Terahertz human body security inspection image optimization method and system
By constructing the correlation matrix between the operating parameters of the equipment and the image quality characteristics and using particle cluster optimization processing, the problem of image quality degradation caused by changes in equipment status in terahertz human security image optimization is solved, and efficient and accurate security image optimization is achieved.
Patent Information
- Application Number
- CN202410524666.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-29
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-04-29
AI Technical Summary
The existing terahertz human body security image optimization methods lack targeted and adaptable, making it difficult to cope with complex and changeable equipment operating status, resulting in a decline in image quality and affecting security efficiency and accuracy.
By collecting real-time device operation status data and historical data, a correlation matrix between device operation parameters and image quality characteristics is constructed, and a pre-trained model and particle cluster can be used to identify device degradation parameters in real time and generate optimized particles to perform targeted image processing.
It realizes intelligent response to complex equipment status, improves image quality and security inspection efficiency, and ensures the accuracy and efficiency of security inspection.
Smart Images

Figure CN118446902B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of security inspection image optimization, and in particular to a terahertz human body security inspection image optimization method and system. Background Art
[0002] With the advancement of science and technology and the improvement of security needs, terahertz technology has been widely used in the field of human security inspection due to its unique non-ionization, strong penetration, and ability to effectively reveal hidden objects. However, the quality of terahertz human security inspection images directly affects the efficiency and accuracy of security inspections. How to effectively optimize such images to ensure the accuracy and efficiency of security inspections has become a key topic in current research and practice.
[0003] Terahertz human body security inspection systems usually contain complex hardware equipment and advanced imaging algorithms. Slight changes in their operating parameters may have a significant impact on the quality of the generated security inspection images. These parameters cover many aspects such as light source intensity, detector sensitivity, scanning speed, signal processing strategy, imaging algorithm, etc. In actual operation, due to factors such as equipment aging, environmental changes, and human operation, the operating status of the security inspection equipment may deviate from the ideal standard, resulting in image brightness imbalance, increased noise, reduced resolution, insufficient contrast, and other problems, which seriously affect the security personnel's identification and judgment of potential threat items. Existing human body security inspection image optimization methods often rely on fixed-mode image post-processing algorithms. This approach lacks pertinence and is difficult to cope with complex and changeable equipment operating conditions; therefore, there is an urgent need to develop an intelligent and adaptive terahertz human body security inspection image optimization method. Summary of the invention
[0004] In order to solve the above technical problems, the present invention provides a terahertz human body security inspection image optimization method and system for improving the quality of security inspection images and security inspection efficiency, and ensuring the accuracy and efficiency of security inspection.
[0005] In a first aspect, the present invention provides a terahertz human body security inspection image optimization method, the method comprising:
[0006] Collect real-time operating status data of security inspection equipment and human body security inspection images to be optimized;
[0007] Obtain standard operating status data information of security inspection equipment;
[0008] Inputting the standard operating status data information of the security inspection equipment and the real-time operating status data information of the security inspection equipment into a pre-trained equipment degradation operating parameter identification model to obtain a security inspection equipment degradation operating parameter set;
[0009] Based on the historical operation status data of the security inspection equipment and the human body security inspection images corresponding to the historical operation status data of the security inspection equipment, a correlation feature matrix between the quality features of the human body security inspection images and the operation parameters of the security inspection equipment is constructed; the correlation feature matrix includes the correlation between each security inspection equipment operation parameter and each human body security inspection image quality feature;
[0010] Taking the security inspection equipment degradation operation parameter set as an index indicator, extracting the human body security inspection image quality feature set corresponding to the security inspection equipment degradation operation parameter set and having a correlation greater than a preset correlation threshold from the correlation feature matrix;
[0011] Input the human body security inspection image quality feature set into the pre-built security inspection image equipment interference optimization particle cluster to obtain the first security inspection image optimization particle corresponding to the security inspection equipment degradation operation parameter set;
[0012] The human body security inspection image to be optimized is optimized according to the first security inspection image optimization particles to obtain a human body security inspection optimized image.
[0013] Furthermore, the method for constructing the equipment degradation operation parameter identification model includes:
[0014] Collect security inspection equipment operating status data, including light source intensity, detector sensitivity, scanning speed, signal processing parameters, and imaging algorithm parameters under different working conditions;
[0015] Label each set of security equipment status data with the corresponding equipment degradation level to form the label data required for supervised learning;
[0016] Select the most discriminating features for the equipment degradation status from the original equipment status data, and standardize and normalize the selected features;
[0017] Selecting a deep learning model as the basic architecture of the equipment degradation operation parameter identification model, wherein the deep learning model includes logistic regression, support vector machine, decision tree, random forest, neural network, convolutional neural network, recurrent neural network and long short-term memory network;
[0018] Divide the processed features into training set, validation set and test set;
[0019] Use the training set and the corresponding degradation labels to update the parameters through the back propagation algorithm, optimize the ability of the equipment degradation operation parameter identification model to identify the equipment degradation status, and regularly evaluate the model performance on the validation set;
[0020] Evaluate the recognition accuracy, recall rate, and F1 score of the equipment degradation operating parameter recognition model on an independent test set;
[0021] The trained equipment degradation operating parameter identification model is encapsulated into an easy-to-call interface and integrated into the system.
[0022] Furthermore, the method for constructing the correlation feature matrix includes:
[0023] Collect the status data of human body security inspection equipment under historical operating status and the corresponding human body security inspection images generated;
[0024] Extract quality features from each historical security inspection image;
[0025] Measure the strength of the correlation between each security inspection equipment operating parameter and each human security inspection image quality feature;
[0026] For each set of equipment operating parameters and image quality features, calculate the correlation coefficient between them;
[0027] The calculated correlation coefficients are organized into a correlation feature matrix according to the correspondence between the equipment operating parameters and the image quality characteristics.
