Method and system for sharing calibration models of multiple micro spectrometers
By constructing a calibration sample set and a spectral transfer sample set, using a standard spectrometer to construct a target property detection model and determine the spectral transfer path, the stability and consistency problems of micro spectrometers are solved, and efficient deployment and accuracy detection of multiple micro spectrometers are achieved.
Patent Information
- Application Number
- CN202411213087.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-08-30
AI Technical Summary
The poor stability and consistency of existing micro spectrometers lead to failure of calibration models when equipment performance changes, limiting the efficient deployment and portability of multiple micro spectrometers.
By constructing the calibration sample set and spectral transfer sample set, a standard spectrometer is used to construct the target property detection model, and the spectral transfer path of the micro spectrometer is determined according to the spectral transfer algorithm, so as to realize the calibration model sharing and standardization processing of multiple micro spectrometers.
It improves the stability and consistency of the micro spectrometer, realizes the efficient deployment of multiple micro spectrometers and multi-scene applications, ensuring the accuracy and reliability of material properties detection.
Smart Images

Figure CN119089204A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of spectrometers, and in particular to a method and system for sharing calibration models of multiple miniature spectrometers. Background Art
[0002] Spectroscopy has been proven to be a feasible solution for simultaneous, accurate and rapid detection of multiple properties in a substance. In recent years, with the development of micro-manufacturing technology, the detection accuracy of micro-spectrometers has basically reached the detection accuracy of laboratory high-performance spectrometers.
[0003] At present, the emergence of micro-spectrometers has greatly reduced the investment cost of material property detection, and the portability brought by the small size of micro-spectrometers has also made it possible to detect material properties on site. However, the small size and low cost of micro-spectrometers will greatly simplify the spectroscopic elements, calibration units and cooling systems, resulting in poor stability of a single micro-spectrometer and poor consistency between different micro-spectrometers.
[0004] In practical applications, the stability of micro-spectrometers and the consistency between different micro-spectrometers are poor. If a calibration model is developed based on a micro-spectrometer, the calibration model will fail as the performance of a single device changes and when it is used between different micro-spectrometers. This greatly limits the application of portable and low-cost micro-spectrometers in actual production and makes it impossible to achieve efficient deployment of multiple micro-spectrometers. Summary of the invention
[0005] The present invention provides a method and system for sharing calibration models of multiple micro-spectrometers, so as to solve the defects of poor feasibility of sharing calibration models of multiple micro-spectrometers and inability to achieve efficient deployment of multiple micro-spectrometers in the prior art, thereby achieving sharing of calibration models of multiple micro-spectrometers and efficient deployment of multiple micro-spectrometers.
[0006] The present invention provides a method for sharing calibration models of multiple micro-spectrometers, which is applied to a micro-spectrometer system, wherein the micro-spectrometer system includes multiple micro-spectrometers and the standard spectrometer; the method includes: constructing a calibration sample set, wherein the calibration sample set includes multiple calibration samples and a target property index reference value corresponding to each calibration sample; obtaining a first spectrum corresponding to each calibration sample collected by the standard spectrometer; constructing the target property detection model by a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample; selecting some calibration samples from the calibration sample set to construct a spectrum transfer sample set; performing standardization processing on each micro-spectrometer to realize the sharing of calibration models of the multiple micro-spectrometers; wherein the standardization processing includes: obtaining a second spectrum corresponding to each calibration sample in the spectrum transfer sample set collected by the micro-spectrometer; and determining the spectrum transfer path corresponding to the micro-spectrometer according to a spectrum transfer algorithm based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set.
[0007] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, after performing standardization processing on each micro-spectrometer to realize sharing of the calibration models of the multiple micro-spectrometers, the method further includes: acquiring an initial spectrum of the sample to be tested acquired by the micro-spectrometer; performing spectral transfer on the initial spectrum of the sample to be tested according to a spectral transfer path corresponding to the micro-spectrometer to obtain a standardized spectrum of the sample to be tested; wherein the spectral transfer path corresponding to the micro-spectrometer is a transfer path required for transferring the spectrum corresponding to the micro-spectrometer to the spectrum corresponding to the standard spectrometer; inputting the standardized spectrum of the sample to be tested into a target property calibration model to obtain a target property index value corresponding to the sample to be tested output by the target property calibration model; wherein the target property calibration model is constructed based on the standard spectrometer.
[0008] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, after constructing a target property detection model by a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample, the method includes: constructing a first verification sample set, the first verification sample set including multiple first verification samples, each first verification sample corresponding to the spectrum of the standard spectrometer, and the target property index reference value corresponding to each first verification sample; verifying the target property detection model based on the first verification sample set.
[0009] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, after determining the spectral transfer path corresponding to the micro-spectrometer according to a spectral transfer algorithm based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set, the method further includes: constructing a second verification sample set, the second verification sample set including multiple second verification samples, each second verification sample corresponding to the spectrum of the standard spectrometer, and each second verification sample corresponding to the spectrum of the micro-spectrometer; and verifying the spectral transfer path corresponding to the micro-spectrometer based on the second verification sample set.
[0010] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, the spectral transfer path corresponding to the micro-spectrometer is determined according to a spectral transfer algorithm based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set, including: based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set, according to the spectral transfer algorithm, calculating the transfer function between the spectrum collected by the micro-spectrometer and the spectrum collected by the standard spectrometer; based on the transfer function between the spectrum collected by the micro-spectrometer and the spectrum collected by the standard spectrometer, determining the spectral transfer path corresponding to the micro-spectrometer.
[0011] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, the method also includes: obtaining multiple samples to be tested, the number of the multiple samples to be tested is consistent with the number of the multiple micro-spectrometers; collecting the multiple samples to be tested by the multiple micro-spectrometers to obtain an initial spectrum corresponding to each sample to be tested; spectrally transferring the initial spectrum corresponding to each sample to be tested based on the spectrum transfer path corresponding to each micro-spectrometer, to obtain a standardized spectrum of each sample to be tested; inputting the standardized spectrum of each sample to be tested into the target property calibration model to obtain a target property index value for each sample to be tested output by the target property calibration model.
[0012] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, the multiple samples to be tested are obtained by preparing different positions of a large sample or a complex sample.
[0013] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, the multiple micro-spectrometers can be arranged at different processing positions of an industrial assembly line.
[0014] According to a method for sharing calibration models of multiple micro-spectrometers provided by the present invention, the multiple micro-spectrometers are used to detect samples to be tested at different stages of a synthesis process or a processing process.
