Agricultural product quality authentication and traceability data management system and method based on amino acid fingerprint spectrum

Abstract

The invention relates to the technical field of data processing, in particular to an agricultural product quality authentication and traceability data management system and method based on an amino acid fingerprint spectrum. Comprising the following steps: collecting amino acid fingerprint data of an agricultural product, and preprocessing the amino acid fingerprint data to obtain preprocessed amino acid fingerprint data; performing feature extraction and optimization processing on the preprocessed amino acid fingerprint spectrum data by using a dynamic nonlinear feature collaborative optimization extraction algorithm to obtain optimized amino acid feature data; performing traceability matching on the optimized amino acid feature data and historical authentication data in the block chain, and when matching succeeds, generating an authentication report; and when the matching fails, carrying out block chain storage and authentication on the optimized amino acid feature data. The technical problem that the accuracy, comprehensiveness and adaptability of data processing are insufficient in agricultural product quality authentication and traceability data management is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an agricultural product quality certification and traceability data management system and method based on amino acid fingerprints. Background Art

[0002] Agricultural product quality certification and traceability systems play an important role in modern agricultural production and supply chain management. As consumers' demand for agricultural product quality and safety continues to increase, traditional quality certification methods have gradually exposed some problems, such as the lack of scientific basis in the certification process, insecure data storage methods, and incomplete traceability chains. These problems make agricultural products vulnerable to counterfeiting, label forgery, and supply chain breakpoints in the production, circulation, and consumption of agricultural products.

[0003] Amino acid fingerprint technology is a high-precision detection method based on amino acid composition and content. It can obtain specific chemical characteristics of agricultural product samples through high-resolution analytical equipment (such as mass spectrometer or liquid chromatograph), and is widely used in food quality assessment, traceability analysis and other fields. However, when using amino acid fingerprints for agricultural product quality certification, current technologies have problems such as complex data processing, large noise interference, insufficient feature extraction and optimization, which limits the accuracy and reliability of quality certification. In addition, due to the wide variety of agricultural products, their amino acid fingerprint features are often manifested as high-dimensional complex data. Traditional feature extraction algorithms are difficult to find a balance between global correlation and local differences, resulting in insufficient adaptability of certification data in dynamic traceability. Summary of the invention

[0004] The present invention provides an agricultural product quality certification and traceability data management system and method based on amino acid fingerprints to solve the technical problems of insufficient accuracy, comprehensiveness and adaptability of data processing in agricultural product quality certification and traceability data management.

[0005] The agricultural product quality certification and traceability data management system and method based on amino acid fingerprint of the present invention specifically include the following technical solutions: The method for agricultural product quality certification and traceability data management based on amino acid fingerprint includes the following steps: S1. Collecting amino acid fingerprint data of agricultural products, and preprocessing the amino acid fingerprint data to obtain preprocessed amino acid fingerprint data; S2. Using a dynamic nonlinear feature collaborative optimization extraction algorithm to extract and optimize the preprocessed amino acid fingerprint data to obtain optimized amino acid feature data; S3. The optimized amino acid feature data is traceably matched with the historical authentication data in the blockchain. When the match is successful, an authentication report is generated; when the match fails, the optimized amino acid feature data is stored and authenticated on the blockchain.

[0006] Preferably, the S1 specifically includes: During the preprocessing process, the amino acid fingerprint data is denoised to obtain denoised amino acid fingerprint data; a comprehensive optimization baseline correction algorithm is introduced to perform baseline correction on the denoised amino acid fingerprint data to obtain optimized baseline correction data; the optimized baseline correction data is subjected to data normalization, feature peak extraction and data formatting to obtain preprocessed amino acid fingerprint data.

[0007] Preferably, the S1 specifically includes: In the process of implementing the comprehensive optimization baseline correction algorithm, the denoised amino acid fingerprint data are processed through five stages: signal smoothing, piecewise adaptive fitting, dynamic residual optimization, disturbance noise suppression, extreme value enhancement and peak correction, to obtain the optimized baseline correction data.

