Stoichiometric model selection by image analysis

Abstract

A method of obtaining chemical composition information of at least one sample (112) by spectral measurement is proposed. The method comprises the following steps: i. Acquiring spectral data of the sample (112) within at least one spatial measurement range (118) of a spectrometer device (114) by using at least one spectrometer device (114); ii. Acquiring image data of a scene (136) within a field of view (138) of the imaging device (134) by using at least one imaging device (134), the scene (136) comprising at least a portion of the sample (112) and at least a portion of a spatial measurement range (118) of the spectrometer device (114), and analyzing the image data and / or the spectral data by using at least one processing device (154), determining at least one item of identification information about the sample (112); and iii. Evaluating the spectroscopic data by using the processing device (154) to obtain at least one item of sample information, where the evaluating comprises processing the spectroscopic data with at least one stoichiometric model, where the stoichiometric model converts the spectroscopic data into chemical composition information, where the chemical composition information comprises at least one sample information. The stoichiometric model is selected according to the identification information.

Description

Technical Field

[0001] The present invention relates to a method for obtaining chemical composition information of at least one sample by spectroscopic measurement and a system for obtaining chemical composition information of at least one sample by spectroscopic measurement. The present invention further relates to a computer program, a computer-readable storage medium, and a non-transitory computer-readable medium. The method and apparatus can be used in particular for obtaining chemical information of a sample, particularly information regarding its chemical composition, and can be used in particular for analyzing inhomogeneous samples. Background Art

[0002] Spectroscopic methods are widely used in research, industry, and consumer applications for a variety of applications, such as optical analysis and / or quality control. For example, use cases can be found in areas such as food production and quality control, agriculture, pharmaceuticals, medical applications, and life sciences. A variety of methods, such as photometry, absorptiometry, fluorescence, and Raman spectroscopy, are available for qualitative and / or quantitative sample analysis. These methods typically involve acquiring spectroscopic data from a sample (also called a specimen) using at least one spectrometer device, which may specifically include at least one wavelength-selective element and at least one detector device.

[0003] In particular, spectroscopic methods (such as near-infrared (NIR) spectroscopy) and chemometric methods can be applied to determine the chemical composition of a sample. In particular, such samples may be inhomogeneous, and their chemical composition may depend significantly on the exact location within the sample. Examples of inhomogeneous samples may include food products, such as fruits and / or vegetables.

[0004] To determine the chemical composition of an object using IR spectroscopy, the IR spectrum of a representative sample can be recorded and processed using a chemometric model. The chemometric model converts the spectral information into chemical composition information. An exemplary chemometric model is described in Celio Pasquini's "Near Infrared Spectroscopy: Fundamentals, Practical Aspects and Analytical Applications," J. Braz. Chem. Soc., Vol. 14, No. 2, pp. 198-219, 2003.

[0005] Typically, each sample requires a separate chemometric model. For example, the determination of starch in wheat and corn requires different chemometric models. Therefore, to perform a measurement, the user must select the type of sample to be measured so that the appropriate chemometric model can be chosen. This approach may be suitable for homogeneous samples such as flowers or oil. However, if many chemometric models are available, it would be more convenient for the user to not have to select a chemometric model, or at least to be able to choose only from a pre-selected list.

[0006] However, the situation is even more challenging for inhomogeneous samples. Samples may contain distinct clumps, such as wheat kernels. The situation becomes even more challenging when measuring a granular mixture, such as corn silage mixed with concentrate. The spectra of the two species are superimposed, and no chemometric model is available for the specific mixture.

[0007] Xu Junli et al.'s "Combining deep learning with chemometrics when it is really needed: A case of real-time object detection and spectral model application for spectral image processing," ANALYTICA CHIMICA ACTA, ELSEVIER, AMSTERDAM, NL, ELSEVIER, Amsterdam, The Netherlands, Vol. 1202, XP087004785, ISSN: 0003-2670, DOI: 10.1016 / J.ACA.2022.339668, describes the early days of deep learning (DL) in the chemometric field of spectral image processing. The paper describes combining DL with chemometrics to process spectral images, even with as few as <100 spectral images.

[0008] US 2013 / 080070 A1 describes systems and methods for identifying and selecting more accurate chemometric models for analyzing specific plant samples via near-infrared spectroscopy.

[0009] Zhu Hongyan et al., "Hyspectral Imaging for Predicting the Internal Quality of Kiwifruits based on Variable Selection Algorithms and Chemometric Models," Scientific Reports, Volume 7, Issue 1, August 10, 2017, XP093061375, DOI: 10.1038 / s41598-017-08509-6, https: / / www.nature.com / articles / s41598-017-08509-6, describes the feasibility and potential of using hyperspectral imaging combined with variable selection methods and calibration models to determine firmness, soluble solids content (SSC), and pH in kiwifruit. Images were acquired using a pushbroom hyperspectral reflectance imaging system covering two spectral ranges. The weighted regression coefficient (BW), successive projections algorithm (SPA), and genetic algorithm-partial least squares (GAPLS) are compared and the selection of effective wavelength is evaluated. In addition, multiple linear regression (MLR), partial least squares regression, and least squares support vector machine (LS-SVM) are described to quantitatively predict quality attributes using effective wavelength.

[0010] Rahman Anisur et al., "Hyperspectral imaging for predicting the allicinand soluble solid content of garlic with variable selection algorithms and chemometric models: Predicting the allicin and soluble solid content of garlic," JOURNAL OF THE SCIENCE OF FOOD ANDAGRICULTURE, volume 98, issue 12, May 14, 2018, pp. 4715–4725, XP093061381, GB, ISSN: 0022-5142, DOI: 10.1002 / jsfa.9006, https: / / api.wiley.com / onlinelibrary / tdm / v1 / articles / 10.1002%2Fjsfa. 9006 describes the acquisition of hyperspectral images of 100 garlic cloves, covering two spectral ranges, from which the average spectrum of each clove was extracted. Calibration models included partial least squares (PLS) and least squares support vector machine (LS-SVM) regression, as well as different spectral preprocessing techniques, from which the highest-performing spectral preprocessing technique and spectral range were selected. Variable selection methods, such as regression coefficients, variable importance in projection (VIP), and the successive projection algorithm (SPA), were evaluated to select effective wavelengths (EWs). Furthermore, PLS and LS-SVM regression methods were applied to quantitatively predict garlic quality attributes using the selected EWs.

[0011] Problem to be solved

[0012] Therefore, it is desirable to provide an apparatus and method that solves the above technical challenges in the field of spectroscopic sample analysis. In particular, an apparatus and method should be provided that allows obtaining accurate spectroscopic data of a sample by taking into account possible local variations and inhomogeneities of the sample. Summary of the Invention

[0013] This problem is solved by a method for obtaining information on the chemical composition of at least one sample by spectroscopic measurements, a system for obtaining information on the chemical composition of at least one sample by spectroscopic measurements, a computer program, and a computer-readable storage medium having the features of the independent claims. Advantageous embodiments are listed in the dependent claims and throughout the description, which can be implemented individually or in any arbitrary combination.

[0014] In a first aspect of the present invention, a method for obtaining chemical composition information of at least one sample by spectroscopic measurement is disclosed. The method comprises the following method steps, which may be performed in the order given. However, a different order is also possible. The method may further comprise additional method steps not listed. Furthermore, one or more, or even all, of the method steps may be performed only once or repeatedly.

[0015] The method comprises the following steps:

[0016] i. acquiring spectral data of the sample by using at least one spectrometer device within at least one spatial measurement range of the spectrometer device;

[0017] ii. acquiring, using at least one imaging device, image data of a scene within a field of view of the imaging device, the scene including at least a portion of the sample and at least a portion of a spatial measurement range of the spectrometer device, and analyzing the image data and / or the spectral data using at least one processing device to determine at least one identification information about the sample;

[0018] iii. evaluating the spectral data using the processing device to obtain at least one item of sample information, wherein the evaluating includes processing the spectral data using at least one chemometric model, wherein the chemometric model converts the spectral data into chemical composition information, wherein the chemometric model is selected based on the identification information.

[0019] As used herein, the term "spectroscopy" is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art, and is not limited to a specific or customized meaning. The term may specifically refer to, but is not limited to, acquiring spectral data about at least one sample. The spectral data may specifically be acquired using at least one spectrometer device. As part of the spectroscopy, the sample may be irradiated with electromagnetic radiation within the infrared spectral range, specifically the near-infrared spectral range. In particular, the electromagnetic radiation may be within the wavelength range of 760 nm to 1000 µm, specifically within the wavelength range of 760 nm to 15 µm, more specifically within the wavelength range of 1 µm to 5 µm, and even more specifically within the wavelength range of 1 µm to 3 µm. Electromagnetic radiation may also be referred to as light, and therefore the two terms are used interchangeably in this document. The spectroscopy may further include receiving incident light after interaction with the sample and generating at least one corresponding signal, which may form part of the spectral data. The spectral data may include information about at least one optical property or optically measurable property of the sample for one or more different wavelengths, the information determined as a function of wavelength. More specifically, spectral data can relate to at least one property characterizing at least one of the transmission, absorption, reflection, and emission of a sample. The at least one optical property can be determined for one or more wavelengths. The spectral data can specifically take the form of signal intensity determined based on the wavelength of the spectrum or a subregion thereof (e.g., a wavelength bin), wherein the signal intensity can preferably be provided as an electrical signal that can be used for further evaluation. Thus, spectral data can be generated as part of a spectral measurement.

[0020] In addition to spectroscopic measurements, the present invention also proposes the use of image data. Combining spectroscopic measurements with image data can allow obtaining accurate chemical composition information of the sample, in particular even in the case of mixtures, local variations and inhomogeneities of the sample.

[0021] As used herein, the term "acquiring spectral data" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art and is not limited to a special or customary meaning. The term may specifically refer to, but is not limited to, any process by which spectrometer equipment at least one of captures, records, and stores spectral data, for example, by measuring at least one of transmission, absorption, reflection, and emission of a sample as a function of wavelength for one or more different wavelengths.

[0022] Spectral measurements, in particular in the infrared spectral range, can be spot measurements. The spot can include an area with any geometric shape. For example, the spot can include an area from 1 mm to 500 mm. 2Image data, particularly images, acquired using an imaging device can cover a larger area than spectral measurements, for example, from 0.01 m to 10 m. Spectral measurements and image data acquisition can occur at different times, T1 and T2. The mixing ratio can be constant at T1 and T2.

[0023] As used herein, the term "spectrometer device" is a broad term and is to be given its ordinary and customary meaning for persons of ordinary skill in the art and is not limited to a special or customary meaning. The term may specifically refer to, but is not limited to, a device configured to acquire spectral data of at least one sample within at least one spatial measurement range. The spectrometer device used in step i. may specifically be a near-infrared spectrometer device. The spectrometer device may specifically be configured to detect electromagnetic radiation in the near-infrared range. The spectrometer device may be configured to perform at least one spectral measurement on the sample. The spectrometer device may specifically include at least one detector device comprising at least one optical element and a plurality of photosensitive elements. The optical element may specifically be configured to separate incident light (specifically, electromagnetic radiation in the near-infrared range) into a spectrum composed of component wavelength components. Each photosensitive element may be configured to receive at least a portion of one of the component wavelength components and to generate a corresponding detector signal based on illumination of the corresponding photosensitive element by the at least a portion of the corresponding component wavelength component. The detector signal, specifically the signal intensity, together with the corresponding wavelength, may form part of the spectral data. The spectrometer device may be or may include a dispersive spectrometer device that can analyze radiation from a sample irradiated with broadband illumination. However, in addition or alternatively, other configurations and / or arrangements of the spectrometer device are possible. As an example, the sample can be irradiated with light of a limited number of different wavelengths, and the spectrometer device can include a broadband detector. In particular, the spectrometer device can be a Fourier transform spectrometer, in particular a Fourier transform infrared spectrometer. Thus, a narrowband light source, such as at least one light emitting diode (LED) and / or at least one laser, can be used to illuminate the sample. In particular, the spectrometer device can be configured to determine a spectrum by measuring and processing an interferogram, in particular by applying at least one Fourier transform to the measured interferogram.

[0024] The spectrometer device can be embodied as a portable spectrometer device. Specifically, the spectrometer device can be part of a mobile device (such as a laptop computer, a tablet computer, or specifically a cellular phone such as a smartphone). Additionally or alternatively, the mobile device can be or include a smartwatch and / or a wearable computer (also referred to as a wearable device), such as a body-mounted computer. Other mobile devices are also possible. The spectrometer device can be at least one of integrated into the mobile device or attached to the mobile device.

[0025] As used herein, the term "spatial measurement range" is a broad term and is given its ordinary and customary meaning to a person of ordinary skill in the art and is not limited to a special or customized meaning. The term may specifically refer to, but is not limited to, a spatially confined segment that can be spectrally inspected by a spectrometer device. As an example, the spatial measurement range may be defined as a solid angle or a three-dimensional angular segment in space, wherein a sample arranged within the solid angle or angular segment can be analyzed by a spectrometer device. As an example, the solid angle or angular segment may be defined by the geometric and / or optical properties of the spectrometer device. Thus, the spatial measurement range may be the field of view of the spectrometer device, within which spectral measurements may be performed. Samples positioned within the spatial measurement range may be spectrally analyzed by the spectrometer device. Specifically, the spectrometer device may be configured to acquire spectral data based on incident light from within the spatial measurement range. The spatial measurement range may in particular be a three-dimensional spatial segment, for example a three-dimensional space, such as a conical spatial segment, the light content of which may be received and analyzed by the spectrometer device. The spectral data acquired by the spectrometer device may include information related to at least one sample located within a spatial measurement range of the spectrometer device. Specifically, to perform spectral analysis on the sample, the spectrometer device may be positioned in close proximity to the sample such that the spatial measurement range at least partially includes the sample, for example, at a distance in the range of 0 mm to 100 mm, specifically in the range of 0 mm to 15 mm, from the object.

