative to the illumination 11 and the camera beam path 12.

[0162] Fig. 3, Fig. 4 and Fig. 5 show exemplary components of the detection device 30 that exhibit deviations from the desired state. The state control device 40 can be configured to detect damage 51 to the measuring window 31 (Fig. 3). The damage 51 can, for example, comprise a scratch or crack. The state control device 40 can alternatively or additionally be configured to detect contamination 52, 53 of the measuring window 31 (Fig. 4). The contamination 52, 53 can, for example, comprise adhesion of a liquid 52, such as a water droplet, and / or adhesion of a foreign substance 53 (for example, soil or eukaryotic material remaining on the measuring window 31). The state control device 40 can alternatively or additionally be configured to detect contamination 54 of the surface of the sample holding device 27 facing the measuring window 31.The contamination 54 may, for example, comprise adhesion of a liquid, such as a drop of water, and / or adhesion of a foreign substance 54 (for example soil or eukaryotic material remaining on the measuring window 31).

[0163] Fig. 6 and Fig. 7 illustrate a depth of field 42 of the camera 41. The camera 41 of the condition monitoring device 40 advantageously has a depth of field 42 that can extend from an inner surface 36 of the measuring window 31 to beyond an outer surface 35 of the measuring window 31. Advantageously, the depth of field 42 extends by a distance 43 beyond the outer surface 35 of the measuring window 31 (against which the sample 19 rests during use), wherein the distance 43 is advantageously at least 1 mm or at least 2 mm. Particularly advantageously, the depth of field 42 extends at least to the surface 28 of the sample holding device 27 that faces the measuring window 31. By means of such configurations, deviations from the desired state of the measuring window 31 (Fig. 6 and Fig. 7) and deviations from the desired state of the surface 28 of the sample holding device 27, which points towards the measuring window 31 (Fig.7), can be detected during image-based condition monitoring. The surface 28 of the sample holding device 27 can be configured as an optical functional surface or can have an optical functional surface, for example, a functional surface with specific absorption properties and / or reflection properties that are tailored to the detection of the optical measurement variable by the detection device 30.

[0164] The image recording device 41 of the condition monitoring device has a first field of view. The detection device 30 can have a second field of view that is entirely contained within the first field of view. Advantageously, the first field of view can be larger than the second field of view. This allows the condition monitoring device 40 to detect deviations from the target state that lie outside the second field of view (i.e., outside the field of view of the detection device 30).

[0165] Fig. 8 shows the first field of view 61 of the camera 41 and the second field of view 62 of the detection device 30. The second field of view 62 may be substantially identical to the first field of view 61 or may be smaller than the first field of view 61. In both cases, the first field of view 61 of the camera 41 includes the second field of view 62 of the detection device 30.

[0166] The device 10 can be designed such that a position and / or size of the second field of view 62 on the outer surface 35 of the measuring window 31 can be changed. Advantageously, the position and / or size of the measuring position at which the information is acquired can then be determined depending on the image-based condition monitoring. The at least one circuit 21 can control the detection device 30 depending on the image-based condition monitoring, optionally also depending on a user input dependent on the near-field image evaluation. As a result, the optical measurement variable can be detected in a region of the measuring window 31 where, according to the image-based condition monitoring, there is no deviation of the measuring window 31 and / or the sample holding device 27 from the desired state.

[0167] The position of the second field of view 62 relative to the first field of view 61 can be determined and known during field use. For example, calibration can be performed at the factory, and the corresponding position information can be stored non-volatilely in the device 10. Alternatively or additionally, the calibration can also be performed repeatedly, for example, after commissioning of the device 10 in order to determine the position information regarding the position of the second field of view 62 relative to the first field of view 61. This position information can be used by the device 10, for example, to capture the optical measurement variable at a location that was determined to be suitable for capturing the optical measurement variable based on the image-based condition monitoring.

[0168] Alternatively or additionally, the position information can also be used to visualize the position of the second field of view 62 relative to the first field of view 61. The position can, for example, be displayed as superimposed information to visualize a measurement spot of the detection device 30. For this purpose, the human-machine interface 25 can be controlled depending on the position information.

[0169] The position of the second field of view 62 relative to the first field of view 61 can be determined in various ways. Corresponding techniques are known to those skilled in the art. For example, the position determination can be based on one of the following techniques:

[0170] Coupling light into a light path of the detection device 30 to illuminate a sample (which may, for example, be a dedicated calibration sample) and detecting the position of the spot (corresponding to the measurement spot of the detection device 30) in the image captured by the camera 41.

[0171] Positioning an optical fiber in front of the measuring window 31 of the device 10 and moving the light exit end of the optical fiber in two dimensions to scan the area of ​​the measuring window 31. Based on the intensities detected by the detection device 30, the position of the measuring spot of the detection device 30 relative to the first field of view 61 can be determined.

[0172] Fig. 9 is a flowchart of a method 70. The method 70 may be performed automatically by the device 10.

[0173] In step 71, an image is captured by the image capture device 41 of the condition monitoring device. The image is captured such that a depth of field of the image capture device 41 sharply images the component or components of the detection device 30 for which the condition monitoring is to be performed. The image can be captured such that at least the measurement window 30 and / or a surface 28 of the sample holder device 27 facing the measurement window 31 are sharply imaged.

[0174] In step 72, the captured image(s) is / are evaluated. The evaluation is performed in such a way that deviations of the component(s) of the detection device 30 from the target state are detected. The evaluation can be performed in various ways, for example, by determining a distance metric of the captured image with a reference image and / or by applying data-driven processing techniques, for example, a trained artificial intelligence model (K1).

[0175] In step 73, the result of the image-based condition check is used. The result can be used in different ways, for example, to control or regulate the acquisition and / or evaluation of the optical measurement variable by the acquisition device 30. Alternatively or additionally, the result of the image-based condition check can be used to control the human-machine interface 25 depending on the image-based condition check. For example, the human-machine interface 25 can be controlled depending on the image-based condition check in such a way that a user input dependent on the image-based condition check is enabled in order to carry out the acquisition of the optical measurement variable.

[0176] In general, and regardless of the specific implementation of the device 10 or the system according to an embodiment, the device, the system, or the method can carry out the image-based state check in different ways. In one embodiment, the image-based state check can be carried out such that an image of the at least one circuit 21 recorded by the image recording device 41 is compared with both a plurality of previous images and a reference image. This makes it possible to detect deviations from the target state that are evident in a plurality of recorded images (i.e., do not only reflect sample-dependent effects), but differ from a reference image. The reference image can have an initial reference image that is captured, for example, when the device 10 is put into operation and / or at predetermined time intervals and / or when an event occurs during use of the device 10.

