Customs entry and exit personnel identity rapid verification and trajectory tracking system

Abstract

The invention relates to the field of identity verification of customs entry and exit personnel, and discloses a rapid identity verification and trajectory tracking system for customs entry and exit personnel, and the system comprises a wearable device module which is used for collecting brain wave signals of the entry and exit personnel; the wireless transmission module is used for receiving the brain wave signals in real time and transmitting the brain wave signals to the central processing system; the signal de-noising module is used for de-noising the received brain wave signal; the feature extraction module is used for extracting personalized features in the EEG signals; the classification module is used for classifying the extracted EEG signal features; and the anomaly detection module is used for detecting abnormal behaviors of entry and exit personnel. By combining EEG signals with RFID and multiple GPS technologies, the efficient and accurate entry and exit personnel identity verification and behavior tracking effect is achieved, individual brain wave characteristics can be recognized through real-time EEG signal analysis, and the risk of disguising and falsely using the identity is effectively avoided.

Description

Technical Field

[0001] The present invention relates to the field of identity verification for customs inbound and outbound personnel, and specifically to a rapid identity verification and trajectory tracking system for customs inbound and outbound personnel. Background Art

[0002] In the context of modern internationalization and globalization, identity verification and behavior monitoring of customs inbound and outbound personnel are particularly important. As the first line of defense of a country's border, customs needs to ensure the accuracy of the identity information of inbound and outbound personnel and at the same time guard against potential security threats. However, with the increase in traffic flow, traditional identity verification methods, such as passport inspection, fingerprint recognition, face recognition, etc., have become inadequate. In certain specific scenarios, such as dense personnel flow, complex environment, and insufficient light.

[0003] Currently, the identity verification of most customs inbound and outbound personnel relies on traditional biometric technologies, such as fingerprint recognition, face recognition, etc. However, these technologies are often limited by external conditions, such as lighting, angle, equipment failure, etc., resulting in a decrease in accuracy and reliability. Face recognition technology often performs poorly in low-light environments, while fingerprint recognition technology may be affected by factors such as wet or damaged fingers, resulting in incorrect recognition. The limitations of these technologies make the application effect of traditional methods in high-density crowds or dynamic scenarios greatly reduced, increasing the recognition error, and may even be breached by forged identities, bringing potential risks to security monitoring.

[0004] Existing trajectory tracking technologies mostly rely on single positioning means such as RFID or GPS. Although RFID technology can track personnel entering specific areas in real time, its limitation is that personnel need to contact specific reading devices, and there are certain limitations in positioning accuracy and range. Although GPS technology can provide a wider coverage, in some indoor or underground environments, the signal is easily lost or interfered, thus affecting the tracking effect. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the present invention provides a rapid identity verification and trajectory tracking system for customs inbound and outbound personnel, which solves the problems of insufficient accuracy of traditional identity verification technology in complex environments and the behavior monitoring and tracking accuracy problems brought by single positioning technology.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A rapid identity verification and trajectory tracking system for customs inbound and outbound personnel, comprising:

[0007] A wearable device module for collecting the electroencephalogram signals of inbound and outbound personnel;

[0008] A wireless transmission module for receiving and transmitting the electroencephalogram signals to the central processing system in real time;

[0009] A signal denoising module for denoising the received electroencephalogram (EEG) signals.

[0010] A feature extraction module for extracting personalized features from EEG signals.

[0011] A classification module for classifying the extracted EEG signal features.

[0012] An anomaly detection module for detecting abnormal behaviors of inbound and outbound passengers.

[0013] A trajectory tracking module for tracking the trajectories of inbound and outbound passengers.

[0014] Preferably, the system includes a data analysis module for aggregating and analyzing the data of each module and generating a decision report.

[0015] A signal transmission module for ensuring the real-time transmission and processing of EEG signals and behavior data, adopting low-latency and high-bandwidth communication technologies.