[0008] Preferably, the S2 specifically includes: In the process of implementing the dynamic nonlinear feature collaborative optimization extraction algorithm, the preprocessed amino acid fingerprint data is defined as an amino acid fingerprint matrix, and the dynamic offset value of the preprocessed amino acid fingerprint data is calculated by dynamic collaborative offset. The calculation formula of the dynamic offset value is: , in, Indicates The first sample of agricultural products Dynamic shift value of amino acid characteristic peak; It is Dynamic weights of agricultural product samples; Indicates The first sample of agricultural products The data intensity of the characteristic peaks of amino acids; Indicates The first sample of agricultural products The global weighted sum of the amino acid characteristic peaks; is the number of agricultural product samples; It is Dynamic weights of agricultural product samples; It is The first sample of agricultural products The data intensity of the characteristic peaks of amino acids.

[0009] Preferably, the S2 specifically includes: In the implementation process of the dynamic nonlinear feature collaborative optimization extraction algorithm, the dynamic offset value is converted into a collaborative eigenvalue through collaborative enhancement processing, and the collaborative feature matrix composed of the collaborative eigenvalues ​​is subjected to nonlinear mapping processing to obtain the enhanced eigenvalue. The specific formula is: , in, It is The first sample of agricultural products The enhanced characteristic value of the amino acid characteristic peak; After the synergy is enhanced The first sample of agricultural products The synergistic characteristic value of the amino acid characteristic peak; It is the sample code of agricultural products; It is a parameter that controls the amplitude of the enhanced eigenvalue change.

[0010] Preferably, the S2 specifically includes: In the process of realizing the dynamic nonlinear feature collaborative optimization extraction algorithm, the enhanced feature matrix composed of enhanced eigenvalues ​​is dynamically penalized to obtain the eigenvalues ​​adjusted by dynamic penalties, and the eigenvalues ​​adjusted by dynamic penalties constitute the feature matrix adjusted by dynamic penalties.

[0011] Preferably, the S2 specifically includes: In the process of implementing the dynamic nonlinear feature collaborative optimization extraction algorithm, the feature matrix after dynamic penalty adjustment is optimized through sparsity analysis and feature selection to obtain optimized eigenvalues; the optimized feature matrix is ​​composed of the optimized eigenvalues, and the optimized feature matrix is ​​used as the optimized amino acid feature data.

[0012] Preferably, the S3 specifically includes: Introduce a dynamic weighted matching formula to trace the optimized amino acid feature data with the historical authentication data in the blockchain and calculate the matching rate , and based on the matching rate Determine whether the traceability match is successful, the matching rate The specific formula is as follows: , in, It is the optimized amino acid feature data after normalization; It is the historical authentication data in the normalized blockchain; is the dynamic weighting factor, ; is the threshold decision function, is the margin of error.

[0013] Preferably, the S3 specifically includes: When the matching rate Preset Threshold , the match is determined to be successful and the authentication report generation phase is entered; at the same time, the logistics node data is collected, the logistics node data is hashed and bound to the authentication data, a unique identifier of the traceability node is generated, and then a block is generated based on the traceability data bound to each logistics node, which is linked to the previous block through the hash value, and finally the block data is written into the distributed storage network of the blockchain, and then the traceability chain is generated based on the logistics node data stored in the blockchain; when the matching rate is Preset Threshold , it is determined that the match fails and the blockchain storage and authentication phase is entered. Based on the unique identifier of the generated optimized amino acid feature data, a blockchain data storage structure is constructed.

[0014] The agricultural product quality certification and traceability data management system based on amino acid fingerprint includes the following parts: Amino acid fingerprint acquisition module, feature extraction and optimization module, traceability and matching module, blockchain storage and authentication module, supply chain dynamic monitoring module; The amino acid fingerprint collection module collects and preprocesses the amino acid fingerprint data of agricultural products to obtain the preprocessed amino acid fingerprint data, and then sends the preprocessed amino acid fingerprint data to the feature extraction and optimization module; The feature extraction and optimization module extracts features from the preprocessed amino acid fingerprint data, optimizes the extracted feature data, obtains optimized amino acid feature data, and sends the optimized amino acid feature data to the traceability and matching module; The traceability and matching module traces and matches the optimized amino acid feature data with the historical authentication data in the blockchain. When the match is successful, a certification report is generated and sent to the supply chain dynamic monitoring module; when the match fails, the optimized amino acid feature data is sent to the blockchain storage and authentication module; The blockchain storage and authentication module uses the unique identifier of the optimized amino acid feature data to construct a blockchain storage structure and provide historical authentication data for the traceability and matching module; The supply chain dynamic monitoring module collects logistics node data of agricultural products during transportation, hashes the logistics node data with the certification data in the certification report, and stores it in the blockchain storage and authentication module to generate a traceability chain.

