Adaptive deformation correction and recognition system for flexible curved strip codes

Abstract

The present invention relates to the field of automated identification technology, and specifically discloses an adaptive deformation correction and identification system for flexible curved barcodes. By real-time monitoring of the three-dimensional shape changes and strain distribution of the barcode area on the flexible surface, the system can accurately capture complex deformation situations, and utilizes a K-means clustering algorithm to calculate curvature eigenvalues, and Haar wavelet transform to analyze strain data to determine deformation eigenvalues, thereby comprehensively evaluating the degree of deformation of the barcode and its impact on the identification process. Based on these eigenvalues, a random forest model is constructed to predict the degree of deformation and output a deformation score to ensure high-precision identification. When severe, uncorrectable deformation is detected, the system automatically triggers an alarm mechanism, including an audible alarm, a visual signal, and a network notification, to promptly inform management personnel and ensure the stability and long-term reliability of the system.

Description

Technical Field

[0001] The present invention relates to the field of automatic identification technology, in particular to a self-adaptive deformation correction and identification system for a flexible curved bar code. Background Art

[0002] With the increasing demand for barcode recognition in industries such as logistics, manufacturing, and retail, traditional flat barcodes can no longer meet the requirements of applications on flexible curved surfaces. The application scenarios of flexible curved barcodes are becoming increasingly extensive, including but not limited to packaging, identification of medical devices and personal electronic products. However, barcodes in these application scenarios often face recognition challenges due to bending, folding or other physical deformations of the substrate material. Therefore, developing a system that can adapt to and correct these deformations has become an important direction of current research. Such a system not only needs to have high-precision three-dimensional shape measurement capabilities, but also needs to be able to monitor and evaluate the strain distribution of the material in real time to ensure accurate barcode recognition even under complex deformation conditions.

[0003] The existing technology has the following deficiencies:

[0004] Although a variety of technologies and methods have been used to address the deformation problem of flexible curved strip codes, they still have significant limitations. Traditional methods usually rely on single-dimensional data acquisition methods, such as using only two-dimensional image processing technology for analysis, which results in incomplete and inaccurate capture of complex deformation patterns. In addition, existing deformation correction algorithms are mostly based on preset models or parameters, which are difficult to adapt to actual changes in different materials and environments, thus affecting the accuracy of recognition. At the same time, there is a lack of effective early warning mechanisms to promptly detect and deal with severe deformations that cannot be corrected, resulting in a slow response of the system in the face of emergencies, increasing the risk of recognition failure. These limitations indicate that existing technologies still have much room for improvement in terms of flexibility, accuracy and reliability, and there is an urgent need for a more comprehensive and intelligent technical solution to address the various challenges in the recognition of flexible curved strip codes. Summary of the Invention

[0005] The object of the present invention is to provide an adaptive deformation correction and recognition system for flexible curved strip codes to solve the above-mentioned problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The adaptive deformation correction and recognition system of flexible curved barcode includes:

[0008] A data acquisition module monitors the three-dimensional shape change of the barcode area on the flexible surface in real time, obtains curvature distribution data, and directly measures the strain distribution on the material to obtain strain data;

[0009] a flexible curved line strip code recognition module, which determines a curvature characteristic value based on the degree of change in the curvature distribution, calculates a deformation characteristic value based on the degree of abnormality in the strain distribution, and identifies the degree of deformation based on the curvature characteristic value and the deformation characteristic value of the flexible curved line strip code;

[0010] The flexible curved strip code correction module determines whether there is any severe deformation that cannot be corrected based on the deformation degree recognition result, and automatically triggers an alarm mechanism if there is any severe deformation that cannot be corrected.

[0011] As a further solution of the present invention, the local deformation characteristics within the barcode area are evaluated according to the degree of change in the curvature distribution, specifically including:

[0012] The curvature distribution data of the flexible curved bar code is collected in real time. The curvature characteristic value is calculated according to the degree of change of the curvature distribution of the flexible curved bar code. It is judged whether the curvature characteristic value of the flexible curved bar code is greater than or equal to a preset threshold. If so, there is local deformation in the bar code area. If not, there is no local deformation in the bar code area.