[0026] As used herein, the term "sample" (also referred to herein as "object") is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not limited to a special or custom meaning. The term may specifically refer to, but is not limited to, any object, such as an animate or inanimate object, that can be imaged by the imaging device and spectrally examined by the spectrometer device. The sample may be a single-component sample or a mixture comprising at least two components. In particular, the sample may be an inhomogeneous sample, for example, a sample whose chemical composition may vary within the sample, for example in a position-dependent manner. However, other samples, particularly homogeneous samples whose chemical composition varies only slightly or not at all, are also feasible. The sample may be any material that shows NIR activity. In this case, a chemometric model may be obtained, as described in more detail below. If no chemometric model is available for a certain material, the method may include selecting a model for another material that is most similar to the material.

[0027] The sample may be a solid sample, such as a powder or solid. However, other samples are also possible, such as a liquid sample.

[0028] The sample may specifically be or include food, such as fruit or vegetables. For example, the sample may be or include at least one element selected from the group consisting of: food; feed; vegetables, such as tomatoes; fruits, such as apples and pears; crops, such as wheat and corn; and waste, such as recycling waste. For example, the sample may specifically be or include a body part, such as skin.

[0029] For example, the method can be used for recycling. The sample can include waste materials, such as plastic. The method can include obtaining at least one piece of identification information about the sample using the image data, for example, at least one piece of identification information related to the sample shape, such as one or more of particle size, color, and surface roughness. This identification information can be used to select a chemometric model for analyzing the spectral data.

[0030] As used herein, the term "imaging device" is a broad term and is to be given its ordinary and customary meaning to those of ordinary skill in the art and is not limited to a special or customary meaning. The term may specifically refer to, but is not limited to, any device configured to record or capture image data and / or capture 2D or 3D spatial information about at least one sample and / or scene. An imaging device may be or include at least one camera having one or more imaging sensors, specifically one or more CCD or CMOS imaging sensors, for acquiring image data. A camera may specifically include at least one camera chip, such as at least one CCD chip and / or at least one CMOS chip configured to record an image. A camera may include a one-dimensional or two-dimensional array of imaging sensors (e.g., pixels), which may be arranged, for example, on a camera chip. By way of example, a camera may include at least 100 pixels in at least one dimension, such as at least 100 pixels in each dimension. By way of example, a camera may include an imaging sensor array comprising at least 100 imaging sensors in each dimension, specifically at least 300 imaging sensors in each dimension. For example, the camera may be a color camera comprising color pixels, each color pixel comprising at least three color sub-pixels sensitive to different colors. For example, the camera may include monochrome pixels and / or color pixels. Color pixels and monochrome pixels may be combined within the camera. The camera may be a camera for a mobile device. The present invention is particularly applicable to cameras typically used in mobile devices, such as laptops, tablet computers, or, more specifically, cell phones such as smartphones. Specifically, the camera may be part of a mobile device that, in addition to at least one camera, includes one or more data processing devices, such as one or more processors. Specifically, the mobile device may have at least one function other than spectroscopy, such as mobile communication, for example, cell phone functionality. However, other camera types are also possible. As mentioned above, the spectrometer device may also be part of a mobile device. In particular, both the camera and the spectrometer device may be part of a mobile device, particularly a smartphone. In addition to at least one camera chip or imaging chip, the camera may include additional components, such as one or more optical elements, for example, one or more lenses. As an example, the camera may be a fixed-focus camera, with at least one lens having a fixed adjustment relative to the camera. Alternatively, however, the camera may include one or more variable lenses that can be adjusted automatically or manually.

[0031] Alternatively or additionally, the imaging device may be or may include at least one LIDAR-based imaging device, where LIDAR stands for Light Detection and Ranging or Light Imaging, Detection and Ranging. The LIDAR-based imaging device may include at least one laser source, for example at least one tunable laser diode, to illuminate the object or at least a portion of the object. The LIDAR-based imaging device may further include at least one positioning unit configured to determine at least one distance of the illuminated portion of the object from the imaging device and / or from at least one further point or position in space. The positioning unit may in particular include at least one sensor element, for example a photodiode, configured to detect at least one laser beam emitted from the laser source and reflected by the object. Determining the distance and thus generating image data may include processing the light beam reflected by the object and / or at least one reference beam and / or a corresponding signal detected by the at least one sensor element.

[0032] As used herein, the term "image data" is a broad term and is to be given its ordinary and customary meaning to those of ordinary skill in the art and is not limited to a special or custom meaning. The term may specifically refer to, but is not limited to, spatially resolved one-dimensional, two-dimensional, or even three-dimensional optical information. Image data may include a plurality of electronic readings from an imaging device, such as from an imaging sensor (e.g., pixels of a camera chip) and / or from sensor elements of a LIDAR-based imaging device. In particular, the image data may include a plurality of numerical values ​​corresponding to the electronic readings from the imaging device. The electronic readings, in particular the numerical values, may relate to at least one optical property of at least one object within the imaging device's field of view. The image data may include at least one array of information values ​​(e.g., grayscale values ​​and / or color information values). Alternatively or additionally, the information values ​​included in the image data may include distance values, each distance value indicating the distance between a portion of an object and at least one reference point (e.g., an imaging device, in particular a LIDAR-based imaging device).

[0033] As used herein, the term "acquiring image data" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art and is not limited to a special or customary meaning. The term may specifically refer to, but is not limited to, any process by which image data is captured or recorded by an imaging device, particularly a camera, for example, in the form of electronic readouts generated by an imaging sensor in response to illumination.

[0034] As used herein, the term "field of view" is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art and is not limited to a special or customized meaning. The term can specifically refer to, but is not limited to, a spatially confined segment whose content can be imaged by an imaging device. In particular, image data generated by an imaging device can include spatially resolved optical information related to objects located within the field of view of the imaging device. The field of view can specifically be a three-dimensional spatial segment accessible to the imaging device. In particular, a scene included in the field of view can be imaged by the imaging device.

[0035] The term "scene" as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not limited to a special or custom meaning. The term may specifically refer to, but is not limited to, the optical content of the field of view of an imaging device. In particular, the scene may include one or more samples, such as the samples mentioned with respect to step i. above, wherein at least one sample in the scene may be imaged by the imaging device. In particular, the scene may include a plurality of objects having a specific arrangement, wherein these objects and their arrangement may be imaged by the imaging device, thereby generating at least one image. As part of method step ii., image data of a scene within the field of view of the imaging device is acquired, the scene including at least a portion of the sample and at least a portion of the spatial measurement range of the spectrometer device. The sample in step i. may be at least partially visible in the image data in step ii. Therefore, the field of view of the imaging device may at least partially overlap with the spatial measurement range of the spectrometer device.

[0036] The spatial relationship between the field of view of the imaging device and the spatial measurement range of the spectrometer device can be known and can be used, for example, in step iii., such as an offset between the field of view of the imaging device and the spatial measurement range of the spectrometer device and / or at least one angle between the field of view of the imaging device and the spatial measurement range of the spectrometer device. Positions and / or objects within the field of view of the imaging device can also be located within the spatial measurement range of the spectrometer device, or vice versa. Specifically, the sample or at least a portion thereof can be located in both the field of view of the imaging device and the spatial measurement range of the spectrometer device. Therefore, the sample or at least a portion thereof can be spectrally inspected by the spectrometer device and at least partially imaged by the imaging device.

[0037] As used herein, the term "identification information" is a broad term and will be given its ordinary and conventional meaning for those of ordinary skill in the art and is not limited to a special or custom meaning. The term specifically may refer to, but is not limited to, any information suitable for identifying a sample obtained from image data and / or spectral data. The identification information may include at least one of the following: at least one information about the degree of inhomogeneity, at least one information about the components of the sample, at least one information about a mixture of these components, at least one information about the mixing ratio of these components. The identification information may be or may include at least one information about at least one of the following: the type of sample, the boundaries of the sample within the scene, the size of the sample, the orientation of the sample. A variety of identification information can be obtained from image data and / or spectral data.

[0038] For example, the identification information may include information about one or more of a crop species, a crop variety, a food or feed type, waste or a waste type (eg, waste in recycling). However, other applications are possible.

[0039] For example, the method can include applying at least one spectral analysis to the spectral data to derive at least one identification information from the spectral data. As an example, the spectral analysis can include analyzing the spectral data to determine at least one peak within the spectral data, the at least one peak reflecting a global or local maximum in transmission, absorption, reflection, and / or emission of the sample. The spectral analysis can further include identifying at least one corresponding wavelength. Furthermore, the spectral analysis can include determining the chemical composition of the object, for example, by comparing the identified peak to at least one predetermined peak or at least one set of predetermined peaks. The spectral analysis of the spectral data can specifically be performed using at least one spectral evaluation algorithm.

[0040] For example, the method may include applying at least one sample identification algorithm to the image data to obtain at least one identification information from the image data. The term "image" as used herein is a broad term and will be given its ordinary and conventional meaning for a person of ordinary skill in the art and is not limited to a special or custom meaning. The term may specifically refer to, but is not limited to, any representation of at least one optically detectable characteristic of the sample, such as a one-dimensional, two-dimensional or three-dimensional representation. In particular, the image may include a graphical representation of a scene within the field of view of an imaging device. The image may specifically be displayed on, for example, a display device, such as a screen of a mobile device (for example, a mobile device that may include an imaging device). The image may specifically include the image data mentioned in step ii. or a portion thereof, and / or may be obtained from the image data or a portion thereof. The image may specifically represent at least one visual characteristic of the sample. The image data may include information of at least one of the following:

[0041] - at least one image obtained from the image data in step ii.;

[0042] at least one item of spatial information about a spatial measurement range within the scene, in particular an indication of the spatial measurement range within the image at which the spectral data were acquired;

[0043] - at least one piece of information about: the type of the object, the boundaries of the object within the scene, the size of the object, the orientation of the object, the color of the object, the texture of the object, the shape of the object, the contrast of the object, the volume of the object, the region of interest of the object;

[0044] at least one item of orientation information about the at least one object, in particular an indication of the orientation of the spectrometer device relative to the at least one object;

[0045] at least one item of direction information, in particular an indication of the direction between the spectrometer device and the at least one object;

[0046] - at least one item of similarity information about the object, in particular similarity information about at least one shared characteristic shared between different regions of the object.

[0047] As used herein, the term "analyzing image data" is a broad term and is to be given its ordinary and customary meaning to those skilled in the art and is not limited to a specific or customary meaning. Specifically, but not limited to, determining at least one item of identification information based at least in part on the image data acquired by the imaging device in step ii. For example, image analysis can be performed on a photograph of the sample to be measured. This analysis can yield one or more of the sample's type, its degree of heterogeneity, and (in the case of a mixture) the mixture's components and their mixing ratio.

[0048] The identification information can be obtained in particular by using at least one sample recognition algorithm (such as an image recognition algorithm and / or a trained model configured to recognize or distinguish samples, for example by using artificial intelligence (such as an artificial neural network). The sample recognition algorithm can specifically include at least one sample recognition algorithm for determining the type of at least one sample. For example, the sample recognition algorithm can identify the type of the sample, such as the class or type of object, such as whether the sample is an apple, an orange, another type of fruit or vegetable, or a part of the human body (such as a hand or face). Other types of samples are also possible, in particular other types of food samples.

[0049] As an example, the image may include information regarding the locations at which the spectral data was acquired and / or the results of an evaluation of the spectral data, such as compositional information derived from the spectral data. Thus, the image may visually indicate a scene or a portion thereof, as well as information derived from the spectral data acquired in step i., optionally with positional information regarding the locations at which the information was acquired. Thus, the image may include an overlap between a sample visible in the scene and one or more locations at which one or more spectral measurements were performed, including the results of the spectral measurements and / or one or more items of information derived from the spectral measurements.

[0050] Identification information can be determined using user selection and / or user feedback. For example, user selection and / or user feedback can be performed using at least one user interface. As used herein, the term "user interface" is a broad term and is to be given its ordinary and conventional meaning to those of ordinary skill in the art and is not limited to a specific or custom meaning. The term may refer to, but is not limited to, an element or device configured to interact with its environment (e.g., for the purpose of exchanging information in a one-way or two-way manner, such as for exchanging one or more of data or commands). For example, a user interface can be configured to share information with a user and receive information provided by the user. A user interface can be a feature that interacts with the user visually (e.g., a display) or a feature that interacts with the user auditorily. As examples, a user interface can include one or more of the following: a human-computer interface (e.g., a display, screen, keyboard, voice interface, touchpad, or port through which a user can provide digital information data), a graphical user interface, or a data interface, such as a wireless and / or wired data interface.

[0051] The identification information may include information about at least one region of interest of the sample. The region of interest can be identified, for example, using an image recognition algorithm and / or a trained model. As an example, the region of interest may be or may include irregularities and / or unexpected features. Other regions of interest are also possible. The region of interest may be, for example, a mole on a section of human skin (e.g., a hand or leg). The method may include providing a user with at least one piece of guidance information indicating the region of interest, for example on a display of a mobile device. The guidance information may specifically prompt the user to perform step i. of the method on the region of interest. Using a dedicated spectral evaluation algorithm, the user may be provided with specific information about the region of interest, such as medical information and / or medical guidance, such as information on cancer diagnosis of a mole.

