Consistency discrimination method for multispectral fusion and related equipment

Abstract

The invention discloses a multispectral fusion consistency discrimination method and related equipment, and the method comprises the steps: synchronously collecting a plurality of spectral bands, and carrying out the preprocessing of the plurality of spectral bands; the spectral bands comprise an ultraviolet band, a visible light band and a near-infrared band; performing multi-type similarity index calculation and similarity weight calculation on each preprocessed spectral band, and calculating the similarity of each spectral band according to each type of similarity index and the corresponding similarity weight; and screening the similarities of the spectral bands according to a similarity preset threshold, and determining a spectral consistency result according to the spectral bands meeting the similarity requirements and a preset accuracy rate. According to the embodiment of the invention, the accuracy of spectrum consistency discrimination can be improved. The method can be widely applied to the technical field of spectral analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of spectral analysis, in particular to a multi-spectrum fusion consistency discrimination method and related equipment. BACKGROUND

[0002] Spectral consistency discrimination refers to a technology of judging whether two samples belong to the same category or have the same properties by comparing the similarity between the spectral characteristics of the samples and reference spectra. In related technologies, the accuracy of spectral consistency discrimination needs to be improved.

[0003] To sum up, the technical problems existing in related technologies need to be improved. SUMMARY

[0004] The main purpose of the embodiments of the present application is to provide a multi-spectrum fusion consistency discrimination method and related equipment, aiming to improve the accuracy of spectral consistency discrimination.

[0005] To achieve the above-mentioned purpose, one aspect of the embodiments of the present application provides a multi-spectrum fusion consistency discrimination method, which comprises: Synchronously collecting a plurality of spectral bands, and respectively pre-processing the plurality of spectral bands; the spectral bands include ultraviolet bands, visible light bands, and near-infrared bands; Respectively calculating a plurality of types of similarity indexes and similarity weights for each of the pre-processed spectral bands, and calculating the similarity of each spectral band according to the various types of similarity indexes and the corresponding similarity weights; According to the similarity pre-set threshold, screening the similarity of each spectral band, and determining the spectral consistency result according to the spectral bands meeting the similarity requirement and the pre-set accuracy rate.

[0006] In some embodiments, the pre-processing of the plurality of spectral bands comprises: Smooth processing is performed on the ultraviolet bands to remove Rayleigh scattering interference; Baseline calibration and vector normalization processing are performed on the visible light bands to compensate for color differences; Standard normal transformation and multivariate scattering correction are performed on the near-infrared bands to remove the influence of particles.

[0007] In some embodiments, the pre-processing of the plurality of spectral bands comprises: The various types of similarity indexes of each spectral band are calculated, and various types of similarity indexes of each spectral band are obtained; the similarity indexes include any multiple of Euclidean distance, dynamic time warping, phase correlation, Mahalanobis distance, or cosine similarity; The various types of similarity indexes of each spectral band are respectively subjected to nonlinear normalization processing to determine the similarity weights of the various types of similarity indexes. The various types of similarity indexes and the corresponding similarity weights are weighted and summed to calculate the similarity of each spectral band.

[0008] In some embodiments, the similarity of each spectral band is screened according to a similarity preset threshold, and a spectral consistency result is determined according to the spectral bands meeting the similarity requirement and a preset accuracy, including: The spectral bands with a similarity greater than the similarity preset threshold are screened from the various spectral bands to obtain the spectral bands meeting the similarity requirement; The spectral bands with an accuracy greater than the preset accuracy are screened from the spectral bands meeting the similarity requirement to obtain the spectral bands meeting the accuracy requirement; The spectral consistency result is determined by weighted sum of the spectral bands meeting the accuracy requirement and the corresponding accuracy.

[0009] In some embodiments, the method further includes: The multiple spectral bands in the overlapping region are fused according to a preset signal-to-noise ratio threshold range to determine the spectrum of the overlapping region.

[0010] In some embodiments, the preset signal-to-noise ratio threshold range includes a high signal-to-noise ratio threshold and a low signal-to-noise ratio threshold, and the multiple spectral bands in the overlapping region are fused according to the preset signal-to-noise ratio threshold range to determine the spectrum of the overlapping region, including: If the correlation coefficient of the two spectral bands in the overlapping region is less than a preset coefficient, the two spectral bands are reacquired; If the correlation coefficient of the two spectral bands in the overlapping region is greater than or equal to the preset coefficient, and the signal-to-noise ratios of the two spectral bands are both greater than or equal to the high signal-to-noise ratio threshold, the two spectral bands are averaged and superimposed; If the correlation coefficient of the two spectral bands in the overlapping region is greater than or equal to the preset coefficient, and the signal-to-noise ratios of the two spectral bands are both less than the low signal-to-noise ratio threshold, the acquisition equipment is rechecked and calibrated; If the correlation coefficient of the two spectral bands in the overlapping region is greater than or equal to the preset coefficient, the signal-to-noise ratio of one spectral band is less than the low signal-to-noise ratio threshold, and the signal-to-noise ratio of the other spectral band is greater than or equal to the low signal-to-noise ratio threshold, the spectral band with the signal-to-noise ratio greater than the low signal-to-noise ratio threshold is taken as the spectrum of the overlapping region; If the correlation coefficient of the two light spectrums in the overlapping region is greater than or equal to a preset coefficient, and the signal-to-noise ratios of the two light spectrums are within the preset signal-to-noise ratio threshold range, the spectrum of the overlapping region is determined according to linear superposition of the signal-to-noise ratios of the two light spectrums.