[0013] As a further solution of the present invention: the process of obtaining the curvature characteristic value is:

[0014] The curvature distribution data of the flexible curved barcode is collected in real time. For each data point in the collected data set, a local quadratic surface is fitted and its principal curvature is extracted to obtain the curvature value of the corresponding point. All the obtained curvature values ​​are integrated into a curvature set. The curvature set is classified using the K-means clustering algorithm. The number of clusters is first selected and the centroid position of each cluster is initialized. Based on the distance between each data point and each centroid, the data point is assigned to the nearest cluster. The centroid position is updated until the algorithm converges. The data points are divided into several categories based on the curvature value to identify the curvature characteristics of different regions.

[0015] For each cluster, the curvature eigenvalue of the corresponding cluster is obtained by adding the curvature values ​​of all data points belonging to the same cluster and dividing it by the total number of data points in the corresponding cluster; the curvature eigenvalues ​​of all clusters are averaged to obtain the curvature eigenvalue of the entire flexible curved strip code.

[0016] As a further solution of the present invention, the physical deformation characteristics of the flexible curved strip code are evaluated according to the abnormality of the strain distribution, specifically including:

[0017] The strain data of the flexible curved strip code is collected in real time. The deformation characteristic value is calculated according to the abnormality of the strain distribution of the flexible curved strip code. It is judged whether the deformation characteristic value of the flexible curved strip code is greater than or equal to a preset threshold. If so, the flexible curved strip code has physical deformation. If not, the flexible curved strip code does not have physical deformation.

[0018] As a further solution of the present invention: the process of obtaining the deformation characteristic value is:

[0019] The strain data of the barcode area on the flexible surface is acquired in real time. The Haar wavelet transform is applied to the strain data to decompose the signal into approximate coefficients and detail coefficients at different scales. After multi-layer decomposition, a set of approximate coefficients and several sets of detail coefficients are obtained. Each set of detail coefficients represents the detail changes at different scales.

[0020] For each group of detail coefficients, calculating its energy distribution, specifically comprising: for each element in each group of detail coefficients, calculating its square value and accumulating all these square values ​​to obtain the energy of the corresponding group of detail coefficients;

[0021] According to the energy distribution of all groups of detail coefficients, this information is integrated in the form of a ratio to calculate the overall deformation eigenvalue. A small positive constant is added to the energy of each group of detail coefficients, all adjusted energy values ​​are multiplied, and divided by the sum of all weight factors to obtain the deformation eigenvalue.

[0022] As a further solution of the present invention: the identification of the deformation degree according to the curvature characteristic value and the deformation characteristic value of the flexible curved strip code specifically includes:

[0023] The curvature eigenvalue and deformation eigenvalue of the flexible curved strip code are obtained, and the curvature eigenvalue and deformation eigenvalue are directly constructed into a comprehensive eigenvector as the input of the machine learning model. The model is trained with minimizing the error between the predicted deformation degree score and the actual deformation degree score as the training goal. According to the trained model, the deformation degree score of the flexible curved strip code is output. The machine learning model is a random forest model.

[0024] As a further solution of the present invention: the training process of the machine learning model is:

[0025] During the model training phase, the constructed comprehensive feature vector is trained using the random forest algorithm. Multiple decision trees are constructed and their results are summarized to improve prediction accuracy and control overfitting. During the training process, the objective function is to minimize the error between the predicted deformation degree recognition accuracy score and the actual deformation degree recognition accuracy score. The parameters of each decision tree are continuously adjusted until the optimal parameter combination is found. The trained random forest model is used to analyze the received new comprehensive feature vector and output the deformation degree score of the flexible curved strip code.

[0026] As a further solution of the present invention, the determination of whether there is an uncorrectable severe deformation specifically includes:

[0027] It is determined whether the deformation degree score of the flexible curved strip code during recognition is greater than or equal to a preset threshold. If so, there is severe deformation that cannot be corrected; otherwise, there is no severe deformation that cannot be corrected.

[0028] As a further solution of the present invention, the determination of whether there is an uncorrectable severe deformation specifically includes:

[0029] It is determined whether the deformation degree score of the flexible curved strip code during recognition is greater than or equal to a preset threshold. If so, there is severe deformation that cannot be corrected; otherwise, there is no severe deformation that cannot be corrected.