[0052] The identification information may include at least one item of similarity information about the sample, specifically, similarity information about at least one shared characteristic shared between different regions of the sample. In particular, the identification information may include information about different regions of the sample that share at least one common characteristic. This characteristic may be a quality identified in the image data (particularly in the image). The shared characteristic may be, for example, a common color shared between different regions of an object, while other regions of the object appear to have a different color. The shared characteristic identified in the image data (e.g., similar image information) may imply shared and / or similar spectral data (e.g., similar spectral information). The method may include predicting spectral data and / or at least spectrally derivable characteristics for regions of the object that are similar to each other with respect to the at least one characteristic of the image data. The method may further include reviewing and / or refining the prediction, for example, by instructing a user to obtain spectral data for other regions that share the shared characteristic. As an example, the sample may be an apple comprising regions of different colors. As part of the method, regions that share a red color may be identified as similar information. Spectral data acquired for one of these regions may indicate a specific sugar content, for example, a sugar content that exceeds that of other regions of a different color (e.g., green). As part of the method, sugar content in other red regions may be predicted. Further, the user may be directed to obtain spectral data for other red regions to check and / or refine the predictions and / or possible additional predictions.

[0053] Between steps i. and ii., at least one of the scene, the field of view, the spatial measurement range, and the object may be modified. Thus, as an example, the scene may change, and / or at least one of the spectrometer device, the imaging device, and a device including both the spectrometer device and the imaging device (e.g., a mobile device) may be moved.

[0054] In particular, the method can generate at least one image of a scene, the at least one image having at least two items of spectral object information and corresponding spatial information about a spatial measurement range within the image for each item of spectral object information. Furthermore, the image obtained from the image data in step ii. can be an image obtained from the repeated image data of step ii., specifically at least one of a combined image and a selected image from the images obtained from the repeated image data of step ii.

[0055] As an example, as described above, the imaging device and / or spectrometer device can be moved between the optional repetitions of step i. and step ii., specifically during these optional repetitions. Specifically, during the initial execution of step ii., image data of a first scene can be acquired at a first distance, wherein, for the repetitions of step ii., the imaging device and / or spectrometer device can be moved closer to the sample so that the imaged scene is a subsection of the first scene. In particular, image data corresponding to a wide image can be acquired during the initial execution of step ii. The wide image can completely or almost completely include the object. For further repetitions of step ii., the distance between the imaging device and / or spectrometer device and the sample can be reduced to at least a second distance, wherein this second distance allows for acquisition of spectral data of the object by performing step i. The second distance can be in the range of 0 mm to 100 mm, specifically, in the range of 0 mm to 15 mm. The image obtained from the image data acquired at the second distance can show a subsection of the image obtained from the image data acquired during the initial execution of step ii. The method may further include tracking movement of the imaging device, such as movement from a first distance to at least one second distance, using the imaging device and motion tracking software. Specifically, a spatial relationship between image data and / or spectral data acquired at the at least one second distance and an image acquired at the first distance may be derived.

[0056] As another example, while performing one or more repetitions of step i. and step ii., the imaging device and / or spectrometer device can be moved over the sample, such as by a fixed distance and / or by a variable distance, for example, along a scan path. By performing step iii., at least one item of sample information can be obtained, wherein the sample information can include multiple items of chemical information corresponding to multiple locations along the scan path. Again, image data of the sample can be acquired, for example, in an initial execution of step ii., wherein the scan path can be included in an image obtained from the image data. In particular, the scan path and / or spectral object information, in particular chemical information, can be indicated in the image. This can allow the retrieval of chemical information along the scan path.

[0057] As generally used herein, the term "processing device" is a broad term and is to be given its ordinary and customary meaning to persons of ordinary skill in the art and is not limited to a special or customary meaning. Specifically, the term may refer to, but is not limited to, any logic circuitry configured to perform the basic operations of a computer or system, and / or generally refers to a device configured to perform computational or logical operations. In particular, a processing device may be configured to process the basic instructions that drive a computer or system. By way of example, a processing device may include at least one arithmetic logic unit (ALU), at least one floating point unit (FPU) (such as a math coprocessor or digital coprocessor), a plurality of registers (specifically, registers configured to supply operands to the ALU and store results of operations), and memory (such as L1 and L2 cache memory). In particular, the processing device may be a multi-core processor. In particular, the processing device may be or include a central processing unit (CPU). Additionally or alternatively, the processing device may be or include a microprocessor, and thus, in particular, the elements of the processing device may be contained on a single integrated circuit (IC) chip. Additionally or alternatively, the processing device may be or may include one or more application specific integrated circuits (ASICs) and / or one or more field programmable gate arrays (FPGAs), etc. The processing device may specifically be configured, such as by software programming, to perform one or more evaluation operations.

[0058] As described above, step iii. includes evaluating the spectral data using a processing device to obtain at least one item of sample information. The evaluation includes processing the spectral data using at least one chemometric model. The chemometric model converts the spectral data into chemical composition information. The chemometric model is selected based on the identification information.

[0059] Specifically, step iii. may include applying at least one spectral evaluation algorithm to the spectral data from step i., wherein the spectral evaluation algorithm is selected based on the identification information, specifically based on the type of the at least one object. Based on the identification information, a chemometric model can be automatically selected, thereby improving the user experience. In the case of a mixture, an appropriate chemometric model can be selected for each component. The mixing ratio can further be used to calculate any measurement of the mixture.

[0060] As used herein, the term "sample information" is a broad term and will be given its common and conventional meaning for those of ordinary skill in the art and is not limited to a special or custom meaning. The term can specifically refer to, but is not limited to, any information related to at least one characteristic of a sample, such as at least one of a chemical, physical, and biological characteristic, for example, the material and / or composition of the sample. Sample information can specifically be determined by considering spectral data of the sample and image data of the sample. Sample information can specifically relate to a characteristic that can vary within the sample, so the characteristic can be a characteristic of a specific location or spatial range within the sample. However, the characteristic can show no change or only slight change throughout the sample. Sample information can describe the characteristic in a qualitative and / or quantitative manner, for example, by describing it through one or more numerical values. Specifically, sample information can include chemical information of the sample, in particular its chemical composition. Sample information can include information about the characteristic as well as spatial information about the specific location or spatial range within the sample at which the characteristic is measured. For example, for wheat or corn, sample information can be one or more of the following: protein content, starch content, fiber content, dry matter content, etc. For example, for apples / tomatoes, the sample information may include one or more of the following: dry matter, sugar content, and acidity. For example, the method may be used for recycling. The sample may include waste materials, such as plastic. The method may include using the image data to obtain at least one item of identification information about the sample, for example, at least one item of identification information related to the sample's shape, such as one or more of particle size, color, and surface roughness. This identification information may be used to select a chemometric model for analyzing the spectral data.

[0061] As used herein, the term "obtaining at least one item of sample information" is a broad term and is to be given its ordinary and customary meaning to those of ordinary skill in the art and is not limited to a special or customary meaning. The term may specifically refer to, but is not limited to, any process for determining at least one item of sample information. Specifically, to determine the sample information, the spectral data in step i. and the image data in step ii. may be considered.

[0062] As used herein, the terms "evaluate spectral data" and "evaluate image information" are broad terms and are to be given their ordinary and customary meanings to those of ordinary skill in the art and are not limited to special or customary meanings. These terms may specifically refer to, but are not limited to, any process of analyzing data and information, respectively, such as by applying at least one analysis step (e.g., an analysis step comprising at least one analysis algorithm applied to the data and / or information). In particular, as part of the analysis step, the data or information may be processed and / or interpreted and / or evaluated, for example, by comparing the data or information, or at least a subset thereof, to at least one predetermined value or identifying at least one global or local maximum or minimum.

[0063] As used herein, the term "chemometric model" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art and is not limited to a special or customized meaning. The term may specifically refer to, but is not limited to, at least one model configured to convert spectral data into chemical composition information. A chemometric model may include mathematical and statistical techniques for extracting relevant information from spectral data. A chemometric model may include the use of one or more of multiple linear regression (MLR), principal component regression (PCR), and partial least squares regression (PLS), modified PLS such as di-PLS (e.g., as described in Ramin Nikzad-Langerodi, Werner Zellinger, Edwin Lughofer, and Susanne Saminger-Platz, Analytical Chemistry 2018 90(11), 6693-6701 DOI: 10.1021 / acs.analchem.8b00498), artificial neural networks (ANNs), support vector machines, random forests, (extreme) gradient boosting regression, and / or classification, among others. The chemometric model can be designed as described in Celio Pasquini, "Near Infrared Spectroscopy: Fundamentals, Practical Aspects and Analytical Applications", J. Braz. Chem. Soc. [Journal of the Brazilian Chemical Society], Vol. 14, No. 2, pp. 198-219, 2003.

[0064] The method may be at least partially computer-implemented, specifically step iii. The computer-implemented steps and / or aspects of the present invention may be performed in particular by using a computer or a computer network. As an example, step iii. of the method may be fully or partially computer-implemented. The evaluation of the spectral data is performed using at least one chemometric model. In addition, as described above, the analysis of the image data may include, for example, analyzing the image information using at least one recognition algorithm. The evaluated spectral data and the analyzed image data may be combined or connected, for example, in a predetermined manner and / or according to a predetermined algorithm to obtain at least one item of sample information. The chemometric model may particularly include at least one trained model. The term "trained model" as used herein is a broad term and will be given its common and conventional meaning for those of ordinary skill in the art and is not limited to a special or custom meaning. The term may specifically refer to, but is not limited to, a mathematical model trained on at least one training data set using one or more of machine learning, deep learning, neural networks, or other forms of artificial intelligence.

[0065] As used herein, the term "chemical composition information" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art and is not limited to a special or customized meaning. The term may specifically refer to, but is not limited to, information regarding one or more of the following: the components of a mixture, the mixing ratio, the presence or absence of at least one chemical component, etc.

[0066] The method may include providing at least one stoichiometric model. The method may include providing multiple stoichiometric models for different identification information. As used herein, the term "providing" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art and is not limited to a special or customized meaning. The term may specifically refer to, but is not limited to, obtaining and / or determining and / or selecting a stoichiometric model.

[0067] As used herein, the term "obtain" is a broad term and is to be given its ordinary and conventional meaning to one of ordinary skill in the art, and is not limited to a special or custom meaning. The term specifically refers to, but is not limited to, the process by which a system (specifically, a computer system) generates and / or obtains data from any data source, such as from a data storage device, from a network, or from another computer or computer system. Specifically, obtaining can occur via at least one computer interface, such as a port (such as a serial or parallel port). Obtaining can include several substeps, such as obtaining one or more primary information and generating secondary information by utilizing the primary information (such as by applying one or more algorithms to the primary information, e.g., using a processor). For example, providing can include obtaining a stoichiometric model from at least one database (e.g., from the cloud). As used herein, the term "database" is a broad term and is to be given its ordinary and conventional meaning to one of ordinary skill in the art, and is not limited to a special or custom meaning. The term specifically refers to, but is not limited to, any collection and / or accumulation of information. Specifically, a database can be or include an accumulation of information that can be sorted and / or linked to facilitate searching for information within the database. The term "ranking" may generally refer to referencing and / or cross-referencing information. The database may include at least one lookup table and / or at least one spreadsheet and / or at least one electronic management chart. The database may include multiple chemometric models. The chemometric models may be accessed via the cloud and may be selected based on the identification information.

[0068] The method may include at least one calibration step. The calibration step includes generating a chemometric model. The calibration step may be performed for each expected single component of the sample.

[0069] The stoichiometric model can be selected automatically. As used herein, the term "automatically" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art and is not limited to a special or customized meaning. Specifically, the term can refer to, but is not limited to, a process that is performed entirely by at least one computer and / or computer network and / or machine, particularly without manual action and / or user interaction.

[0070] In the case of mixtures, a stoichiometric model can be selected for each component.

[0071] In the case of a mixture, the image data may include information about the mixture ratio. The mixture ratio can be used to determine sample information of the mixture in the following ways:

[0072] - polymerizing these stoichiometric models according to the mixing ratio, wherein the polymerizing comprises adding and weighing according to the mixing ratio, and / or

[0073] - using the mixing ratio as an input parameter of at least one forward model.

[0074] A forward model can be designed to use input parameters and predict and / or simulate spectra. For example, a forward model can use the mixing ratio as an input parameter.

[0075] The method may include, if no chemometric model is available for the determined identification information, selecting a chemometric model for identification information similar to the determined identification information.

[0076] Step iii) of the method may further include considering at least one additional sensor information when obtaining the at least one item of sample information. The additional sensor information may, for example, include gyroscope information and / or GPS information. The additional sensor (specifically, the gyroscope) may be part of the mobile device. Additionally or alternatively, the additional sensor information, such as GPS information, may be provided by the mobile device. The additional sensor information may be considered, for example, by inspecting, verifying, or evaluating image data.

[0077] The method may further include providing at least one output based on the sample information and / or the selected chemometric model using at least one user interface. The method may further include providing at least one item of sample information about the at least one sample, specifically by optically providing the at least one item of sample information about the at least one sample via a display device. Specifically, the sample information may be displayed, for example, on a display device, such as a screen of a mobile device (e.g., a mobile device that may include an imaging device and / or a spectrometer device).