[0015] The beneficial effects of the technical solution of the present invention are: 1. Through comprehensive optimization of the baseline correction algorithm, the denoised amino acid fingerprint data is subjected to multi-stage fine processing, including signal smoothing, segmented adaptive fitting, dynamic residual optimization, disturbance noise suppression, extreme value enhancement and peak correction, so as to correct and optimize the baseline data; effectively remove noise interference, strengthen key features, and ensure that the amino acid feature data extracted in the process of agricultural product quality certification is highly accurate and reliable, laying a solid foundation for subsequent traceability and matching.

[0016] 2. Through the dynamic nonlinear feature collaborative optimization extraction algorithm, the agricultural product quality certification and traceability data management system can extract features from the preprocessed amino acid fingerprint data, and improve the global correlation and distribution of features through dynamic collaborative offset and nonlinear mapping; in addition, the feature data is further optimized through dynamic penalty processing and sparsity analysis to effectively filter outliers and strengthen important features; this efficient feature extraction and optimization processing enables the agricultural product quality certification and traceability data management system to adapt to the characteristics of different agricultural products and provide high-dimensional and evenly distributed feature support for traceability and certification.

[0017] 3. Introduce a dynamic weighted matching formula to trace the optimized amino acid feature data with the historical authentication data in the blockchain; emphasize feature differences through dynamic weighting factors, and accurately evaluate the matching results in combination with the threshold judgment function, so as to effectively distinguish between successful and unsuccessful matching of agricultural product samples; after a successful match, the agricultural product quality certification and traceability data management system generates a detailed certification report and builds a complete traceability chain to ensure that the traceability data has high accuracy and full coverage. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a structural diagram of the agricultural product quality certification and traceability data management system based on amino acid fingerprints according to the present invention; Figure 2 The present invention provides a flowchart of the method for agricultural product quality authentication and traceability data management based on amino acid fingerprint. DETAILED DESCRIPTION

[0019] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all 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.

[0020] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0021] The specific scheme of the agricultural product quality authentication and traceability data management system and method based on amino acid fingerprint provided by the present invention is described in detail below with reference to the accompanying drawings.

[0022] See attached Figure 1 , which shows a structural diagram of an agricultural product quality certification and traceability data management system based on amino acid fingerprint provided by an embodiment of the present invention, the system includes the following parts: Amino acid fingerprint acquisition module, feature extraction and optimization module, traceability and matching module, blockchain storage and authentication module, supply chain dynamic monitoring module; The amino acid fingerprint collection module collects the amino acid fingerprint data of agricultural products, pre-processes the collected fingerprint data to obtain the pre-processed amino acid fingerprint data, and then sends the pre-processed amino acid fingerprint data to the feature extraction and optimization module; The feature extraction and optimization module extracts features from the preprocessed amino acid fingerprint data, optimizes the extracted feature data, obtains optimized amino acid feature data, and sends the optimized amino acid feature data to the traceability and matching module; The traceability and matching module performs traceability matching on the optimized amino acid feature data and the historical authentication data in the blockchain. If the match is successful, a certification report is generated and sent to the supply chain dynamic monitoring module; if the match fails, the optimized amino acid feature data is sent to the blockchain storage and authentication module; the authentication report includes basic information such as the production site, authentication time, and testing agency, as well as authentication data such as matching rate, matching feature data summary (amino acid type and content), and fingerprint spectrum; The blockchain storage and authentication module uses the hash algorithm to generate a unique identifier for the optimized amino acid feature data, constructs a blockchain storage structure based on the unique identifier, implements tamper-proof storage and authentication of the data, and provides basic historical authentication data for the traceability and matching modules; The supply chain dynamic monitoring module dynamically collects logistics node data of agricultural products during transportation, and hashes the logistics node data with the certification data in the certification report, and stores it in the blockchain storage and authentication module to generate a traceability chain that includes the complete path of production, certification, transportation, and consumption, providing a traceability chain for subsequent queries.