[0015] The present invention also provides a device for sharing calibration models of multiple miniature spectrometers, comprising a first construction module for constructing a calibration sample set, wherein the calibration sample set includes multiple calibration samples and a target property index reference value corresponding to each calibration sample; a first acquisition module for obtaining a first spectrum corresponding to each calibration sample acquired by the standard spectrometer; a second construction module for constructing the target property detection model by a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample; a selection module for selecting some calibration samples from the calibration sample set to construct a spectrum transfer sample set; a processing module for performing standardization processing on each miniature spectrometer to realize the sharing of calibration models of the multiple miniature spectrometers; wherein the standardization processing includes: obtaining a second spectrum corresponding to each calibration sample in the spectrum transfer sample set acquired by the miniature spectrometer; and determining the spectrum transfer path corresponding to the miniature spectrometer according to a spectrum transfer algorithm based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set.
[0016] The present invention also provides a micro-spectrometer system, comprising a plurality of micro-spectrometers, a standard spectrometer and a calibration model sharing device for the plurality of micro-spectrometers.
[0017] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for sharing calibration models of multiple micro-spectrometers as described in any one of the above is implemented.
[0018] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for sharing calibration models of multiple micro-spectrometers as described in any one of the above is implemented.
[0019] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for sharing calibration models of multiple micro-spectrometers as described in any one of the above is implemented.
[0020] The method and system for sharing calibration models of multiple micro-spectrometers provided by the present invention construct a calibration sample set; obtain the first spectrum corresponding to each calibration sample collected by a standard spectrometer; construct a target property detection model through a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample; select some calibration samples from the calibration sample set to construct a spectrum transfer sample set; and perform standardization processing for each micro-spectrometer. The scheme of the present invention can realize the sharing of calibration models of multiple micro-spectrometers, can realize the efficient deployment of multiple micro-spectrometers, and provide support for multi-scenario applications of micro-spectrometer systems. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0022] Figure 1 This is a schematic structural diagram of the micro-spectrometer system provided by the present invention.
[0023] Figure 2 This is one of the flow charts of the method for sharing calibration models of multiple micro-spectrometers provided by the present invention.
[0024] Figure 3 This is the second flow chart of the method for sharing calibration models of multiple micro-spectrometers provided by the present invention.
[0025] Figure 4 It is a structural schematic diagram of a calibration model sharing device for multiple miniature spectrometers provided by the present invention.
[0026] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION
[0027] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0028] The following specific embodiments are used to describe in detail the technical solution of the present application and how the technical solution of the present application solves the above technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. Figures 1 to 3 The present invention describes a method for sharing calibration models of multiple micro-spectrometers.
[0029] The method for sharing calibration models of multiple micro-spectrometers provided in this embodiment is applied to a micro-spectrometer system. Figure 1 The schematic diagram of the structure of the micro-spectrometer system provided by the present invention is as follows: Figure 1 As shown, the micro-spectrometer system includes multiple micro-spectrometers and a standard spectrometer.
[0030] In practical applications, the execution subject of the method for sharing calibration models of multiple micro-spectrometers can be multiple micro-spectrometer calibration model sharing devices, and there are many ways to implement the multiple micro-spectrometer calibration model sharing devices, for example, it can be implemented by computer programs, such as application software, etc.; or, for example, chips, etc. It can also be implemented as a medium storing relevant computer programs, such as a U disk, a cloud disk, etc.; or it can also be implemented by a physical device integrated or installed with relevant computer programs, such as a server, etc.
[0031] Figure 2 This is one of the flow charts of the method for sharing calibration models of multiple micro-spectrometers provided by the present invention, such as Figure 2 As shown, the method includes steps 101 to 105.
[0032] Step 101: construct a calibration sample set, where the calibration sample set includes a plurality of calibration samples and a target property index reference value corresponding to each calibration sample.
[0033] Step 102: Acquire a first spectrum corresponding to each calibration sample acquired by a standard spectrometer.
[0034] Specifically, multiple representative samples of the target substance are collected, and the multiple representative samples are pre-processed to obtain multiple calibration samples. The target property index reference value corresponding to each calibration sample is measured to construct a calibration sample set. Furthermore, the first spectrum corresponding to each calibration sample in the calibration sample set is collected by a standard spectrometer.
[0035] Step 103: construct a target property detection model by a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample.
[0036] Among them, the target property calibration model is a spectral analysis model. According to the first spectrum corresponding to each calibration sample collected by the standard spectrometer, the true value of the target property index corresponding to each calibration sample can be output. In practical applications, the accuracy of the target property detection model can be verified by calculating the error between the true value of the target property index corresponding to each calibration sample and the reference value of the target property index corresponding to each calibration sample. If the error is large, the parameters of the target property detection model can be adjusted until the error meets the requirements, that is, the required target property detection model is obtained. It can be understood that in the constructed target property detection model, the error between the true value of the target property index corresponding to each calibration sample input and the reference value of the target property index corresponding to each calibration sample is within a certain range.
[0037] Step 104: Select some calibration samples from the calibration sample set to construct a spectrum transfer sample set.
[0038] In practical applications, some calibration samples can be selected from the calibration sample set according to the first spectrum corresponding to each calibration sample in the calibration sample set and the property differences of multiple calibration samples. In the present invention, the method for selecting some calibration samples is not specifically limited. For example, the selection method includes but is not limited to the concentration gradient method and the spectral distance method.
[0039] Step 105: For each micro-spectrometer, perform standardization processing to achieve the sharing of calibration models of multiple micro-spectrometers. The standardization processing includes: obtaining the second spectrum corresponding to each calibration sample in the spectrum transfer sample set collected by the micro-spectrometer. Based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set, determine the spectrum transfer path corresponding to the micro-spectrometer according to the spectrum transfer algorithm.
[0040] In combination with the above description, the spectrum transfer path corresponding to the micro-spectrometer is the transfer path required for the spectrum corresponding to the micro-spectrometer to be transferred to the spectrum corresponding to the standard spectrometer. Therefore, determining the spectrum transfer path corresponding to the micro-spectrometer means determining the correspondence between the spectrum corresponding to the micro-spectrometer and the spectrum corresponding to the standard spectrometer.
[0041] Specifically, the first spectrum and the second spectrum corresponding to each calibration sample are collected by the standard spectrometer and the micro spectrometer, respectively. Therefore, in this embodiment, based on the first spectrum and the second spectrum corresponding to each calibration sample, the spectrum transfer path corresponding to the micro spectrometer is determined according to the spectrum transfer algorithm.