[0016] Preferably, the denoising process of the signal denoising module includes performing multi-resolution time-frequency analysis on the signal using wavelet transform to remove high-frequency noise, and smoothing the signal through the heat conduction equation to eliminate low-frequency noise.

[0017] Preferably, the feature extraction module includes a Hurst exponent calculation unit based on fractal geometry. The Hurst exponent evaluates the self-similarity of the signal by calculating the ratio of the range to the standard deviation of the signal, thereby extracting the personalized EEG features of inbound and outbound passengers. The feature extraction module includes a Shannon entropy calculation unit based on information theory. Shannon entropy is used to quantify the complexity of EEG signals, and further identify the individual behavior patterns of inbound and outbound passengers.

[0018] Preferably, the classification module classifies the EEG signal features through the support vector machine (SVM) algorithm. SVM uses a Gaussian kernel function to perform non-linear mapping on EEG signals and classifies the data according to the maximum margin principle.

[0019] Preferably, the anomaly detection module compares the time series similarity between the real-time EEG signal and the preset normal behavior pattern through the dynamic time warping algorithm. If the similarity is lower than the preset threshold, it is determined as an abnormal behavior.

[0020] Preferably, the trajectory tracking module real-time tracks the positions of inbound and outbound passengers through RFID tags and GPS positioning systems, and further analyzes the behavior patterns through EEG signals to determine whether the person stays in an unauthorized area.

[0021] Preferably, the data analysis module processes and comprehensively analyzes the signal, behavior and trajectory data in real time based on the big data analysis platform, generates a decision report, and helps customs management personnel to respond in real time.

[0022] Preferably, the wavelet transform in the signal denoising module performs noise suppression through multi-scale analysis and adaptive threshold selection.

[0023] Preferably, the trajectory tracking module can update the trajectory of personnel in real time by integrating RFID, GPS and EEG signals, and provide early warning for abnormal activities in combination with behavior pattern analysis.

[0024] The present invention provides a system for rapid identity verification and trajectory tracking of customs entry and exit personnel. It has the following beneficial effects:

[0025] 1. The present invention combines EEG signals with RFID and multiple GPS technologies to achieve efficient and accurate identity verification and behavior tracking of people entering and leaving the country. Compared with the traditional biometric identification methods in the prior art, the system can identify the brain wave characteristics of individuals through real-time EEG signal analysis, effectively avoiding the risks of disguise and identity fraud, and solving the problem of low recognition accuracy of traditional methods in complex environments.

[0026] 2. The present invention optimizes the quality of EEG signals and removes high-frequency noise and low-frequency interference through a hybrid denoising scheme of wavelet transform and partial differential equations. Compared with the traditional single denoising technology, the present invention provides a more efficient denoising effect, especially in complex noise environments, and can better retain effective signals, thereby improving the accuracy of subsequent classification and behavior analysis.

[0027] 3. The present invention combines the support vector machine algorithm to classify EEG signals, and uses the Gaussian kernel function to effectively process nonlinear features, achieving accurate identity recognition and behavior classification. Compared with other classification methods in the prior art, SVM not only has stronger robustness, but also can find the optimal decision boundary in the high-dimensional feature space, thereby achieving higher classification accuracy.

[0028] 4. The present invention combines RFID, GPS and EEG signals in the trajectory tracking module to achieve real-time, multi-dimensional personnel positioning and behavior monitoring. Compared with the traditional positioning system that only uses RFID or GPS, the present invention not only provides more accurate location information through multi-source data fusion, but also can analyze the behavior patterns of personnel in real time, detect potential abnormal behaviors in time, and enhance the security and monitoring capabilities of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a system framework diagram of the present invention. DETAILED DESCRIPTION

[0030] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0031] Please refer to the attached Figure 1 , the embodiment of the present invention provides a rapid identity verification and trajectory tracking system for customs entry and exit personnel, including:

[0032] A wearable device module for collecting the electroencephalogram (EEG) signals of entry and exit personnel;

[0033] A wireless transmission module for receiving and transmitting the EEG signals to the central processing system in real time;

[0034] A signal denoising module for denoising the received EEG signals;

[0035] A feature extraction module for extracting personalized features in the EEG signals;

[0036] A classification module for classifying the extracted EEG signal features;

[0037] An anomaly detection module for detecting abnormal behaviors of entry and exit personnel;

[0038] A trajectory tracking module for tracking the trajectories of entry and exit personnel.