[0030] Beneficial effects of the present invention:

[0031] (1) By integrating advanced three-dimensional scanning technology and strain measurement technology, the present invention realizes real-time and accurate monitoring of the barcode area on the flexible surface, not only capturing its complex three-dimensional shape changes, but also directly measuring the strain distribution on the material, fully reflecting the real physical state of the environment in which the barcode is located. The curvature data is classified and the curvature eigenvalues ​​are calculated using the K-means clustering algorithm. At the same time, the Haar wavelet transform is applied to analyze the strain detail coefficient energy at multiple scales to determine the deformation eigenvalues, thereby systematically evaluating the local and overall deformation of the barcode and its impact on the recognition process. This multi-level data processing and analysis method significantly improves the accuracy and robustness of deformation recognition, providing a solid foundation for subsequent correction measures. Even in the face of complex and unpredictable deformation conditions, the system can still ensure high-precision barcode recognition, greatly enhancing the adaptability and reliability of the system, and ensuring stable and efficient barcode reading and decoding in various application scenarios. In addition, through in-depth analysis of the collected data, the present invention also supports the triggering of an automatic warning mechanism, further ensuring the long-term stable operation and management efficiency of the system.

[0032] (2) When uncorrectable severe deformation is detected, the present invention can intelligently and automatically trigger a multi-level alarm mechanism, including immediate sound alarms, visual signals such as flashing warning lights, and notifications to relevant managers via the network, ensuring that potential problems are responded to in the first place. In addition, based on the constructed random forest model, the system conducts in-depth analysis of the received new comprehensive feature vectors and outputs an accurate deformation degree score, so that potential problems can be identified and handled at an early stage, effectively preventing recognition failures or errors caused by deformation. This innovation not only significantly improves the system's response speed and processing efficiency, but also optimizes resource management through continuous data monitoring and analysis, minimizing losses caused by deformation. The design concept of the system emphasizes comprehensive protection and efficient management. By integrating advanced data acquisition, processing technology and intelligent early warning mechanisms, it ensures excellent stability and long-term reliability even in complex and changing application environments, providing users with a highly reliable and easy-to-manage barcode recognition solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below with reference to the accompanying drawings.

[0034] Figure 1 It is a flow chart of the adaptive deformation correction and recognition system of the flexible curved strip code of the present invention. DETAILED DESCRIPTION

[0035] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0036] See also Figure 1 As shown, the present invention is an adaptive deformation correction and recognition system for flexible curved strip codes, comprising:

[0037] A data acquisition module monitors the three-dimensional shape change of the barcode area on the flexible surface in real time, obtains curvature distribution data, and directly measures the strain distribution on the material to obtain strain data;

[0038] a flexible curved line strip code recognition module, which determines a curvature characteristic value based on the degree of change in the curvature distribution, calculates a deformation characteristic value based on the degree of abnormality in the strain distribution, and identifies the degree of deformation based on the curvature characteristic value and the deformation characteristic value of the flexible curved line strip code;

[0039] The flexible curved strip code correction module determines whether there is any severe deformation that cannot be corrected based on the deformation degree recognition result, and automatically triggers an alarm mechanism if there is any severe deformation that cannot be corrected.

[0040] In the data acquisition module, the three-dimensional shape changes of the barcode area on the flexible surface are monitored in real time to obtain curvature distribution data. The strain distribution on the material is directly measured to obtain strain data, including:

[0041] First, a 3D scanning device monitors the 3D shape changes of the barcode area on a flexible surface in real time, acquiring precise spatial coordinate information. These devices project structured light or laser light and analyze the reflected pattern to construct a 3D model of the barcode surface, extracting the location of each data point and its corresponding curvature distribution. This process ensures high-precision capture of the complex geometry of the barcode surface, providing the foundational data for subsequent analysis.

[0042] At the same time, the strain distribution on the material is directly measured to obtain strain data using strain gauges. Strain gauges are directly attached to the surface of the material and can sensitively respond to changes in the material's expansion and contraction, outputting corresponding electrical signals to represent the local strain.

[0043] In the flexible curved strip code recognition module, the curvature characteristic value is determined according to the degree of change in the curvature distribution, and the deformation characteristic value is calculated according to the abnormality of the strain distribution. The deformation degree is identified based on the curvature characteristic value and deformation characteristic value of the flexible curved strip code, specifically including:

[0044] The curvature distribution data of the flexible curved bar code is collected in real time. The curvature characteristic value is calculated according to the degree of change of the curvature distribution of the flexible curved bar code. It is judged whether the curvature characteristic value of the flexible curved bar code is greater than or equal to a preset threshold. If so, there is local deformation in the bar code area. If not, there is no local deformation in the bar code area.