[0042] As an example, for a method of determining a spectral transfer path corresponding to a micro-spectrometer, in one example, the spectral transfer path corresponding to the micro-spectrometer is determined based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set according to a spectral transfer algorithm, including: based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set, according to the spectral transfer algorithm, calculating a transfer function between a spectrum collected by the micro-spectrometer and a spectrum collected by a standard spectrometer; based on the transfer function between a spectrum collected by the micro-spectrometer and a spectrum collected by the standard spectrometer, determining the spectral transfer path corresponding to the micro-spectrometer.
[0043] It should be noted that the present invention does not specifically limit the spectrum transfer algorithm, and the spectrum transfer algorithm includes but is not limited to the spectrum difference correction method, the direct standardization method and the slope intercept correction method.
[0044] In practical applications, multiple micro-spectrometers can be deployed according to needs. On the one hand, based on the solution of the present invention, multiple micro-spectrometers can be calibrated based on a standard spectrometer, saving the cost of material property detection. On the other hand, based on the solution of the present invention, accurate calibration of multiple micro-spectrometers can be achieved, improving the accuracy and reliability of material detection.
[0045] In this embodiment, a calibration sample set is constructed; a first spectrum corresponding to each calibration sample collected by a standard spectrometer is obtained; a target property detection model is constructed by a chemometric method based on the first spectrum corresponding to each calibration sample and the reference value of the target property index corresponding to each calibration sample; some calibration samples are selected from the calibration sample set to construct a spectral transfer sample set; for each micro-spectrometer, standardization processing is performed to ensure accurate calibration of multiple micro-spectrometers, improve the stability of the micro-spectrometer and the consistency between different micro-spectrometers, realize the sharing of calibration models of multiple micro-spectrometers, enable efficient deployment of multiple micro-spectrometers, and provide support for multi-scenario applications of micro-spectrometer systems.
[0046] As an example, Figure 3 FIG. 2 is a flow chart of a method for sharing calibration models of multiple micro-spectrometers provided by the present invention. Figure 3 As shown, after the above step 105, the above method for sharing calibration models of multiple micro-spectrometers further includes: step 201, step 202 and step 203.
[0047] Step 201: Acquire an initial spectrum of the sample to be tested collected by a micro-spectrometer.
[0048] Among them, the micro-spectrometer is the core component of the spectral measurement system. The micro-spectrometer is a basic device that uses optical technology, electronic technology and computer technology to analyze and measure the composition and structure of substances. The specific modularization and high-speed acquisition characteristics of the micro-spectrometer have been widely used in system integration and on-site detection. Combined with light sources, optical fibers, and measurement accessories, it can be combined into various optical measurement systems. Due to its small size, it is easy to flexibly build a spectral system and is widely used in environmental monitoring, industrial control, chemical analysis, food quality testing, material analysis, clinical testing, aerospace remote sensing, and scientific education.
[0049] In practical applications, micro-spectrometers can be used for transmission absorption measurement, reflection measurement, laser measurement, fluorescence measurement, oxygen content measurement, Raman spectroscopy measurement, laser-induced breakdown spectroscopy (LIBS) measurement, light-emitting diode (LED) measurement, etc.
[0050] Combined with specific application scenarios, the micro-spectrometer can quickly detect multiple properties in a substance, such as material composition analysis. In this embodiment, the sample to be tested refers to a sample of the substance to be tested, and can be any part selected from the substance to be tested. As an example, the micro-spectrometer can be used to detect the carbon and nitrogen content in the soil. In this scenario, the sample to be tested is a sample of the soil to be tested. In practical applications, after obtaining the sample to be tested, the sample to be tested can be pre-processed, for example, to remove interfering substances, remove unrepresentative samples, etc.
[0051] In practical applications, the sample to be tested is placed on the path of the incident light. The sample to be tested absorbs, reflects or scatters the incident light. The micro-spectrometer collects the reflected light of the sample to be tested and determines the initial spectrum of the sample to be tested. It is understandable that the initial spectrum of the sample to be tested can characterize the properties of the sample to be tested. Furthermore, the properties of the sample to be tested can be determined by analyzing the initial spectrum of the sample to be tested.
[0052] Step 202: according to the spectrum transfer path corresponding to the micro-spectrometer, the initial spectrum of the sample to be tested is spectrum transferred to obtain the standardized spectrum of the sample to be tested.
[0053] The spectrum transfer path corresponding to the micro-spectrometer is a transfer path required for transferring the spectrum corresponding to the micro-spectrometer to the spectrum corresponding to the standard spectrometer. The spectrum transfer path corresponding to the micro-spectrometer characterizes the relationship between the spectrum corresponding to the micro-spectrometer and the spectrum corresponding to the standard spectrometer. As an example, the spectrum transfer path includes a spectrum transfer method, spectrum transfer parameters, etc. In practical applications, the spectrum transfer path corresponding to the micro-spectrometer can be predetermined according to a spectrum transfer algorithm.
[0054] In practical applications, multiple micro-spectrometers can be set up, and the multiple micro-spectrometers are calibrated by a standard spectrometer. Specifically, each micro-spectrometer corresponds to a spectral transfer path. For example, three micro-spectrometers are set up, namely micro-spectrometer 1, micro-spectrometer 2, and micro-spectrometer 3. Micro-spectrometer 1 corresponds to spectral transfer path 1, micro-spectrometer 2 corresponds to spectral transfer path 2, and micro-spectrometer 3 corresponds to spectral transfer path 3. For example, according to the spectral transfer path 1 corresponding to micro-spectrometer 1, the initial spectrum 1 of the sample to be tested collected by micro-spectrometer 1 can be spectrally transferred to obtain the standardized spectrum 1 of the sample to be tested. For another example, according to the spectral transfer path 2 corresponding to micro-spectrometer 2, the initial spectrum 2 of the sample to be tested collected by micro-spectrometer 2 can be spectrally transferred to obtain the standardized spectrum 2 of the sample to be tested. It can be understood that according to the spectral transfer path corresponding to the micro-spectrometer, the initial spectrum of the sample to be tested can be spectrally transferred to obtain the spectrum corresponding to the standard spectrometer, that is, the standardized spectrum of the sample to be tested.
[0055] It should be noted that when a new micro-spectrometer needs to be deployed, the spectral transfer path corresponding to the micro-spectrometer needs to be determined so that the initial spectrum of the sample to be tested collected by the micro-spectrometer can be subsequently spectrally transferred to obtain a standardized spectrum of the sample to be tested.
[0056] Furthermore, after obtaining the standardized spectrum of the sample to be tested, the standardized spectrum of the sample to be tested can be analyzed based on a standard spectrometer to achieve detection of target properties.
[0057] Step 203: input the standardized spectrum of the sample to be tested into the target property calibration model to obtain the target property index value corresponding to the sample to be tested output by the target property calibration model.