[0078] For example, the method may be executed as follows:

[0079] In the first step, a chemometric model can be generated. For example, to determine the acidity in apples or the sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. Next, an imaging device can capture an image of the sample, for example using a camera. Image recognition can be performed on the sample. In a parallel series of steps, a spectrometer device can perform IR spectroscopy on the same sample. Spectral data for the sample can be obtained from the IR spectroscopy. Next, identification of one or more of the crop species, crop variety, or food / feed type can be completed based solely on the image recognition results. Information about one or more of the crop species, crop variety, or food / feed type can be used to select a chemometric model. This selected chemometric model can then be applied to the spectral data. Finally, based on the selected chemometric model, sample information specific to the (single-component) sample can be obtained. For example, for wheat or corn, this sample information could include one or more of the following: protein content, starch content, fiber content, dry matter content, etc. For apples / tomatoes, this sample information could include one or more of the following: dry matter content, sugar content, and acidity.

[0080] For example, the method may be executed as follows:

[0081] A chemometric model can be generated in a separate calibration step. For example, in the case of acidity in apples or sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. Next, an imaging device can capture an image of the sample, for example using a camera. Image recognition can be performed on the sample. In a parallel series of steps, a spectrometer device can perform IR spectroscopy on the same sample. Spectral data for the sample can be obtained from the IR spectroscopy. Identification of one or more of the crop species, crop variety, or food / feed type can be accomplished based on the image recognition results and / or the IR spectrum. There is no specific order for performing IR spectroscopy or image recognition first. In some cases (for example, wheat or apples), crop species can be distinguished using IR. In other cases (apples or pears, both of which have similar IR spectra), image recognition can be used to distinguish crop species. Information about one or more of the crop species, crop variety, or food / feed type can be used to select a chemometric model. This selected chemometric model can then be applied to the spectral data. Finally, sample-specific information can be obtained based on the selected chemometric model.

[0082] For example, the method may be executed as follows:

[0083] A chemometric model can be generated in a separate calibration step. For example, in the case of acidity in apples or sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. Next, an imaging device can capture an image of the sample, for example using a camera. Image recognition can be performed on the sample. In a parallel series of steps, a spectrometer device can perform IR spectroscopy on the same sample. Spectral data for the sample can be obtained from the IR spectroscopy. Identification of one or more of the crop species, crop variety, or food or feed type can be performed first based on the image recognition results and / or IR spectrum, and then additionally based on user selection or user feedback. Specifically, when multiple options are available, especially where the crop species / variety is difficult to identify via image recognition and / or IR spectroscopy (for example, when the sample is a white powder that could be wheat flour or plastic powder; for example, apples and pears are quite similar), the user can input additional knowledge and select the optimal option. For example, the user can select "apple" instead of "pear" by clicking in the user interface. Information about one or more of the crop species, crop variety, and food or feed type can be used to select a chemometric model. This selected chemometric model can then be applied to the spectral data. Finally, also depending on the chosen stoichiometric model, sample-specific sample information can be obtained.

[0084] For example, the method may be executed as follows:

[0085] A chemometric model can be generated in a separate calibration step. For example, in the case of acidity in apples or sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. Calibration can be performed for individual components. An imaging device can capture an image of a sample, for example, using a camera. Next, image recognition can be performed on the sample. In a parallel series of steps, a spectrometer device performs IR spectroscopy on the same sample. Spectral data for the sample is obtained from the IR spectroscopy. Identification of one or more of the crop species, crop variety, or food / feed type can be performed based solely on the image recognition results. Simultaneously with or in conjunction with identifying the crop species, crop variety, or food / feed type, the mixing ratio can also be determined based on the image recognition results. A chemometric model can be selected based on one or more of the crop species, crop variety, or food / feed type. The chemometric model (data model) can be aggregated based on the mixing ratio information obtained from image recognition. For example, if the proportions of rye and wheat in a sample are 30% and 70%, respectively, the method can include aggregating individual models based on the mixing ratio by addition and weighing (e.g., a 30% rye chemometric model plus a 70% wheat chemometric model). For different species and / or materials, such as plastic and wheat, the method can include a "backward calculation" of the individual components within the mixture. Additionally or alternatively, a forward model can be used, for example, using the determined mixing ratio as an input parameter. Finally, depending on the selected stoichiometric model, sample-specific (mixture) information can be obtained. This sample information can include information about the mixing ratio. For different species and / or materials, such as plastic and wheat, the method can include a "reverse mixing of the model" (e.g., a reverse mixing of wheat and green wheat kernels with a pesticide).

[0086] For example, the method may be executed as follows:

[0087] In a separate calibration step, a chemometric model can be generated. For example, in the case of acidity in apples or sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. Calibration can be performed for a single component. The imaging device can, for example, capture an image of the sample via a camera. Next, image recognition can be performed on the sample. In a series of parallel steps, the spectrometer device performs IR spectroscopy on the same sample. Spectral data of the sample is obtained from the IR spectroscopy. The results of image recognition and / or the IR spectrum can be used to perform identification of one or more of the crop species, crop variety, or food or feed type. Whether IR spectroscopy or image recognition is performed first, no specific order is required. In some cases (e.g., wheat or apples), crop species can be distinguished via IR. In other cases (e.g., apples or pears, both of which have similar IR spectra), crop species can be distinguished via image recognition. Subsequent steps can be performed as described with respect to the previous example.

[0088] For example, the method may be executed as follows:

[0089] In a separate calibration step, a chemometric model can be generated. For example, in the case of acidity in apples or sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. Calibration can be performed for individual components. An imaging device can, for example, capture an image of a sample using a camera. Next, image recognition can be performed on the sample. In a parallel series of steps, a spectrometer device performs IR spectroscopy on the same sample. Spectral data for the sample is obtained from the IR spectroscopy. Image-based IR measurements of the mixture can be used to identify one or more of the following: crop species and / or variety. Specifically, identification of the crop species, crop variety, or food / feed type is performed first based on the results of image recognition and / or IR spectroscopy, and then (additionally) based on user selection or user feedback. In the case of multiple options, especially crop species / variety identification that may be difficult to perform via image recognition and / or IP spectroscopy (for example, if a sample contains white powder that could be wheat flour or plastic powder; for example, apples and pears are quite similar), the user can input additional knowledge and select the best option. For example, the user can select "apple" instead of "pear" by clicking in the user interface. Subsequent steps may be performed as described with respect to the previous example.

[0090] As mentioned above, in addition to spectroscopic measurements, the present invention also proposes the use of image data. Combining spectroscopic measurements with image data can allow accurate chemical composition information to be obtained, particularly even in the presence of mixtures, local variations in the sample, and inhomogeneities. This method can be used for a number of applications. The following illustrative examples are provided for illustrative purposes and should not be considered limiting in scope.

[0091] For example, in the case of a silage-concentrate mixture, sample A and sample B can be collected, for example, sample A is silage and sample B is concentrate. For example, a farmer wants to know the energy content of the feed that will be fed to the cows later. The farmer can optimize the silage-concentrate mixture. First, the energy content of component A and the energy content of component B can be measured. Therefore, the energy content of the final feed mixture can be predicted via calculation. However, it is usually impossible to accurately determine the energy content of the final feed mixture. The present invention can allow the determination of the mixing ratio determined via image recognition plus the protein content determined via IR spectroscopy.

[0092] For example, in the case of wheat-barley or wheat-rye mixtures, different stoichiometric models can be used. Image recognition can be used to determine the wheat-barley or wheat-rye ratio. For example, wheat and rye have different protein contents. Image recognition can be used to determine the mixing ratio and, therefore, the protein content. Image recognition can provide information on whether the mixture is wheat-barley, wheat-rye, or another mixture.

[0093] For example, for foods (e.g., placed on a plate), such as a vegetable stew or meat sauce mixture that cannot be separated, this method can measure the mixture. Thus, for example, the protein content of the dish can be determined not only as an average, but also as a total value. Specifically, for different apple species / varieties (e.g., for apple juice), a total value can be measured.

[0094] In another aspect of the present invention, a system for obtaining chemical composition information of at least one sample by spectroscopic measurement is disclosed. The system comprises:

[0095] I. at least one spectrometer device, the at least one spectrometer device being configured to acquire spectral data of the sample by using at least one spectrometer device within at least one spatial measurement range of the spectrometer device;

[0096] II. at least one imaging device configured to acquire image data of a scene within the field of view of the imaging device, the scene including at least a portion of the sample and at least a portion of the spatial measurement range of the spectrometer device;

[0097] III. At least one processing device configured to analyze the image data and / or the spectral data to determine at least one identification information about the sample, wherein the processing device is configured to evaluate the spectral data to obtain at least one item of sample information, wherein the evaluation includes processing the spectral data using at least one chemometric model, wherein the chemometric model is configured to convert the spectral data into chemical composition information, wherein the chemometric model is selected based on the identification information.

[0098] The system can be specifically used to perform the method for obtaining at least one item of sample information according to the present invention (e.g., according to any of the above embodiments and / or according to any of the embodiments described further below). Therefore, with respect to terms and definitions, reference can be made to the description of the method for obtaining at least one item of sample information as given above.

[0099] As used herein, the term "system" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art and is not limited to a special or customary meaning. The term may specifically refer to, but is not limited to, a group of interacting components or a collection of interacting components that can interact to achieve at least one common function. At least two components may be processed independently, or may be coupled or connectable.

[0100] The imaging device may include at least one camera having one or more imaging sensors, specifically one or more CCD or CMOS imaging sensors, for acquiring image data of the scene. The imaging device may specifically include a one-dimensional or two-dimensional array of imaging sensors (e.g., pixels), which may be arranged, for example, on a camera chip. Additionally or alternatively, the imaging device may be or include at least one LIDAR-based imaging device. A LIDAR-based imaging device may include at least one laser source for illuminating an object and at least one localization unit. The localization unit may include at least one sensor element configured to detect at least one laser beam emitted from the laser source and reflected by the object. The localization unit may be configured to determine at least one distance from the illuminated portion of the object to at least one reference point. Determining the distance and generating image data accordingly may include processing the beam reflected by the object and / or at least one reference beam and / or corresponding signals detected by the at least one sensor element. For further options and / or optional details, reference may be made to the description of the imaging device provided above.

[0101] The spectrometer device may include at least one detector device comprising at least one optical element and a plurality of photosensitive elements, wherein the at least one optical element is configured to separate incident light into a spectrum consisting of component wavelength components, wherein each photosensitive element is configured to receive at least a portion of one of the component wavelength components and to generate a corresponding detector signal based on the illumination of the corresponding photosensitive element by the at least a portion of the corresponding component wavelength component. Thus, the spectrometer device can analyze the incident light after it interacts with the object and generate at least one corresponding detector signal, which can form part of the spectral data. The optical element may include at least one wavelength-selective element. The wavelength-selective element may specifically be selected from the group consisting of: a prism; a grating; a linear gradient filter; an optical filter, specifically a narrow bandpass filter. The detector device may further include a plurality of photosensitive elements arranged in a linear array, wherein the photosensitive element array includes a number of 10 to 1000, specifically a number of 100 to 500, specifically a number of 200 to 300, and more specifically a number of 256 photosensitive elements. Each photosensitive element can be selected from the group consisting of: a pixelated inorganic camera element, particularly a pixelated inorganic camera chip, more particularly a CCD chip or a CMOS chip; a monochrome camera element, particularly a monochrome camera chip; and at least one photoconductor, particularly an inorganic photoconductor, more particularly an inorganic photoconductor comprising Si, PbS, PbSe, Ge, InGaAs, extended InGaAs, InSb, or HgCdTe. Each photosensitive element can be sensitive to electromagnetic radiation in the wavelength range from 760 nm to 1000 µm, particularly in the wavelength range from 760 nm to 15 µm, more particularly in the wavelength range from 1 µm to 5 µm, and even more particularly in the wavelength range from 1 µm to 3 µm. The spectrometer device can be or include a dispersive spectrometer device that can analyze the radiation profile of an object illuminated with broadband illumination, such as described above. However, other configurations and / or arrangements of the spectrometer device are also possible, which can particularly affect its components, such as the detector and / or illumination source used. As an example, the object can be illuminated with light of a limited number of different wavelengths. The spectrometer device can include a broadband detector. In particular, the spectrometer device can be a Fourier transform spectrometer, specifically a Fourier transform infrared spectrometer. Thus, the object can be illuminated using a narrowband light source, such as at least one light-emitting diode (LED) and / or at least one laser. Specifically, the spectrometer device can be configured to determine a spectrum by measuring and processing an interferogram, in particular by applying at least one Fourier transform to the measured interferogram.

[0102] The spectrometer device and the imaging device may have a known, specifically fixed, orientation relative to each other. In particular, the spectrometer device and the imaging device may have a known, specifically fixed, spatial relationship relative to each other. Furthermore, the spatial measurement range of the spectrometer device and the field of view of the imaging device may have a fixed spatial relationship relative to each other.

[0103] The system may further comprise at least one light source configured to emit electromagnetic radiation in a wavelength range from 760 nm to 1000 μm, in particular in a wavelength range from 760 nm to 15 μm, more particularly in a wavelength range from 1 μm to 5 μm, more particularly in a wavelength range from 1 μm to 3 μm. The spectrometer device may in particular be referred to as a "near-infrared spectrometer device".