[0039] Wearable device module

[0040] The system collects the EEG signals of entry and exit personnel through wearable devices. These devices can be smart glasses, head-mounted devices, bracelets, etc., integrated with EEG sensors. The collection of EEG signals needs to ensure a high sampling rate (for example, 250 Hz to 1000 Hz) to capture sufficiently delicate EEG activity information. The signal collection process does not rely on any contact method. Wearing the device can eliminate the need for traditional contact-based verification methods, which makes the user experience more convenient.

[0041] These EEG signals can be transmitted to the background system in real time, ensuring the rapidity and real-time nature of data transmission. The system uses a low-power wireless transmission module (such as Bluetooth or Wi-Fi) to ensure that data can be stably and quickly transmitted to the central processing system.

[0042] Signal denoising module

[0043] After the brain wave signals are collected, they first enter the signal denoising module. EEG signals are usually affected by interference such as eye movement, electromyogram, and environmental noise. These noises will affect the effectiveness and accuracy of the signals. Therefore, denoising processing is crucial.

[0044] The system uses a hybrid optimization algorithm of wavelet transform and partial differential equation (PDE) for denoising. Wavelet transform extracts the local features of the signal in the time domain and frequency domain through multi-resolution analysis, and is especially good at dealing with instantaneous noise in EEG signals. The partial differential equation (PDE) is used to smooth the EEG signal, simulating the expansion and attenuation of the signal through the heat conduction equation to eliminate high-frequency noise and low-frequency interference.

[0045] Feature extraction module

[0046] EEG signals are highly personalized. Therefore, effective feature extraction is one of the cores of the system. The system extracts signal features through fractal geometry and information theory methods. By calculating the Hurst exponent, the system can evaluate the self-similarity of the signal. This is because each person's brain wave signal has its unique structure and pattern. The Hurst exponent can quantify its self-similarity by analyzing the range and standard deviation of the signal.

[0047] In addition, the Shannon entropy calculation unit is used to quantify the complexity of the EEG signal. The higher the entropy, the greater the complexity of the signal. Through the calculation of the entropy value, the system can further extract the unique behavior patterns of the inbound and outbound personnel.

[0048] Classification module

[0049] The EEG signals after feature extraction are input into the classification module. The classification module adopts the support vector machine (SVM) algorithm. This algorithm is a powerful machine learning method, especially suitable for the classification of high-dimensional data. Through the Gaussian kernel function, SVM can map the original EEG features to a high-dimensional space and find the optimal hyperplane for classification.

[0050] The advantage of SVM is that it can handle non-linear classification problems and has high robustness, which is particularly effective in the recognition of personnel behavior patterns. In this way, the system can accurately and quickly verify the identities of inbound and outbound personnel.

[0051] Anomaly detection module

[0052] The anomaly detection module performs real-time behavior monitoring through the dynamic time warping (DTW) algorithm. DTW is an algorithm used to calculate the similarity between two time series. The system judges whether there is abnormal behavior by comparing the time series difference between the real-time EEG signal and the preset normal behavior pattern.

[0053] If the matching degree of the real-time signal with the normal mode is lower than the preset threshold, the system will determine this behavior as abnormal. For example, if a person attempts to disguise their identity or exhibits unconventional behavior during customs inspection, the DTW algorithm can effectively detect such deviations and issue an alarm for further manual verification.

[0054] Trajectory Tracking Module

[0055] This module real-time tracks the behavior trajectories of inbound and outbound passengers by integrating RFID tags and GPS positioning systems. The RFID tags can be embedded in the passports, luggage, or wearable devices of inbound and outbound passengers. By reading the RFID tags, the system can quickly obtain the location information of the passengers. The GPS module is used to determine the specific location of the passengers in real-time over a large area.