[0045] The process of obtaining the curvature characteristic value is as follows:

[0046] The curvature distribution data of the flexible curved barcode is collected in real time. For each data point in the collected data set, a local quadratic surface is fitted and its principal curvature is extracted to obtain the curvature value of the corresponding point. All the obtained curvature values ​​are integrated into a curvature set. The curvature set is classified using the K-means clustering algorithm. The number of clusters is first selected and the centroid position of each cluster is initialized. Based on the distance between each data point and each centroid, the data point is assigned to the nearest cluster. The centroid position is updated until the algorithm converges. The data points are divided into several categories based on the curvature value to identify the curvature characteristics of different regions.

[0047] For each cluster, the curvature eigenvalue of the corresponding cluster is obtained by adding the curvature values ​​of all data points belonging to the same cluster and dividing it by the total number of data points in the corresponding cluster; the curvature eigenvalues ​​of all clusters are averaged to obtain the curvature eigenvalue of the entire flexible curved strip code.

[0048] The strain data of the flexible curved strip code is collected in real time. The deformation characteristic value is calculated according to the abnormality of the strain distribution of the flexible curved strip code. It is judged whether the deformation characteristic value of the flexible curved strip code is greater than or equal to a preset threshold. If so, the flexible curved strip code has physical deformation. If not, the flexible curved strip code does not have physical deformation.

[0049] The process of obtaining the deformation eigenvalue is as follows:

[0050] The strain data of the barcode area on the flexible surface is acquired in real time. The Haar wavelet transform is applied to the strain data to decompose the signal into approximate coefficients and detail coefficients at different scales. After multi-layer decomposition, a set of approximate coefficients and several sets of detail coefficients are obtained. Each set of detail coefficients represents the detail changes at different scales.

[0051] For each set of detail coefficients, calculate its energy distribution. Specifically, for each element in each set of detail coefficients, take its square value and accumulate all these square values ​​to obtain the energy of the set of detail coefficients. This process helps us identify the intensity of the change in strain data at each scale.

[0052] Based on the energy distribution of all groups of detail coefficients, this information is synthesized in the form of ratios to calculate the overall deformation eigenvalue. A small positive constant is added to the energy of each group of detail coefficients. Then, all the adjusted energy values ​​are multiplied and divided by the sum of all weight factors to obtain the deformation eigenvalue. This method not only takes into account the deformation information at different scales, but also emphasizes the deformation characteristics at certain specific scales by introducing weight factors.

[0053] It should be noted that the overall deformation eigenvalue reflects the overall deformation degree and anomaly characteristics of the strain data. It is derived by adjusting, multiplying, and averaging the energies of all groups of detail coefficients. It accurately captures local anomalies caused by material deformation in flexible curved barcodes. It is suitable for barcode recognition in complex deformation scenarios and provides reliable input features for subsequent correction algorithms, improving the sensitivity and accuracy of deformation assessment.

[0054] It should be noted that in the flexible curved bar code recognition module, the curvature distribution data of the bar code area is first collected in real time through a three-dimensional scanning or stereo vision system, and the K-means clustering algorithm is applied to calculate the curvature eigenvalue to identify local deformation. At the same time, strain gauges are used to obtain strain distribution data on the material, and the Haar wavelet transform is used to analyze the detail coefficient energy at each scale. The deformation eigenvalue is comprehensively calculated to evaluate the physical deformation. Based on the comparison of the calculated curvature and deformation eigenvalues ​​with the preset threshold, it is determined whether there is severe and uncorrectable deformation, and the corresponding alarm mechanism is triggered. This innovative method can not only accurately capture the complex deformation of the bar code on the flexible surface, but also provide reliable data support for subsequent correction measures. It significantly improves the accuracy and stability of bar code recognition in complex environments, reflecting a high degree of technical innovation and practicality.

[0055] In the flexible curved strip code correction module, based on the deformation degree recognition results, it is determined whether there is severe deformation that cannot be corrected. If there is severe deformation that cannot be corrected, an alarm mechanism is automatically triggered, specifically including:

[0056] The curvature eigenvalue and deformation eigenvalue of the flexible curved strip code are obtained, and the curvature eigenvalue and deformation eigenvalue are directly constructed into a comprehensive eigenvector as the input of the machine learning model. The model is trained with minimizing the error between the predicted deformation degree score and the actual deformation degree score as the training goal. According to the trained model, the deformation degree score of the flexible curved strip code is output. The machine learning model is a random forest model.