[0104] The processing device of the system is configured to obtain at least one item of sample information about at least one sample. The system may include at least one display device configured to provide the at least one item of sample information about the at least one sample. The system may further include at least one mobile device, wherein the mobile device includes the at least one spectrometer device and the at least one imaging device. Thus, both the spectrometer device and the imaging device (such as at least one camera) may be integrated into a mobile device (such as a smartphone). As used herein, the term "mobile device" is a broad term and is to be given its ordinary and conventional meaning to those of ordinary skill in the art and is not limited to a specific or customary meaning. The term may specifically refer to, but is not limited to, a mobile electronic device, more specifically a mobile communication device such as a cellular phone or smartphone. Additionally or alternatively, a mobile device may also refer to a tablet computer or other type of portable computer with at least one camera. The mobile device may specifically include at least one display device, specifically a screen, configured to display object information.

[0105] The system may further include at least one control unit. As used herein, the term "control unit" is a broad term and is to be given its ordinary and customary meaning to those skilled in the art, and is not limited to a special or customary meaning. The term may specifically refer to, but is not limited to, a device or combination of devices capable of and / or configured to perform at least one computing operation and / or control at least one function of at least one other device (such as at least one other component of the system for obtaining at least one item of object information). The control unit may specifically control at least one function of a spectrometer device, such as the acquisition of spectral data. The control unit may specifically control at least one function of an imaging device, such as the acquisition of image data. The control unit may specifically control a processing device, such as the evaluation of spectral and / or image data. Specifically, the at least one control unit may be embodied as and / or may include at least one processor, wherein the processor may be configured to perform one or more operations, specifically through software programming. As used herein, the term "processor" is a broad term and is to be given its ordinary and customary meaning to those skilled in the art, and is not limited to a special or customary meaning. Specifically, the term may refer to, but is not limited to, any logic circuitry configured to perform the basic operations of a computer or system, and / or generally refers to a device configured to perform computational or logical operations. In particular, a processor may be configured to process the basic instructions that drive a computer or system. By way of example, a processor may include at least one arithmetic logic unit (ALU), at least one floating-point unit (FPU), such as a math coprocessor or digital coprocessor, multiple registers (specifically, registers configured to provide operands to the ALU and store operation results), and memory such as L1 and L2 cache memory. In particular, the processor may be a multi-core processor. In particular, the processor may be or include a central processing unit (CPU). Additionally or alternatively, the processor may be or include a microprocessor, and thus, in particular, the components of the processor may be contained within a single integrated circuit (IC) chip. Additionally or alternatively, the processor may be or include one or more application-specific integrated circuits (ASICs) and / or one or more field-programmable gate arrays (FPGAs) and / or one or more tensor processing units (TPUs) and / or one or more chips, such as specialized machine learning optimized chips. The processor may specifically be configured, such as by software programming, to control and / or perform one or more evaluation operations.

[0106] In another aspect, a computer program is disclosed. The computer program includes instructions that, when executed by a control unit of a system as disclosed herein (e.g., according to any of the aforementioned embodiments and / or any of the embodiments described in further detail below), cause the system to perform a method as disclosed herein (e.g., according to any of the aforementioned embodiments and / or any of the embodiments described in further detail below). Thus, by way of example, the computer program may cause the system or trigger the system to acquire spectral data using a spectrometer device according to step i., cause the system or trigger the system to acquire image data of a scene using an imaging device according to step ii., and provide instructions to the system to perform analysis and / or evaluation according to step iii. The computer program may specifically include instructions for causing the system to perform step iii. of the method. The computer program may further include instructions for causing the system to perform steps i. and ii. of the method based on its own motion or in response to at least one user action, such as initiating the acquisition of spectral data in step i. and / or the acquisition of image data in step ii., such as a user interaction such as pressing a start button. The computer program may also include instructions for causing the system or triggering the system to prompt the user to provide specific input. Thus, as an example, the user may be prompted to start acquiring spectral data in step i. and / or may be prompted to start acquiring image data in step ii. In particular, the computer program may be stored on a computer-readable data carrier and / or a computer-readable storage medium.

[0107] On the other hand, a computer-readable storage medium is disclosed, which includes instructions that, when executed by a control unit of a system as disclosed herein (e.g., according to any one of the above embodiments and / or according to any one of the embodiments described in further detail below), cause the control unit to perform a method as disclosed herein (e.g., according to any one of the above embodiments and / or according to any one of the embodiments described in further detail below). As used herein, the term "computer-readable storage medium" may specifically refer to a non-transitory data storage device, such as a hardware storage medium having computer-executable instructions stored thereon. A computer-readable data carrier or storage medium may specifically be or include a storage medium such as a random access memory (RAM) and / or a read-only memory (ROM).

[0108] This document further discloses and proposes a computer program product having program code means for performing the method according to the present invention in accordance with one or more embodiments of the present invention when the program is executed on a computer or computer network. Specifically, the program code means can be stored on a computer-readable data carrier and / or a computer-readable storage medium.

[0109] This document further discloses and proposes a data carrier having a data structure stored thereon, which, after being loaded into a computer or a computer network (for example, into a working memory or main memory of the computer or the computer network), can execute the method according to one or more embodiments disclosed herein.

[0110] This document further discloses and proposes a computer program product having program code means stored on a machine-readable carrier, so that when the program is executed on a computer or computer network, it performs a method according to one or more embodiments disclosed herein. As used herein, a computer program product refers to a program that is a tradable product. The product can generally be in any format, such as paper format, or on a computer-readable data carrier and / or computer-readable storage medium. Specifically, the computer program product can be distributed via a data network.

[0111] Finally, a modulated data signal containing instructions readable by a computer system or a computer network for executing a method according to one or more embodiments disclosed herein is disclosed and proposed herein.

[0112] With reference to the computer-implemented aspects of the present invention, one or more or even all of the method steps of one or more of the methods according to the embodiments disclosed herein can be performed using a computer or computer network. Thus, generally, any of the method steps involving the provision and / or manipulation of data can be performed using a computer or computer network. Generally, these method steps can include any method steps, except for method steps that typically require manual work, such as providing samples and / or performing certain aspects of the actual measurement.

[0113] Specifically, this article further discloses:

[0114] - a computer or a computer network comprising at least one processor, wherein the processor is adapted to perform a method according to one of the embodiments described in this specification,

[0115] - a computer-loadable data structure adapted to perform a method according to one of the embodiments described in this specification when the data structure is executed on a computer,

[0116] - a computer program, wherein the computer program is adapted to perform a method according to one of the embodiments described in this description when the program is executed on a computer,

[0117] - a computer program comprising program means for carrying out the method according to one of the embodiments described in this description when the computer program is executed on a computer or on a computer network,

[0118] - a computer program comprising the program means according to the preceding embodiment, wherein the program means is stored on a computer-readable storage medium,

[0119] a storage medium on which a data structure is stored and wherein the data structure is suitable for carrying out a method according to one of the embodiments described in this specification after being loaded into a main storage device and / or a working storage device of a computer or a computer network, and

[0120] A computer program product having program code means, wherein the program code means can be stored or stored on a storage medium for performing a method according to one of the embodiments described in this description when the program code means are executed on a computer or a computer network.

[0121] As used herein, the terms "having," "comprises," or "includes," or any grammatical variations thereof, are used in a non-exclusive manner. Thus, these terms can refer both to the absence of additional features in the entity described in that context, in addition to the features introduced by these terms, and to the presence of one or more additional features. By way of example, the expressions "A has B," "A includes B," and "A comprises B" can refer both to the absence of additional elements in A besides B (i.e., A consists solely and solely of B), and to the presence of one or more additional elements in entity A besides B (e.g., element C, elements C and D, or even additional elements).

[0122] Furthermore, it should be noted that the terms "at least one", "one or more", or similar expressions indicating that a feature or element may occur once or more than once are typically used only once when introducing the corresponding feature or element. In most cases, the expression "at least one" or "one or more" is not repeated when referring to the corresponding feature or element, although the corresponding feature or element may occur once or more than once.

[0123] Further, as used herein, the terms "preferably", "more preferably", "particularly", "more particularly", "specifically", "more specifically" or similar terms are used in conjunction with optional features, without limiting the possibilities of alternatives. Therefore, the features introduced by these terms are optional features and are not intended to limit the scope of the claims in any way. As the skilled person will recognize, the present invention can be carried out through the use of alternative features. Similarly, features introduced by "in an embodiment of the invention" or similar expressions are intended to be optional features, without any limitation on alternative embodiments of the invention, without any limitation on the scope of the invention, and without any limitation on the possibility of combining features introduced in this manner with other optional or non-optional features of the invention.

[0124] In summary and without excluding further possible embodiments, the following embodiments may be envisaged:

[0125] Embodiment 1: A method for obtaining chemical composition information of at least one sample by spectroscopic measurement, the method comprising:

[0126] i. acquiring spectral data of the sample by using at least one spectrometer device within at least one spatial measurement range of the spectrometer device;

[0127] ii. acquiring, using at least one imaging device, image data of a scene within a field of view of the imaging device, the scene including at least a portion of the sample and at least a portion of a spatial measurement range of the spectrometer device, and analyzing the image data and / or the spectral data using at least one processing device to determine at least one identification information about the sample;

[0128] iii. evaluating the spectral data using the processing device to obtain at least one item of sample information, wherein the evaluating includes processing the spectral data using at least one chemometric model, wherein the chemometric model converts the spectral data into chemical composition information, wherein the chemometric model is selected based on the identification information.

[0129] Embodiment 2: A method according to the previous embodiment, wherein the image data includes at least one of the following items: at least one identification information about the at least one sample, at least one information about the degree of heterogeneity, at least one information about the components of the sample, at least one information about a mixture composed of these components, and at least one information about the mixing ratio of these components.

[0130] Embodiment 3: The method according to any one of the preceding embodiments, wherein the method comprises providing at least one stoichiometric model.

[0131] Embodiment 4: The method according to the preceding embodiment, wherein the providing comprises obtaining the stoichiometric model from at least one database.

[0132] Embodiment 5: The method according to any one of the two preceding embodiments, wherein the method comprises at least one calibration step, wherein the calibration step comprises generating the chemometric model.

[0133] Embodiment 6: The method according to the preceding embodiment, wherein the calibration step is performed for each expected single component of the sample.

[0134] Embodiment 7: The method according to any one of the preceding embodiments, wherein the method includes providing multiple chemometric models for different identification information.

[0135] Embodiment 8: The method according to any one of the preceding embodiments, wherein the stoichiometric model is automatically selected.

[0136] Embodiment 9: The method according to any one of the preceding embodiments, wherein, in the case of a mixture, the stoichiometric model is selected for each component.

[0137] Embodiment 10: The method according to any one of the preceding embodiments, wherein the method comprises: in a case where no chemometric model is available for the determined identification information, selecting a chemometric model for identification information similar to the determined identification information.

[0138] Embodiment 11: The method according to any one of the preceding embodiments, wherein the image data includes information about a mixture ratio, wherein the mixture ratio is used to determine sample information of the mixture by:

[0139] - polymerizing these stoichiometric models according to the mixing ratio, wherein the polymerizing comprises adding and weighing according to the mixing ratio, and / or

[0140] - using the mixing ratio as an input parameter of at least one forward model.

[0141] Embodiment 12: A method according to any one of the preceding embodiments, wherein the imaging device is or includes at least one camera having one or more imaging sensors for acquiring the image data, wherein the imaging device is or includes at least one CCD imaging sensor or at least one CMOS imaging sensor.

[0142] Embodiment 13: The method according to any one of the preceding embodiments, wherein the method comprises applying at least one sample identification algorithm to the image data to obtain at least one item of identification information from the image data.

[0143] Embodiment 14: The method according to any one of the preceding embodiments, wherein the method comprises applying at least one spectral analysis to the spectral data to obtain the at least one identification information from the spectral data.

[0144] Embodiment 15: A method according to any one of the preceding embodiments, wherein the identification information is determined using user selection and / or user feedback.

[0145] Embodiment 16: The method according to any one of the preceding embodiments, wherein the spectrometer device is configured to detect electromagnetic radiation in the near infrared range.

[0146] Embodiment 17: The method according to any one of the preceding embodiments, wherein the method further comprises providing at least one output based on the sample information and / or the selected chemometric model using at least one user interface.

[0147] Embodiment 18: A system for obtaining chemical composition information of at least one sample by spectroscopic measurement, the system comprising:

[0148] I. at least one spectrometer device, the at least one spectrometer device being configured to acquire spectral data of the sample by using at least one spectrometer device within at least one spatial measurement range of the spectrometer device;

[0149] II. at least one imaging device configured to acquire image data of a scene within the field of view of the imaging device, the scene including at least a portion of the sample and at least a portion of the spatial measurement range of the spectrometer device;

[0150] III. At least one processing device configured to analyze the image data and / or the spectral data to determine at least one identification information about the sample, wherein the processing device is configured to evaluate the spectral data to obtain at least one item of sample information, wherein the evaluation includes processing the spectral data using at least one chemometric model, wherein the chemometric model is configured to convert the spectral data into chemical composition information, wherein the chemometric model is selected based on the identification information.

[0151] Embodiment 19: The system according to the preceding embodiment, wherein the system is configured to perform the method according to any one of the preceding method-related embodiments.

[0152] Embodiment 20: The system according to any of the preceding embodiments relating to systems, further comprising at least one user interface configured to provide at least one output based on the sample information and / or the selected chemometric model.