[0056] The combination of these two technologies enables continuous monitoring of inbound and outbound passengers globally, especially when passengers cross multiple checkpoints, ensuring seamless docking of all trajectory information.

[0057] Data Analysis Module

[0058] The Data Analysis Module integrates and analyzes all the collected EEG signals, trajectory data, and behavior data, and generates a decision report. Based on big data analysis technology, this module real-time summarizes the data from each module to help customs management personnel make decisions quickly.

[0059] The system will display the status of each inbound and outbound passenger through data visualization, marking information such as abnormal behaviors, verification results, and location trajectories, so as to provide a detailed real-time report for staff reference.

[0060] Module 1: Wearable Device Module

[0061] The present invention provides a rapid identity verification and trajectory tracking system for inbound and outbound passengers at customs. The wearable device module, as the first link of the system, is responsible for collecting the electroencephalogram (EEG) signals of inbound and outbound passengers. The wearable device module is one of the core components of the entire system. Its main function is to real-time obtain the EEG activity signals of inbound and outbound passengers through sensors and transmit the signals to subsequent modules for processing and analysis. This module can convert the EEG signals into digital data and transmit it to the central processing unit of the system through wireless transmission technology.

[0062] Under normal circumstances, the devices worn by inbound and outbound passengers include smart glasses, smart head-mounted devices, or smart bracelets, etc. These devices are integrated with EEG sensors and can real-time capture EEG signals without contact. Different wearable devices can be selected according to the scenario requirements to ensure that they do not affect the normal activities of passengers and can accurately collect EEG signals.

[0063] In this embodiment, the EEG sensor in the wearable device module can stably collect brain wave signals for a long time. To ensure the accuracy of data collection, the sampling frequency of the EEG signal is usually between 250 Hz and 1000 Hz. This frequency range can ensure that delicate brain wave activity information is collected, which is not only applicable to identification in a static state but also can be effectively identified in a dynamic environment. During the EEG signal collection process, the device senses the brain electrical activity on the scalp through the embedded sensor and converts these signals into electrical signals in real time for processing.

[0064] In a possible implementation, the wearable device module realizes wireless data transmission by adopting low-power Bluetooth or Wi-Fi technology. The adoption of this wireless technology can not only reduce the physical connection between devices, improve the flexibility and mobility of the devices, but also ensure the stability and real-time nature of data transmission. Whenever the inbound and outbound personnel wear the device and enter the system range, the EEG signal will be uploaded to the central processing system in real time, and the system can immediately process the data, judge the identity of the personnel, and provide support for subsequent steps.

[0065] As an option, the sensor design of the wearable device module can be flexibly customized according to different scenario requirements. For example, if it is necessary to collect signals in a high-density environment, an EEG sensor with high anti-interference ability can be selected; if the device is required to be more lightweight, a more simplified sensor layout can be used to ensure the comfort and wearability of the device.

[0066] Specifically, during the process of the device collecting brain wave signals, the EEG sensor in the device can detect the electrical activity of the cerebral cortex through the electrode array. After these electrical activities are converted into electrical signals by the sensor, through analog signal processing, and then the signal is converted into a digital signal by the digital conversion module. This process is crucial for the accuracy and stability of the signal, ensuring that ideal results can be obtained in subsequent signal processing and analysis.

[0067] In some embodiments, to ensure the accurate transmission of the signal, the device can also adopt a multi-channel signal collection design to increase the coverage range of EEG signal collection and improve the collection quality of brain wave signals. This multi-channel design can collect multi-dimensional information at different electrode positions, improve the integrity of the signal, and reduce the interference of noise on the signal.

[0068] In this embodiment, the wireless transmission of EEG signals adopts a low-latency and high-bandwidth communication technology to ensure efficient and stable data transmission between the device and the central processing system. This transmission method ensures that the EEG signals collected by the device can be uploaded to the background system in real time for subsequent signal processing modules to analyze. The high-bandwidth characteristic of the wireless transmission module ensures that no information is lost during data transmission, thus avoiding recognition errors caused by transmission interruptions or signal attenuation.