[0057] During the model training phase, the constructed comprehensive feature vector is trained using the random forest algorithm. Multiple decision trees are constructed and their results are summarized to improve prediction accuracy and control overfitting. During the training process, the objective function is to minimize the error between the predicted deformation degree recognition accuracy score and the actual deformation degree recognition accuracy score. The parameters of each decision tree are continuously adjusted until the optimal parameter combination is found. The trained random forest model is used to analyze the received new comprehensive feature vector and output the deformation degree score of the flexible curved strip code.

[0058] It is determined whether the deformation degree score of the flexible curved strip code during recognition is greater than or equal to a preset threshold. If so, there is severe deformation that cannot be corrected; otherwise, there is no severe deformation that cannot be corrected.

[0059] When the flexible curved barcode is detected to have severe and uncorrectable deformation, the system will automatically trigger an early warning mechanism, including sound alarms and visual signals such as flashing warning lights, and immediately notify relevant managers through the network.

[0060] The present invention operates by comprising a data acquisition module, a flexible curved barcode recognition module, and a flexible curved barcode correction module. The data acquisition module utilizes a 3D scanning device to monitor the 3D shape changes of the barcode area in real time, acquiring precise spatial coordinate information. This information then extracts the location of each data point and its corresponding curvature distribution data, ensuring high-precision capture of the complex geometric morphology of the barcode surface. Simultaneously, the strain distribution on the material is directly measured to obtain strain data. Strain gauges are used to sensitively respond to changes in material expansion and contraction, outputting corresponding electrical signals representing the local strain.

[0061] In the flexible curved strip code recognition module, based on curvature distribution data collected in real time, the curvature value of each data point is calculated by fitting a local quadratic surface and extracting its principal curvature. These curvature values ​​are then classified into clusters using the K-means clustering algorithm. The curvature eigenvalue is determined based on the average curvature value of the data points within each cluster. Furthermore, the Haar wavelet transform is applied to the strain data to analyze the detail coefficient energy at each scale and comprehensively calculate the deformation eigenvalue. By comparing the curvature and deformation eigenvalues ​​with preset thresholds, it is determined whether there is severe, uncorrectable deformation, providing a basis for decision-making in subsequent processing.

[0062] The flexible curved bar code correction module performs further operations based on the above recognition results. If the recognition results show that there is a serious deformation that cannot be corrected, the alarm mechanism is automatically triggered, including sound alarms, visual signals such as flashing warning lights, and immediate notification of relevant management personnel through the network. Specifically, the curvature eigenvalues ​​and deformation eigenvalues ​​are constructed into a comprehensive feature vector as the input of the random forest model. The model improves the prediction accuracy by summarizing the results of multiple decision trees, and is trained with the objective function of minimizing the error between the predicted deformation degree score and the actual deformation degree score. The trained model can analyze the new comprehensive feature vector received and output the deformation degree score of the flexible curved bar code, thereby evaluating the possibility of the bar code being correctly recognized under the current conditions.

[0063] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0064] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0065] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0066] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0067] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. Adaptive deformation correction and recognition system for flexible curved strip codes, characterized by: include: A data acquisition module monitors the three-dimensional shape change of the barcode area on the flexible surface in real time, obtains curvature distribution data, and directly measures the strain distribution on the material to obtain strain data; a flexible curved line strip code recognition module, which determines a curvature characteristic value based on the degree of change in the curvature distribution, calculates a deformation characteristic value based on the degree of abnormality in the strain distribution, and identifies the degree of deformation based on the curvature characteristic value and the deformation characteristic value of the flexible curved line strip code; The process of obtaining the deformation eigenvalue is as follows: The strain data of the barcode area on the flexible surface is acquired in real time. The Haar wavelet transform is applied to the strain data to decompose the signal into approximate coefficients and detail coefficients at different scales. After multi-layer decomposition, a set of approximate coefficients and several sets of detail coefficients are obtained. Each set of detail coefficients represents the detail changes at different scales. For each group of detail coefficients, calculating its energy distribution, specifically comprising: for each element in each group of detail coefficients, calculating its square value and accumulating all these square values ​​to obtain the energy of the corresponding group of detail coefficients; According to the energy distribution of all groups of detail coefficients, this information is integrated in the form of ratios to calculate the overall deformation eigenvalue. A small positive constant is added to the energy of each group of detail coefficients, all adjusted energy values ​​are multiplied, and then divided by the sum of all weight factors to obtain the deformation eigenvalue. The flexible curved strip code correction module determines whether there is any severe deformation that cannot be corrected based on the deformation degree recognition result, and automatically triggers an alarm mechanism if there is any severe deformation that cannot be corrected.