[0177] Fig. 10 illustrates such a mode of operation of the state control device 40. A currently recorded image 81 with an intensity distribution I t (where I t an intensity in at least one color channel spatially resolved for several pixels or voxels) is compared by the state control device 40 with one or more further images 83, 84 taken by the image recording device 41 with intensity distributions / t-1; / t-2 , ... are compared. The comparison can be made pixel-wise or voxel-wise. The state control device 40 is further configured to compare the currently recorded image 81 with an initial reference image 82 with intensity distribution I Qpixel- or voxel-wise to identify currently existing deviations from the target state. For example, a scratch and / or a foreign body (e.g., dust or soil adhesion) appears on the measurement window 41 as a constant object 85 between the images 85, 84 (intensity distributions I t and / t -i), but not in the initial reference image 82 (intensity distribution I Q ). Thus, it can be effectively detected that undesirable changes in the measuring apparatus, in particular a deviation of at least one component of the detection device 30 from its desired state, are present.

[0178] The device 10 or the system can thus be configured such that, for image-based condition monitoring, an image 81 currently captured by the image recording device 41 is compared pixel-by-pixel or voxel-by-voxel with at least one previous comparison image 84 and with at least one initial reference image 82 in order to detect a deviation of at least one component of the recording device 30 from the desired state.

[0179] Alternatively or additionally, the device 10 or the system can be configured for image analysis for image-based condition monitoring using at least one model of the AI. Parameters of the AI ​​model can be stored non-volatilely in the memory 22 or in the memory of a computing device different from the device 10. The parameters can define the AI ​​model and can, for example, specify weights, activation thresholds, or other parameters of the AI ​​model.

[0180] Fig. 11 schematically shows a Kl model 90. The Kl model 90 has an input 91, an output 92, and a plurality of hidden layers 93. The Kl model 90 is configured to receive pixel or voxel values ​​of the current image 81 recorded with the image recording device 41 at the input 91. Alternatively, the input 91 can be configured to receive input data determined by preprocessing from pixel or voxel values ​​of the current image 81 recorded with the image recording device 41. The preprocessing can, in particular, comprise local operations and can be carried out in such a way that an assignment of input data to assigned pixels or voxels of the image 81 remains possible. The Kl model 90 is configured to output data at the output 92 that indicate a deviation from the target state. Various embodiments of the Kl model 90 are possible.For example, the Kl model 90 can be configured to output a value at output 92 that indicates a probability that a deviation from the target state exists. The Kl model 90 can be configured to output spatially resolved values ​​at output 92, each of which indicates a probability that a deviation from the target state exists at the corresponding pixel or voxel. The Kl model 90 can be configured to determine the type of deviation from the target state and to output data at output 92 that indicates information about the type of deviation from the target state. For example, probabilities that the respective type of deviation exists can be output for several different types of deviation. The types of deviations are determined by training the Kl model 90.The types of deviations can be selected from a group that includes or consists of the following deviations: scratches in a component of the detection device 30; other damage to the component of the detection device 30; adhesion of a foreign substance (for example a foreign body and / or a liquid) to the component of the detection device 30; displacement of the component of the detection device 30 relative to its target position.

[0181] In one embodiment, the state control device 40 can be configured such that the current image 81 (intensity distribution I) recorded with the image recording device 41 t) for the presence of dirt, dust, and liquids using a deep learning segmentation model. For this purpose, a SegFormer model can be used, the implementation of which is familiar to those skilled in the art from relevant literature. The SegFormer can initially be trained on a collection of pixel-by-pixel annotated training data.

[0182] To train the AI ​​model, training data can be used, which are advantageously acquired with the device 10 for which the image-based state control is to be carried out. Alternatively, the training data can be acquired with an identical device 10 or a similar device 10. At least 100, advantageously at least 200 different images can be used to provide the training data. The learning data can be divided into training data (e.g., 70% of the learning data), test data (e.g., 15% of the learning data), and validation data (e.g., 15% of the learning data). The training of the AI ​​model can be implemented as supervised, semi-supervised, or unsupervised training. Corresponding training methods, for example, using gradient descent techniques, are familiar to the person skilled in the art for different AI models due to their general specialist knowledge.

[0183] Fig. 12 is a flowchart of a method 75. The method 75 may be executed automatically by the device 10 or the system. The method 75 may be executed to implement step 72 of the method 70.

[0184] In step 76, the image captured by the image capture device 41 is segmented. Segmentation can be performed by comparing it with multiple images previously captured by the image capture device 41 and / or using a KI model for segmentation and / or using other segmentation techniques.

[0185] In step 77, the type of deviation from the target state is determined for at least one deviation from the target state. The type can be determined automatically based on an image. The type of deviation from the target state can, for example, be selected from a group consisting of or comprising: damage (e.g., scratches and / or other damage) to the component of the detection device 30; adhesion of a foreign substance (e.g., a foreign body such as a dust particle or pollen and / or a liquid) to the component of the detection device 30; displacement of the component of the detection device 30 relative to its target position. The type of deviation can be determined inherently with the segmentation 76, for example, using a KI model for segmentation.

[0186] In step 78, it is determined which control action is to be executed. The determination of the control action to be executed can be determined depending on the type of deviation and / or a location of the deviation. For example, if there is only local damage and / or adhesion of a foreign substance to the measuring window 31, the control action can consist of the detection of the optical measurement variable by the detection device 30 at a location suitable for detection, spaced apart from the local damage and / or the adhering foreign substance. Alternatively or additionally, the device 10 or the system can be configured to selectively perform the detection of the optical measurement variable depending on a dedicated operator confirmation.For this purpose, the device 10 or the system can control the human-machine interface 25 in such a way that information about the existing deviation from the target state is output and operator confirmation is enabled. Alternatively or additionally, the device 10 or the system can be configured to store data dependent on the detected deviation from the target state in association with the detected optical measurement variable and / or the sample analysis results derived therefrom.

[0187] Fig. 13 is a flowchart of a method 100 according to one embodiment. The method 100 may be executed automatically by the device 10 or the system.

[0188] In step 101, an image is captured by the image capture device 41 of the status control device 40. The image shows at least one component of the detection device 30.

[0189] In step 102, a change in a state of the detection device 30 is optionally initiated. Inducing the change may involve moving the sample holding device 27 toward the measurement window 31. The change may be initiated automatically, for example, by actuating an actuator, or by user guidance.

[0190] In step 103, another image is captured by the image capture device 41 of the status control device 40. The additional image shows at least one additional component of the detection device that is different from the component captured in step 101. Advantageously, the additional component is farther away from the image capture device 41 than the component for which an image is captured in step 101.

[0191] In step 104, the image captured in step 101 and the additional image captured in step 103 are processed for image-based status monitoring. The image-based status monitoring can be performed systematically by first checking whether the component of the detection device 30 positioned closer to the image recording device 41 exhibits a deviation from its target state. Subsequently, it can be automatically checked whether the additional component of the detection device 30 positioned further away from the image recording device 41 exhibits a deviation from its target state.As a result, several components of the detection device 30 can be systematically checked to determine whether their condition corresponds to the target condition, without the condition check of the component of the detection device 30 positioned further away from the image recording device 41 being impaired by a deviation from the target condition in the component of the detection device positioned closer to the image recording device 41. In step 105, the result of the condition check of the component and the further component is used to influence the detection and / or evaluation of the optical measurement variable by the detection device 30 and / or to annotate the detected optical measurement variable and / or a sample analysis result derived therefrom based on the result of the condition check of the component and the further component.