[0011] To achieve the above object, another aspect of the embodiment of the present application proposes a spectrum fusion consistency discrimination device, which comprises: The acquisition module is configured to synchronously acquire a plurality of spectral bands, and pre-process the plurality of spectral bands respectively; the spectral bands include ultraviolet bands, visible light bands and near-infrared bands; The similarity calculation module is configured to calculate a plurality of types of similarity indexes and similarity weights for each of the pre-processed spectral bands, and calculate the similarity of each spectral band according to the types of similarity indexes and the corresponding similarity weights; The discrimination module is configured to filter the similarity of each spectral band according to a preset similarity threshold, and determine the spectrum consistency result according to the spectral bands meeting the similarity requirement and a preset accuracy rate.

[0012] To achieve the above object, another aspect of the embodiment of the present application proposes an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0013] To achieve the above object, another aspect of the embodiment of the present application proposes a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above method.

[0014] To achieve the above object, another aspect of the embodiment of the present application proposes a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the above method.

[0015] The embodiments of the present application at least have the following beneficial effects: The present application provides a multi-spectrum fusion consistency discrimination method, device, electronic equipment and storage medium, which feeds back different characteristics of different test samples by adopting a multi-spectrum fusion scheme, can more fully reflect the component and content characteristics of the sample, adopts a wave band preprocessing scheme, can specifically improve the signal-to-noise ratio, retain the feature integrity under different wave bands, facilitate dynamic optimization of the calculation efficiency, and improve the anti-interference ability; adopts a composite spectrum similarity fusion algorithm, can comprehensively extract the traditional similarity algorithm, retains the advantages of extracting different features in different calculation modes, and has clearer physical meaning after weighted fusion through the similarity weight, and is better for fusion and multi-layer judgment; adopts a multi-level threshold value judgment, first performs threshold suppression on the composite spectrum similarity results of different wave bands, reduces the interference of data fluctuation of low correlation wave bands on the results, and then performs weighted fusion through the accuracy, can better comprehensively the detection results of different wave bands, so as to improve the accuracy of the spectrum consistency discrimination. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of a multi-spectrum fusion consistency discrimination method provided by the embodiments of the present application; Figure 2 is a flowchart of a multi-spectrum preprocessing method provided by the embodiments of the present application; Figure 3 is a flowchart of a multi-spectrum similarity calculation method provided by the embodiments of the present application; Figure 4 is a flowchart of a multi-spectrum fusion consistency discrimination method provided by the embodiments of the present application; Figure 5 is a flowchart of another multi-spectrum fusion consistency discrimination method provided by the embodiments of the present application; Figure 6 is a structural schematic diagram of a multi-spectrum fusion consistency discrimination device provided by the embodiments of the present application; Figure 7 is a hardware structural schematic diagram of an electronic equipment provided by the embodiments of the present application. DETAILED DESCRIPTION

[0017] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not intended to limit the present application. When the following description refers to the accompanying drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with embodiments of the present application. They are merely examples of apparatuses and methods consistent with some aspects of the embodiments of the present application as detailed in the appended claims.

[0018] It can be understood that the terms "first", "second" and the like used in the present application can be used herein to describe various concepts, but unless specifically stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiments of the present application, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "when" or "upon" or "in response to determining".

[0019] The terms "at least one", "multiple", "each", "any" and the like used in the present application include one, two or more than two, multiple includes two or more than two, each refers to each of the corresponding plurality, and any refers to any one of the plurality.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0021] Before the embodiments of the present application are described in detail, first, some nouns and terms involved in the embodiments of the present application are described, and the nouns and terms involved in the embodiments of the present application are applicable to the following explanations.