[0153] Embodiment 21: A computer program comprising instructions, which, when executed by a processing device of a system according to any one of the preceding embodiments relating to systems, causes the system to perform a method according to any one of the preceding embodiments relating to methods.

[0154] Embodiment 22: A computer-readable storage medium comprising instructions, which, when executed by a processing device of a system according to any one of the aforementioned embodiments involving systems, causes the system to perform a method according to any one of the aforementioned embodiments involving methods.

[0155] Embodiment 23: A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform a method according to any one of the preceding embodiments involving methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0156] Additional optional features and embodiments will be disclosed in more detail in the subsequent description of the embodiments, preferably in conjunction with the dependent claims. As will be appreciated by those skilled in the art, the respective optional features can be implemented independently and in any feasible combination. The scope of the present invention is not limited by the preferred embodiments. The embodiments are schematically depicted in the accompanying drawings. Identical reference numerals in these drawings represent identical or functionally equivalent elements.

[0157] In the attached figure:

[0158] Figure 1 A schematic diagram showing a system for obtaining chemical composition information of at least one sample and a sample;

[0159] Figure 2 A flow chart illustrating an embodiment of a method for obtaining chemical composition information of at least one sample;

[0160] Figure 3A and Figure 3B each presents further possible features of a method of obtaining information on the chemical composition of at least one sample; and

[0161] Figures 4A to 4D Each shows a flow chart of an embodiment of a method of obtaining chemical composition information of at least one sample. DETAILED DESCRIPTION

[0162] Figure 1A system 110 is shown for obtaining chemical composition information of at least one sample 112 by spectroscopic measurement. The system 110 includes at least one spectrometer device 114 configured to acquire spectroscopic data of the sample 112 within at least one spatial measurement range 118 of the spectrometer device 114 using the at least one spectrometer device 114.

[0163] Figure 1 A system 110 and an apple 120 as an exemplary sample 112 placed within a spatial measurement range 118 of a spectrometer device 114 of the system 110 are shown. As an example, the spatial measurement range 118 may be defined as a solid angle or a three-dimensional angular segment in space, wherein the sample 112 arranged within the solid angle or angular segment may be analyzed by the spectrometer device 114. The sample 112 positioned within the spatial measurement range 118 may be spectrally analyzed by the spectrometer device 114. In particular, the spectrometer device 114 may be configured to acquire spectral data based on incident light 122 from within the spatial measurement range 118. The spatial measurement range 118 may in particular be a three-dimensional spatial segment, for example a three-dimensional space, such as a conical spatial segment, the light content of which may be received and analyzed by the spectrometer device 114. The spectral data acquired by the spectrometer device 114 may include information related to at least one sample 112 located within the spatial measurement range 118 of the spectrometer device 114. As Figure 1 As shown, in order to perform spectral analysis on the sample 112, the spectrometer device 114 can be positioned in close proximity to the sample 112 so that the spatial measurement range 118 at least partially includes the sample 112, for example, positioned at a distance in the range of 0 mm to 100 mm from the sample, specifically in the range of 0 mm to 15 mm.

[0164] like Figure 1 As shown in FIG, the spectrometer device 114 may include at least one detector device 124 including at least one optical element 126 and a plurality of photosensitive elements 128. The at least one optical element 126 may be configured to separate the incident light 122 into a spectrum consisting of component wavelength components. Each photosensitive element 128 may be configured to receive at least a portion of one of the component wavelength components and to generate a corresponding detector signal based on illumination of the corresponding photosensitive element 128 by at least a portion of the corresponding component wavelength component. Thus, the spectrometer device 114 may analyze the incident light 122 after it interacts with the sample 112 and generate at least one corresponding detector signal, which may form part of the spectral data.

[0165] like Figure 1As shown, the optical element 126 may include at least one wavelength selection element 130. By way of example, the wavelength selection element 130 may be selected from the group consisting of: a prism; a grating; a linear gradient filter; an optical filter, particularly a narrow bandpass filter. The detector device 124 may further include a plurality of photosensitive elements 128 arranged in a linear array, wherein the array of photosensitive elements 128 includes a number of 10 to 1000, particularly a number of 100 to 500, particularly a number of 200 to 300, and more particularly a number of 256 photosensitive elements 128. Each photosensitive element 128 can be selected from the group consisting of: a pixelated inorganic camera element, in particular a pixelated inorganic camera chip, more particularly a CCD chip or a CMOS chip; a monochrome camera element, in particular a monochrome camera chip; at least one photoconductor, in particular an inorganic photoconductor, more particularly an inorganic photoconductor comprising Si, PbS, PbSe, Ge, InGaAs, extended InGaAs, InSb, or HgCdTe. Each photosensitive element 128 can be sensitive to electromagnetic radiation 122 in the wavelength range from 760 nm to 1000 μm, in particular in the wavelength range from 760 nm to 15 μm, more particularly in the wavelength range from 1 μm to 5 μm, more particularly in the wavelength range from 1 μm to 3 μm.

[0166] Spectrometer device 114 may be or include a dispersive spectrometer device that can analyze radiation 122 from sample 112 illuminated with broadband illumination, such as described above. However, other configurations and / or arrangements of spectrometer device 114 are also possible, which may particularly affect its components, such as the detector device 124 and / or illumination source used. By way of example, sample 112 may be illuminated with a limited number of different wavelengths of light 122. Spectrometer device 114 may include a broadband detector. In particular, spectrometer device 114 may be a Fourier transform spectrometer, specifically a Fourier transform infrared spectrometer. Thus, sample 112 may be illuminated using a narrowband light source, such as at least one light-emitting diode (LED) and / or at least one laser. Spectrometer device 114 may be configured to determine spectrum 132 by measuring and processing an interferogram, in particular by applying at least one Fourier transform to the measured interferogram.

[0167] like Figure 1 As shown, the system 110 includes at least one imaging device 134 configured to acquire image data of a scene 136 within a field of view 138 of the imaging device 134. The scene 136 includes at least a portion of the sample 112 and at least a portion of the spatial measurement range 118 of the spectrometer device 114, such as from Figure 1As will be apparent in the figure, the imaging device 134 may be or include at least one camera 140 having one or more imaging sensors 142 for acquiring image data. The camera 140 may specifically include at least one camera chip, such as at least one CCD chip and / or at least one CMOS chip configured to record images. By way of example, the camera 140 may include an array of imaging sensors 142, the imaging sensor array including at least 100 imaging sensors 142 in each dimension, and specifically at least 300 imaging sensors 142 in each dimension. For example, the camera 140 may be a color camera 140 including color pixels, each color pixel including at least three color sub-pixels sensitive to different colors. For example, the camera 140 may include black and white pixels and / or color pixels. The color pixels and black and white pixels may be combined within the camera 140. In addition to the at least one camera chip or imaging chip, the camera 140 may include additional components, such as one or more optical elements, for example, one or more lenses (not shown). By way of example, the camera 140 may be a fixed-focus camera 140, wherein the adjustment of at least one lens relative to the camera 140 is fixed. Alternatively, however, the camera 140 may also include one or more variable lenses that may be adjusted automatically or manually.

[0168] like Figure 1 As depicted, imaging device 134 may specifically be camera 140 of mobile device 144. Figure 1 As further shown in FIG, the spectrometer device 114 can be embodied as a portable spectrometer device 114. Specifically, the spectrometer device 114 can be part of a mobile device 144, such as a laptop computer, a tablet computer, or specifically a cellular phone such as a smart phone 146. Additionally or alternatively, the mobile device 144 can be or include a smartwatch and / or a wearable computer (also referred to as a wearable device), for example, a body-mounted computer. Other mobile devices 144 are also possible. Figure 1As shown, the spectrometer device 114 can be integrated into a mobile device 144. Additionally or alternatively, the spectrometer device 114 can be attachable to the mobile device. Therefore, both the camera 140 and the spectrometer device 114 can be part of the mobile device 144, in particular the smartphone 146. The present invention should be applicable in particular to a camera 140 as typically used in a mobile device 144, such as a notebook computer, a tablet computer or a cellular phone such as in particular the smartphone 146. The smartphone 146 can further include a housing 148, wherein the spectrometer device 114 and the imaging device 134, in particular the camera 140, can be integrally contained within the housing 148. The smartphone 146 can in particular include a front camera 150 and a rear camera 152. In particular, the field of view 138 of the front camera 150 can at least partially overlap with the spatial measurement range 118 of the spectrometer device 114, as Figure 1 . In addition to at least one camera 140, the mobile device 144 may also include one or more data processing devices 154, such as one or more processors 156. The mobile device 144 may specifically have at least one function other than spectroscopy, such as a mobile communication function, for example, a cellular phone function. However, other cameras 140 are also possible.

[0169] In particular, imaging device 134 may be or may include at least one LIDAR-based imaging device 134, where LIDAR stands for Light Detection and Ranging or Light Imaging, Detection, and Ranging (not shown). LIDAR-based imaging device 134 may include at least one laser source, such as at least one tunable laser diode, for illuminating a sample or at least a portion of the sample. LIDAR-based imaging device 134 may further include at least one positioning unit configured to determine at least one distance of the illuminated portion of the sample from imaging device 134 and / or from at least one other point or location in space. The positioning unit may specifically include at least one sensor element, such as a photodiode, configured to detect at least one laser beam emitted from the laser source and reflected by sample 112. Determining the distance and thereby generating image data may include processing the beam reflected by sample 112 and / or at least one reference beam and / or corresponding signals detected by the at least one sensor element.

[0170] like Figure 1As shown in FIG, the spectrometer device 114 and the imaging device 134 can have a known orientation relative to each other, specifically a fixed orientation. In particular, the spectrometer device 114 and the imaging device 134 can have a known, specifically fixed, spatial relationship relative to each other. Further, the spatial measurement range 118 of the spectrometer device 114 and the field of view 138 of the imaging device 134 can have a fixed spatial relationship relative to each other.

[0171] The system 110 may further include at least one light source 158, such as Figure 1 As shown. The light source 158 can be specifically configured to emit electromagnetic radiation 122 in a wavelength range from 760 nm to 1000 μm, specifically in a wavelength range from 760 nm to 15 μm, more specifically in a wavelength range from 1 μm to 5 μm, more specifically in a wavelength range from 1 μm to 3 μm. The spectrometer device 114 can be specifically referred to as a “near-infrared spectrometer device”. The light source 158 can be specifically configured to irradiate the sample 112, such as from Figure 1 In particular, a narrowband light source 158, such as at least one light emitting diode (LED) and / or at least one laser, may be used to illuminate the sample 112.

[0172] System 110 includes at least one processing device 154 configured to analyze image data and / or spectral data to determine at least one item of identification information about sample 112. Processing device 154 is configured to evaluate the spectral data to obtain at least one item of sample information, wherein the evaluation includes processing the spectral data using at least one chemometric model. The chemometric model is configured to convert the spectral data into chemical composition information. The chemometric model is selected based on the item of identification information.

[0173] The processing device 154 of the system 110 is configured to obtain at least one item of sample information about the at least one sample 112. The system 110 may include at least one display device 160, which is configured to provide at least one item of sample information about the at least one sample 112. Specifically, for example, as described above or described in further detail below, the system 110 may include at least one mobile device 144, wherein the mobile device 144 includes at least one spectrometer device 114 and at least one imaging device 134. Therefore, the spectrometer device 114 and the imaging device 134 (such as at least one camera 140) can be integrated into the mobile device 144 (such as a smart phone 146). Figure 1 As illustrated in FIG, the mobile device 144 may particularly have at least one display device 160, specifically a screen 162, configured to display object information.

[0174] like Figure 1 As depicted, system 110 can further include at least one control unit 164. Control unit 164 can be configured to perform at least one computing operation and / or to control at least one function of at least one other component of system 110 to obtain chemical composition information for at least one sample 112. Control unit 164 can specifically control at least one function of spectrometer device 114, such as the acquisition of spectral data. Control unit 164 can specifically control at least one function of imaging device 134, such as the acquisition of image data. Control unit 164 can specifically control processing device 154, such as the evaluation of spectral data and / or image data. Specifically, at least one control unit 164 can be embodied as at least one processor 156 and / or can include at least one processor 156, wherein processor 156 can be specifically configured to perform one or more operations through software programming.

[0175] The system 110 may further include at least one user interface 166. The user interface 166 may, for example, be configured to share information with a user and receive information provided by a user. As an example, the user interface 166 may include one or more of the following: a human-machine interface (such as the display 160, screen 162, keyboard, voice interface, touchpad, or a portion of a port via which a user may provide digital information data), a graphical user interface; a data interface, such as a wireless and / or wired data interface. Figure 1 In the illustrated embodiment of system 110, user interface 166 is screen 162 of display device 160. At least one user interface 166 may be specifically configured to provide at least one output based on sample information and / or a selected chemometric model.

[0176] System 110 can be specifically configured to perform a method for obtaining chemical composition information of at least one sample 112 by spectroscopic measurement, as described above or in greater detail below. The method includes the following method steps, which can be performed in the order given. However, a different order is also possible. The method can further include additional method steps not listed. Furthermore, one or more of the method steps, or even all of the steps, can be performed only once or repeatedly. The method steps are as follows:

[0177] i. Acquiring spectral data of the sample 112 by using at least one spectrometer device 114 within at least one spatial measurement range 118 of the spectrometer device 114;

[0178] ii. acquiring, using at least one imaging device 134, image data of a scene 136 within a field of view 138 of the imaging device 134, the scene 136 including at least a portion of the sample 112 and at least a portion of the spatial measurement range 118 of the spectrometer device 114, and analyzing the image data and / or the spectral data using at least one processing device 154 to thereby determine at least one identification information about the sample 112; and

[0179] iii. evaluating the spectral data using the processing device 154 to obtain at least one item of sample information, wherein the evaluating includes processing the spectral data using at least one chemometric model, wherein the chemometric model converts the spectral data into chemical composition information, wherein the chemometric model is selected based on the identification information.