[0192] For illustration, the device 10 or the system can be configured such that, when executing the method 100, a status check for the measurement window 31 is first performed, followed by an image-based status check for the sample holder 27. In this case, inducing the change in step 102 can involve moving the sample holder 27 from a first to a second position relative to the measurement window 31, wherein the second position is closer to the measurement window 31 than the first position.

[0193] Active illumination can be provided during image acquisition by the image acquisition device 41 of the condition monitoring device 40. For this purpose, for example, the illumination device 33 of the measuring system 32 can be controlled. In a further embodiment, the condition monitoring device 40 can have at least one illumination source different from the illumination device 33 of the measuring system 32, which can be controlled to acquire one or more images for the image-based condition monitoring.

[0194] Fig. 14 shows an embodiment of a device 10 in which the condition monitoring device 40 has an illumination source 44, which may be different from the illumination device 33 of the measuring system 32. The device 10 is configured to control the illumination source 44 for image acquisition for image-based condition monitoring. Controlling the illumination source 44 may include selective activation, temporal modulation of the light output, and / or control of a direction and / or wavelength of the illumination emitted by the illumination source 44. Performing the image-based condition monitoring using such controllable illumination may be particularly advantageous for detecting various deviations from the desired state and / or defects and / or foreign matter positioned in various ways.

[0195] Fig. 15 shows an embodiment in which an illumination source 44 is configured such that illumination can be provided in different directions during image acquisition for image-based condition monitoring. For this purpose, the illumination source 44 can, for example, comprise a plurality of controllable light emitting elements 45 that are positioned around an axis 46 of a beam path of the detection device 30 and / or the image recording device 41. Such an embodiment of the illumination source 44, in which a direction of illumination is controllable, can be used both when the illumination is provided by the illumination device 33 of the measuring system 32, and when the illumination for image acquisition for image-based condition monitoring is provided by a different illumination source 42 of the

[0196] Condition control device 40.

[0197] The device 10 can be configured such that the evaluation of the image captured by the image recording device 41 takes place locally in the device 10. Alternatively or additionally, the device 10 can be configured such that an evaluation of the image captured by the image recording device 41 for image-based condition monitoring and / or an evaluation of the optical measurement variable captured by the detection device 30 takes place in a computing system separate from the device 10.

[0198] Fig. 16 shows a schematic representation of a system 110 comprising the device 10 and a computing system 111. The device 10 can be configured to provide the at least one image acquired by the image recording device 41 to the computing system 111 for evaluation. Alternatively or additionally, the device 10 can be configured to provide the optical measurement variable acquired by the acquisition device 30 to the computing system 111 for evaluation via the data interface 42. The device 10 can be configured to receive a result of the image-based condition check and / or a result of the sample analysis from the computing system 111 via the data interface 24. The device 10 can be configured to control the human-machine interface 25 depending on the received result of the image-based condition check and / or the sample analysis.This is advantageous in order to be able to provide feedback to the operator while he is still in field use with the device 10.

[0199] The computing system 111 can be configured to receive the image captured by the image recording device 41 for image-based condition monitoring and / or the optical measurement value captured by the detection device 30 via a communication interface. The computing system 111 can be configured to have a unidirectional or advantageously bidirectional communication connection with the device 10 via a communication system, which may, for example, be a wide area network 113 and / or a cellular network. The computing system 111 can have a further human-machine interface 112 to output a result of the image-based condition monitoring and / or a result of the sample analysis based on the optical measurement value.

[0200] The image recording device 41 can advantageously be integrated into the housing 20 of the device 10. However, this is not mandatory. The image recording device 41, for example, a camera, can be coupled to the housing 20 in a non-destructive, reversible manner.

[0201] Fig. 17 is a schematic representation of a system 120 comprising a device 10, wherein the detection device 30 is arranged in or on a housing 20 of the device 10. The image recording device 41 of the condition control device can be non-destructively detachably connected to the TI

[0202] Housing 20 can be coupled. For this purpose, the image recording device 41 can have coupling elements 122 for non-destructive, releasable engagement with corresponding coupling elements 121 of the housing 20. The image recording device 41 can also be integrated, for example, into a communications terminal, such as a smartphone or other mobile communications device. This can be coupled to the housing 20 via a suitable mechanical and / or optical coupling arrangement in order to record at least one image for the image-based condition monitoring.

[0203] The device 10 or system may be configured to perform a sample analysis for a sample comprising a soil sample.

[0204] Fig. 18 is a schematic representation of a device 10 configured for sample analysis of a soil sample. The sample holder 27 has a receptacle 130 for receiving the soil sample 132. One or more surfaces of the receptacle 130 can be provided with an optical functional layer 131, for example, an optical functional layer 131 with defined reflection properties and / or absorption properties. The optical functional layer 131 can be configured to enable the detection of the optical measurement variable in a reflection arrangement and / or a transflection arrangement and / or a transmission arrangement.

[0205] The device 10 or the system can be configured to perform image-based condition monitoring to read at least one sensor. This allows deviations from a target range in parameters that influence the optical measurement, such as relative or absolute humidity, to be detected.

[0206] Fig. 19 shows a device 10 in which the detection device 30 has at least one sensor 39 for detecting at least one parameter of the detection device 30. The at least one sensor 39 can comprise, without being limited thereto, a humidity sensor and / or a temperature sensor. The at least one sensor 39 can be configured to indicate a result of the measurement in a visually perceptible manner. The condition monitoring device 40 can be configured to exclude the at least one sensor 39 for image-based condition monitoring on an image-based basis, wherein an image captured by the image recording device 41 shows the visually perceptible measurement of the sensor 39.

[0207] The detection device 30 can be designed in different ways.

[0208] Detection device 30 can be configured for image-based sample analysis.

[0209] Detection device 30 can alternatively or additionally be configured to process a wavelength-resolved optical measurement variable. For this purpose, detection device 30 can comprise a spectrometer or another spectral analysis measuring system. Fig. 20 shows a schematic representation of the device 10, in which detection device 30 comprises a spectral analysis detection device. The spectral analysis detection device comprises an illumination device 140 (which can, for example, comprise a laser, a laser diode, or another coherent light source) and a spectrometer 142. The object and reference beams can be guided to the spectrometer 142 via a beam splitter 141. The spectral analysis detection device comprises an optics 143, which optionally enables adjustment of a measurement spot relative to the measurement window 31.