[0022] The core of the spectral consistency discrimination is to convert the complex optical information into quantifiable difference indicators. In the related art, the spectral consistency discrimination technology mainly includes three parts of spectral data feature, similarity measurement and consistency judgment. The spectral data feature includes spectral feature extraction, dimension reduction processing, etc.; the similarity measurement mainly includes Euclidean distance, cosine similarity, dynamic time warping, Mahalanobis distance, spectral angle, etc.; the consistency judgment is usually through the threshold judgment rule established by the pre-acquired data set. There are the following problems: the single-band spectrum (such as only ultraviolet or near-infrared) information is insufficient, and it is difficult to identify complex samples; a single spectral similarity calculation method (such as correlation coefficient, Euclidean distance, spectral angle) has certain limitations, and the distance dimensions of different methods are different, which cannot be directly superimposed and calculated; a single spectral preprocessing algorithm has certain limitations, and depends on manual selection, and it is difficult to establish a unified adaptive selection mechanism; the conventional method uses a fixed threshold, which cannot adapt to the feature differences of different spectral intervals; the weight distribution lacks a dynamic adjustment mechanism when multiple spectral data are fused.

[0023] Therefore, in the embodiments of the present application, a multi-spectral fusion consistency discrimination method and related equipment are provided. The scheme can more fully reflect the component and content characteristics of the sample by feeding back the different characteristics of different test samples through the multi-spectral fusion scheme; the signal-to-noise ratio can be improved, the feature integrity in different bands can be preserved, the calculation efficiency can be dynamically optimized, and the anti-interference ability can be improved through the band-by-band preprocessing scheme; the traditional similarity algorithms can be comprehensively extracted through the composite spectral similarity fusion algorithm, the advantages of extracting different features through different calculation methods are retained, and the composite spectral similarity fusion algorithm has clearer physical meaning and better fusion and multi-level judgment after weighted fusion through the similarity weight; the multi-level threshold judgment is used to suppress the threshold of the composite spectral similarity result of different bands, reduce the interference of data fluctuation of low correlation bands on the result, and then the accuracy is weighted and fused, so that the detection results of different bands can be better integrated, and the accuracy of spectral consistency discrimination is improved.

[0024] The consistency discrimination method for multispectral fusion provided in the embodiments of the present application relates to the field of information technology. The consistency discrimination method for multispectral fusion provided in the embodiments of the present application can be applied to a terminal, can also be applied to a server, and can further be software running in the terminal or the server. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, a vehicle-mounted terminal, and the like, but is not limited thereto; the server end can be configured as a stand-alone physical server, can also be configured as a server cluster or a distributed system formed by multiple physical servers, can further be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and big data and artificial intelligence platform, and the server can also be a node server in a blockchain network; and the software can be an application implementing the consistency discrimination method for multispectral fusion, and the like, but is not limited to the above forms.

[0025] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as a program module. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0026] It should be noted that in each specific embodiment of the present application, when relevant processing needs to be performed according to user information, user behavior data, user historical data, and user location information, and the like related to the identity or characteristics of the user, the user's permission or consent will be obtained first, and the collection, use, and processing of the data will comply with relevant laws, regulations, and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the separate permission or separate consent of the user will be obtained through a pop-up window or a jump to a confirmation page, and after obtaining the separate permission or separate consent of the user, the necessary user-related data for enabling the embodiments of the present application to normally operate will be obtained.

[0027] Figure 1is an optional flowchart of a consistency discrimination method of multi-spectrum fusion provided by the embodiment of the present application, Figure 1 The method in the method can include but is not limited to steps S101 to S103.

[0028] In step S101, a plurality of spectral bands are synchronously collected, and the plurality of spectral bands are respectively preprocessed; the spectral bands include ultraviolet bands, visible light bands and near-infrared bands.

[0029] Hardware synchronous collection of ultraviolet-visible band (200-1100nm) and near-infrared (1000-2500nm) dual-band spectra is performed on the sample to be measured based on a dual-sensor cooperative working system of silicon-based and InGaAs detectors. Among them, a deuterium tungsten combined light source is used for the ultraviolet-visible band, a grating spectrometer system (1200 lines / mm) and a PMT detector. A halogen light source (900-2500nm) is used for the near-infrared band, a CT monochromator and a TE refrigeration type InGaAs detector. FPGA chip realizes time synchronization of dual-spectrum collection.

[0030] It should be noted that the preprocessing methods of different spectral bands are different.

[0031] In step S102, a plurality of types of similarity indexes and similarity weight calculations are performed on each spectral band after preprocessing, and the similarity of each spectral band is calculated according to the various types of similarity indexes and the corresponding similarity weight.

[0032] A plurality of types of similarity index calculations are performed on each spectral band respectively, and the corresponding similarity weight is calculated according to the plurality of types of similarity indexes. Finally, the similarity of each spectral band is calculated by weighted summation according to the various types of similarity indexes and the corresponding similarity weight.

[0033] In step S103, the similarity of each spectral band is screened according to a preset similarity threshold, and the spectral consistency result is determined according to the spectral band meeting the similarity requirement and the preset accuracy.