[0023] See attached Figure 2 , which shows a flow chart of a method for agricultural product quality authentication and traceability data management based on amino acid fingerprint provided by an embodiment of the present invention, the method comprising the following steps: S1. Collecting amino acid fingerprint data of agricultural products, and preprocessing the amino acid fingerprint data to obtain preprocessed amino acid fingerprint data; The amino acid fingerprint data of agricultural products are collected using existing amino acid fingerprint technology, and the amino acid fingerprint data of agricultural products are preprocessed such as denoising, baseline correction, data normalization, characteristic peak extraction, data formatting, etc. to obtain preprocessed amino acid fingerprint data. The denoising, data normalization, characteristic peak extraction, data formatting, etc. in the preprocessing all use existing technologies and will not be elaborated here.

[0024] In the above preprocessing process, the amino acid fingerprint data is denoised to obtain the denoised amino acid fingerprint data; for baseline correction, a comprehensive optimization baseline correction algorithm is introduced to perform baseline correction on the denoised amino acid fingerprint data. The comprehensive optimization baseline correction algorithm is a comprehensive algorithm that combines five stages: signal smoothing, segmented adaptive fitting, dynamic residual optimization, disturbance noise suppression, extreme value enhancement and peak correction. It is designed to maximize the accuracy of baseline correction and eliminate noise through dynamic optimization by smoothing, segmented fitting, dynamic residual optimization, disturbance noise suppression, extreme value enhancement and peak correction of the denoised amino acid fingerprint data. The specific implementation process of the comprehensive optimization baseline correction algorithm is as follows: First, the denoised amino acid fingerprint data was fitted with a cubic polynomial and smoothed with exponential weighting to obtain the smoothed data. : , in, is the position of the current data point; are other data points within the smoothing range; is the smoothing window width; are the coefficients of the fitting polynomial, generated by the fitting process of the local data; is an exponential decay factor used to control the distance from the current data point during data smoothing. The farther the data points are, the greater the impact on the smoothing result. Will make it move away from the current data point The data points have a weaker influence on the smoothing result. For a smaller smoothing window, Take a larger value (such as 1 to 10) to ensure that the local characteristics of the signal can be strongly preserved. For a larger smoothing window, Usually a smaller value (such as 0.01 to 0.5) is used to ensure that the entire smoothing process can affect a wider range of data points; It is an exponential decay function, which is used to control the distance from the current data point during the data smoothing process. The farther away the data points are, the less influence they have on the results. The output of this step It is a smoothed signal that retains the approximate peak shape while removing high-frequency noise.

[0025] Further, the smoothed data is divided into The segmented data is obtained, and then the segmented data is processed using the existing fitting technology to obtain an independent baseline representation for each segment. ,in, Indicates Then, according to the expert experience method, the weight coefficient is introduced to represent the baseline of each segment. Fitting to global baseline data through weight coefficients ; Furthermore, in order to optimize the global baseline data, dynamic residual optimization processing is introduced. By dynamically adjusting the residual, the fitting error is gradually reduced and the adaptability of the global baseline data is improved. The residual is defined as : , in, is the denoised amino acid fingerprint data; It is a factor that adjusts the sensitivity of residuals to outliers. It is determined based on expert experience and has a value range of By performing gradient descent update on the global baseline data, the optimized baseline data is obtained: , in, is the learning rate, which is used to control the step size of the update; It is Baseline data after iterative optimization; It is After the number of optimization iterations preset according to the expert experience method or when the difference between the baselines after two iterations is less than the threshold preset according to the expert experience method, the iteration is stopped and the baseline data after the last iteration is used as the baseline data after dynamic residual optimization. .