[0028] Furthermore, the method for obtaining the optimized human body security inspection image includes:
[0029] Optimize particles based on the first security inspection image and determine the image processing strategy;
[0030] According to the first security inspection image optimization particles, the parameters of the selected image processing strategy are adjusted and optimized;
[0031] Monitor the quality changes of processed images in real time and adjust the image processing strategy based on the feedback information;
[0032] Conduct quality assessment and verification on the optimized images after processing;
[0033] The final optimized human body security inspection image is output and recorded.
[0034] Furthermore, the real-time operation status data information includes light source intensity, detector sensitivity, scanning speed, signal processing strategy and imaging algorithm parameters.
[0035] Furthermore, the methods for acquiring the standard operating status data information of the security inspection equipment include equipment manuals and technical documents, equipment calibration and certification, and experience accumulation.
[0036] Furthermore, factors influencing the setting of the preset related thresholds include security requirements, image quality requirements, range of changes in equipment operating status, historical data analysis, and performance balance.
[0037] On the other hand, the present application also provides a terahertz human body security inspection image optimization system, the system comprising:
[0038] Real-time data acquisition module, used to collect real-time operating status data information of security inspection equipment and human body security inspection images to be optimized;
[0039] The equipment degradation operation parameter identification module is used to obtain the standard operation status data information of the security inspection equipment, input the standard operation status data information of the security inspection equipment and the real-time operation status data information of the security inspection equipment into the pre-trained equipment degradation operation parameter identification model, and obtain the security inspection equipment degradation operation parameter set;
[0040] A correlation feature matrix construction module is used to construct a correlation feature matrix between human body security inspection image quality features and security inspection equipment operation parameters based on the historical operation status data of the security inspection equipment and the human body security inspection images corresponding to the historical operation status data of the security inspection equipment; the correlation feature matrix includes the correlation between each security inspection equipment operation parameter and each human body security inspection image quality feature;
[0041] An image screening module is used to use the security inspection equipment degradation operation parameter set as an index indicator, and extract the human body security inspection image quality feature set corresponding to the security inspection equipment degradation operation parameter set and having a correlation greater than a preset correlation threshold from the correlation feature matrix;
[0042] An optimization particle generation module is used to input the human body security inspection image quality feature set into a pre-built security inspection image equipment interference optimization particle cluster to obtain a first security inspection image optimization particle corresponding to a security inspection equipment degradation operation parameter set;
[0043] The image optimization processing module is used to optimize the human body security inspection image to be optimized according to the first security inspection image optimization particles to obtain the human body security inspection optimized image.
[0044] In a third aspect, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any one of the above methods.
[0045] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.
[0046] Compared with the prior art, the present invention has the following beneficial effects: by collecting real-time operating status data information of security inspection equipment and obtaining standard operating status data information, and applying a pre-trained equipment degradation operating parameter recognition model, the method can monitor and identify the operating status changes of security inspection equipment in real time, and adjust the image optimization strategy according to the real-time situation to maintain the adaptability of the system;
[0047] By constructing a correlation feature matrix based on the historical operating status data of security inspection equipment, the relationship between the operating parameters of security inspection equipment and the quality characteristics of human body security inspection images can be comprehensively considered, thereby ensuring the comprehensiveness and accuracy of image quality optimization;
[0048] By using the pre-built security inspection image equipment to interfere with the optimized particle cluster, it is possible to intelligently generate optimized particles and optimize the image, thereby improving image quality and security inspection results;
[0049] By extracting image feature sets that are highly correlated with the equipment degradation operating parameter set, the images that need to be optimized can be efficiently screened out. At the same time, the use of pre-trained models and clusters ensures the reliability and stability of the method.
[0050] This method can flexibly adjust the optimization strategy according to the specific operating status of the security inspection equipment and image characteristics to meet the needs of different scenarios, and has strong pertinence and flexibility;
[0051] In summary, the terahertz human body security inspection image optimization method can effectively cope with the complex and changeable environment and equipment status changes in the security inspection field, improve the quality of security inspection images and security inspection efficiency, and ensure the accuracy and efficiency of security inspections. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] Figure 1 is a flow chart of the present invention;
[0053] Figure 2 It is a flow chart of a method for constructing a model for identifying equipment degradation operating parameters;
[0054] Figure 3 It is a structural diagram of the terahertz human body security inspection image optimization system. DETAILED DESCRIPTION
[0055] In the description of this application, those skilled in the art should know that this application can be implemented as a method, an apparatus, an electronic device, and a computer-readable storage medium. Therefore, this application can be specifically implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), a combination of hardware and software. In addition, in some embodiments, this application can also be implemented in the form of a computer program product in one or more computer-readable storage media, and the computer-readable storage medium contains computer program code.
[0056] The above-mentioned computer-readable storage medium may adopt any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories, flash memories, optical fibers, optical disc read-only memories, optical storage devices, magnetic storage devices, or any combination of the above. In the present application, computer-readable storage media can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.
[0057] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws.
[0058] The present application describes the provided methods, devices, and electronic devices through flowcharts and / or block diagrams.
[0059] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer-readable program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified by the boxes in the flowchart and / or block diagram.
[0060] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to work in a specific manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction device product including functions / operations specified in the blocks in the flowchart and / or block diagram.
[0061] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable data processing apparatus to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.
[0062] The present application is described below in conjunction with the drawings in the present application.