[0034] It should be noted that the preset similarity threshold and the preset accuracy are determined according to actual application, and the embodiment does not make specific limitation. The spectrum with a similarity greater than the preset similarity threshold is selected from the spectral bands, and the screened spectrum meets the accuracy greater than the preset accuracy. The spectral consistency result is determined by weighted summation according to the similarity and the accuracy of the spectrum after the final screening processing. The spectral consistency result includes but is not limited to quantitative indexes or qualitative indexes, the quantitative indexes represent specific similarity probability, and the qualitative indexes represent whether it is consistent.

[0035] In some embodiments, reference can be made to Figure 2The multiple spectral bands are respectively preprocessed, including steps S201 to S203.

[0036] In step S201, the ultraviolet band is smoothed to remove Rayleigh scattering interference. In step S202, the visible light band is subjected to baseline correction and vector normalization to compensate for color differences. In step S203, the near-infrared band is subjected to standard normal transformation and multivariate scattering correction to remove particle effects.

[0037] For the ultraviolet spectral band region (200-400nm), the Rayleigh scattering intensity in the ultraviolet region is large, and the overall signal of ultraviolet light is weak. SG (polynomial) smoothing and first derivative are used as the basic preprocessing method to mainly eliminate the interference of Rayleigh scattering. For the visible light spectral band region (400-700nm), the color difference of the object will cause the reflectivity of the visible light band to change. This band usually has a wide absorption band, which can easily mask the subtle feature differences. Baseline correction and vector normalization are used to compensate for color differences. For the near-infrared spectral band region (700-2500nm), the particle scattering in this band is serious, and the overlapping absorption peak is difficult to analyze. Standard normal transformation and multivariate scattering correction are used to eliminate the particle effects.

[0038] In some embodiments, reference is made to Figure 3 The multiple types of similarity indexes and similarity weightings are calculated for each preprocessed spectral band, and the similarity of each spectral band is calculated according to the various types of similarity indexes and corresponding similarity weightings, including steps S301 to S303.

[0039] In step S301, multiple types of similarity indexes are calculated for each preprocessed spectral band to obtain multiple types of similarity indexes for each spectral band. The similarity indexes include any multiple of Euclidean distance, dynamic time warping, phase correlation, Mahalanobis distance, or cosine similarity. In step S302, the multiple types of similarity indexes for each spectral band are respectively subjected to nonlinear normalization processing to determine the similarity weightings of the multiple types of similarity indexes. In step S303, the similarity of each spectral band is calculated by weighted summation according to the various types of similarity indexes and corresponding similarity weightings.

[0040] For the test sample similarity calculation, the embodiment adopts a fusion algorithm covering different wavebands. After pre-processing the different waveband spectrums respectively, various similarity calculations are performed on each waveband, and various spectral similarity indexes are calculated in the ultraviolet, visible, and near-infrared three wavebands, including Euclidean distance, dynamic time warping, phase correlation, Mahalanobis distance, or cosine similarity; then, various similarity indexes of each waveband are subjected to nonlinear normalization processing, including exponential transformation, power function transformation, sigmoid transformation, and the like, wherein the parameters are dynamically set according to different detection targets, for example, for a scene more sensitive to similarity, exponential transformation should be selected and the decay rate should be increased to achieve the purpose of rapid decay, and for a scene less sensitive to similarity, power function transformation should be selected and the parameter compressing high values should be set. After completing the nonlinear normalization, the different similarity indexes can be fused by weighting, and the weight values are fitted by using a logistic regression with non-negative coefficients, and the fitting target is the determination accuracy of the sample set; after the fitting is completed, the composite spectral similarity under each waveband can be obtained.

[0041] In a specific embodiment, taking the consistency detection of a fragrance solution as an example, five brands of fragrances are prepared, and the full spectrum information of each brand is recorded as a standard spectrum database; the fragrance samples of each brand are diluted by 2% to 20% at an interval of 2%, and the diluted samples are used as “negative samples” of each brand of fragrance, and then 50 times of sampling without replacement is performed in each original fragrance solution to obtain “positive samples”, thereby obtaining a total of 50 positive samples and 50 negative samples.

[0042] The similarity fusion calculation is performed on each sample and the standard sample, and the calculation is performed in the ultraviolet, visible, and near-infrared three wavebands. Taking the ultraviolet waveband as an example, four indexes of Euclidean distance, Mahalanobis distance, cosine similarity, and dynamic time warping are calculated, and are denoted as D 紫-欧 , D 紫-马 , D 紫-余弦 , and D 紫-动态 .

[0043] The nonlinear transformation is performed on each original distance index, and taking the sigmoid function as an example, the normalization calculation method is as follows:

[0044] wherein p is a self-defined distance offset, and p has differences under various distances, for example, the Euclidean distance p can be set to 0.9, and the Mahalanobis distance p can be set to 0.3; D represents the similarity, and Dnorm represents the normalized value after the normalization processing.