[0058] The target property calibration model is constructed based on a standard spectrometer. The standard spectrometer is a calibration model for the micro spectrometer and can be a laboratory high-performance spectrometer. In this embodiment, the model, size, weight and performance parameters of the standard spectrometer and the micro spectrometer are not specifically limited.
[0059] It can be understood that the target property calibration model is a spectral analysis model, and the target property index value corresponding to the sample to be tested can be obtained according to the standardized spectrum of the input sample to be tested. In practical applications, the target property calibration model is constructed based on a standard spectrometer. Specifically, multiple representative samples of the target substance are collected, and the target property index value of each representative sample is determined; the spectrum of each representative sample collected by the standard spectrometer is obtained. In one example, based on the target property index value of each representative sample and the spectrum of each representative sample, a target property calibration model is established by a chemometric method. In another example, an initial target property calibration model is constructed based on machine learning, a training set is constructed based on the target property index value of each representative sample and the spectrum of each representative sample, and the initial target property calibration model is trained until the target property calibration model is obtained.
[0060] In this embodiment, according to the spectrum transfer path corresponding to the micro-spectrometer, the initial spectrum of the sample to be tested collected by the micro-spectrometer is spectrally transferred to obtain the spectrum under the standard spectrometer corresponding to the initial spectrum of the sample to be tested, that is, the standardized spectrum of the sample to be tested. Among them, the standard spectrometer is the calibration model of the micro-spectrometer, which is a laboratory high-performance spectrometer. Further, the standardized spectrum of the sample to be tested is input into the target property calibration model constructed based on the standard spectrometer to obtain the target property index value corresponding to the sample to be tested. Therefore, the scheme of this embodiment realizes the accurate calibration of the micro-spectrometer, improves the stability of the micro-spectrometer and the consistency between different micro-spectrometers, thereby improving the accuracy of material property detection.
[0061] In addition, in order to further improve the accuracy of material property detection, an independent verification sample set with target property index values can be established to evaluate the performance of the target material property calibration model and the performance of the micro-spectrometer in detecting the target material property. The evaluation index is not specifically limited in the present invention, and includes but is not limited to the determination coefficient and the root mean square error value.
[0062] As an example, in a possible implementation, after the above step 103, the above method for sharing calibration models of multiple micro-spectrometers also includes: constructing a first verification sample set, the first verification sample set including multiple first verification samples, each first verification sample corresponding to the spectrum of a standard spectrometer, and a target property index reference value corresponding to each first verification sample; verifying the target property detection model based on the first verification sample set.
[0063] In practical applications, the target property detection model is verified based on the first verification sample set. Specifically, the difference between the actual value of the target property index corresponding to each first verification sample output by the target property detection model and the target property reference value can be calculated, for example, the determination coefficient, the root mean square error value, etc. If the difference value meets the requirements, the target property detection model is determined to be accurate; if the difference value does not meet the requirements, the target property detection model is adjusted until the target property detection model meets the requirements.
[0064] In this implementation, the target property detection model is verified based on the first verification sample set, which improves the stability and accuracy of the target property detection model, thereby improving the accuracy of material property detection.
[0065] As an example, in a possible implementation, after determining the spectral transfer path corresponding to the micro-spectrometer according to the spectral transfer algorithm based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set, the above-mentioned method for sharing calibration models of multiple micro-spectrometers also includes: constructing a second verification sample set, the second verification sample set including multiple second verification samples, each second verification sample corresponding to the spectrum of the standard spectrometer, and each second verification sample corresponding to the spectrum of the micro-spectrometer; verifying the spectral transfer path corresponding to the micro-spectrometer based on the second verification sample set.
[0066] In practical applications, the spectral transfer path corresponding to the micro-spectrometer is verified based on the second verification sample set. Specifically, the difference between the spectrum obtained by spectrally transferring the spectrum of each second verification sample corresponding to the standard spectrometer according to the spectral transfer path corresponding to the micro-spectrometer and the spectrum of each second verification sample corresponding to the micro-spectrometer can be compared to verify the accuracy of the spectral transfer path corresponding to the micro-spectrometer.
[0067] In this implementation, the spectrum transfer path corresponding to the micro-spectrometer is verified based on the second verification sample set, which improves the accuracy of the spectrum transfer path corresponding to the micro-spectrometer, thereby improving the accuracy of material property detection.
[0068] In combination with the above description, the micro-spectrometer system includes multiple micro-spectrometers. Figure 1 As shown, the micro-spectrometer system includes multiple micro-spectrometers, a standard spectrometer, and a target property calibration model, which can realize the efficient deployment of multiple micro-spectrometers. In one example, multiple micro-spectrometers can detect different samples in parallel to improve the efficiency of material detection.
[0069] In another example, in order to accurately detect the properties of the sample to be tested, the sample to be tested can be tested by multiple micro-spectrometers respectively, and the target property index value corresponding to the sample to be tested can be determined based on the detection results obtained by the multiple micro-spectrometers.
[0070] As an example, in one possible implementation, the above-mentioned method for sharing calibration models of multiple micro-spectrometers is applied to a micro-spectrometer system, and the micro-spectrometer system includes multiple micro-spectrometers and a standard spectrometer; the above-mentioned method for sharing calibration models of multiple micro-spectrometers also includes: obtaining multiple samples to be tested, and the number of the multiple samples to be tested is consistent with the number of the multiple micro-spectrometers; collecting the multiple samples to be tested by multiple micro-spectrometers to obtain an initial spectrum corresponding to each sample to be tested; based on the spectrum transfer path corresponding to each micro-spectrometer, spectrally transferring the initial spectrum corresponding to each sample to be tested to obtain a standardized spectrum of each sample to be tested; inputting the standardized spectrum of each sample to be tested into the target property calibration model to obtain the target property index value of each sample to be tested output by the target property calibration model.
[0071] In practical applications, there is a need to perform material detection on multiple samples to be tested. For example, for large samples or complex samples, it is necessary to perform material detection on multiple locations separately. In one example, multiple samples to be tested are obtained by preparing different locations of a large sample or a complex sample, and multiple micro-spectrometers in a micro-spectrometer system are used to perform material property detection on multiple samples to be tested, thereby improving the efficiency of material property detection.
[0072] For another example, during the operation of an industrial assembly line, it is necessary to detect the material properties of the materials obtained in different steps. In one example, multiple micro-spectrometers can be set at different processing positions of the industrial assembly line, so that the products at different processing positions of the industrial assembly line can be detected, thereby improving the efficiency of material property detection.
[0073] In one example, multiple micro-spectrometers are used to detect samples at different stages of a synthesis or processing process, thereby improving the efficiency of detecting material properties.