[0210] Regardless of the specific implementation of the device 10 or the system, the device 10 and / or the system can be configured such that the image-based condition monitoring and / or the sample analysis is carried out in such a way that an assignment to the respective position of the device 10 in a cultivation area (for example, an agricultural open field and / or a greenhouse) remains possible. For this purpose, the device 10 can have a position sensor to enable logging of the respective position in relation to the at least one image captured by the image recording device 41 and / or the optical measurement variable captured by the detection device 30. The position sensor can be configured as a receiver 149 of a global navigation satellite system (GNSS).

[0211] Fig. 21 is a flowchart of a method 150. The method 150 may be performed automatically by the device 10 or the system according to one embodiment.

[0212] In step 151, an image-based condition check is carried out to check for a deviation from a target condition of at least one component of the detection device 30 for detecting an optical measurement variable.

[0213] Depending on the result of the image-based condition check, one or more of the following actions can be performed automatically: controlling a human-machine interface to issue a recommended action (step 152), controlling the acquisition and / or evaluation of the optical measurement variable by the acquisition device (step 153), and / or annotating the optical measurement variable and / or a sample analysis result derived therefrom (step 154). Which of the actions is performed can be determined automatically by the device 10 or the system, for example, depending on the type of deviation from the target state and / or the position at which the deviation occurs in the acquisition device 30 (for example, laterally or centrally in the measurement window 31). The logic according to which the action to be performed is determined can be fixed or user-configurable.

[0214] Fig. 22 is a flowchart of a method 160. The method 160 may be performed by the device

[0215] 10 or the system to perform a condition analysis of a plant. In step 161, an image-based condition check is performed. The image-based condition check includes a check to determine whether a component of the detection device 30 for detecting an optical measurement variable deviates from its target state.

[0216] In step 162, a spectral analysis detection device 30 is activated to detect an optical measurement variable with wavelength resolution. Detection can be performed with both spatial and wavelength resolution.

[0217] In step 163, the optical measurement value is evaluated. This evaluation may include a quantitative determination of nutrient concentration and / or an analysis of the plant's health status and / or oversupply or undersupply. The spectral analytical information can be processed to determine the plant's health status in various ways. For example, the techniques described in WO 2019 / 169434 A1, US Pat. No. 7,804,588 B2, and US Pat. No. 1,132,307 B2 can be used.

[0218] At step 164, a result of the sample analysis is provided. Providing the result may include providing a recommended course of action to enable better care for the plant. The recommended course of action may include spatially resolved recommendations for multiple locations within a cultivation area.

[0219] Below, various further modifications and features are described that can optionally be used to support the technical effects already described.

[0220] Color channels: The image recording device 41 of the condition monitoring device 40 can have a camera that captures grayscale images. Such grayscale images enable image-based condition monitoring. A color camera image (e.g., RGB color channels) from the camera 41 of the condition monitoring device 40 can be advantageous, for example, to improve image-based condition monitoring (e.g., by detecting additional faults and / or more precise detection). The camera image from the camera 41 can have multiple color channels in the visible spectral range. The camera image from the camera 41 can have multiple color channels in the visible spectral range and one or more color channels in the near-infrared spectral range.

[0221] Sensitive wavelength range of the image recording device 41: The image recording device 41 (for example, a near-field camera 41) can have at least one color channel in the near-infrared spectral range. Image recording in the near-infrared range could make it possible to improve the contrast, for example, for water and organic compounds compared to a pure RGB camera. This can be particularly advantageous for checking for contamination of the measurement window 31 or the sample holder 27.

[0222] Field of view: The first field of view 61 may substantially correspond to the second field of view 62.

[0223] A larger field of view or field of view 61 of the camera 41 compared to the measuring range of the detection device 30 has the advantage that the illuminated area of ​​the spectrometer can be unintentionally moved during the measurement and yet the (more global) camera analysis is still valid.

[0224] Positioning of the camera of the condition control device 40 in the housing 20: The image recording device 41 (for example a near-field camera 41) is positioned or positionable in such a way that it covers one or more measurement-relevant components of the detection device

[0225] 30. Advantageously, the image recording device 41 is structurally integrated into the housing 20 for this purpose. As a result, the relative position of the image recording device 41 to the component(s) of the detection device 30 can be assumed to be known. In particular, the image recording device 41 can be positioned such that it illuminates the spectrometer illumination device

[0226] 31 and / or does not have direct reflected images. Another possible implementation is the design of a system 120 with an external camera 41, for example a smartphone camera, which can be coupled into the housing 20 via an optical path 29, as shown schematically in Fig. 17.

[0227] Frequency band filter: If the image recording device 41 has a camera with sensitivity in the visible and near-infrared spectral range, it is advantageous for the image recording device 41 to have a filter that is configured to reduce stray light and / or heating.

[0228] Realization of the required depth of field: As already explained and schematically shown in Fig. 6 and Fig. 7, the image recording device 41 is advantageously designed such that a depth of field range 42 extends from a plane 36 of the internal referencing to at least a glass flange plane 35 and advantageously a distance 43 beyond (in order to also view curved leaves sharply) or even to the plane 28 of the sample holding device 27. Such a depth of field range can be achieved, for example, in that the image recording device 41 has an aperture that is dimensioned so small that the desired depth of field is achieved.

[0229] Lighting for image acquisition for image-based condition control: The lighting for image acquisition for image-based condition control can be done in different ways, e.g. by

[0230] Use of ambient light (e.g. a useful implementation in bright greenhouses);

[0231] Use of the illumination device 33, which can be activated for the detection of the optical measurement variable;

[0232] Use of an additional illumination source 42, which can be activated for image acquisition for image-based condition monitoring (and optionally can only be used for this purpose). If an additional illumination source 42 is installed in addition to the already existing illumination device 33, this illumination source 42 can be advantageously designed with regard to image-based condition monitoring. For example, the illumination source 42 can be configured to illuminate obliquely against the measurement window 31, whereby unevenness on the glass and / or cracks in the glass are more clearly visible than with more orthogonal illumination.Particularly advantageous for the analysis is the use of an illumination source 44 which has more than one light source, for example at least four diametrically arranged LED light sources 45, whereby at least two images with angle-selective illumination are acquired for the image-based condition control and the at least two images can be converted into digital phase contrasts, virtual dark field images, and / or similar advantageous representations.

[0233] The illumination source 42 can be designed as an ultraviolet (UV) light source (e.g., a UV light-emitting diode (LED)). This allows for better visualization of chemical traces or organic compounds left behind by plants compared to a visible light source. This allows for a reliable analysis of the condition of the measurement window 31 and / or the sample holder 27.

[0234] (Reference) images for image-based condition monitoring: The detection of a deviation of at least one component of the detection device 30 from the target state can be carried out by comparing an image 81 acquired on a sample 19, 132 when the device 10 is used. The image-based condition monitoring can comprise a comparison with one or more further images 83, 84 also acquired on samples 19, 132 when the device 10 is used. The device 10 or the system can alternatively or additionally be configured such that the image-based condition monitoring is carried out based on an image recording of a reference object (instead of analyzing the sample recording). The device 10 can comprise a white standard (for example, a reflective fluoropolymer) for adjusting the channel gain factors (white balance) and adjusting the exposure time. The white standard can be pivotable.Since a white standard contains no sample information, any interference signal can be directly traced back to a deviation of a component of the detection device 30 from the target state. It is also possible to include two standards, e.g., an external and an internal white standard, which makes the detection of a deviation of a component of the detection device 30 from the target state even more reliable.