[0045] The normalized distance is subjected to a logistic regression of non-negative coefficients, and is converted into a binary classification problem by a sigmoid function, and the accuracy of sample set classification is taken as an evaluation index, and cross entropy is taken as a loss function, and the optimal coefficient is found. The coefficient is used to combine various distance indicators, and the indicator obtained by the sigmoid function is used as the composite spectrum similarity in the ultraviolet band of the sample.

[0046] In some embodiments, referring to Figure 4 , the similarity of each spectrum band is screened according to a preset similarity threshold, and the spectrum consistency result is determined according to the spectrum bands meeting the similarity requirement and a preset accuracy rate, including steps S401 to S403.

[0047] Step S401, screening the spectrum bands with a similarity greater than a preset similarity threshold from the spectrum bands, to obtain the spectrum bands meeting the similarity requirement; Step S402, screening the spectrum bands with an accuracy rate greater than a preset accuracy rate from the spectrum bands meeting the similarity requirement, to obtain the spectrum bands meeting the accuracy requirement; Step S403, performing weighted summation on the spectrum bands meeting the accuracy requirement and the corresponding accuracy rate, to determine the spectrum consistency result.

[0048] After obtaining the composite spectrum similarity in the ultraviolet-visible-near infrared three bands, the similarity of the three bands is first subjected to threshold suppression, and the calculation method is as follows:

[0049] Wherein, S band (t) represents the similarity result, t represents the similarity of a certain band, and Thresh represents the preset similarity threshold. According to the threshold, for the case that the similarity of each band is lower than a certain threshold, it needs to be directly set to 0, and the purpose is to suppress the noise interference of the bands with poor correlation effect on the result; for the case that it is higher than the threshold, it is retained. The band threshold should be determined according to the scene, for example, for the scene that needs to detect the color consistency, the suppression threshold of ultraviolet and near infrared band can be set higher to reduce its influence on the result determination.

[0050] Finally, the similarity of the remaining bands is subjected to weight fusion, the weight of which is referred to the accuracy rate of the similarity calculated by each band, if the accuracy rate is lower than the preset accuracy rate, it is directly discarded, and the remaining is subjected to weighted fusion according to the accuracy; according to the finally calculated similarity, the threshold set by the last layer is used for determination.

[0051] In a specific embodiment, the consistency of the essence solution is detected. Through the foregoing steps, the composite spectrum similarities of the ultraviolet, visible, and near-infrared bands have been obtained respectively, and the prediction accuracy of each band similarity fitted by the training set is 65%, 90%, and 80%. Assuming that the three-band similarities of a sample and a standard library with a certain number are 0.45, 0.6, and 0.7 respectively, and the inhibition thresholds of the three bands are 0.6, 0.3, and 0.3 respectively. According to the set rule, if the accuracy is less than 60%, it is directly discarded, the similarity of the ultraviolet band will be directly inhibited to 0, and the visible band and the near-infrared band will be retained, and the weighted fusion is performed according to the band accuracy during training (that is, 9:8), so the comprehensive similarity calculated after weighting is:

[0052] Finally, if the quantitative index of the similarity needs to be output, 64.7% can be directly returned; if the qualitative index of the similarity needs to be output (that is, consistent or inconsistent), the threshold can be directly determined according to the set threshold. In this example, if the determination threshold is 60%, the result is "consistent", and if the determination threshold is 68%, the result is "inconsistent".

[0053] In some embodiments, the method further comprises step S501.

[0054] Step S501, fusing the multiple spectral bands in the overlapping region according to a preset signal-to-noise ratio threshold range to determine the spectrum of the overlapping region.

[0055] The multiple spectral bands can have multiple overlapping regions, and adjacent two spectral bands can have an overlapping region. Each overlapping region is fused respectively. The overlapping region is fused according to the signal-to-noise ratio of the spectral band.

[0056] In some embodiments, the preset signal-to-noise ratio threshold range includes a high signal-to-noise ratio threshold and a low signal-to-noise ratio threshold, and the fusing the multiple spectral bands in the overlapping region according to the preset signal-to-noise ratio threshold range to determine the spectrum of the overlapping region comprises: Step S601, if the correlation coefficient of the two spectral bands in the overlapping region is less than a preset coefficient, the two spectral bands are reacquired; Step S701, if the correlation coefficient of the two spectral bands in the overlapping region is greater than or equal to the preset coefficient, and the signal-to-noise ratios of the two spectral bands are both greater than or equal to the high signal-to-noise ratio threshold, the two spectral bands are averaged and superimposed; Step S702, if the correlation coefficient of the two spectral bands in the overlapping region is greater than or equal to the preset coefficient, and the signal-to-noise ratios of the two spectral bands are both less than the low signal-to-noise ratio threshold, the acquisition device is rechecked and calibrated; Step S703, if the correlation coefficient of the two light spectrums in the overlapping region is greater than or equal to a preset coefficient, the signal-to-noise ratio of one light spectrum is less than a signal-to-noise ratio low threshold, and the signal-to-noise ratio of one light spectrum is greater than or equal to the signal-to-noise ratio low threshold, the light spectrum with the signal-to-noise ratio greater than the signal-to-noise ratio low threshold is taken as the light spectrum of the overlapping region; Step S704, if the correlation coefficient of the two light spectrums in the overlapping region is greater than or equal to a preset coefficient, and the signal-to-noise ratios of the two light spectrums are within a preset signal-to-noise ratio threshold range, the light spectrum of the overlapping region is determined according to the signal-to-noise ratios of the two light spectrums.