[0063] Embodiment 1: Figure 1 to Figure 2 As shown, the terahertz human body security inspection image optimization method of the present invention specifically includes the following steps:
[0064] S1. Collect real-time operating status data of security inspection equipment and human body security inspection images to be optimized;
[0065] Step S1 obtains comprehensive and accurate real-time operating status data of security inspection equipment and human security inspection images to be optimized, providing necessary data support for identifying equipment degradation parameters, building correlation feature matrix, generating optimized particles and performing image optimization processing in subsequent steps;
[0066] The real-time operation status data information covers the key parameters of the terahertz security inspection system during operation, including:
[0067] Light source intensity: record the output power of the terahertz light source currently in use to ensure that the light source is within a stable and appropriate range to generate high-quality terahertz waves; too strong or too weak a light source will cause the image to be overexposed or underexposed, affecting the presentation of image details;
[0068] Detector sensitivity, monitoring the responsiveness and signal-to-noise ratio of the terahertz detector under current working conditions; highly sensitive detectors can more effectively capture weak terahertz echo signals and reduce the impact of noise on image quality;
[0069] Scanning speed, which records the speed of the device when performing terahertz scanning on the human body; too fast or too slow scanning speed will cause motion blur of the image or too long acquisition time, affecting the clarity of the image and the overall imaging efficiency;
[0070] Signal processing strategy, collecting the pre-processing parameters such as signal filtering, gain control, background suppression, etc. used by the equipment; a reasonable signal processing strategy can effectively reduce environmental interference, improve the image signal-to-noise ratio, and enhance the contrast between hidden objects and the background;
[0071] Imaging algorithm parameters, recording the reconstruction algorithm used by the system and its related parameters, which directly affect the resolution, contrast and noise level of the image;
[0072] The human body security inspection images to be optimized are directly generated by the terahertz security inspection system during the security inspection task. These images are presented in two-dimensional and three-dimensional forms, and contain the terahertz reflection or transmission information of the inspected person's body surface and under the clothes; image acquisition follows the following principles:
[0073] Complete coverage ensures that the image covers the entire body of the inspected person, with no obvious missing areas, so as to comprehensively check for potential threatening items;
[0074] Appropriate resolution: The image resolution should meet the needs of identifying small and hidden objects, while taking into account data processing and transmission efficiency;
[0075] Standardized format: Image data should be stored in a unified standard format to facilitate subsequent image analysis and optimization processing;
[0076] Timestamp tagging adds an accurate timestamp to each image and associates it with the corresponding device operating status data to facilitate tracking of the device status when the image was generated.
[0077] In this step, by collecting the key parameter data of the security inspection equipment in real time, the operating status of the equipment can be fully understood, which is helpful to timely discover the abnormal operation or failure of the equipment; by obtaining the human body security inspection image to be optimized in real time, the body structure and the items carried by the inspected person can be intuitively displayed, providing basic data for subsequent image optimization processing, which is helpful to improve the clarity, contrast and resolution of the image; by analyzing the real-time operating status data information, the possible degradation parameters of the equipment can be identified, and the target parameters can be provided for the subsequent optimization processing; by associating the real-time operating status data with the human body security inspection image, it is helpful to establish a correlation feature matrix, so as to analyze the degree of correlation between the equipment operating parameters and the image quality, and provide guidance for the subsequent image optimization; by adding an accurate timestamp to each image and associating it with the corresponding equipment operating status data, it is possible to track and manage the equipment status when the image is generated, which is helpful for the subsequent data analysis and optimization processing;
[0078] In summary, the beneficial effects of step S1 include comprehensive understanding of equipment status, optimization of image quality, accurate identification of degradation parameters, establishment of correlation feature matrix, and data tracking and management, which provide necessary data support and guidance for subsequent terahertz human body security inspection image optimization.
[0079] S2. Obtaining standard operating status data information of security inspection equipment;
[0080] The standard operating status data information refers to the recommended range of various key operating parameters of the terahertz security inspection equipment under ideal conditions; the standard operating status data information of the security inspection equipment is consistent with the real-time operating status data information, including:
[0081] Light source intensity standards, based on the technical specifications provided by the equipment manufacturer and the most appropriate light source output power determined by laboratory tests, ensure that image quality is maintained while avoiding discomfort caused by excessive exposure to the human body;
[0082] Detector sensitivity standards, the best responsivity and signal-to-noise ratio measured when the equipment leaves the factory, are used to evaluate whether the current detector performance meets expectations and whether calibration or maintenance is required;
[0083] Scanning speed standard: the appropriate scanning rate is set according to the equipment design principle and the human body movement speed factor, which can not only ensure the integrity of image acquisition, but also avoid image blur or long acquisition time caused by too fast or too slow speed;
[0084] Standard signal processing strategy, the best signal filtering, gain control, background suppression and other preprocessing parameter combinations recommended by equipment manufacturers or verified by a large number of experiments, aiming to minimize noise, improve signal-to-noise ratio, while keeping image details clear;
[0085] Standard parameters of imaging algorithms, reconstruction algorithm parameters optimized for specific device models and application scenarios, aiming to ensure that images with high resolution, good contrast and low noise can be generated under different conditions;
[0086] The methods for obtaining the standard operating status data information of the security inspection equipment include:
[0087] Equipment manuals and technical documents: Obtain standard operating parameters from official materials such as user manuals, technical white papers, and performance reports provided by equipment manufacturers;
[0088] Equipment calibration and certification: During the equipment installation and commissioning phase, professional organizations conduct on-site calibration and performance testing to obtain the standard operating status data that the equipment actually achieves in the current environment;
[0089] Based on accumulated experience, combined with the actual performance of equipment in long-term operation, user feedback and the latest research results, the standard operating status data is continuously revised and improved to ensure that it keeps pace with the times and reflects best practices.