[0069] Module 2: Wireless Transmission Module

[0070] In the rapid verification of the identity and trajectory tracking system for customs entry and exit personnel of the present invention, the wireless transmission module is a key link for realizing data communication between the device and the central processing system. This module ensures the stable transmission of EEG signals through wireless network technology and supports real-time signal processing and subsequent behavior analysis. The efficiency and stability of the wireless transmission module directly affect the performance of the entire system, ensuring that the system can seamlessly complete identity verification and trajectory tracking in the real-time scenario of personnel entry and exit.

[0071] Generally, the wireless transmission module uses a low-power and high-bandwidth communication technology to ensure the stable and efficient transmission of EEG signals and related behavior data to the central processing system. With the development of technology, the transmission rate and stability of the wireless communication module have been greatly improved, especially the adaptability to real-time application scenarios has been greatly enhanced. The wireless transmission module in the system of the present invention mainly adopts Bluetooth, Wi-Fi or other low-power wide area network (LPWAN) technologies, enabling the device to exchange data with the central processing system efficiently and stably for a long time.

[0072] In this embodiment, the wireless transmission module adopts low-power Bluetooth or Wi-Fi technology, which can ensure that the EEG signals are wirelessly transmitted to the background processing system immediately after being collected. The key to this technology lies in the characteristics of low latency and high bandwidth, ensuring the real-time and integrity of EEG signal data. In a high-density crowd or complex environment, the system can still effectively transmit data without being affected by environmental factors.

[0073] As an option, in a scenario with a longer distance, the system can also adopt a wider area communication method, such as 4G / 5G network or LPWAN technology. These technologies can maintain a lower latency over a longer distance and ensure that data transmission is not restricted by distance. For scenarios that require cross-regional or cross-border communication, adopting such high-bandwidth and long-distance wireless transmission technologies will greatly improve the applicability and flexibility of the system.

[0074] Specifically, the design of the wireless transmission module takes into account that the system needs to quickly process a large amount of EEG signal data. Therefore, high-bandwidth communication technology is adopted, which can transmit a large amount of brain wave data in a short time, avoiding data delay or loss. After the EEG signals are collected, these signals will be transmitted to the wireless communication unit in the module through sensors, and then uploaded to the background system in real time after digital conversion. During the transmission process, the system automatically encrypts the signals to ensure the security of data transmission and prevent data from being maliciously tampered with.

[0075] In some embodiments, to address issues such as network bandwidth or instability during transmission, the system also adopts a data compression and buffering mechanism. After being processed, the EEG signal data will be compressed according to a certain algorithm to reduce the amount of data transmitted, and will be temporarily stored in the buffer when the network is unstable, and then uploaded after the network returns to normal. This mechanism ensures the stability of the system and guarantees the complete transmission of data under any network conditions.

[0076] In a possible implementation, the wireless transmission module can ensure seamless connection of EEG signals and other behavioral data in real-time scenarios by using low-latency communication technologies such as low-power Bluetooth or Wi-Fi 6. Its low-power feature also enables the device to operate for a long time without frequent charging, increasing the service life of the device. Wi-Fi 6, as a relatively advanced wireless communication technology at present, can support more devices to access simultaneously and provide a higher data transmission rate, which is especially suitable for systems that need to process large-scale data.

[0077] In an extended implementation, the system can also integrate intelligent routers or relay devices to ensure stable signal transmission within the coverage area. For some complex environments or areas with signal obstacles, the intelligent router can automatically adjust the signal frequency band and transmission mode to ensure the stable transmission of EEG signals and other data. This intelligent routing method can provide a stable wireless network connection during the process of personnel entering and leaving the country.