[0057] For the overlapping region, a dynamic adaptive weighted fusion based on the signal-to-noise ratio is adopted. Calibration and signal-to-noise ratio test are performed before each sample collection, and a dynamic weight is set according to the comprehensive signal-to-noise ratio of different light paths in the overlapping region. In a specific embodiment, the configuration rule is to set a high threshold Tmax and a low threshold Tmin of the signal-to-noise ratio. The correlation coefficient R of the two light spectrums is calculated, and when R<0.95, the two light spectrums are re-collected. However, when the correlation coefficient R of the two light spectrums is greater than 0.95, the following processing is performed. When the signal-to-noise ratios of the two light spectrums are both greater than Tmax, the weights are equal, and the overlapping region can be directly averaged; when the signal-to-noise ratios of the two light spectrums are both less than Tmin, it is determined that an abnormality occurs, and the instrument needs to be rechecked and calibrated; when the signal-to-noise ratio of one light spectrum is less than Tmin, the signal of the other light spectrum is selected as the light spectrum of the overlapping region; and when the signal-to-noise ratios of the two light spectrums are both between Tmin and Tmax, a linear weight is set according to the signal-to-noise ratio (the higher the signal-to-noise ratio, the higher the weight).

[0058] The present application performs fusion of the light spectrum of the overlapping region based on the signal-to-noise ratio, which can distinguish the effectiveness of the signal and better fuse the detection results of different instruments.

[0059] Next, specific application examples of the consistency discrimination of the spectrum fusion are combined for description. Figure 5 The scheme of the embodiment of the present application is described in detail as follows: S1, according to the sample to be detected, standard samples, positive samples and negative samples are prepared respectively, the positive sample refers to a sample with high similarity to the standard sample, and the negative sample refers to a sample with low similarity to the standard sample; the positive sample and the negative sample should be not less than 30; S2, the full spectrum of each sample is collected respectively; S3, each waveband of the full spectrum is preprocessed separately; S4, a plurality of classical spectral similarity indexes are calculated in the ultraviolet, visible and near-infrared three wavebands respectively, including Euclidean distance, dynamic time warping, phase correlation, Mahalanobis distance and cosine similarity. S5, respectively, each waveband of various similarity indicators are nonlinearly normalized, including exponential transformation, logarithmic transformation, sigmoid transformation, etc., wherein the parameters should be set according to the different dynamic of the detection target; S6, the normalized multi-index of each waveband is fused respectively, the weight is fitted by non-negative unary linear regression, and the fitting target is the determination accuracy of the sample set; S7, the similarity of the three wavebands is threshold suppressed, for each waveband, if the similarity is lower than a certain threshold, it is directly set to 0, and if it is higher than the threshold, it is retained; S8, the similarity of the remaining wavebands is fused by weight, the weight is referred to the accuracy of the similarity calculated by each waveband, if the accuracy is lower than 60%, it is directly discarded, and the remaining is fused by weight according to the accuracy; S9, according to the final calculation of the comprehensive similarity, if the quantitative index needs to be output, the comprehensive similarity can be directly returned; if the qualitative index (i.e. consistent or inconsistent) needs to be output, the threshold set by the last layer is used for determination, and the output result is output.

[0060] Please refer to Figure 6 The embodiment of the application also provides a spectrum fusion consistency discrimination device, which can realize the above method, and the device comprises: The acquisition module is used for synchronously acquiring a plurality of spectral wavebands, and pre-processing the plurality of spectral wavebands; the spectral wavebands include ultraviolet wavebands, visible light wavebands and near-infrared wavebands; The similarity calculation module is used for calculating a plurality of types of similarity indicators and similarity weights for each spectral waveband after pre-processing, and calculating the similarity of each spectral waveband according to the plurality of types of similarity indicators and the corresponding similarity weights; The discrimination module is used for screening the similarity of each spectral waveband according to a preset similarity threshold, and determining the spectral consistency result according to the spectral wavebands meeting the similarity requirement and a preset accuracy.