[0090] In this step, obtaining the standard operating status data information of the security inspection equipment can provide a reference range for actual operation; this data can not only serve as a benchmark for equipment performance, but also help operators understand the expected performance of the equipment under ideal conditions, so as to better operate and maintain the equipment; the standard operating status data information provides guidance for the parameter setting of the equipment; by understanding the optimal parameter configuration recommended by the equipment manufacturer, the operator can more accurately adjust the equipment to meet the requirements of security inspection while ensuring image quality and safety; the standardized setting of parameters such as light source intensity and detector sensitivity helps to ensure that the security inspection equipment will not cause discomfort or harm to the inspected person during operation; by following the standardized operating status data information, the safety of the security inspection process can be guaranteed to the greatest extent; standardized signal processing strategies and imaging algorithm parameters can improve the quality of the image; by using the verified optimal parameter combination, the noise in the image can be minimized and the clarity and details of the image can be maintained, thereby improving the accuracy and efficiency of security inspection; experience accumulation and continuous optimization enable the standard operating status data information to keep pace with the times; with the continuous development of technology and the accumulation of practical experience, the standard parameters can be revised and updated at any time to reflect best practices and continuously improve the performance and efficiency of security inspection equipment;
[0091] In summary, by obtaining the standard operating status data information of security inspection equipment, the operating efficiency and safety of the equipment can be improved, while the image quality can be optimized to provide more reliable support for security inspections.
[0092] S3, inputting the standard operation status data information of the security inspection equipment and the real-time operation status data information of the security inspection equipment into a pre-trained equipment degradation operation parameter identification model to obtain a set of security inspection equipment degradation operation parameters;
[0093] Step S3 uses the pre-trained equipment degradation operation parameter identification model to accurately identify and quantify the performance degradation parameters of the equipment under the current working conditions by comparing the real-time operation status of the security inspection equipment with the standard operation status data, providing a key basis for the subsequent targeted optimization of the security inspection image;
[0094] The method for constructing the equipment degradation operation parameter identification model includes:
[0095] S31. Collect a large amount of security inspection equipment operating status data, including light source intensity, detector sensitivity, scanning speed, signal processing parameters and imaging algorithm parameters under different working conditions;
[0096] S32, labeling the corresponding equipment degradation level for each set of security inspection equipment status data to form label data required for supervised learning;
[0097] S33, selecting the most discriminating features for the equipment degradation status from the original equipment status data, and standardizing and normalizing the selected features to ensure that different features are on the same scale;
[0098] S34, selecting a deep learning model as the infrastructure of the equipment degradation operation parameter identification model, wherein the deep learning model includes logistic regression, support vector machine, decision tree, random forest, neural network, convolutional neural network, recurrent neural network and long short-term memory network;
[0099] S35, dividing the processed features into a training set, a validation set and a test set;
[0100] S36. Use the training set and the corresponding degradation labels to update the parameters through the back propagation algorithm to optimize the ability of the equipment degradation operation parameter identification model to identify the equipment degradation state, and regularly evaluate the model performance on the validation set to prevent overfitting;
[0101] S37: Evaluate the recognition accuracy, recall rate and F1 score of the equipment degradation operation parameter recognition model on an independent test set to comprehensively evaluate the performance of the model;
[0102] S38. Encapsulate the trained equipment degradation operation parameter identification model into an easy-to-call interface to facilitate receiving real-time equipment status data and quickly output degradation parameter identification results in actual applications.
[0103] In this step, by comparing the real-time operating status of the security inspection equipment with the standard operating status data, the model can accurately identify and quantify the performance degradation parameters of the equipment under the current working conditions, which helps to timely discover the operating problems of the equipment and reduce the degradation of the security inspection image quality and the reduction of security inspection efficiency; by obtaining the degraded operating parameter set of the security inspection equipment, it can provide a key basis for the subsequent targeted optimization of the security inspection image; understanding the degradation state of the equipment can help adjust parameters such as imaging algorithms and signal processing strategies, thereby improving the quality of security inspection images and improving the accuracy and efficiency of security inspections; using a deep learning model as the basic architecture, it can effectively identify the degradation state of the equipment with high accuracy and generalization ability; it helps to improve the accuracy and stability of recognition, thereby more reliably guiding the subsequent image optimization work; the trained model is encapsulated into an easy-to-call interface, which can easily receive real-time equipment status data and quickly output degradation parameter identification results; this makes the model easily applicable to actual security inspection systems, providing timely support and guidance for security inspection work;
[0104] In summary, the equipment degradation operation parameter identification model constructed in step S3 can accurately and efficiently identify the degradation state of the equipment, provide important basis and support for subsequent security inspection image optimization, and help improve security inspection efficiency and accuracy.
[0105] S4. Based on the historical operation status data of the security inspection equipment and the human body security inspection images corresponding to the historical operation status data of the security inspection equipment, a correlation feature matrix between the human body security inspection image quality features and the security inspection equipment operation parameters is constructed; the correlation feature matrix includes the correlation between each security inspection equipment operation parameter and each human body security inspection image quality feature;
[0106] The constructed correlation feature matrix clearly shows the correlation between each security inspection equipment operating parameter and each human body security inspection image quality feature; the correlation feature matrix is the basis for determining key image quality features, generating optimized particles and performing image optimization in the subsequent steps; the correlation feature matrix reflects the inherent connection between the equipment operating status and the security inspection image quality, which helps to quickly locate the image quality features that are highly correlated with the equipment degraded operating parameter set, thereby achieving targeted image optimization;
[0107] The method for constructing the correlation feature matrix includes:
[0108] S41. Collect the status data of human body security inspection equipment under historical operating status and the corresponding human body security inspection images generated; ensure that the data covers various states of normal operation, slight degradation and severe degradation of the equipment, so as to reflect the impact of changes in equipment parameters on image quality;
[0109] S42, extracting quality features of each historical security inspection image, including brightness distribution, contrast, signal-to-noise ratio, resolution, edge clarity, texture features, grayscale histogram, and frequency domain features;
[0110] S43, selecting a correlation measurement method to measure the strength of association between each security inspection equipment operating parameter and each human body security inspection image quality feature; the correlation measurement method includes Pearson correlation coefficient, Spearman rank correlation coefficient, Kendall tau coefficient and distance correlation;
[0111] S44, for each set of device operating parameters and image quality features, using the selected correlation measurement method, calculate the correlation coefficient between them; for each device parameter, obtain a vector associated with all image quality features; conversely, for each image quality feature, also obtain a vector associated with all device parameters;
[0112] S45. Organize the calculated correlation coefficients into a two-dimensional matrix, namely, a correlation feature matrix, according to the corresponding relationship between the device operating parameters and the image quality characteristics.