[0074] It should be noted that the above-mentioned multiple samples to be tested may not come from one sample, and the multiple samples to be tested may also be prepared from different positions of a large sample or a complex sample. In this embodiment, multiple micro-spectrometers can be used to achieve parallel material detection of multiple samples to be tested, thereby improving the efficiency of material detection.
[0075] The method and device for sharing calibration models of multiple micro-spectrometers provided in this embodiment are as follows: constructing a calibration sample set; obtaining the first spectrum corresponding to each calibration sample collected by a standard spectrometer; constructing a target property detection model through a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample; selecting some calibration samples from the calibration sample set to construct a spectrum transfer sample set; and performing standardization processing for each micro-spectrometer. The scheme of this embodiment can realize the sharing of calibration models of multiple micro-spectrometers, can realize the efficient deployment of multiple micro-spectrometers, and provide support for multi-scenario applications of micro-spectrometer systems.
[0076] The following is a description of a calibration model sharing device for multiple micro-spectrometers provided by the present invention. The calibration model sharing device for multiple micro-spectrometers described below and the calibration model sharing method for multiple micro-spectrometers described above can refer to each other.
[0077] Figure 4 Schematic diagram of the structure of the calibration model sharing device for multiple micro-spectrometers provided by the present invention. Figure 4 As shown, the device for sharing calibration models of multiple micro-spectrometers includes: a first construction module 41, a first acquisition module 42, a second construction module 43, a selection module 44 and a processing module 45.
[0078] The first construction module 41 is used to construct a calibration sample set, which includes a plurality of calibration samples and a target property index reference value corresponding to each calibration sample.
[0079] The first acquisition module 42 is used to acquire a first spectrum corresponding to each calibration sample acquired by the standard spectrometer.
[0080] Specifically, multiple representative samples of the target substance are collected, and the multiple representative samples are pre-processed to obtain multiple calibration samples. The target property index reference value corresponding to each calibration sample is measured to construct a calibration sample set. Furthermore, the first spectrum corresponding to each calibration sample in the calibration sample set is collected by a standard spectrometer.
[0081] The second construction module 43 is used to construct a target property detection model through a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample.
[0082] Among them, the target property calibration model is a spectral analysis model. According to the first spectrum corresponding to each calibration sample collected by the standard spectrometer, the true value of the target property index corresponding to each calibration sample can be output. In practical applications, the accuracy of the target property detection model can be verified by calculating the error between the true value of the target property index corresponding to each calibration sample and the reference value of the target property index corresponding to each calibration sample. If the error is large, the parameters of the target property detection model can be adjusted until the error meets the requirements, that is, the required target property detection model is obtained. It can be understood that in the constructed target property detection model, the error between the true value of the target property index corresponding to each calibration sample input and the reference value of the target property index corresponding to each calibration sample is within a certain range.
[0083] The selection module 44 is used to select some calibration samples from the calibration sample set to construct a spectrum transfer sample set.
[0084] In practical applications, some calibration samples can be selected from the calibration sample set according to the first spectrum corresponding to each calibration sample in the calibration sample set and the property differences of multiple calibration samples. In the present invention, the method for selecting some calibration samples is not specifically limited. For example, the selection method includes but is not limited to the concentration gradient method and the spectral distance method.
[0085] The processing module 45 is used to perform standardization processing on each micro-spectrometer to realize the sharing of calibration models of multiple micro-spectrometers. The standardization processing includes: obtaining the second spectrum corresponding to each calibration sample in the spectrum transfer sample set collected by the micro-spectrometer. Based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set, the spectrum transfer path corresponding to the micro-spectrometer is determined according to the spectrum transfer algorithm.
[0086] In combination with the above description, the spectrum transfer path corresponding to the micro-spectrometer is the transfer path required for the spectrum corresponding to the micro-spectrometer to be transferred to the spectrum corresponding to the standard spectrometer. Therefore, determining the spectrum transfer path corresponding to the micro-spectrometer means determining the correspondence between the spectrum corresponding to the micro-spectrometer and the spectrum corresponding to the standard spectrometer.
[0087] Specifically, the first spectrum and the second spectrum corresponding to each calibration sample are collected by the standard spectrometer and the micro spectrometer, respectively. Therefore, in this embodiment, the processing module 45 determines the spectrum transfer path corresponding to the micro spectrometer based on the first spectrum and the second spectrum corresponding to each calibration sample according to the spectrum transfer algorithm.
[0088] As an example, for a method of determining a spectral transfer path corresponding to a micro-spectrometer, in one example, the processing module 45 is used to determine the spectral transfer path corresponding to the micro-spectrometer based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set and according to a spectral transfer algorithm. Specifically, it is used to: calculate a transfer function between a spectrum collected by the micro-spectrometer and a spectrum collected by the standard spectrometer based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectral transfer sample set and according to the spectral transfer algorithm; determine the spectral transfer path corresponding to the micro-spectrometer based on the transfer function between the spectrum collected by the micro-spectrometer and the spectrum collected by the standard spectrometer.
[0089] It should be noted that the present invention does not specifically limit the spectrum transfer algorithm, and the spectrum transfer algorithm includes but is not limited to the spectrum difference correction method, the direct standardization method and the slope intercept correction method.
[0090] In practical applications, multiple micro-spectrometers can be deployed according to needs. On the one hand, based on the solution of the present invention, multiple micro-spectrometers can be calibrated based on a standard spectrometer, saving the cost of material property detection. On the other hand, based on the solution of the present invention, accurate calibration of multiple micro-spectrometers can be achieved, improving the accuracy and reliability of material detection.
[0091] In this embodiment, a calibration sample set is constructed; a first spectrum corresponding to each calibration sample collected by a standard spectrometer is obtained; a target property detection model is constructed by a chemometric method based on the first spectrum corresponding to each calibration sample and the reference value of the target property index corresponding to each calibration sample; some calibration samples are selected from the calibration sample set to construct a spectral transfer sample set; for each micro-spectrometer, standardization processing is performed to ensure accurate calibration of multiple micro-spectrometers, improve the stability of the micro-spectrometer and the consistency between different micro-spectrometers, realize the sharing of calibration models of multiple micro-spectrometers, enable efficient deployment of multiple micro-spectrometers, and provide support for multi-scenario applications of micro-spectrometer systems.
[0092] In a possible implementation manner, the device for sharing calibration models of multiple micro-spectrometers further includes: an acquisition module, a conversion module and an input module.
[0093] The acquisition module is used to acquire the initial spectrum of the sample to be tested collected by the micro-spectrometer.