[0061] It can be understood that the contents in the above method embodiments are applicable to the device embodiments, the device embodiments specifically realize the functions of the above method embodiments, and the beneficial effects achieved by the device embodiments are the same as the beneficial effects achieved by the above method embodiments.

[0062] The embodiment of the application also provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor realizes the above method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0063] It can be understood that the contents in the above method embodiments are all applicable to the present device embodiments, the present device embodiments specifically implement the functions same as those of the above method embodiments, and achieve the same beneficial effects as those of the above method embodiments.

[0064] Please refer to Figure 7 , Figure 7 The hardware structure of the electronic device of another embodiment is illustrated, and the electronic device comprises: The processor 701 can be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, and is used to execute related programs to implement the technical solutions provided by the embodiments of the present application. The memory 702 can be implemented in the form of a ROM (Read Only Memory), a static storage device, a dynamic storage device, or a RAM (Random Access Memory). The memory 702 can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are stored in the memory 702 and are called and executed by the processor 701 to implement the above-mentioned methods of the embodiments of the present application. The input / output interface 703 is used to realize information input and output. The communication interface 704 is used to realize the communication interaction between the present device and other devices. The communication can be realized by a wired manner (for example, a USB, a network cable, etc.) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, etc.). The bus 705 is used to transmit information between the components (for example, the processor 701, the memory 702, the input / output interface 703, and the communication interface 704) of the device. The processor 701, the memory 702, the input / output interface 703, and the communication interface 704 are connected to each other through the bus 705 for internal communication in the device.

[0065] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the above-mentioned method.

[0066] It can be understood that the contents in the above method embodiments are all applicable to the present storage medium embodiments, the present storage medium embodiments specifically implement the same functions as the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0067] The present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the above method.

[0068] It can be understood that the contents in the above method embodiments are all applicable to the present program product embodiments, the present program product embodiments specifically implement the same functions as the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0069] The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory can include a high-speed random access memory and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory that is remotely arranged with respect to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0070] The present application provides a multi-spectral fusion consistency discrimination method and related equipment, which can more fully reflect the composition and content characteristics of the sample by feeding back the different characteristics of different test samples through the multi-spectral fusion scheme; the signal-to-noise ratio can be improved, the feature integrity under different wavebands can be preserved, the calculation efficiency can be dynamically optimized, and the anti-interference ability can be improved through the waveband preprocessing scheme; the traditional similarity algorithm can be comprehensively extracted through the composite spectral similarity fusion algorithm, the advantages of extracting different features through different calculation methods are retained, and after weighted fusion through the similarity weight, the physical meaning is clearer, and the fusion and multi-level judgment are better; the multi-level threshold judgment is adopted to suppress the threshold of the composite spectral similarity result of different wavebands, reduce the interference of data fluctuation of low correlation wavebands on the result, and then the accuracy is weighted and fused, so that the detection results of different wavebands can be better integrated, and the accuracy of spectral consistency discrimination is improved.

[0071] The embodiments described in the present application are used to more clearly illustrate the technical solutions of the present application, and do not constitute a limitation on the technical solutions provided by the present application. Those skilled in the art can know that with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the present application are also applicable to similar technical problems.

[0072] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and can include more or fewer steps than the figures, or combine certain steps, or different steps.

[0073] The apparatus embodiments described above are merely illustrative, and units described as separate components can or can not be physically separate, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0074] Those skilled in the art can understand that all or some of the steps in the above disclosed method, the functions of the modules / units in the system and the device can be implemented as software, firmware, hardware and their appropriate combinations.

[0075] The terms "first", "second", "third", "fourth" and the like used in the description of the present application and the above-described drawings (if any) are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0076] It should be understood that in the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the relationship between the associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that there are three cases: only A, only B, and A and B at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0077] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented by other manners. For example, the apparatus embodiments described above are merely illustrative, for example, the division of the above units is merely a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, apparatuses or units, and can be electrical, mechanical or other forms.

[0078] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they can be located in one place or distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0079] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0080] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art that makes a contribution or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method of each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.

[0081] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, which are not limited to the scope of the embodiments of the present application. Any modifications, equivalent replacements and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the embodiments of the present application.

Claims

1. A multi-spectral fusion consistency discrimination method, characterized in that, The method comprises the following steps: Synchronously collecting a plurality of spectral bands, and respectively pre-processing the plurality of spectral bands; the spectral bands comprise ultraviolet bands, visible light bands, and near-infrared bands; Respectively calculating a plurality of types of similarity indexes and similarity weightings for the pre-processed spectral bands, and calculating the similarity of each spectral band according to the various types of similarity indexes and the corresponding similarity weightings; According to a similarity preset threshold, screening the similarity of each spectral band, and determining a spectral consistency result according to the spectral bands meeting the similarity requirement and a preset accuracy.