[0113] In this step, by constructing a correlation feature matrix, the correlation between each security inspection equipment operating parameter and each human body security inspection image quality feature is systematically quantified, forming a structured knowledge base that intuitively displays the complex relationship network between the equipment operating status and the security inspection image quality; the correlation feature matrix provides a basis for determining the key image quality features that are highly correlated with the equipment degradation operating parameters in the subsequent steps; when facing a specific equipment degradation condition, the image features that are most affected can be quickly located by querying the matrix, ensuring that the optimization work focuses on the most critical problem points and improving the pertinence and efficiency of the optimization; based on the correlation feature matrix, the optimization strategy can be dynamically adjusted according to the real-time equipment operating status, and optimized particles adapted to the current equipment status can be generated to realize the intelligent and adaptive security inspection image. Optimization; Enhanced the adaptability of the system, enabling it to effectively cope with equipment status fluctuations caused by factors such as equipment aging, environmental changes, and human operations, and ensure the stability and accuracy of security inspection image quality; Through in-depth mining and analysis of historical data, the constructed correlation feature matrix not only serves the immediate image optimization task, but also provides a basis for long-term knowledge accumulation and technology iteration; With the continuous accumulation and updating of data, the matrix can be continuously improved, further improving the understanding and modeling capabilities of equipment image relationships, and promoting the development of security inspection image optimization technology; The correlation feature matrix provides data-driven decision support for security inspection image optimization, reduces reliance on subjective experience, and improves the objectivity and scientificity of decision-making; It is of great significance to improve the overall performance of the security inspection system and ensure the accuracy and efficiency of security inspection work;
[0114] In summary, constructing a correlation feature matrix is a key link in the terahertz human body security inspection image optimization method, which realizes the deep correlation analysis between the equipment operation status and the security inspection image quality, provides strong data support and decision-making basis for subsequent targeted optimization, and promotes the precise control and efficient improvement of the security inspection image quality.
[0115] S5. Using the security inspection equipment degradation operation parameter set as an index indicator, extracting a human body security inspection image quality feature set corresponding to the security inspection equipment degradation operation parameter set and having a correlation greater than a preset correlation threshold from the correlation feature matrix;
[0116] The goal of step S5 is to select the image quality feature set that is highly correlated with the identified security inspection equipment degradation operating parameter set and has a significant impact on the image quality from the correlation feature matrix; this ensures the pertinence of subsequent image optimization work, so that the optimization measures can directly act on the image characteristics that are most significantly affected by equipment degradation, thereby more effectively improving image quality and enhancing security inspection accuracy;
[0117] Based on the preset correlation threshold, the image quality features with significant correlation with the equipment degradation operation parameter set are screened out;
[0118] For each parameter in the degraded operating parameter set, find the corresponding row or column in the correlation feature matrix to obtain the correlation coefficient between the parameter and all image quality features;
[0119] Screen out image quality features with correlation coefficients greater than a preset correlation threshold. These features have a strong correlation with the degradation parameters of the current device, that is, changes in device parameters have a greater impact on them.
[0120] All the screened image quality features with correlations higher than the threshold are merged to form a "human body security inspection image quality feature set" that is closely related to the current equipment degradation operating parameter set; the set represents the image attributes that need the most attention and optimization under the current equipment status;
[0121] The factors affecting the setting of the preset relevant thresholds include:
[0122] Security requirements. The setting of preset relevant thresholds should take security requirements into consideration to ensure that the identification and judgment of potential threat items are accurate and reliable enough. If the relevant threshold is set too high, it will lead to missed detection or misjudgment, while if it is set too low, it will introduce too many interference factors, affecting the efficiency and accuracy of security inspection.
[0123] Image quality requirements. The setting of preset relevant thresholds should also take into account the requirements for image quality. If the relevant threshold is set too low, too many images will be excluded from the optimization range, affecting the effect of image optimization. If it is set too high, too much noise or interference images will be included, affecting the quality of the optimized image.
[0124] The range of changes in the equipment's operating status. Considering that the operating status of security inspection equipment is affected by a variety of factors, the setting of preset relevant thresholds needs to consider the range of changes in the equipment's operating status. If the relevant thresholds are set outside the range of changes in the equipment's operating status, it will lead to the inability to effectively identify and adapt to changes in the equipment's operating status, affecting the effect of image optimization;
[0125] Historical data analysis: by analyzing the historical operating status data of security inspection equipment and the corresponding human body security inspection images, the actual correlation between different operating parameters and image quality can be understood; the setting of preset relevant thresholds can refer to the analysis results of historical data to ensure its good applicability and effectiveness in practical applications;
[0126] Performance balance: The setting of preset relevant thresholds requires a trade-off and balance between factors such as security, image quality requirements, and the range of changes in the equipment operating status, in order to achieve the best effect of optimizing security inspection images.