[0026] Furthermore, disturbance noise suppression processing is performed to eliminate the influence of high-frequency jitter or abnormal points by performing nonlinear mapping on the baseline data after dynamic residual optimization and the amino acid fingerprint data after denoising. The disturbance noise suppression formula is: , in, is the baseline data after disturbance noise suppression; is the disturbance noise suppression amplitude control factor, which is used to control the amplitude of the disturbance noise suppression operation. It is a positive real number with a value range of ; It is a nonlinear frequency factor used to adjust the relationship between the signal and noise in the disturbance noise suppression process. The nonlinear frequency factor controls the nonlinear characteristics of disturbance noise suppression and mainly affects the frequency response characteristics when removing noise.

[0027] Finally, the baseline data after disturbance noise suppression is subjected to extreme value enhancement and peak correction to obtain the final optimized baseline correction data. : , in, It is the enhancement factor, which is used to control the enhancement degree of the signal peak. It is an adjustment parameter that determines the influence of the difference between the signal and the baseline on the correction result. It is determined by expert experience. It is the nonlinear enhancement order, which strengthens the peak feature by amplifying the difference between the denoised amino acid fingerprint data and the baseline data after disturbance noise suppression.

[0028] S2. Using a dynamic nonlinear feature collaborative optimization extraction algorithm to extract and optimize the pre-processed amino acid fingerprint data to obtain optimized amino acid feature data; After extracting the features of the pre-processed amino acid fingerprint data using a dynamic nonlinear feature collaborative optimization extraction algorithm, the extracted feature data is optimized to obtain optimized amino acid feature data; the specific implementation process is as follows: First, the preprocessed amino acid fingerprint data is defined as the amino acid fingerprint matrix , whose dimensions are ,in is the number of agricultural product samples, is the number of characteristic peaks of amino acids. is an element in the amino acid fingerprint matrix, indicating the The first sample of agricultural products The data intensity of the characteristic peaks of amino acids. The agricultural product sample number matrix is ​​defined as , used to identify the number of each agricultural product sample. In the initial state, the weight vector of all agricultural product samples , let the mean weight be , to balance the impact of all agricultural product samples on the calculation.

[0029] First, the dynamic offset value of the preprocessed amino acid fingerprint data is calculated by dynamic synergistic shift. The core purpose of dynamic synergistic shift is to analyze the offset characteristics of each amino acid characteristic peak in the agricultural product sample while considering the global synergistic characteristics. The calculation formula of the dynamic offset value is: , in, Indicates The first sample of agricultural products Dynamic shift value of amino acid characteristic peak; It is The dynamic weight of each agricultural product sample is used to reflect the importance of each agricultural product sample in the calculation. It is usually adjusted according to the performance of the agricultural product sample or other influencing factors. The agricultural product sample with a higher weight value contributes more to the calculation of the dynamic offset value. At initialization, the weight values ​​of all agricultural product samples are set to be equal, but as the dynamic nonlinear feature collaborative optimization extraction algorithm proceeds, the dynamic weight will be dynamically adjusted according to indicators such as the differences between agricultural product samples using the expert experience method; Indicates The first sample of agricultural products The global weighted sum of the amino acid characteristic peaks; It is Dynamic weights of agricultural product samples; It is The first sample of agricultural products The data intensity of the characteristic peaks of amino acids.

[0030] Furthermore, the dynamic offset value is transformed into a collaborative feature value through collaborative enhancement processing to enhance its global correlation. The formula is: , in, After the synergy is enhanced The first sample of agricultural products The synergistic characteristic value of the amino acid characteristic peaks reflects the interaction strength and synergistic change characteristics of the amino acid characteristic peaks in different agricultural product samples. The synergistic characteristic value is obtained by enhancing the dynamic offset value and the synergistic effect, which not only reflects the change of a single characteristic peak, but also considers the mutual influence between multiple characteristic peaks. is the smoothing factor, which is used to adjust the influence weight of the preprocessed amino acid fingerprint data. The smoothing factor determines The decay rate of , thereby smoothing the signal and avoiding the excessive dominance of the strong signal on the eigenvalue, the value range is ; It is used to amplify the dynamic offset value by cubic power, while combining exponential weighting to weaken the absolute dominance of strong signals; Indicates The first sample of agricultural products The dynamic shift value of the amino acid characteristic peak.