[0235] Sequence of checking the status of multiple components: The imaging and evaluation of a component (e.g., the sample holder 27) located further away from the image acquisition device 41 (e.g., the near-field camera 41) could be distorted if defects exist along the optical path from the image acquisition device 41 (e.g., a deviation of the measurement window 31 from the desired state). An advantageous sequence for checking the status of the detection device 30 is to begin with the evaluation of the component closest to the image acquisition device 41, followed by components further away from the image acquisition device 41.

[0236] An example of a possible sequence is (where only some of the checks mentioned can be carried out):

[0237] Checking for the presence of dust in the housing of the spectral analytical detection device 30;

[0238] Verification of an internal calibration standard;

[0239] Checking the inner surface of the measuring window 31;

[0240] Checking the outer surface of the measuring window 31;

[0241] Checking the surface 28 of the sample holder 27.

[0242] Another example of such a procedure may include the following: When the illumination source 33 of the detection device 30 is switched on, the device 10 or the system may start the component test. In a final step, the homogeneity of the illumination source 33 may be checked. In this way, the image-based condition check of the other components of the detection device 30 may be performed while the illumination source 33 of the detection device 30 is warming up (which may take several minutes for a halogen lamp, for example), so that time loss for the condition check may be eliminated or reduced. The light homogeneity of the illumination source 33 of the detection device 30 is checked at a time when the illumination source 33 already has stable lighting conditions.

[0243] In this way, misinterpretations when analyzing component status can be reduced.

[0244] Feedback to the operator. The device 10 or the system can be configured to output a result of the image-based condition monitoring to an operator. For example, in response to the detection of a quantity of dust and / or dirt, a recommended action for cleaning the measuring window 31 can be issued. This can include a visualization of the affected areas in the image. This can provide the operator with information on how to correct the cause, for example, by prompting them to clean specifically, remove dust, etc.

[0245] The device 10 or the system can be configured to issue a request to replace the measuring window 31 in the event of irreversible and detrimental damage to the measuring window 31 (for example due to scratches).

[0246] The device 10 or the system can be configured to inform the user of necessary maintenance in the event of malfunctions or detrimental damage to the sample holder 27 or a shutter. Selective activation and / or deactivation of image-based condition monitoring: Image-based condition monitoring takes time. Some operators may prefer to take the risk of measuring with the detection device 30 not fully intact in order to achieve faster measurement times. The device 10 or the system can be configured to enable selective activation and / or deactivation of image-based condition monitoring.

[0247] Triggering the image-based condition check: The device 10 or the system can be configured to perform the image-based condition check whenever a trigger criterion is met. The trigger criterion can be time-based (e.g., a condition check once per specified time interval, e.g., once per hour) and / or usage-based (e.g., a condition check once after a specified number of sample analyses, e.g., every 30 sample analyses).

[0248] Status check of the image recording device 41: The image recording device 41 (e.g., a near-field camera 41) can also be faulty. This can lead to misleading diagnoses. To check the functionality of the image recording device 41 before analyzing the detection device 30, the device 10 or the system can be configured to systematically evaluate images acquired at a specific location within the device 10 (e.g., at a location with a fine and easily recognizable feature).

[0249] Alternatively or additionally, the device 10 can be designed to introduce features with a suitable pattern and a suitable size into one or various components of the detection device 30 (example: at various levels such as the shutter, the measuring window 31, and the sample holding device 27, etc., e.g. by means of laser engraving) that do not hinder the spectroscopic analysis.

[0250] Alternatively or additionally, the device 10 can be designed to examine recorded images of the reference objects used for systematic camera errors, such as constant image pixels, image distortion, or image stripes, without being limited thereto.

[0251] Any deviation of these reference images can be tracked and identified to detect a malfunction of the image pickup device 41.

[0252] Design and mounting of the device 10: The device 10 can be designed as a portable, in particular manually held device 10.

[0253] Fig. 23 schematically shows an embodiment of the device 10 as a manually held device having a structure 18 for holding the device 10.

[0254] Fig. 24 shows a further embodiment in which a vehicle 170 according to one embodiment has one or more devices 10 mounted thereon. Accordingly, the device 10 or the devices 10 can be configured with a support structure for attachment to an agricultural vehicle. Fig. 25 shows a further embodiment in which a robot 180 has the device 10.

[0255] A base 182 of the robot 180 can be stationary or movable, for example, along a rail system. A controller 181 can have one or more actuators for positioning the

[0256] Control device 10.

[0257] Fig. 26 shows a further embodiment in which a flying object 190 comprises the device 10. The flying object 190 can be remotely controllable and / or configured for autonomous flight operation.

[0258] In yet another embodiment, the device 10 can be configured as a stationary system. The device 10 can be mounted on a stationary support near a conveyor on which plants, other eukaryotic material, or soil samples are transported past the device 10.

[0259] Design of the detection device 30: The detection device 30 can comprise a spectrometer. However, other designs are possible that enable the detection of an optical measurement variable at two or more (for example, at least ten or more) wavelengths. For this purpose, filter-based systems (configured to block light components), single-wavelength diodes, or other designs can be used as an alternative or in addition to the use of a spectrometer.

[0260] Types of image-based condition checks: The device 10 or the system can be configured such that the image-based condition check can detect scratches, dust on the inner and / or outer surface of the measuring window 31, and / or deviations of the sample holder 27 from the desired condition. The device 10 or the system can be configured such that the image-based condition check can alternatively or additionally detect other types of deviations from the desired condition, for example, one, several, or all of the following deviations:

[0261] Components in the optical path: Components inadvertently moved into the optical path inside the detection device 30 can impact the detection of the optical measurement variable (e.g., components that have been moved from their original position or loose parts, such as broken glass from an optical component). In order to perform a condition check to detect such deviations from the target state, one or more reference images can be recorded that show or show the installation space in the target state. A reconstruction model, e.g., an autoencoder, can be trained on the basis of the recorded reference images that show the target state. In the application phase, one or more images that show the actual state can be compared with the at least one reference image.For example, subtraction and smoothing can be performed, followed by segmentation, which identifies (significant) differences in the differential image as anomalies. Reconstruction models are a class of machine learning models. Segmentation can be performed through conventional processing (e.g., using thresholds) and / or data-driven learning with machine learning segmentation models. The device 10 or the system can be configured to detect one, several, or all of the following deviations in this way:

[0262] • cables protruding into the optical path;

[0263] • broken pieces of plastic, for example from a cooler;

[0264] • broken or bent pieces of metal, for example from a passive cooling device;

[0265] • broken pieces of glass, for example after the glass of an optical component breaks.