2. The method of claim 1, wherein, The pre-processing of the plurality of spectral bands respectively comprises: Performing smoothing processing on the ultraviolet bands to remove Rayleigh scattering interference; Performing baseline calibration and vector normalization processing on the visible light bands to compensate for color differences; Performing standard normal transformation and multivariate scattering correction on the near-infrared bands to remove the influence of particles.

3. The method of claim 1, wherein, The calculation of the plurality of types of similarity indexes and the similarity weightings for the pre-processed spectral bands respectively, and the calculation of the similarity of each spectral band according to the various types of similarity indexes and the corresponding similarity weightings, comprise: Respectively calculating a plurality of types of similarity indexes for the pre-processed spectral bands, to obtain a plurality of types of similarity indexes for each spectral band; the similarity indexes comprise any multiple of Euclidean distance, dynamic time warping, phase correlation, Mahalanobis distance, or cosine similarity; Respectively performing nonlinear normalization processing on the plurality of types of similarity indexes of each spectral band, to determine the similarity weightings of the plurality of types of similarity indexes; According to the weighted sum of the various types of similarity indexes and the corresponding similarity weightings, the similarity of each spectral band is calculated.

4. The method of claim 1, wherein, The screening of the similarity of each spectral band according to a similarity preset threshold, and the determination of a spectral consistency result according to the spectral bands meeting the similarity requirement and a preset accuracy, comprise: From each spectral band, a spectral band with a similarity greater than the similarity preset threshold is screened, to obtain a spectral band meeting the similarity requirement; From the spectral band meeting the similarity requirement, a spectral band with an accuracy greater than the preset accuracy is screened, to obtain a spectral band meeting the accuracy requirement; According to the weighted sum of the spectral band meeting the accuracy requirement and the corresponding accuracy, a spectral consistency result is determined.

5. The method according to any one of claims 1 to 4, characterized in that, The method further comprises: According to a preset signal-to-noise ratio threshold range, fusing the plurality of spectral bands in the overlapping area, to determine the spectrum of the overlapping area.

6. The method of claim 5, wherein, The preset signal-to-noise ratio threshold range comprises a high signal-to-noise ratio threshold and a low signal-to-noise ratio threshold, and the fusing of the plurality of spectral bands in the overlapping area according to the preset signal-to-noise ratio threshold range, to determine the spectrum of the overlapping area, comprises: If the correlation coefficient of the two spectral bands in the overlapping area is less than a preset coefficient, the two spectral bands are recollected; If the correlation coefficient of the two spectral bands in the overlapping area is greater than or equal to the preset coefficient, and the signal-to-noise ratios of the two spectral bands are both greater than or equal to the high signal-to-noise ratio threshold, the two spectral bands are averaged and superimposed. If the correlation coefficient of the two light spectrums in the overlapping area is greater than or equal to a preset coefficient, and the signal-to-noise ratios of the two light spectrums are both less than the low signal-to-noise ratio threshold, the acquisition device is rechecked and calibrated; If the correlation coefficient of the two light spectrums in the overlapping area is greater than or equal to a preset coefficient, the signal-to-noise ratio of one light spectrum is less than the low signal-to-noise ratio threshold, and the signal-to-noise ratio of the other light spectrum is greater than or equal to the low signal-to-noise ratio threshold, the light spectrum with the signal-to-noise ratio greater than the low signal-to-noise ratio threshold is taken as the light spectrum of the overlapping area; If the correlation coefficient of the two light spectrums in the overlapping area is greater than or equal to a preset coefficient, and the signal-to-noise ratios of the two light spectrums are both within the preset signal-to-noise ratio threshold range, the light spectrum of the overlapping area is determined according to linear superposition of the signal-to-noise ratios of the two light spectrums.

7. A spectral fusion consistency discriminator apparatus characterized by comprising: The device comprises: The acquisition module is configured to synchronously acquire a plurality of spectral bands, and pre-process the plurality of spectral bands respectively; the spectral bands include ultraviolet bands, visible light bands and near-infrared bands; The similarity calculation module is configured to calculate a plurality of types of similarity indexes and similarity weights for each of the pre-processed spectral bands, and calculate the similarity of each spectral band according to the types of similarity indexes and the corresponding similarity weights; The discrimination module is configured to filter the similarity of each spectral band according to a preset similarity threshold, and determine a spectral consistency result according to the spectral bands meeting the similarity requirement and a preset accuracy rate.

8. An electronic device, comprising: comprise: at least one processor; at least one memory configured to store at least one program; when the at least one program is executed by the at least one processor, the at least one processor implements the method of any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. The computer program is executed by the processor to implement the method of any one of claims 1 to 6.

10. A computer program product, characterised in that, The computer program is executed by the processor to implement the method of any one of claims 1 to 6. The computer program is executed by the processor to implement the method of any one of claims 1 to 6.

Images