[0031] Furthermore, for the cooperative eigenvalue Synergy Feature Matrix Nonlinear mapping is performed to further enhance the high-dimensional expression ability of features using nonlinear transformation, while introducing dynamic changes through the agricultural product sample numbers. The nonlinear mapping formula is as follows: , in, It is The first sample of agricultural products The enhanced characteristic value of the amino acid characteristic peak; It is the sample code of agricultural products; It is a parameter that controls the amplitude of the enhancement eigenvalue change, and is used to smooth high-intensity signals, thereby limiting the speed at which the enhancement eigenvalue increases; and Agricultural product sample numbers were introduced The periodic transformation of the square synergy eigenvalue avoids the linear correlation of the features. The enhanced feature matrix composed of A set of more distributed high-dimensional feature representations is generated.

[0032] The enhanced eigenvalues ​​contain abnormal peaks or noise signals, so dynamic penalty processing is needed to eliminate the influence of abnormal agricultural product samples. The core of dynamic penalty processing is to calculate the weight adjustment value of each enhanced eigenvalue, and the formula is: , in, It is The first sample of agricultural products The characteristic value of the amino acid characteristic peak after dynamic penalty adjustment, is the penalty coefficient, which is used to control the intensity of dynamic penalty and is determined according to expert experience; is the enhanced feature matrix Middle The mean of the column-enhanced features; It represents the absolute difference between the enhanced feature value and the mean of the enhanced feature of its corresponding column, and on this basis, the influence of outliers is reduced by exponential penalty; is an exponential decay function that is used to weight the eigenvalues ​​according to their deviation from the mean.

[0033] Further, The dynamic penalty adjusted feature matrix Perform feature optimization processing. By adjusting the feature matrix after dynamic penalty Sparsity analysis is performed to screen important features. The sparsity weight calculation formula is: , in, It is The first sample of agricultural products The sparsity weight of the amino acid characteristic peak reflects the characteristic value after dynamic penalty adjustment contribution to the overall distribution in the agricultural product sample; is a sparsity adjustment factor used to adjust the sensitivity of the exponential penalty. It is a hyperparameter set manually. The appropriate value is selected based on experimental experience, such as 0.1 or 0.5. middle, is an exponential decay term, which exponentially decays the eigenvalue after dynamic penalty adjustment, aiming to penalize features with small values ​​to reduce their impact on the final result. The sparsity weight increases the weight of high-contribution features by combining the square and exponential functions. After the sparsity analysis is completed, feature selection is performed based on the sparsity weight, and the feature matrix after dynamic penalty adjustment is reconstructed. The optimized eigenvalue reconstruction formula is: , In the formula, It is The first sample of agricultural products The final optimized characteristic value corresponding to the characteristic peak of amino acids; is the nonlinear amplification factor, which is used to control the contribution weight of the nonlinear optimization term and adjust Strength; It is a nonlinear enhancement factor, which is used to control the nonlinear enhancement amplitude of the eigenvalue after dynamic penalty adjustment. Adjustment can be enlarged value, making its influence in the final optimized features more significant.

[0034] Ultimately, it will be The optimized feature matrix composed of As the optimized amino acid feature data.

[0035] S3. The optimized amino acid feature data is traceably matched with the historical authentication data in the blockchain. When the match is successful, an authentication report is generated; when the match fails, the optimized amino acid feature data is stored and authenticated in the blockchain; Introduce a dynamic weighted matching formula to trace the optimized amino acid feature data with the historical authentication data in the blockchain and calculate the matching rate , the formula is as follows: , in, It is the optimized amino acid feature data after normalization; It is the historical authentication data in the normalized blockchain; It is a dynamic weighting factor, which is used to dynamically adjust the weight in the calculation according to the difference between two feature data. The dynamic weighting factor is used to emphasize the features with large differences. The larger the difference, the larger the dynamic weighting factor, and vice versa, the smaller the dynamic weighting factor, to ensure a more sensitive response when the feature difference is large. ; is the threshold decision function, is the allowable error range.