[0266] Tilting of the illumination device 33 of the detection device 30: The device 10 or the system can be configured to check, during the image-based status check, whether an orientation of the illumination device 33 illuminates a measurement spot through the measurement window 31 in the desired manner or is tilted in an undesired manner, such that, for example, areas of the housing 20 are illuminated. For this purpose, the illumination device 33 can be activated, an image can be captured and recorded by the image recording device 41, and compared with a desired position of the light cone. Similarly, the alignment of one or more camera sensors can be checked.

[0267] Glass crack or glass breakage: The device 10 or the system can be configured to check whether a glass crack or glass breakage is present during the image-based condition inspection. Optionally, this can include checking whether a glass crack is continuous in a thickness direction of the measurement window 31. Optionally, images can be captured and evaluated at different depths, for example, to examine the inner and outer surfaces of the measurement window 31 and / or to estimate the depth of a crack.

[0268] Illumination inhomogeneities: The device 10 or the system can be configured to detect inhomogeneities in the illumination generated by the illumination source 33 of the detection device 30. For this purpose, the device 10 or the system can be configured such that the illumination source 33 is switched on and the at least one image acquired for image-based status monitoring of the illumination source 33 is only recorded when stabilized illumination conditions exist (e.g., the illumination source 33 has warmed up). The device 10 or the system can be configured to detect changes in the illumination (e.g., changes in homogeneity due to lamp aging or lamp position dependency, without being limited thereto) from the recorded image of an external (e.g., homogeneous) sample or an internal reference.The device 10 or the system can be configured to evaluate these changes with regard to their criticality for the acquisition of the optical measurement variable. The use of the internal standard offers the advantage of allowing more reproducible conditions and reducing the influence of other factors (e.g., the state of the measurement window 31).

[0269] State of the sample holding device: The device 10 or the system can be configured to perform a state check of the sample holding device 27. The image-based state check can include a check of the state of the surface (surface roughness). For this purpose, at least one image can be acquired with the sample holding device closed using internal lighting 33, 42. Light inhomogeneity in the image caused by a change in the surface 28 of the sample holding device can be detected based on the image. This is advantageously carried out after checking the state of the measuring window 31 in order to exclude its influence. Alternatively or additionally, the image-based state check can include a check of the closed state of the sample holding device. For this purpose, at least one image can be acquired with the internal lighting 33, 42 switched off. Parasitic ambient light can be detected based on the image.For this purpose, a comparison can be made with a previously taken reference image that was captured in the dark or with the internal aperture closed.

[0270] Condition of an air cooling system and / or temperature-critical components: The device 10 or the system can be configured to perform image-based monitoring of the condition of an air cooling system and / or temperature-critical components. A defective air cooling system can lead to a temperature increase inside the housing 20. The temperature increase can exceed a permissible maximum temperature. This can lead to a temporary malfunction or permanent damage to the components of the detection device 30. In order to perform image-based monitoring of the condition of the air cooling system and / or temperature-critical components, the device 10 or the system can be configured to check the internal temperature of the detection device 30. This can be done by using an (optionally irreversible) thermometer strip (e.g., sensor 39 in Fig.19) within the housing 20, which can be glued, for example, near the measuring window 31. Before each use of the detection device 30, the image recording device 41 can image the thermometer strip. The strip can be localized in the image (for example, using a trained detection algorithm or conventional image processing). The image section can optionally be converted into a standardized representation (for example, rectified). The device 10 or the system can be configured to automatically evaluate, based on the image section, whether a temperature has been reached that lies above a given threshold and is thus to be classified as a critical temperature.In this case, the device 10 or system can control the human-machine interface 25 to initiate a check of the cooling system and possibly other components that may be damaged by excessive temperature.

[0271] Detection of foreign particles (e.g., dust) in the housing: The device 10 or the system can be configured to detect dust particles. A non-dust-tight housing 20 results in dust particles being present in the beam path of the detection device 30, which either deposit on optical components and / or float in the air. These particles can lead to light absorption and / or scattering and impair the quality of the measurement of the optical measurement variable. Therefore, their presence should be checked before detecting the optical measurement variable. To detect dust particles, the device 10 or the system can be configured such that, with the illumination 33 and / or 42 switched on, a camera image of an optical component or a part of the housing 20 located near the measurement window 31 (e.g., dark shutter) is recorded.The corresponding image is compared with a reference image (taken in the absence of dust) to highlight the light scattering emanating from these particles or to check for the difference in dust coverage on the surface under investigation. Optionally, the device 10 or system can be configured to activate an air fan (e.g., a cooling system) before and during image acquisition to circulate the air (and particles) inside the housing.

[0272] Carrying out the image analysis for image-based condition monitoring: The device 10 or the system is configured to evaluate an image 81 captured by the image recording device 41 or a plurality of images 80, 81 captured by the image recording device 41 in order to carry out the image-based condition monitoring. Several image analysis techniques have already been explained with reference to Fig. 10 and Fig. 11, in particular the use of image processing techniques for detecting changes (which may include averaging, edge extraction, smoothing, etc.) and data-driven techniques, which may include AI-based techniques. Data-driven techniques can be used in various embodiments, all of which are usable for the solution presented here of a camera-based condition analysis of the detection device 30 and each of which is associated with different advantages.The data-driven techniques may differ in the model output and / or the annotation used during the training phase.

[0273] Type of model output: The device 10 or the system can be configured as follows: Image classification can be used to determine the membership of the captured image to one or more known classes (e.g., "Contains pollen" and "Does not contain pollen");

[0274] Semantic segmentation can be used to determine pixel-precise assignment to one or more known classes;

[0275] Instance segmentation can be used to determine pixel-precise assignments to one or more classes while distinguishing between different instances of the same semantic category;

[0276] Object detection can be used to determine a rough localization of instances of different semantic categories, for example in the form of bounding boxes;

[0277] Anomaly detection can be used to determine the probability of the image belonging to the set of good images;

[0278] Anomaly localization can be used to spatially determine the pixels in the image that are atypical for the pixels in the set of good images;

[0279] All mentioned outputs can simultaneously be provided with the estimated uncertainty of the prediction in order to enable a realistic assessment of the actual occurrence of the detected image features in the image.

[0280] Type of annotation: Depending on the model type chosen (and especially on the model output), different annotations can be used for training.

[0281] For anomaly detection and anomaly localization, only the provision of good images without further annotations is necessary.

[0282] For image classification, the defect classes contained in each image must be provided as labels. In the context of image classification, this is referred to as fully supervised training. Using other training methods such as semi-supervised learning, combinations of labeled and unlabeled images can also be used.