[0127] In this step, by extracting the image quality feature set that is highly correlated with the security inspection equipment degradation operating parameter set from the correlation feature matrix, the pertinence of the image optimization work is ensured; it can directly act on the image characteristics that are most significantly affected by equipment degradation, thereby more effectively improving image quality and enhancing security inspection accuracy; by optimizing the image characteristics that are most significantly affected by equipment degradation, it can effectively reduce the image quality degradation caused by changes in the equipment operation state; it helps to improve the clarity, contrast and resolution of security inspection images, thereby improving the accuracy of security inspectors in identifying and judging potential threat items; by merging image quality features that are highly correlated with the equipment degradation operating parameter set, a closely related "human body security inspection image quality feature set" is formed, which can be optimized in a targeted manner to significantly improve image quality; when setting the preset relevant thresholds, the safety requirements, image quality requirements, equipment operation state change range, historical data analysis and performance balance are comprehensively considered to ensure that the preset relevant thresholds have good applicability and effectiveness in practical applications; it helps to balance the relationship between security inspection accuracy and efficiency, as well as the trade-off between the effect and cost of image quality optimization;
[0128] In summary, the beneficial effects of step S5 are reflected in improving security inspection accuracy, optimizing image quality, improving targeted optimization effects, and comprehensively considering various factors, which helps to improve the effect and practicality of terahertz human body security inspection image optimization.
[0129] S6. Input the human body security inspection image quality feature set into the pre-built security inspection image equipment interference optimization particle cluster to obtain the first security inspection image optimization particle corresponding to the security inspection equipment degradation operation parameter set;
[0130] Step S6 uses the pre-built security inspection image equipment interference optimization particle cluster to generate a highly targeted optimization strategy that adapts to changes in equipment status based on key image quality features under equipment degradation. This overcomes the limitations of traditional fixed-mode image post-processing algorithms and achieves intelligent response to complex and changeable equipment status, meeting the needs of the security inspection image optimization field for accurate and efficient optimization technology.
[0131] The image quality feature set is used as input and sent to the security inspection image equipment interference optimization particle cluster;
[0132] The particle cluster for optimizing interference of security image equipment starts the internal optimization algorithm based on these characteristics, and through iterative calculation and information sharing, seeks the optimization strategy that can improve the image quality to the greatest extent. This process involves adjusting the parameters of image processing technologies such as image enhancement, denoising, sharpening, equalization, and resolution improvement to correct image defects caused by equipment degradation.
[0133] The security inspection image equipment interference optimization particle cluster generates a first security inspection image optimization particle that can optimize the image according to the current equipment degradation state;
[0134] The method for constructing the security inspection image equipment interference optimization particle cluster includes:
[0135] Define the specific image quality improvement goals to be achieved by the optimization cluster, and construct objective evaluation indicators suitable for terahertz human body security inspection images based on the optimization goals. The evaluation indicators include signal-to-noise ratio, contrast entropy, structural similarity index and peak signal-to-noise ratio, which are used to quantitatively evaluate the optimization effect;
[0136] Select an optimization algorithm, which includes particle swarm optimization, genetic algorithm, simulated annealing, and deep reinforcement learning algorithm. When selecting, factors such as the adaptability, convergence speed, and ease of implementation of the algorithm should be considered to ensure that it is suitable for terahertz human body security inspection image optimization scenarios;
[0137] Determine how particles represent image optimization strategies, with each dimension of the particle representing a set of parameter values; define how to convert particle encoding into actual image processing operations, design a decoding function, and map particle vectors to a specific set of image processing parameters for application to the image to be optimized;
[0138] Generate an initial particle swarm based on the dimension of the optimization problem, where each particle represents a possible optimization strategy;
[0139] Set appropriate control parameters to affect the particle search range and convergence speed, and set termination criteria to prevent the algorithm from falling into local over-iteration;
[0140] The evaluation index is embedded in the optimization algorithm to form a fitness function. In each iteration, the optimization effect of each particle is evaluated according to the fitness function to guide the particle update.
[0141] Use a representative terahertz human body security inspection image sample set and the corresponding standard optimization results to train and verify the constructed optimization cluster; adjust the algorithm parameters, optimization objectives or evaluation indicators to optimize the cluster performance and ensure that it can effectively improve the image quality under various equipment degradation conditions.
[0142] In this step, by generating a highly targeted optimization strategy based on the key image quality features under the equipment degradation state, the optimized particle cluster can better adapt to the equipment state changes; it is more flexible and adaptable than the traditional fixed-mode image post-processing algorithm; through the iterative calculation and information sharing of the internal optimization algorithm, the optimized particle cluster can intelligently find the optimization strategy that can maximize the improvement of image quality; it can better adapt to the complex and changeable equipment state, thereby improving the quality and accuracy of security inspection images; the optimized particle cluster optimizes the image according to the equipment degradation state, and corrects the image defects caused by equipment degradation through parameter adjustment of image processing technologies such as image enhancement, denoising, sharpening, equalization, and resolution improvement; it can improve the quality of security inspection images, thereby improving security inspection efficiency and accuracy; by defining specific image quality improvement goals and objective evaluation indicators applicable to terahertz human body security inspection images, the optimization effect can be quantitatively evaluated; it helps to more accurately evaluate the effect of the optimization algorithm and guide the subsequent optimization process; considering the characteristics of the terahertz human body security inspection image optimization scenario, a variety of optimization algorithms are selected, and the most suitable optimization algorithm can be selected according to the actual situation to achieve better optimization effect;
[0143] In summary, the construction method of the security inspection image equipment interference optimization particle cluster in S6 can effectively improve the quality of terahertz human body security inspection images, thereby improving security inspection efficiency and accuracy, and meeting the needs of the security inspection image optimization field for precise and efficient optimization technology.