[0036] If the matching rate ( The authentication report includes basic information such as the place of production, authentication time, and testing agency, as well as authentication data such as the matching rate, matching feature data summary (amino acid type and content), and fingerprint spectrum. At the same time, the IoT device is used to dynamically collect logistics node data, including location data, time data, environmental data, node identification, etc. The collected logistics node data is organized into a structured data packet, and the logistics node data and the authentication data are hashed and bound using the existing hash algorithm to generate a unique identification of the traceability node, that is, the binding result of the logistics node data and the authentication data. A block is then generated based on the traceability data bound to each logistics node, and the block is linked to the previous block through a hash value. Finally, the block data is written into the distributed storage network of the blockchain to ensure tamper-proofness. Then, according to the logistics node data stored in the blockchain, each block is read in chronological order. Starting from the starting authentication data (place of production), all logistics node data are connected in series into a complete chain to generate a traceability chain containing the complete path of production, authentication, transportation, and consumption.

[0037] If the matching rate , then the match is judged to have failed, and the blockchain storage and authentication phase is entered. The existing hash algorithm is used to generate the unique identifier of the optimized amino acid feature data, and the blockchain storage structure is constructed based on the unique identifier to achieve tamper-proof storage and authentication of the data. The blockchain storage structure includes the block header (timestamp, previous block hash value), the data body (feature summary hash value, production information, detection information) and the pointer hash to ensure the traceability and non-tamperability of the new data.

[0038] In summary, the agricultural product quality certification and traceability data management system and method based on amino acid fingerprint was completed.

[0039] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0040] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0041] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such 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, and should be included in the protection scope of the present invention.

Claims

1. A method for agricultural product quality certification and traceability data management based on amino acid fingerprint, characterized in that: The following steps are involved: S1. Collecting amino acid fingerprint data of agricultural products, and preprocessing the amino acid fingerprint data to obtain preprocessed amino acid fingerprint data; S2. Using a dynamic nonlinear feature collaborative optimization extraction algorithm to extract and optimize the preprocessed amino acid fingerprint data to obtain optimized amino acid feature data; S3. The optimized amino acid feature data is traceably matched with the historical authentication data in the blockchain. When the match is successful, an authentication report is generated; when the match fails, the optimized amino acid feature data is stored and authenticated on the blockchain.

2. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 1, characterized in that: The S1 specifically includes: During the preprocessing process, the amino acid fingerprint data is denoised to obtain denoised amino acid fingerprint data; a comprehensive optimization baseline correction algorithm is introduced to perform baseline correction on the denoised amino acid fingerprint data to obtain optimized baseline correction data; the optimized baseline correction data is subjected to data normalization, feature peak extraction and data formatting to obtain preprocessed amino acid fingerprint data.

3. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 2, characterized in that: The S1 specifically includes: In the process of implementing the comprehensive optimization baseline correction algorithm, the denoised amino acid fingerprint data are processed through five stages: signal smoothing, piecewise adaptive fitting, dynamic residual optimization, disturbance noise suppression, extreme value enhancement and peak correction, to obtain the optimized baseline correction data.

4. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 1, characterized in that: The S2 specifically includes: In the process of implementing the dynamic nonlinear feature collaborative optimization extraction algorithm, the preprocessed amino acid fingerprint data is defined as an amino acid fingerprint matrix, and the dynamic offset value of the preprocessed amino acid fingerprint data is calculated by dynamic collaborative offset. The calculation formula of the dynamic offset value is: ， in, Indicates The first sample of agricultural products Dynamic shift value of amino acid characteristic peak; It is Dynamic weights of agricultural product samples; Indicates The first sample of agricultural products The data intensity of the characteristic peaks of amino acids; Indicates The first sample of agricultural products The global weighted sum of the amino acid characteristic peaks; is the number of agricultural product samples; It is Dynamic weights of agricultural product samples; It is The first sample of agricultural products The data intensity of the characteristic peaks of amino acids.

5. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 4, characterized in that: The S2 specifically includes: In the implementation process of the dynamic nonlinear feature collaborative optimization extraction algorithm, the dynamic offset value is converted into a collaborative eigenvalue through collaborative enhancement processing, and the collaborative feature matrix composed of the collaborative eigenvalues ​​is subjected to nonlinear mapping processing to obtain the enhanced eigenvalue. The specific formula is: ， in, It is The first sample of agricultural products The enhanced characteristic value of the amino acid characteristic peak; After the synergy is enhanced The first sample of agricultural products The synergistic characteristic value of the amino acid characteristic peak; It is the sample code of agricultural products; It is a parameter that controls the amplitude of the enhanced eigenvalue change.

6. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 5, characterized in that: The S2 specifically includes: In the process of realizing the dynamic nonlinear feature collaborative optimization extraction algorithm, the enhanced feature matrix composed of enhanced eigenvalues ​​is dynamically penalized to obtain the eigenvalues ​​adjusted by dynamic penalties, and the eigenvalues ​​adjusted by dynamic penalties constitute the feature matrix adjusted by dynamic penalties.

7. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 6, characterized in that: The S2 specifically includes: In the process of implementing the dynamic nonlinear feature collaborative optimization extraction algorithm, the feature matrix after dynamic penalty adjustment is optimized through sparsity analysis and feature selection to obtain optimized eigenvalues; the optimized feature matrix is ​​composed of the optimized eigenvalues, and the optimized feature matrix is ​​used as the optimized amino acid feature data.

8. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 1, characterized in that: The S3 specifically includes: Introduce a dynamic weighted matching formula to trace the optimized amino acid feature data with the historical authentication data in the blockchain and calculate the matching rate , and based on the matching rate Determine whether the traceability match is successful, the matching rate The specific formula is as follows: ， in, It is the optimized amino acid feature data after normalization; It is the historical authentication data in the normalized blockchain; is the dynamic weighting factor, ; is the threshold decision function, is the margin of error.

9. The method for agricultural product quality authentication and traceability data management based on amino acid fingerprint according to claim 8, characterized in that: The S3 specifically includes: When the matching rate Preset Threshold , the match is determined to be successful and the authentication report generation phase is entered; at the same time, the logistics node data is collected, the logistics node data is hashed and bound to the authentication data, a unique identifier of the traceability node is generated, and then a block is generated based on the traceability data bound to each logistics node, which is linked to the previous block through the hash value, and finally the block data is written into the distributed storage network of the blockchain, and then the traceability chain is generated based on the logistics node data stored in the blockchain; when the matching rate is Preset Threshold , it is determined that the match fails and the blockchain storage and authentication phase is entered. Based on the unique identifier of the generated optimized amino acid feature data, a blockchain data storage structure is constructed.

10. The agricultural product quality authentication and traceability data management system based on amino acid fingerprint is applied to the agricultural product quality authentication and traceability data management method based on amino acid fingerprint as claimed in claim 1, characterized in that: Includes the following parts: Amino acid fingerprint acquisition module, feature extraction and optimization module, traceability and matching module, blockchain storage and authentication module, supply chain dynamic monitoring module; The amino acid fingerprint collection module collects and preprocesses the amino acid fingerprint data of agricultural products to obtain the preprocessed amino acid fingerprint data, and then sends the preprocessed amino acid fingerprint data to the feature extraction and optimization module; The feature extraction and optimization module extracts features from the preprocessed amino acid fingerprint data, optimizes the extracted feature data, obtains optimized amino acid feature data, and sends the optimized amino acid feature data to the traceability and matching module; The traceability and matching module traces and matches the optimized amino acid feature data with the historical authentication data in the blockchain. When the match is successful, a certification report is generated and sent to the supply chain dynamic monitoring module; when the match fails, the optimized amino acid feature data is sent to the blockchain storage and authentication module; The blockchain storage and authentication module uses the unique identifier of the optimized amino acid feature data to construct a blockchain storage structure and provide historical authentication data for the traceability and matching module; The supply chain dynamic monitoring module collects logistics node data of agricultural products during transportation, hashes the logistics node data with the certification data in the certification report, and stores it in the blockchain storage and authentication module to generate a traceability chain.

Images