[0283] For localization and segmentation, corresponding boxes and / or masks typically need to be annotated. In the context of localization and segmentation, this is also referred to as fully supervised training. Using other training methods such as weakly supervised, semi-supervised, or semi-weakly supervised, simpler annotation types can also be utilized, even in combination, for example, one, several, or all of: labels; points; scribbles; polygons; boxes; or masks for segmentation. Relationship between image-based state control and sample analysis: The result of image-based state control can be used in different ways for the acquisition of the optical measurement and / or sample analysis based on the optical measurement:

[0284] Use of image-based condition monitoring to decide whether and / or when the optical measurement variable is recorded: In this case, at least one image can initially be recorded with the image recording device 41 and evaluated continuously or in response to a trigger signal in order to assess the presence of errors. Only when the evaluation shows that the recording device 30 is in a proper condition is the recording of the optical measurement variable carried out. The result can be output, for example via an optical, acoustic and / or tactile signal. Recording at least one image by the image recording device 41 before recording the optical measurement variable can save time if a deviation from the target state is detected, since the recording of the optical measurement variable can be omitted.

[0285] Using image-based condition monitoring to trigger user confirmation: If the image-based condition monitoring reveals that a problem exists, the human-machine interface 25 can be triggered to enable user confirmation. The optical measurement value can be selectively acquired only if the user confirms via the human-machine interface that the optical measurement value should be acquired despite the problem.

[0286] Image-based condition monitoring after acquisition of the optical measurement value: In this embodiment, the device 10 or the system is configured such that the optical measurement value is acquired independently of the image-based condition monitoring. An evaluation of the optical measurement value can also optionally be performed independently of the image-based condition monitoring. The sample analysis based on the optical measurement value and the image-based condition monitoring can be performed independently of one another. Both results can be provided to the operator, for example, as the result of the sample analysis in combination with an indication of the expected reliability of the sample analysis determined from the image-based condition monitoring. For this purpose, a traffic light-like system can be used, for example, to output the reliability via the human-machine interface 25 or display device 112.Alternatively or additionally, a visualization of the problem found in the detection device 30 (for example, "glass crack in the measuring window") can be performed.

[0287] Use image-based condition monitoring at a later time: Image-based condition monitoring can also be used after the optical measurement has been taken, or even determined. This allows, for example, the subsequent filtering or marking of invalid measurements from statistics for historical analyses or for analyses that aggregate results from individual measurements.

[0288] Areas of application: The disclosed devices, evaluation devices, evaluation systems, systems, and methods can be used for various applications. In particular, the devices, evaluation devices, evaluation systems, systems, and methods can be used on crops or plants, products derived therefrom, and / or soil samples. The plants on which the techniques can be used can, in particular, also include forestry plants (e.g., trees), hybrids, and / or genetically modified plants.

[0289] Examples of such plants can be:

[0290] • Crops

[0291] • Fodder plants

[0292] • Fiber plants

[0293] • Oil plants

[0294] • Ornamental plants

[0295] • Industrial crops, such as o Oilseeds o Tobacco o Hemp o Hops o Aromatic, culinary and medicinal plants o Seeds for herbaceous oil plants o Seeds for linseed (and consequently fibre flax) o Energy crops o Plants used for the production of feedstocks for renewable energy production

[0296] Sample analysis may include, but is not limited to, the qualitative and / or quantitative determination of nutrient concentrations. For example, sample analysis may include analyses related to one, several, or all of the following plant conditions:

[0297] Nutrient supply problems, for example o Nutrient deficiency o Toxicity biotic stress, for example due to o Nematodes o Insects o Arachnids o Fungi o Bacteria o Viruses

[0298] Abiotic stress, for example due to o drought o excessive water supply and / or waterlogging o salinity o temperature-related stress (cold, frost, heat) o UV light o metal toxicity o mechanical stress (e.g. pressure, squeezing, pressing).

[0299] The results of image-based condition monitoring and sample analysis based on optical measurements can be used to improve plant conditions. This can be achieved through automatic or semi-automatic adjustment of artificially applied substances (such as approved fertilizers or approved insecticides) and artificially applied water. Numerous other applications are possible, for example, for monitoring plant health with regard to environmental and regulatory issues.

[0300] Further areas of application: Even if embodiments have been described in the context of a sample analysis of plants and / or soil samples, the devices, evaluation devices and evaluation systems and methods can also be used elsewhere, for example for sample analysis of eukaryotic material from plant-like eukaryotes such as algae.

[0301] Fig. 27 is a flowchart of a method 200 that may be performed using the device 10 or system.

[0302] At 201, an image-based condition check and sample analysis are performed using the device 10 or the system. At 202, a result of the sample analysis is used to determine measures to improve conditions for the plants. The measures can then be implemented to improve conditions.

[0303] The disclosed devices, systems, and methods provide various technical effects. In particular, they reduce the risk of erroneous sample analysis when using a device 10 for detecting an optical measurement variable. For this purpose, an image recording device 41, for example, a near-field camera 41, is provided to image measurement-relevant parts of a detection device 30 (for example, a spectrometer system). A status check of one or more components of the detection device 30 can be performed in an image-based manner.

[0304] Embodiments of the invention can be used in particular in connection with plant cultivation in agriculture, which is a central economic and industrial sector of humanity. Global trends such as a growing population and a simultaneous decrease in the availability of agricultural land lead to a constantly growing demand for crop yields per area. At the same time, the unsustainable exploitation of soils leads to an increased need for targeted fertilization in order to maintain or even increase previous yields. Furthermore, a growing awareness of sustainable management and increasing regulatory pressure are creating new requirements for precisely tailored fertilization, in particular to avoid over-fertilization.

[0305] Of particular relevance for the growth of any crop is the appropriate supply of nitrogen (N), as this is the most important element of chlorophyll and thus essential for optimal metabolism. In addition to N, other macronutrients are also relevant, such as phosphorus (P) and potassium (K), which together with N can often be added to the soil as so-called NPK fertilizers, as well as calcium (Ca), magnesium (Mg), and sulfur (S). Mg is the central element in the chlorophyll ring. Alternatively or additionally, possibly depending on the plant type, other nutrients may also be relevant for optimal growth and yield; however, these are often required in much smaller quantities and are therefore also referred to as micronutrients. These include, for example, boron (B), molybdenum (Mo), copper (Cu), manganese (Mn), zinc (Zn), iron (Fe), and chlorine (Cl).

[0306] Embodiments of the invention can contribute to ensuring optimal conditions with regard to nutrient supply throughout the entire growth phase of a plant. Nutrient deficiencies or nutrient oversupply can be responded to appropriately by adapted fertilizer applications (for example, through a specific fertigation program in the substrate or soil growth, or as additional soil and / or foliar fertilization). Embodiments of the invention allow for regular monitoring of plants. Compared to laboratory-based leaf analysis, sample analysis can be performed quickly. This is particularly desirable for fast-growing plants and plant species with rapid fruit ripening (for example, strawberries and lettuce).Embodiments of the invention thus also address the objective that there should not be a delay of several days between measurement and availability of the evaluation result, as otherwise the measurement result would already be invalid and would not allow for any meaningful adjustment of the nutrient application. Rapid evaluation is also desirable when environmental factors change rapidly (e.g., temperature, precipitation, etc.), as otherwise the condition would change significantly between the measurement and the time of analysis.