[0144] S7, optimizing the human body security inspection image to be optimized according to the first security inspection image optimization particles to obtain a human body security inspection optimized image;
[0145] In step S7, based on the image processing strategy indicated by the first security inspection image optimization particle, the optimized terahertz human body security inspection image is optimized to help security inspectors identify potential threat items more accurately and efficiently, thereby improving the overall efficiency of the security inspection system.
[0146] The method for obtaining the human body security inspection optimized image comprises:
[0147] S71. Determine an image processing strategy applicable to the current security inspection image according to the optimization strategy provided by the first security inspection image optimization particle;
[0148] S72, adjusting and optimizing the parameters of the selected image processing strategy according to the parameter values contained in the first security inspection image optimization particle; adjusting the parameters according to specific image features and optimization goals to ensure that the image quality is improved to the greatest extent;
[0149] S73, in the process of processing the human body security inspection image to be optimized, monitoring the quality change of the processed image in real time, and adjusting the image processing strategy according to the feedback information;
[0150] S74. Perform quality assessment and verification on the optimized image after processing to ensure that the improvement in image quality meets expectations and can meet security inspection requirements;
[0151] S75: Output and record the finally optimized human body security inspection image for use in subsequent security inspection tasks.
[0152] In this step, by performing targeted processing and parameter optimization on the optimized terahertz human body security inspection image according to the optimization strategy indicated by the first security inspection image optimization particle, the clarity, contrast and detail display of the image can be significantly improved, thereby enhancing the security inspectors' ability to identify potential threat items; the optimized human body security inspection image can help security inspectors detect and judge potential threat items more accurately and quickly, reduce missed detections and false detections, and improve the overall efficiency and work effectiveness of the security inspection system; according to the optimization strategy provided by the first security inspection image optimization particle, the image processing strategy is adjusted in real time to cope with different situations The security inspection image features and optimization requirements of the system ensure the adaptability and flexibility of the processing method; during the processing, the quality changes of the processed images are monitored in real time, and the processing strategy is adjusted according to the feedback information, which can ensure the real-time processing and the accuracy of the effect; the quality of the optimized images is evaluated and verified to ensure that the improvement of image quality meets expectations and meets the security inspection requirements, which enhances the credibility and practicality of the optimization method; the final optimized human body security inspection images are output and recorded, which provides reliable image resources for subsequent security inspection tasks and provides a basis for the traceability and optimization of the processing process;
[0153] In summary, the beneficial effect of step S7 is that it improves the quality and recognition accuracy of security inspection images, enhances the overall effectiveness and work efficiency of the security inspection system, and also ensures the flexibility and credibility of the processing method.
[0154] Embodiment 2: Figure 3 As shown, the terahertz human body security inspection image optimization system of the present invention specifically includes the following modules:
[0155] Real-time data acquisition module, used to collect real-time operating status data information of security inspection equipment and human body security inspection images to be optimized;
[0156] The equipment degradation operation parameter identification module is used to obtain the standard operation status data information of the security inspection equipment, input the standard operation status data information of the security inspection equipment and the real-time operation status data information of the security inspection equipment into the pre-trained equipment degradation operation parameter identification model, and obtain the security inspection equipment degradation operation parameter set;
[0157] A correlation feature matrix construction module is used to construct a correlation feature matrix between human body security inspection image quality features and security inspection equipment operation parameters based on the historical operation status data of the security inspection equipment and the human body security inspection images corresponding to the historical operation status data of the security inspection equipment; the correlation feature matrix includes the correlation between each security inspection equipment operation parameter and each human body security inspection image quality feature;
[0158] An image screening module is used to use the security inspection equipment degradation operation parameter set as an index indicator, and extract the human body security inspection image quality feature set corresponding to the security inspection equipment degradation operation parameter set and having a correlation greater than a preset correlation threshold from the correlation feature matrix;
[0159] An optimization particle generation module is used to input the human body security inspection image quality feature set into a pre-built security inspection image equipment interference optimization particle cluster to obtain a first security inspection image optimization particle corresponding to a security inspection equipment degradation operation parameter set;
[0160] The image optimization processing module is used to optimize the human body security inspection image to be optimized according to the first security inspection image optimization particles to obtain the human body security inspection optimized image.
[0161] The system can obtain the operating status data of security inspection equipment in a timely manner through the real-time data acquisition module and the equipment degradation operating parameter identification module, and automatically adjust the optimization strategy based on this data to maintain the system's adaptability to changes in equipment status;
[0162] By using the correlation feature matrix construction module and the image screening module, it is possible to accurately screen out high-quality human security inspection images that are related to the equipment degradation operating parameters, thereby improving the accuracy and efficiency of security inspection images;
[0163] With the help of the optimized particle generation module and the image optimization processing module, it is possible to generate optimized particles according to actual conditions and perform intelligent processing on images, effectively improving image quality and security inspection effects;
[0164] The system considers multiple factors such as equipment operation status, image quality characteristics, relevance, etc., and conducts comprehensive analysis and processing, making the optimization method more comprehensive and integrated, and able to cope with complex and changing actual scene requirements;
[0165] The system uses pre-trained models and clusters, has high reliability and stability, and can continuously and effectively optimize security inspection images to ensure the smooth completion of security inspection tasks;
[0166] Combining these advantages, the terahertz human body security inspection image optimization system can effectively cope with challenges such as changes in equipment status in real security inspection environments, improve security inspection image quality and security inspection efficiency, and ensure the accuracy and efficiency of security inspections.
[0167] The various variations and specific embodiments of the terahertz human body security inspection image optimization method in the aforementioned embodiment 1 are also applicable to the terahertz human body security inspection image optimization system of this embodiment. Through the aforementioned detailed description of the terahertz human body security inspection image optimization method, those skilled in the art can clearly know the implementation method of the terahertz human body security inspection image optimization system of this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.
[0168] In addition, the present application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned method for controlling output data is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.