[0180] Figure 2 A method of obtaining chemical composition information of at least one sample 112 by spectroscopic measurement is presented. Specifically, step i. is represented by reference numeral 168 ; step ii. is represented by reference numeral 170 ; and step ii. is represented by reference numeral 172 .

[0181] In addition to spectroscopic measurements, the method also proposes the use of image data. Combining spectroscopic measurements with image data can allow obtaining accurate chemical composition information of the sample 112, in particular even in the case of mixtures, local variations and inhomogeneities of the sample 112. Figure 1 The sample 112 shown within the spatial measurement range 118 of the spectrometer device 114 is an apple 120. However, a variety of samples 112 are also feasible. The sample 112 can be a single-component sample 112 or a mixture comprising at least two components. In particular, the sample 112 can be an inhomogeneous sample 112, for example a sample 112 whose chemical composition can vary within the sample 112, for example in a position-dependent manner. However, other samples 112, in particular homogeneous samples 112 whose chemical composition varies only slightly or not at all, are also feasible. The sample 112 can be any material that shows NIR activity. In this case, a chemometric model can be obtained, as described in more detail above and / or as described in more detail below. If no chemometric model is available for a certain material, a model of another material that is most similar to the material can be used.

[0182] The sample 112 may be a solid sample 112, such as a powder or a solid. However, other samples 112 are also possible, such as a liquid sample 112. The sample 112 may specifically be or include a food, such as a fruit 174 or a vegetable. For example, the sample 112 may be or include at least one element selected from the list consisting of: food; feed; vegetables, such as tomatoes; fruit 174, such as apples 120, pears; crops, such as wheat, corn; waste, such as waste from recycling. For example, the sample 112 may specifically be or include a body part 176, such as skin 178. For example, the method may be used for recycling. The sample may include waste material, such as plastic. The method may include obtaining at least one identification information about the sample by using the image data, for example, at least one identification information related to the shape of the sample, such as one or more of particle size, color, and surface roughness. The chemometric model for analyzing the spectral data may be selected taking into account the identification information.

[0183] Figure 3A and Figure 3B Three possible samples 112 are shown, specifically an apple 120, a banana 180, and the skin 178 of a human hand 182 and arm 183, respectively. Figure 3A and Figure 3B Spectral data in the form of spectrum 132 acquired using spectrometer device 114 as part of the spectral measurement of step i. of the method is further illustrated. As part of the spectral measurement, sample 112 may be illuminated with electromagnetic radiation 122 within the infrared spectral range, specifically the near-infrared spectral range. Specifically, the electromagnetic radiation may be within a wavelength range from 760 nm to 1000 µm, specifically within a wavelength range from 760 nm to 15 µm, more specifically within a wavelength range from 1 µm to 5 µm, and even more specifically within a wavelength range from 1 µm to 3 µm. The spectral measurement may further include receiving incident light after interaction with the sample and generating at least one corresponding signal, which may form part of the spectral data. The spectral data may include information regarding at least one optical property or optically measurable property of sample 112 for one or more different wavelengths, the information determined as a function of wavelength. More specifically, the spectral data may relate to at least one property characterizing at least one of transmission, absorption, reflection, and emission of the sample. The spectral data may specifically take the form of a signal intensity determined as a function of the wavelength of the spectrum or a subregion thereof (e.g., a wavelength interval), wherein the signal intensity may preferably be provided as an electrical signal that can be used for further evaluation. The spectral data may be graphically represented, for example, in the form of a spectral curve 184, wherein the signal intensity I plotted on the y-axis 186 is shown as a function of the wavelength λ plotted on the x-axis 188, as shown in FIG. Figure 3A and Figure 3BSpecifically, the signal intensity I may correspond to the intensity of the reflected electromagnetic radiation 122 (eg, electromagnetic radiation 122 in the infrared spectral range) that may illuminate the sample 112. The spectral curve 184 may show the reflected intensity I as a function of the wavelength λ, as shown in FIG. Figure 3A B and Figure 3B As shown in .

[0184] As part of step ii. of the method, image data of a scene within the field of view of an imaging device is acquired. The image data may include a plurality of electronic readings from the imaging device, such as from an imaging sensor (e.g., pixels of a camera chip) and / or from sensor elements of a LIDAR-based imaging device. In particular, the image data may include a plurality of numerical values ​​corresponding to the electronic readings from the imaging device. Figure 3A and Figure 3B Image data of a scene 136 within a field of view 138 of an imaging device, specifically a camera 140, is shown in the form of a graphical representation, specifically in the form of an image 200. Figure 3A As is apparent from the above, the scene 136 may include a plurality of objects having a particular arrangement, wherein these objects and their arrangement may be imaged by the imaging device 134, thereby generating at least one image 200. Specifically, Figure 3A The image 200 shown in shows a scene 136 including an apple 120 and a banana 180 arranged on a board 202 . Figure 3B The image 200 shown in FIG. 1 shows a scene 136 including a portion of a human hand 182 and an arm 183. The sample in step i. may be at least partially visible in the image data in step ii. Thus, the field of view of the imaging device and the spatial measurement range of the spectrometer device may at least partially overlap.

[0185] like Figure 3A and Figure 3B As shown, the sample or at least a portion thereof can be located in both the field of view 138 of the imaging device 134 and the spatial measurement range 118 of the spectrometer device 114. The sample or at least a portion thereof can thus be spectroscopically inspected by the spectrometer device 114 and at least partially imaged by the imaging device 134. Figure 3A and Figure 3B As indicated, the spectral measurement can be, for example, a spot measurement of a spot 204 having a diameter of 8 mm to 2 cm. The spot can include an area having any geometric shape. For example, the spot can include an area from 1 mm to 500 mm 2The image data acquired by the imaging device 134, in particular the image 200, can cover a larger area than the spectral measurement, for example, an area from 0.01 m to 10 m. The spectral measurement and the acquisition of the image data can occur at different time points T1 and T2. The mixing ratio of the components of the sample 112 can be constant at T1 and T2. Figure 3A and Figure 3B As depicted in FIG. , image 200 may include information regarding the location at which the spectral data was acquired (particularly light spot 204) and / or information regarding the evaluation results of the spectral data, such as composition information derived from the spectral data. Thus, image 200 may visually indicate scene 136 or a portion thereof, as well as information derived from the spectral data acquired in step i., and optionally, positional information regarding the location at which the information was acquired.

[0186] Step ii. of the method includes analyzing the image data and / or spectral data using at least one processing device 154 to determine at least one piece of identification information about the sample 112. A variety of identification information can be obtained from the image data and / or spectral data. The identification information can include at least one of the following: at least one piece of information about the degree of inhomogeneity, at least one piece of information about the components of the sample 112, at least one piece of information about a mixture of the components, and at least one piece of information about a mixing ratio of the components. The identification information can be or include at least one piece of information about at least one of the following: the type of sample 112, the boundaries of the sample 112 within the scene 136, the size of the sample 112, and the orientation of the sample 112.

[0187] In particular, at least one spectral evaluation algorithm may be applied to the spectral data as part of the spectral analysis to obtain at least one item of identification information. The spectral analysis may include, for example, Figure 3A and Figure 3BThe chemical composition of the sample is determined by comparing one or more identified peaks 206 of the labeled spectral data with at least one predetermined peak 206 or at least one set of predetermined peaks 206. Additionally or alternatively, the method may include applying at least one sample identification algorithm to the image data to derive at least one item of identification information from the image data. The identification information may be derived, in particular, by using at least one sample identification algorithm (e.g., an image recognition algorithm and / or a trained model configured to identify or distinguish sample 112, for example, by using artificial intelligence (e.g., an artificial neural network). For example, the sample identification algorithm may identify the type of sample 112, e.g., the class or type of object, such as whether the sample 112 is an apple 120, an orange, another type of fruit 174 or vegetable, or a human body part 176 (e.g., a hand 182 or a face). Other types of samples 112 are also possible, in particular other types of food samples 112. By way of example, the identification information may be determined using user selection and / or user feedback (e.g., using at least one user interface 166).

[0188] As a further example, at least one of scene 136, field of view 138, sample 112, and spatial measurement range 118 may be modified between steps i. and a possible repetition of step ii. In particular, scene 136 may change, and / or at least one of spectrometer device 114, imaging device 134, and a device including both spectrometer device 114 and imaging device 134 (such as mobile device 144) may be moved. Thus, as Figure 3A and Figure 3B As shown in FIG. , the method can generate at least one image 200 of the scene 136, the at least one image having at least two items of spectral object information and corresponding spatial information about the spatial measurement range 118 within the image 200 for each item of spectral object information. Further, the image 200 obtained from the image data in step ii. can be an image 200 obtained from the repeated image data of step ii., specifically, at least one of a combined image 200 and a selected image 200 among the images 200 obtained from the repeated image data of step ii. Figure 3BAs shown in , the imaging device and / or spectrometer device can be moved along a scan path 208 over the sample 112 while performing one or more repetitions of steps i. and ii. By performing step iii., at least one item of sample information can be obtained, wherein the sample information can include multiple items of chemical information corresponding to multiple locations along the scan path 208. Again, image data of the sample 112 can be acquired, for example, during an initial execution of step ii., wherein the scan path 208 can be included in an image 200 derived from the image data. Specifically, the scan path 208 and / or the spectroscopic object information, in particular the chemical information, can be indicated in the image 200. This can allow the retrieval of chemical information along the scan path 208.

[0189] Step iii. includes evaluating the spectral data using processing device 154 to obtain at least one item of sample information. The evaluation includes processing the spectral data using at least one chemometric model. The chemometric model converts the spectral data into chemical composition information. The chemometric model is selected based on the identification information. Step iii. may include applying at least one spectral evaluation algorithm to the spectral data from step i., wherein the spectral evaluation algorithm is selected based on the identification information, specifically, based on the type of at least one object 112.

[0190] The sample information may in particular be determined by taking into account spectral data of the sample as well as image data of the sample. The sample information may in particular relate to a property that may vary within the sample 112, so that the property may be characteristic about a specific location or spatial range within the sample 112. However, the property may also show no variation or only slight variation throughout the sample 112. The sample information may describe the property in a qualitative and / or quantitative manner, for example by one or more numerical values. In particular, the sample information may comprise chemical information of the sample 112, in particular the chemical composition. The sample information may comprise information about the property as well as spatial information about the specific location or spatial range within the sample 112 where the property is measured. As an example, for Figure 3A For example, for apple 120, sample information may include one or more of the following: dry matter, sugar content, and acidity. For wheat or corn, sample information may include one or more of the following: protein content, starch content, fiber content, dry matter, etc. For recycling waste, sample information may include one or more of the following: particle size, color, and surface roughness.

[0191] The chemometric model converts the spectral data into chemical composition information, which may relate to information about one or more of the following: the components of the mixture, the mixing ratio, the presence or absence of at least one chemical component, etc. The method may include providing at least one chemometric model, specifically providing multiple chemometric models for different identification information. Specifically, the chemometric model can be obtained from at least one database, such as from the cloud. The chemometric model can be accessed via the cloud and can be selected based on the identification information. The chemometric model can be selected automatically. As indicated above, the sample 112 can be a single-component sample 112 or a mixture containing at least two components. In the case of a mixture, a chemometric model can be selected for each component. In the case of a mixture, the image data may include information about the mixing ratio. The mixing ratio can be used to determine sample information of the mixture in the following manner:

[0192] - polymerizing these stoichiometric models according to the mixing ratio, wherein the polymerizing comprises adding and weighing according to the mixing ratio, and / or

[0193] - using the mixing ratio as an input parameter of at least one forward model.

[0194] A forward model can be designed to use input parameters and predict and / or simulate spectra. For example, a forward model can use the mixing ratio as an input parameter.

[0195] The method may include, if no chemometric model is available for the determined identification, selecting a chemometric model for identification information similar to the determined identification information. The method may be at least partially computer-implemented, specifically step iii. The chemometric model may include mathematical and statistical techniques for extracting relevant information from the spectral data. The chemometric model may specifically include at least one trained model. Furthermore, as described above, the analysis of the image data may include, for example, analyzing the image information using at least one recognition algorithm.

[0196] The method may include at least one calibration step. The calibration step includes generating a chemometric model. The calibration step may be performed for each expected single component of the sample 112.

[0197] Hereinafter, the following will be described in an exemplary manner. Figures 4A to 4D Four embodiments of the method are illustrated in the flowchart shown.