2. The adaptive deformation correction and recognition system for flexible curved strip codes according to claim 1, characterized in that: Based on the degree of change in the curvature distribution, the local deformation characteristics within the barcode area are evaluated, including: The curvature distribution data of the flexible curved bar code is collected in real time. The curvature characteristic value is calculated according to the degree of change of the curvature distribution of the flexible curved bar code. It is judged whether the curvature characteristic value of the flexible curved bar code is greater than or equal to a preset threshold. If so, there is local deformation in the bar code area. If not, there is no local deformation in the bar code area.

3. The adaptive deformation correction and recognition system for flexible curved strip codes according to claim 1, characterized in that: The process of obtaining the curvature characteristic value is as follows: The curvature distribution data of the flexible curved barcode is collected in real time. For each data point in the collected data set, a local quadratic surface is fitted and its principal curvature is extracted to obtain the curvature value of the corresponding point. All the obtained curvature values ​​are integrated into a curvature set. The curvature set is classified using the K-means clustering algorithm. The number of clusters is first selected and the centroid position of each cluster is initialized. Based on the distance between each data point and each centroid, the data point is assigned to the nearest cluster. The centroid position is updated until the algorithm converges. The data points are divided into several categories based on the curvature value to identify the curvature characteristics of different regions. For each cluster, the curvature characteristic value of the corresponding cluster is obtained by adding the curvature values ​​of all data points belonging to the same cluster and dividing it by the total number of data points in the corresponding cluster; The curvature eigenvalues ​​of all clusters are averaged to obtain the overall curvature eigenvalue of the flexible curved strip code.

4. The adaptive deformation correction and recognition system for flexible curved strip codes according to claim 1, characterized in that: According to the abnormality of strain distribution, the physical deformation characteristics of the flexible curved strip code are evaluated, including: The strain data of the flexible curved strip code is collected in real time. The deformation characteristic value is calculated according to the abnormality of the strain distribution of the flexible curved strip code. It is judged whether the deformation characteristic value of the flexible curved strip code is greater than or equal to a preset threshold. If so, the flexible curved strip code has physical deformation. If not, the flexible curved strip code does not have physical deformation.

5. The adaptive deformation correction and recognition system for flexible curved strip codes according to claim 1, characterized in that: The identification of the deformation degree according to the curvature characteristic value and the deformation characteristic value of the flexible curved strip code specifically includes: The curvature eigenvalue and deformation eigenvalue of the flexible curved strip code are obtained, and the curvature eigenvalue and deformation eigenvalue are directly constructed into a comprehensive eigenvector as the input of the machine learning model. The model is trained with minimizing the error between the predicted deformation degree score and the actual deformation degree score as the training goal. According to the trained model, the deformation degree score of the flexible curved strip code is output. The machine learning model is a random forest model.

6. The adaptive deformation correction and recognition system for flexible curved strip codes according to claim 5, characterized in that: The training process of the machine learning model is: During the model training phase, the constructed comprehensive feature vector is trained using the random forest algorithm. Multiple decision trees are constructed and their results are summarized to improve prediction accuracy and control overfitting. During the training process, the objective function is to minimize the error between the predicted deformation degree recognition accuracy score and the actual deformation degree recognition accuracy score. The parameters of each decision tree are continuously adjusted until the optimal parameter combination is found. The trained random forest model is used to analyze the received new comprehensive feature vector and output the deformation degree score of the flexible curved strip code.

7. The adaptive deformation correction and recognition system for flexible curved strip codes according to claim 1, characterized in that: The determination of whether there is severe, uncorrectable deformation specifically includes: It is determined whether the deformation degree score of the flexible curved strip code during recognition is greater than or equal to a preset threshold. If so, there is severe deformation that cannot be corrected; otherwise, there is no severe deformation that cannot be corrected.

8. The adaptive deformation correction and recognition system for flexible curved strip codes according to claim 1, characterized in that: If there is severe, uncorrectable deformation, the alarm mechanism is automatically triggered, specifically including: When the flexible curved barcode is detected to have severe and uncorrectable deformation, the system will automatically trigger an early warning mechanism, including an audible alarm and a visual signal of a flashing warning light, and will immediately notify relevant management personnel through the network.

Images