[0169] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A terahertz human body security inspection image optimization method, characterized in that: The method comprises: Collect real-time operating status data of security inspection equipment and human body security inspection images to be optimized; Obtain standard operating status data information of security inspection equipment; Inputting the standard operating status data information of the security inspection equipment and the real-time operating status data information of the security inspection equipment into a pre-trained equipment degradation operating parameter identification model to obtain a security inspection equipment degradation operating parameter set; Based on the historical operation status data of the security inspection equipment and the human body security inspection images corresponding to the historical operation status data of the security inspection equipment, a correlation feature matrix between the quality features of the human body security inspection images and the operation parameters of the security inspection equipment is constructed; the correlation feature matrix includes the correlation between each security inspection equipment operation parameter and each human body security inspection image quality feature; Taking the security inspection equipment degradation operation parameter set as an index indicator, extracting the human body security inspection image quality feature set corresponding to the security inspection equipment degradation operation parameter set and having a correlation greater than a preset correlation threshold from the correlation feature matrix; Input the human body security inspection image quality feature set into the pre-built security inspection image equipment interference optimization particle cluster to obtain the first security inspection image optimization particle corresponding to the security inspection equipment degradation operation parameter set; The human body security inspection image to be optimized is optimized according to the first security inspection image optimization particles to obtain a human body security inspection optimized image.
2. The terahertz human body security inspection image optimization method according to claim 1, characterized in that: The method for constructing the equipment degradation operation parameter identification model includes: Collect security inspection equipment operating status data, including light source intensity, detector sensitivity, scanning speed, signal processing parameters, and imaging algorithm parameters under different working conditions; Label each set of security equipment status data with the corresponding equipment degradation level to form the label data required for supervised learning; Select the most discriminating features for the equipment degradation status from the original equipment status data, and standardize and normalize the selected features; Selecting a deep learning model as the basic architecture of the equipment degradation operation parameter identification model, wherein the deep learning model includes logistic regression, support vector machine, decision tree, random forest, neural network, convolutional neural network, recurrent neural network and long short-term memory network; Divide the processed features into training set, validation set and test set; Use the training set and the corresponding degradation labels to update the parameters through the back propagation algorithm, optimize the ability of the equipment degradation operation parameter identification model to identify the equipment degradation status, and regularly evaluate the model performance on the validation set; Evaluate the recognition accuracy, recall rate, and F1 score of the equipment degradation operating parameter recognition model on an independent test set; The trained equipment degradation operating parameter identification model is encapsulated into an easy-to-call interface and integrated into the system.
3. The terahertz human body security inspection image optimization method according to claim 1, characterized in that: The method for constructing the correlation feature matrix includes: Collect the status data of human body security inspection equipment under historical operating status and the corresponding human body security inspection images generated; Extract quality features from each historical security inspection image; Measure the strength of association between each security inspection equipment operating parameter and each human security inspection image quality feature; For each set of equipment operating parameters and image quality features, calculate the correlation coefficient between them; The calculated correlation coefficients are organized into a correlation feature matrix according to the correspondence between the equipment operating parameters and the image quality characteristics.
4. The terahertz human body security inspection image optimization method according to claim 1, characterized in that: The method for obtaining the human body security inspection optimized image comprises: Optimize particles based on the first security inspection image and determine the image processing strategy; According to the first security inspection image optimization particles, the parameters of the selected image processing strategy are adjusted and optimized; Monitor the quality changes of processed images in real time and adjust the image processing strategy based on the feedback information; Conduct quality assessment and verification on the optimized images after processing; The final optimized human body security inspection image is output and recorded.
5. The terahertz human body security inspection image optimization method according to claim 1, characterized in that: The real-time operation status data information includes light source intensity, detector sensitivity, scanning speed, signal processing strategy and imaging algorithm parameters.
6. The terahertz human body security inspection image optimization method according to claim 1, characterized in that: The methods for obtaining the standard operating status data information of the security inspection equipment include equipment manuals and technical documents, equipment calibration and certification, and experience accumulation.
7. The terahertz human body security inspection image optimization method according to claim 1, characterized in that: The factors affecting the setting of the preset relevant thresholds include security requirements, image quality requirements, range of changes in equipment operating status, historical data analysis, and performance balance.
8. A terahertz human body security inspection image optimization system, characterized in that: The system comprises: Real-time data acquisition module, used to collect real-time operating status data information of security inspection equipment and human body security inspection images to be optimized; The equipment degradation operation parameter identification module is used to obtain the standard operation status data information of the security inspection equipment, input the standard operation status data information of the security inspection equipment and the real-time operation status data information of the security inspection equipment into the pre-trained equipment degradation operation parameter identification model, and obtain the security inspection equipment degradation operation parameter set; A correlation feature matrix construction module is used to construct a correlation feature matrix between human body security inspection image quality features and security inspection equipment operation parameters based on the historical operation status data of the security inspection equipment and the human body security inspection images corresponding to the historical operation status data of the security inspection equipment; the correlation feature matrix includes the correlation between each security inspection equipment operation parameter and each human body security inspection image quality feature; An image screening module is used to use the security inspection equipment degradation operation parameter set as an index indicator, and extract the human body security inspection image quality feature set corresponding to the security inspection equipment degradation operation parameter set and having a correlation greater than a preset correlation threshold from the correlation feature matrix; An optimization particle generation module is used to input the human body security inspection image quality feature set into a pre-built security inspection image equipment interference optimization particle cluster to obtain a first security inspection image optimization particle corresponding to a security inspection equipment degradation operation parameter set; The image optimization processing module is used to optimize the human body security inspection image to be optimized according to the first security inspection image optimization particles to obtain the human body security inspection optimized image.
9. A terahertz human body security inspection image optimization electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.