[0198] like Figure 4A As shown in , the method can be performed as follows:

[0199] During the calibration step, a chemometric model can be generated. For example, when determining the acidity in apples 120 or the sugar content in tomatoes, it may be necessary to perform a separate calibration for the crop parameter combination. The calibration step is indicated by reference numeral 210. Next, the imaging device 134 may capture an image 200 of the sample 112, for example, via the camera 140. This step is indicated by reference numeral 212. Image recognition may be performed on the sample 112, as indicated by reference numeral 214. In a parallel series of steps, the spectrometer device 114 may perform an IR spectrum measurement on the same sample 112, as indicated by reference numeral 216. Spectral data for the sample may be obtained from the IR spectrum measurement, as indicated by reference numeral 218. Next, one or more of the crop species, crop variety, or food / feed type may be identified based solely on the results of the image recognition. This step is indicated by reference numeral 220. Information regarding one or more of the crop species, crop variety, or food / feed type may be used to select a chemometric model, as indicated by reference numeral 222. The selected chemometric model can then be applied to the spectral data, as indicated by reference numeral 224. Finally, sample information specific to the (single-component) sample can be obtained, also based on the selected chemometric model. For example, for wheat or corn, the sample information could be one or more of the following: protein content, starch content, fiber content, dry matter, etc. For example, for apples or tomatoes, the sample information could be one or more of the following: dry matter, sugar content, acidity. The step of obtaining sample information is indicated by reference numeral 226.

[0200] like Figure 4B As shown in , the method can be performed as follows:

[0201] In a separate calibration step, a chemometric model can be generated. For example, in the case of acidity in apples 120 or sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. The calibration step is indicated by reference numeral 210. Next, the imaging device 134 may capture an image 200 of a sample 212, for example, via camera 140. This step is indicated by reference numeral 212. Image recognition can be performed on the sample 112, as indicated by reference numeral 214. In a parallel series of steps, the spectrometer device 114 may perform an IR spectrum measurement on the same sample 112, as indicated by reference numeral 216. Spectral data for the sample 112 can be obtained from the IR spectrum measurement, as indicated by reference numeral 218. Identification of one or more of the following: crop species, crop variety, or food / feed type can be accomplished based on the image recognition results and / or the IR spectrum. There is no particular order in which to perform IR spectroscopy or image recognition first. In some cases (for example, for wheat or apples 120), crop species can be distinguished using IR. In other cases (apples 120 or pears, both of which have similar IR spectra), crop species can be distinguished via image recognition. The identification step based on the results of image recognition and / or IR spectra is indicated by reference numeral 228. Information regarding one or more of the crop species, crop variety, and food / feed type can be used to select a chemometric model, as indicated by reference numeral 230. This selected chemometric model can then be applied to the spectral data, as indicated by reference numeral 224. Finally, also based on the selected chemometric model, sample information specific to sample 112 can be obtained. The step of obtaining sample information is indicated by reference numeral 226.

[0202] like Figure 4C As shown in , the method can be performed as follows:

[0203] In a separate calibration step, a chemometric model can be generated. For example, in the case of the acidity in apples 120 or the sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. The calibration step is indicated by reference numeral 210. Next, the imaging device 134 may capture an image 200 of the sample 112, for example, via the camera 140. This step is indicated by reference numeral 212. Image recognition may be performed on the sample 112, as indicated by reference numeral 214. In a parallel series of steps, the spectrometer device 112 may perform an IR spectrum measurement on the same sample 112, as indicated by reference numeral 216. Spectral data for the sample 112 may be obtained from the IR spectrum measurement, as indicated by reference numeral 218. Identification of one or more of the crop species, crop variety, or food or feed type may be performed first based on the image recognition results and / or IR spectrum (as indicated by reference numeral 228), and then additionally based on user selection or user feedback (as indicated by reference numeral 232). Specifically, in the case of multiple options, especially where the crop species / variety is difficult to identify via image recognition and / or IP spectroscopy (for example, where sample 112 is a white powder that could be wheat flour or plastic powder; for example, apple 120 and pear, which are quite similar), the user can input additional knowledge and select the best option. For example, the user can select "apple" 120 instead of "pear" by clicking in user interface 166. Information regarding one or more of the crop species, crop variety, and food or feed type can be used to select a chemometric model, as indicated by reference numeral 234. This selected chemometric model can then be applied to the spectral data, as indicated by reference numeral 224. Finally, based on the selected chemometric model, sample information specific to sample 112 can be obtained. The step of obtaining sample information is indicated by reference numeral 226.

[0204] like Figure 4D As shown in , the method can be performed as follows:

[0205] In a separate calibration step, a chemometric model can be generated. For example, in the case of acidity in apples or sugar content in tomatoes, it may be necessary to perform a separate calibration for a combination of crop parameters. Calibration can be performed for a single component. The calibration step is indicated by reference numeral 210. The imaging device 134 can, for example, capture an image 200 of a sample 212 via a camera 140. This step is indicated by reference numeral 212. Next, image recognition can be performed on the sample 112, as indicated by reference numeral 214. In a series of parallel steps, the spectrometer device 114 performs an IR spectrum measurement on the same sample 112, as indicated by reference numeral 216. Spectral data of the sample is obtained from the IR spectrum measurement, as indicated by reference numeral 218. Identification of one or more of the crop species, crop variety, or food / feed type can be completed based solely on the results of the image recognition. This step is indicated by reference numeral 220. Simultaneously with or in parallel with identifying the crop species, crop variety, or food / feed type, a mixing ratio can also be determined based on the results of the image recognition. This step is indicated by reference numeral 236. One or more stoichiometric models can be selected based on one or more of the following: crop species, crop variety, and food / feed type, as indicated in step 234. The stoichiometric models (data models) can be aggregated based on the mixing ratio information derived from image recognition. For example, if the proportions of rye and wheat in a sample are 30% and 70%, respectively, the method can include aggregating individual models based on the mixing ratio by adding and weighing (e.g., a 30% rye stoichiometric model plus a 70% wheat stoichiometric model). For different species and / or materials, such as plastic and wheat, the method can include performing a "backward calculation" of the individual components within the mixture. Additionally or alternatively, a forward model can be used, for example, using the determined mixing ratio as an input parameter. The step of selecting a stoichiometric model based on the aggregated stoichiometric model or based on the forward model is indicated by reference numeral 238. Finally, sample information specific to the (mixture) sample 112 can be obtained, also based on the selected stoichiometric model. The sample information can include information about the mixing ratio. For species and / or materials that are different from each other, e.g., plastic and wheat, the method may include a “reverse mixing of models” (e.g., reverse mixing of wheat and green wheat kernels with pesticides). The step of obtaining sample information specific to the (mixed) sample 112 is indicated by reference numeral 240 .

[0206] List of Reference Numerals

[0207] 110 system

[0208] 112 samples

[0209] 114 Spectrometer Equipment

[0210] 118 spatial measurement range

[0211] 120 apples

[0212] 122 Light

[0213] 124 detector equipment

[0214] 126 optical elements

[0215] 128 photosensitive elements

[0216] 130 wavelength selection elements

[0217] 132 Spectrum

[0218] 134 Imaging Equipment

[0219] 136 scenes

[0220] 138 field of view

[0221] 140 cameras

[0222] 142 imaging sensor

[0223] 144 mobile devices

[0224] 146 smartphones

[0225] 148 shell

[0226] 150 front camera

[0227] 152 rear camera

[0228] 154 processing equipment

[0229] 156 processors

[0230] 158 light sources

[0231] 160 display devices

[0232] 162 screens

[0233] 164 control unit

[0234] 166 User Interface

[0235] 168Step i.

[0236] 170Step ii.

[0237] 172Step iii.

[0238] 174 Fruit

[0239] 176 body parts

[0240] 178 Skins

[0241] 180 bananas

[0242] 182 Hands

[0243] 183 arms

[0244] 184 spectral curves

[0245] 186y-axis

[0246] 188x axis

[0247] 200 images

[0248] 202 board

[0249] 204 light spots

[0250] 206 Peaks

[0251] 208 scan path

[0252] 210 "Generate chemometric models in separate calibration steps (e.g., acidity in apples or sugar content in tomatoes, i.e., you need to perform separate calibrations for crop parameter combinations)"

[0253] 212 “Capturing an image of the sample (e.g., via a camera)”

[0254] 214 “Image recognition of samples”

[0255] 216 "IR spectroscopy measurement of samples"

[0256] 218 "Obtaining Spectral Data from IR"

[0257] 220 “Identification of crop species / variety or type of food / feed based on image recognition”

[0258] 222 “Selecting a chemometric model (data model) based on image recognition results”

[0259] 224 "Applying Chemometric Models to Spectroscopic Data (Data Models)"

[0260] 226 "Results of single components: e.g., for wheat / corn: protein content, starch content, fiber content, dry matter, etc.; e.g., for apples / tomatoes: dry matter, sugar content, acidity"

[0261] 228 "Results of image recognition and / or IR spectra: Crop species / variety (e.g., wheat or rye) or type of food / feed; Details: Whether IR spectroscopy or image recognition comes first, in no particular order; In some cases (wheat or apples), they can be distinguished via IR; in other cases (apples or pears, both of which have similar IR spectra), they can only be distinguished via image recognition."

[0262] 230 “Selection of a chemometric model (data model) based on image recognition results and / or IR spectra”

[0263] 232 “User choice / user feedback: in situations where there are multiple options (e.g., it is a white powder and could be wheat flour or plastic powder; e.g., apple vs. pear, both are quite similar), but the user knows what it is and can choose the best option; relevant to the B2C sector.”

[0264] 234 “Selection of a chemometric model (data model) based on the results of image recognition and / or IR spectrum and / or user selection / user feedback”

[0265] 236 "Image Recognition Result 2: Mixture Ratio"

[0266] 238 "Subvariant A: Aggregation of >1 chemometric models (data models) based on mixing ratio information from image recognition; Subvariant B: (most preferably) forward model, where the mixing ratio is also required as input"

[0267] 240 "a) results for mixtures (e.g. a wheat / barley mixture with protein content X; mixing ratio) and / or b) for species / materials that are different from each other (e.g. plastic and wheat): results for the individual components within the mixture based on "back calculation" in the case of subvariant B or "back mixing of the model" in the case of subvariant A".

Claims

1. A method for obtaining chemical composition information of at least one sample (112) by spectroscopic measurement, the method comprising: i. acquiring spectral data of the sample (112) within at least one spatial measurement range (118) of the spectrometer device (114) by using at least one spectrometer device (114); ii. acquiring, using at least one imaging device (134), image data of a scene (136) within a field of view (138) of the imaging device (134), the scene (136) including at least a portion of the sample (112) and at least a portion of a spatial measurement range (118) of the spectrometer device (114), and determining at least one item of identification information about the sample (112) by analyzing the image data and / or the spectral data using at least one processing device (154), wherein the image data includes information about a mixing ratio; and iii. evaluating the spectral data using the processing device (154) to obtain at least one item of sample information, wherein the evaluating includes processing the spectral data using at least one chemometric model, wherein the chemometric model converts the spectral data into chemical composition information, wherein the chemometric model is selected based on the identification information, wherein, in the case of a mixture, the chemometric model is selected for each component, and wherein the mixing ratio is used to determine the sample information of the mixture by: - polymerizing the stoichiometric models according to the mixing ratio, wherein the polymerizing comprises adding and weighing according to the mixing ratio, and / or - using the mixing ratio as an input parameter of at least one forward model.

2. The method according to the preceding claim, wherein The image data includes at least one of the following: at least one identification information about the at least one sample (112), at least one information about the degree of inhomogeneity, at least one information about the components of the sample (112), at least one information about a mixture composed of these components, and at least one information about a mixing ratio of these components.

3. A method according to any one of the preceding claims, wherein The method includes providing at least one stoichiometric model.

4. The method according to the preceding claim, wherein The method comprises at least one calibration step, wherein the calibration step comprises generating the stoichiometric model.

5. A method according to any one of the preceding claims, wherein The method includes providing a plurality of chemometric models for different identification information.

6. A method according to any one of the preceding claims, wherein The method includes selecting a chemometric model for identification information similar to the determined identification information if no chemometric model is available for the determined identification information.

7. A method according to any one of the preceding claims, wherein The identification information is determined using user selections and / or user feedback.

8. A method according to any one of the preceding claims, wherein The method further includes providing at least one output based on the sample information and / or the selected chemometric model using at least one user interface (166).

9. A system (110) for obtaining chemical composition information of at least one sample (112) by spectroscopic measurement, the system (110) comprising: I. at least one spectrometer device (114), the at least one spectrometer device being configured to acquire spectral data of the sample (112) within at least one spatial measurement range (118) of the spectrometer device (114) by using the at least one spectrometer device (114); II. at least one imaging device (134), the at least one imaging device being configured to acquire image data of a scene (136) within a field of view (138) of the imaging device (134), the scene (136) including at least a portion of the sample (112) and at least a portion of a spatial measurement range (118) of the spectrometer device (114); as well as III. At least one processing device (154) configured to analyze the image data and / or the spectral data to determine at least one identification information about the sample (112), wherein the processing device (154) is configured to evaluate the spectral data to obtain at least one item of sample information, wherein the evaluation includes processing the spectral data using at least one chemometric model, wherein the chemometric model is configured to convert the spectral data into chemical composition information, wherein the chemometric model is selected based on the identification information.

10. System (110) according to the preceding claim, wherein The system (110) is configured to perform a method according to any of the preceding method-related claims.

11. A computer program comprising instructions which, when executed by a processing device (154) of a system (110) according to any of the preceding claims relating to a system (110), cause the system (110) to perform a method according to any of the preceding claims relating to a method.

12. A computer-readable storage medium comprising instructions which, when executed by a processing device (154) of a system (110) according to any one of the preceding claims relating to a system (110), cause the system (110) to perform a method according to any one of the preceding claims relating to a method.

13. A non-transitory computer readable medium comprising instructions which, when executed by one or more processors (156), cause the one or more processors (156) to perform the method according to any one of the preceding method-related claims.

Images