[0094] Among them, the micro-spectrometer is the core component of the spectral measurement system. The micro-spectrometer is a basic device that uses optical technology, electronic technology and computer technology to analyze and measure the composition and structure of substances. The specific modularization and high-speed acquisition characteristics of the micro-spectrometer have been widely used in system integration and on-site detection. Combined with light sources, optical fibers, and measurement accessories, it can be combined into various optical measurement systems. Due to its small size, it is easy to flexibly build a spectral system and is widely used in environmental monitoring, industrial control, chemical analysis, food quality testing, material analysis, clinical testing, aerospace remote sensing, and scientific education.
[0095] In combination with specific application scenarios, the micro-spectrometer can quickly detect multiple properties in a substance, such as material composition analysis. In this embodiment, the sample to be tested refers to a sample of the substance to be tested, and can be any part selected from the substance to be tested.
[0096] In practical applications, the sample to be tested is placed on the path of the incident light. The sample to be tested absorbs, reflects or scatters the incident light. The micro-spectrometer collects the reflected light of the sample to be tested and determines the initial spectrum of the sample to be tested. It is understandable that the initial spectrum of the sample to be tested can characterize the properties of the sample to be tested. Furthermore, the properties of the sample to be tested can be determined by analyzing the initial spectrum of the sample to be tested.
[0097] The conversion module 42 is used to perform spectral transfer on the initial spectrum of the sample to be tested according to the spectral transfer path corresponding to the micro-spectrometer to obtain a standardized spectrum of the sample to be tested.
[0098] The spectrum transfer path corresponding to the micro-spectrometer is a transfer path required for transferring the spectrum corresponding to the micro-spectrometer to the spectrum corresponding to the standard spectrometer. The spectrum transfer path corresponding to the micro-spectrometer characterizes the relationship between the spectrum corresponding to the micro-spectrometer and the spectrum corresponding to the standard spectrometer. As an example, the spectrum transfer path includes a spectrum transfer method, spectrum transfer parameters, etc. In practical applications, the spectrum transfer path corresponding to the micro-spectrometer can be predetermined according to a spectrum transfer algorithm.
[0099] In practical applications, multiple micro-spectrometers can be set up, and the multiple micro-spectrometers are calibrated by a standard spectrometer. Specifically, each micro-spectrometer corresponds to a spectrum transfer path. It can be understood that according to the spectrum transfer path corresponding to the micro-spectrometer, the initial spectrum of the sample to be tested can be spectrally transferred to obtain the spectrum corresponding to the standard spectrometer, that is, the standardized spectrum of the sample to be tested.
[0100] It should be noted that when a new micro-spectrometer needs to be deployed, the spectral transfer path corresponding to the micro-spectrometer needs to be determined so that the initial spectrum of the sample to be tested collected by the micro-spectrometer can be subsequently spectrally transferred to obtain a standardized spectrum of the sample to be tested.
[0101] Furthermore, after obtaining the standardized spectrum of the sample to be tested, the standardized spectrum of the sample to be tested can be analyzed based on a standard spectrometer to achieve detection of target properties.
[0102] The above-mentioned input module is used to input the standardized spectrum of the sample to be tested into the target property calibration model to obtain the target property index value corresponding to the sample to be tested output by the target property calibration model; wherein the target property calibration model is constructed based on a standard spectrometer.
[0103] The standard spectrometer is a calibration model of the micro spectrometer, and can be a laboratory high-performance spectrometer. In this embodiment, the model, size, weight and performance parameters of the standard spectrometer and the micro spectrometer are not specifically limited.
[0104] It can be understood that the target property calibration model is a spectral analysis model, and the target property index value corresponding to the sample to be tested can be obtained according to the standardized spectrum of the input sample to be tested. In practical applications, the target property calibration model is constructed based on a standard spectrometer. Specifically, multiple representative samples of the target substance are collected, and the target property index value of each representative sample is determined; the spectrum of each representative sample collected by the standard spectrometer is obtained. In one example, based on the target property index value of each representative sample and the spectrum of each representative sample, a target property calibration model is established by a chemometric method. In another example, an initial target property calibration model is constructed based on machine learning, a training set is constructed based on the target property index value of each representative sample and the spectrum of each representative sample, and the initial target property calibration model is trained until the target property calibration model is obtained.
[0105] In this embodiment, the conversion module performs spectral transfer on the initial spectrum of the sample to be tested acquired by the micro-spectrometer acquired by the acquisition module according to the spectrum transfer path corresponding to the micro-spectrometer, and obtains the spectrum under the standard spectrometer corresponding to the initial spectrum of the sample to be tested, that is, the standardized spectrum of the sample to be tested. Among them, the standard spectrometer is the calibration model of the micro-spectrometer, which is a laboratory high-performance spectrometer. Furthermore, the input module inputs the standardized spectrum of the sample to be tested into the target property calibration model constructed based on the standard spectrometer, and obtains the target property index value corresponding to the sample to be tested. Therefore, the scheme of this embodiment realizes the accurate calibration of the micro-spectrometer, improves the stability of the micro-spectrometer and the consistency between different micro-spectrometers, thereby improving the accuracy of material property detection.
[0106] In addition, in order to further improve the accuracy of material property detection, an independent verification sample set with target property index values can be established to evaluate the performance of the target material property calibration model and the performance of the micro-spectrometer in detecting the target material property. The evaluation index is not specifically limited in the present invention, and includes but is not limited to the determination coefficient and the root mean square error value.
[0107] As an example, in one possible implementation, the above-mentioned device for sharing calibration models of multiple micro-spectrometers also includes: a first verification module, used to construct a first verification sample set, the first verification sample set including multiple first verification samples, each first verification sample corresponding to the spectrum of a standard spectrometer, and a target property index reference value corresponding to each first verification sample; verifying the target property detection model based on the first verification sample set.
[0108] In practical applications, the first verification module verifies the target property detection model based on the first verification sample set. Specifically, the difference between the actual value of the target property index corresponding to each first verification sample output by the target property detection model and the target property reference value can be calculated, for example, the determination coefficient, the root mean square error value, etc. If the difference value meets the requirements, the target property detection model is determined to be accurate; if the difference value does not meet the requirements, the target property detection model is adjusted until the target property detection model meets the requirements.
[0109] In this implementation, the first verification module verifies the target property detection model based on the first verification sample set, thereby improving the stability and accuracy of the target property detection model, thereby improving the accuracy of material property detection.
[0110] As an example, in a possible implementation, the above-mentioned device for sharing calibration models of multiple micro-spectrometers also includes: a second verification module, used to construct a second verification sample set, the second verification sample set includes multiple second verification samples, each second verification sample corresponds to the spectrum of a standard spectrometer, and each second verification sample corresponds to the spectrum of the micro-spectrometer; based on the second verification sample set, the spectral transfer path corresponding to the micro-spectrometer is verified.