[0307] While exemplary embodiments have been described with reference to the figures, modifications may be implemented in further exemplary embodiments. For example, the modifications already explained may be used cumulatively or alternatively.

[0308] While embodiments have been described that can be used in crops or cultivated plants, the disclosed techniques can also be used in other fields of application.

[0309] The present disclosure also encompasses embodiments with any combination of features mentioned or shown for different embodiments. It also encompasses individual features in the figures, even if they are shown there in connection with other features and / or are not mentioned above or below. Furthermore, the alternative embodiments described in the figures and the description and individual alternative features thereof may be excluded from the subject matter of the invention or from the disclosed subject matter.

[0310] The terms "comprising" and "having" and derivatives thereof indicate a non-exhaustive relationship and do not exclude the presence of other elements or steps. The indefinite article "a" or "an" and derivatives thereof do not exclude the presence of a plurality of the corresponding elements. The functions of several features listed in the claims may be fulfilled by one unit or step.

[0311] A machine-readable instruction code that can be executed by a programmable circuit to perform methods according to embodiments may be stored and / or distributed on a suitable medium, such as an optical storage medium or a solid-state medium provided together with or as part of other hardware. The instruction code may also be distributed in another form, such as a modulated data signal sequence.

[0312] Embodiments of the invention provide improved techniques for sample analysis of

[0313] Plants, soil samples and / or samples containing eukaryotic material.

Claims

CLAIMS 1. Device (10) or system (110; 120) for sample analysis of a sample (19; 132) comprising eukaryotic material (19) and / or a soil sample (132), wherein the device (10) or the system (110; 120) comprises: a detection device (30) which is configured to detect at least one optical measurement variable on the sample (19; 132), wherein the optical measurement variable can be evaluated for sample analysis, and a state control device (40) which is configured to control a state of the detection device (30), wherein the state control device (40) has an image recording device (41) for image-based state control of at least one component of the detection device (30).

2. Device (10) or system (110; 120) according to claim 1, wherein the image recording device (41) has a first field of view (61) and wherein the detection device (30) has a second field of view (62) which is completely contained in the first field of view (61).

3. Device (10) or system (110; 120) according to claim 1 or claim 2, wherein the detection device (30) has a housing (20) and a measuring window (31) arranged on the housing (20), which has an outer surface (35) for contact with the sample (19; 132) when detecting the at least one optical measurement variable and an inner surface (36), and wherein the image recording device (41) has a depth of field which covers at least one depth of field range (42) from the inner surface (36) to the outer surface (35) of the measuring window (31).

4. Device (10) or system (110; 120) according to claim 3, wherein the image-based condition control comprises an image-based detection of a deviation of the measurement window (31) from a desired state.

5. Device (10) or system (110; 120) according to claim 3 or claim 4, wherein the image recording device (41) is arranged to capture at least one image (81) of an area which is arranged offset relative to the measuring window (31) along an optical axis (46) of the detection device (40), wherein the image-based condition control one, several or all of the following condition checks based on the at least one image: Detecting foreign substances (52, 53) in or on the detection device (30), Detecting a shift between actual and target positions of the components (31) of the detection device (30), Checking at least one parameter based on an image-based readout of at least one sensor (39) of the detection device (30) which is configured to detect the at least one parameter, Condition control of a sample holding device (27) for holding the sample (19; 132), detection of spatial and / or temporal inhomogeneities of an illumination source (33) of the detection device (30).

6. Device (10) or system (110; 120) according to one of claims 3 to 5, wherein the image recording device (41) is mounted in the housing (20).

7. Device (10) or system (110; 120) according to one of the preceding claims, wherein the condition control device (40) has at least one evaluation circuit (21) for evaluating images (80, 81) captured by the image recording device (41) and at least one memory (22), wherein the at least one evaluation circuit (21; 111) is set up to carry out an evaluation of the images (80, 81) captured by the image recording device (41) for the image-based condition control.

8. Device (10) or system (110; 120) according to claim 7, wherein parameters influencing the evaluation are stored in the memory (22) and retrieved by the at least one evaluation circuit (21; 111) and used for the evaluation, wherein the parameters comprise model parameters of a model (90) of the artificial intelligence, K1.

9. Device (10) or system (110; 120) according to claim 7 or claim 8, wherein the evaluation comprises a comparison of an image (81) acquired on a sample when used with at least one comparison image (82-84) acquired with the image recording device.

10. Device (10) or system (110; 120) according to claim 9, wherein the at least one comparison image (82-84) comprises a reference image (82) acquired on a reference object and stored in the memory and / or a sample image (83, 84) acquired on another sample when used.

11. Device (10) or system (110; 120) according to one of the preceding claims, wherein the condition control device (40) has one or more controllable light sources (44, 45) which can be selectively activated for image recording by the image recording device (41).

12. Device (10) or system (110; 120) according to claim 11, wherein a plurality of controllable light sources (45) are arranged distributed around an optical axis (46) of the detection device (30) and / or wherein the state control device (40) is configured for a time-sequential activation of the plurality of controllable light sources (45).

13. Device (10) or system (110; 120) according to one of the preceding claims, wherein the state control device (40) is configured to carry out the image-based state control for a plurality of components (27, 31) of the detection device in a sequence in which the image-based state control for a component (31) positioned closer to the image recording device (41) takes place before the image-based state control for a component (27) positioned further away from the image recording device (41).

14. Device (10) or system (110; 120) according to one of the preceding claims, wherein the image recording device (41) has a fixed aperture and a fixed focusing optics.

15. Device (10) or system (110; 120) according to one of the preceding claims, wherein the condition control device (40) is configured to influence the detection of the at least one optical measurement variable depending on the image-based condition control, to output a result of the image-based condition control via a human-machine interface (25) and / or to mark the at least one optical measurement variable or a sample analysis result derived therefrom based on the result of the image-based condition control.

16. Device (10) or system (110; 120) according to one of the preceding claims, wherein the condition control device (40) is configured to carry out the image-based condition control depending on at least one trigger criterion.

17. Device (10) or system (110; 120) according to one of the preceding claims, wherein the device (10) or the system (110; 120) comprises at least one processing circuit (21; 111) which is configured to carry out the sample analysis based on the detected optical measurement variable.

18. A method for sample analysis of a sample (19; 132) comprising eukaryotic material (19) and / or a soil sample (132), the method comprising: Detecting at least one optical measurement variable on the sample (19; 132) by a detection device (30), wherein the optical measurement variable can be evaluated for the condition analysis of the sample, and Checking a state of the detection device (30) by a state control device (40) which has an image recording device (41) for an image-based state control of at least one component (27, 31) of the detection device (30).

Images