[0111] In actual applications, the second verification module verifies the spectral transfer path corresponding to the micro-spectrometer based on the second verification sample set. Specifically, the spectrum obtained by spectrally transferring the spectrum of each second verification sample corresponding to the standard spectrometer according to the spectral transfer path corresponding to the micro-spectrometer can be compared with the difference between the spectrum of each second verification sample corresponding to the micro-spectrometer, so as to verify the accuracy of the spectral transfer path corresponding to the micro-spectrometer.
[0112] In this implementation, the second verification module verifies the spectral transfer path corresponding to the micro-spectrometer based on the second verification sample set, thereby improving the accuracy of the spectral transfer path corresponding to the micro-spectrometer and thus improving the accuracy of material property detection.
[0113] In combination with the above description, the micro-spectrometer system includes multiple micro-spectrometers. Figure 1 As shown, the micro-spectrometer system includes multiple micro-spectrometers, a standard spectrometer, and a target property calibration model, which can realize the efficient deployment of multiple micro-spectrometers. In one example, multiple micro-spectrometers can detect different samples in parallel to improve the efficiency of material detection.
[0114] In another example, in order to accurately detect the properties of the sample to be tested, the sample to be tested can be tested by multiple micro-spectrometers respectively, and the target property index value corresponding to the sample to be tested is determined based on the test results obtained by the multiple micro-spectrometers. As an example, in a possible implementation, the above-mentioned multiple micro-spectrometer calibration model sharing device is applied to a micro-spectrometer system, and the micro-spectrometer system includes multiple micro-spectrometers and a standard spectrometer; the above-mentioned multiple micro-spectrometer calibration model sharing device also includes: a preparation module, which is used to obtain multiple samples to be tested, and the number of the multiple samples to be tested is consistent with the number of the multiple micro-spectrometers; the above-mentioned acquisition module is also used to collect multiple samples to be tested through multiple micro-spectrometers to obtain the initial spectrum corresponding to each sample to be tested; the above-mentioned conversion module is also used to perform spectral transfer on the initial spectrum corresponding to each sample to be tested based on the spectral transfer path corresponding to each micro-spectrometer, and obtain the standardized spectrum of each sample to be tested; the above-mentioned input module is also used to input the standardized spectrum of each sample to be tested into the target property calibration model to obtain the target property index value of each sample to be tested output by the target property calibration model.
[0115] In practical applications, there is a need to perform material detection on multiple samples to be tested. For example, for large samples or complex samples, it is necessary to perform material detection on multiple locations separately. In one example, multiple samples to be tested are obtained by preparing different locations of a large sample or a complex sample, and multiple micro-spectrometers in a micro-spectrometer system are used to perform material property detection on multiple samples to be tested, thereby improving the efficiency of material property detection.
[0116] For another example, during the operation of an industrial assembly line, it is necessary to detect the material properties of the materials obtained in different steps. In one example, multiple micro-spectrometers can be set at different processing positions of the industrial assembly line, so that the products at different processing positions of the industrial assembly line can be detected, thereby improving the efficiency of material property detection.
[0117] It should be noted that the above-mentioned multiple samples to be tested may not come from one sample, and the multiple samples to be tested may also be prepared from different positions of a large sample or a complex sample. In this embodiment, multiple micro-spectrometers can be used to achieve parallel material detection of multiple samples to be tested, thereby improving the efficiency of material detection.
[0118] The present embodiment provides a device for sharing calibration models of multiple micro-spectrometers. The first construction module constructs a calibration sample set; the first acquisition module obtains the first spectrum corresponding to each calibration sample acquired by the standard spectrometer; the second construction module constructs a target property detection model through a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample; the selection module selects some calibration samples from the calibration sample set to construct a spectrum transfer sample set; the processing module performs standardization processing on each micro-spectrometer. The solution of the present invention can realize the sharing of calibration models of multiple micro-spectrometers, can realize the efficient deployment of multiple micro-spectrometers, and provide support for multi-scenario applications of micro-spectrometer systems.
[0119] In addition, the present invention also provides a micro-spectrometer system, which includes a plurality of micro-spectrometers, a standard spectrometer, and a calibration model sharing device for a plurality of micro-spectrometers.
[0120] Figure 5 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530 and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call the logic instructions in the memory 530 to execute a method for sharing calibration models of multiple micro-spectrometers, the method comprising: constructing a calibration sample set, the calibration sample set comprising multiple calibration samples and a target property index reference value corresponding to each calibration sample; obtaining a first spectrum corresponding to each calibration sample collected by a standard spectrometer; constructing a target property detection model through a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample; selecting some calibration samples from the calibration sample set to construct a spectrum transfer sample set; performing standardization processing for each micro-spectrometer to realize the sharing of calibration models of multiple micro-spectrometers; wherein the standardization processing comprises: obtaining a second spectrum corresponding to each calibration sample in the spectrum transfer sample set collected by the micro-spectrometer; determining a spectrum transfer path corresponding to the micro-spectrometer according to a spectrum transfer algorithm based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set.
[0121] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.
[0122] On the other hand, the present invention further provides a computer program product, the computer program product comprising a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute the calibration model sharing method for multiple micro-spectrometers provided by the above methods, the method comprising: Construct a calibration sample set, the calibration sample set includes multiple calibration samples and a reference value of a target property index corresponding to each calibration sample; obtain a first spectrum corresponding to each calibration sample acquired by a standard spectrometer; construct a target property detection model through a chemometric method based on the first spectrum corresponding to each calibration sample and the reference value of the target property index corresponding to each calibration sample; select some calibration samples from the calibration sample set to construct a spectrum transfer sample set; perform standardization processing for each micro-spectrometer to realize the sharing of calibration models of multiple micro-spectrometers; wherein the standardization processing includes: obtaining a second spectrum corresponding to each calibration sample in the spectrum transfer sample set acquired by the micro-spectrometer; based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set, according to a spectrum transfer algorithm, determine the spectrum transfer path corresponding to the micro-spectrometer.
[0123] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the method for sharing calibration models of multiple micro-spectrometers provided by the above methods, the method comprising: Construct a calibration sample set, the calibration sample set includes multiple calibration samples and a reference value of a target property index corresponding to each calibration sample; obtain a first spectrum corresponding to each calibration sample acquired by a standard spectrometer; construct a target property detection model through a chemometric method based on the first spectrum corresponding to each calibration sample and the reference value of the target property index corresponding to each calibration sample; select some calibration samples from the calibration sample set to construct a spectrum transfer sample set; perform standardization processing for each micro-spectrometer to realize the sharing of calibration models of multiple micro-spectrometers; wherein the standardization processing includes: obtaining a second spectrum corresponding to each calibration sample in the spectrum transfer sample set acquired by the micro-spectrometer; based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set, according to a spectrum transfer algorithm, determine the spectrum transfer path corresponding to the micro-spectrometer.
[0124] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.
[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for sharing calibration models of multiple micro-spectrometers, characterized in that: Applied to a micro-spectrometer system, the micro-spectrometer system includes a plurality of micro-spectrometers and a standard spectrometer; the method includes: Constructing a calibration sample set, wherein the calibration sample set includes a plurality of calibration samples and a target property index reference value corresponding to each calibration sample; Acquire a first spectrum corresponding to each calibration sample acquired by the standard spectrometer; Based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample, a target property detection model is constructed by a chemometric method; Selecting some calibration samples from the calibration sample set to construct a spectrum transfer sample set; For each micro-spectrometer, a standardization process is performed to achieve the sharing of the calibration models of the multiple micro-spectrometers; wherein the standardization process includes: obtaining a second spectrum corresponding to each calibration sample in the spectrum transfer sample set collected by the micro-spectrometer; based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set, according to a spectrum transfer algorithm, determining the spectrum transfer path corresponding to the micro-spectrometer.
2. The method for sharing calibration models of multiple micro-spectrometers according to claim 1, characterized in that: After performing standardization processing on each micro-spectrometer to achieve sharing of calibration models of the multiple micro-spectrometers, the method further includes: Obtaining an initial spectrum of the sample to be tested acquired by the micro-spectrometer; According to the spectrum transfer path corresponding to the micro-spectrometer, the initial spectrum of the sample to be tested is spectrum-transferred to obtain a standardized spectrum of the sample to be tested; wherein the spectrum transfer path corresponding to the micro-spectrometer is a transfer path required for transferring the spectrum corresponding to the micro-spectrometer to the spectrum corresponding to the standard spectrometer; The standardized spectrum of the sample to be tested is input into the target property calibration model to obtain the target property index value corresponding to the sample to be tested output by the target property calibration model; wherein the target property calibration model is constructed based on the standard spectrometer.
3. The method for sharing calibration models of multiple micro-spectrometers according to claim 1, characterized in that: After constructing a target property detection model based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample by a chemometric method, the method includes: Constructing a first verification sample set, the first verification sample set comprising a plurality of first verification samples, a spectrum of each first verification sample corresponding to the standard spectrometer, and a target property index reference value corresponding to each first verification sample; The target property detection model is verified based on the first verification sample set.
4. The method for sharing calibration models of multiple micro-spectrometers according to claim 1, characterized in that: After determining the spectrum transfer path corresponding to the micro-spectrometer based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set according to the spectrum transfer algorithm, the method further includes: Constructing a second verification sample set, wherein the second verification sample set includes a plurality of second verification samples, each second verification sample corresponds to a spectrum of the standard spectrometer, and each second verification sample corresponds to a spectrum of the miniature spectrometer; The spectral transfer path corresponding to the micro-spectrometer is verified based on the second verification sample set.
5. The method for sharing calibration models of multiple micro-spectrometers according to claim 1, characterized in that: The determining, based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set and according to a spectrum transfer algorithm, a spectrum transfer path corresponding to the micro-spectrometer comprises: Based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set, a transfer function between the spectrum collected by the micro-spectrometer and the spectrum collected by the standard spectrometer is calculated according to a spectrum transfer algorithm; Based on a transfer function between a spectrum acquired by the micro-spectrometer and a spectrum acquired by a standard spectrometer, a spectrum transfer path corresponding to the micro-spectrometer is determined.
6. The method for sharing calibration models of multiple micro-spectrometers according to claim 1, characterized in that: The method further comprises: Acquire a plurality of samples to be tested, wherein the number of the plurality of samples to be tested is consistent with the number of the plurality of micro-spectrometers; The plurality of samples to be tested are collected by the plurality of micro-spectrometers to obtain an initial spectrum corresponding to each sample to be tested; Based on the spectrum transfer path corresponding to each micro-spectrometer, the initial spectrum corresponding to each sample to be tested is spectrum transferred to obtain a standardized spectrum of each sample to be tested; The standardized spectrum of each sample to be tested is input into the target property calibration model to obtain the target property index value of each sample to be tested output by the target property calibration model.
7. The method for sharing calibration models of multiple micro-spectrometers according to claim 6, characterized in that: The multiple samples to be tested are obtained by preparing different positions of a large sample or a complex sample.
8. The method for sharing calibration models of multiple micro-spectrometers according to any one of claims 1 to 7, characterized in that: The multiple micro-spectrometers can be arranged at different processing positions of an industrial assembly line.
9. The method for sharing calibration models of multiple micro-spectrometers according to any one of claims 1 to 7, characterized in that: The multiple micro-spectrometers are used to detect samples to be tested at different stages of the synthesis process or the processing process.
10. A micro-spectrometer system, characterized in that: The micro-spectrometer system includes a plurality of micro-spectrometers, a standard spectrometer, and a plurality of micro-spectrometer calibration model sharing devices, wherein the plurality of micro-spectrometer calibration model sharing devices include: A first construction module is used to construct a calibration sample set, wherein the calibration sample set includes a plurality of calibration samples and a target property index reference value corresponding to each calibration sample; A first acquisition module, used for acquiring a first spectrum corresponding to each calibration sample acquired by the standard spectrometer; A second construction module is used to construct the target property detection model by a chemometric method based on the first spectrum corresponding to each calibration sample and the target property index reference value corresponding to each calibration sample; A selection module, used for selecting some calibration samples from the calibration sample set to construct a spectrum transfer sample set; A processing module is used to perform standardization processing on each micro-spectrometer to achieve sharing of calibration models of the multiple micro-spectrometers; wherein the standardization processing includes: obtaining a second spectrum corresponding to each calibration sample in the spectrum transfer sample set collected by the micro-spectrometer; based on the first spectrum and the second spectrum corresponding to each calibration sample in the spectrum transfer sample set, determining the spectrum transfer path corresponding to the micro-spectrometer according to a spectrum transfer algorithm.
Citation Information
Patent Citations
Rapid detection method of diesel engine nitrogen oxide reducing agent
CN110927099A
Spectrum transmission method and system, medium and electronic equipment
CN113592020A
Evaluation model construction method for near-infrared spectrometer measurement spectrum prediction
CN115326746A
Method and system for migration and standardization between heterologous laser-induced breakdown spectroscopy
CN115436343A