[0078] Module 3: Signal Denoising Module

[0079] In the customs entry and exit personnel identity rapid verification and trajectory tracking system of the present invention, the signal denoising module plays a crucial role in the entire system. The collection of EEG signals is often accompanied by various noises and interferences, such as eye movements, electromyographic activities, external electromagnetic interferences, etc. These noises will seriously affect the signal quality and thus the accuracy of identity verification and behavior analysis. Therefore, the signal denoising module optimizes the EEG signals through effective denoising techniques to ensure that the subsequent processing module can obtain clean and accurate signal data.

[0080] In general, EEG signals are inevitably disturbed by various noises during the acquisition process. Traditional denoising methods often have certain limitations when dealing with noises in specific frequency bands. Especially in real-time data processing, how to efficiently and accurately remove noises while retaining the effective information in the signals is a technical challenge. The signal denoising module of the present invention adopts a hybrid optimization algorithm of wavelet transform and partial differential equation (PDE), which can fully exploit the time-frequency characteristics in the signals, removing both high-frequency instantaneous noises and smoothing low-frequency interference components.

[0081] In this embodiment, the signal denoising module first performs preliminary time-frequency analysis using wavelet transform. Through localization, wavelet transform can independently analyze different frequency bands of the signal. Specifically, wavelet transform decomposes the signal into multiple sub-signals of different scales, extracts the high-frequency part and low-frequency part of the signal through wavelet basis functions of different scales, and then identifies and removes noises. The mathematical expression of wavelet transform is:

[0082]

[0083] where ψ(t) is the mother wavelet, a is the scale factor, and b is the translation factor. By adjusting a and b, multi-resolution analysis of the signal in the time domain and frequency domain can be achieved. High-frequency noises usually manifest as instantaneous changes in the signal, so denoising in the high-frequency components can effectively eliminate eye movement and electromyogram interferences; low-frequency noises are usually related to environmental interferences. The signal denoising module effectively removes these influences through smoothing the low-frequency components.

[0084] As an option, partial differential equation PDE is used to further optimize the smoothness of the signal. By processing the EEG signal with the heat conduction equation, some low-frequency random interferences can be eliminated. In this embodiment, the form of the heat conduction equation adopted by PDE is:

[0085]

[0086] where u(t,x) is the change of the signal in time and space, α is the heat conduction coefficient, is the Laplace operator, and f is the noise source term. By adjusting the value of α, the smoothness degree of the signal can be adjusted at different scales. This process helps to remove low-frequency systematic noises, especially the background noises caused by sensors or electromagnetic interferences. Specifically, in the heat conduction equation, represents the second-order spatial derivative of the signal. Through the effect of this term, the high-frequency part of the signal is suppressed, making the signal smoother and removing low-frequency random noises. By this method, the low-frequency part of the signal is suppressed, thus avoiding the influence caused by external factors such as electromagnetic interferences.

[0087] In a possible implementation, the signal denoising module of the present invention also introduces an adaptive threshold selection mechanism. During the wavelet transform process, the system uses an adaptive threshold to distinguish the noise part and the effective components in the signal. By setting an appropriate threshold, the system can automatically adjust the denoising intensity according to the specific characteristics of the signal, avoiding excessive denoising and losing the effective signal. The adaptive threshold method ensures that in different environments, the system can adaptively adjust its denoising effect and optimize the performance of the system in various complex scenarios.

[0088] In some embodiments, to improve the denoising accuracy, the system can also adopt a combined processing method of multi-level wavelet transform and PDE algorithm. The signal is first preliminarily denoised through multi-level wavelet transform, and then the residual noise is further removed by PDE smoothing processing. This multi-level denoising method effectively combines the advantages of the two technologies and further improves the denoising effect.

[0089] In an extended implementation, the signal denoising module can select different wavelet basis functions and PDE coefficients according to the specific noise type. For example, when there is strong periodic noise in the EEG signal, a wavelet basis function with high time-frequency localization ability can be selected to perform special processing on this noise. In addition, the system can dynamically adjust the parameters to cope with the change of the noise level in the environment and ensure the stability of the signal quality.

[0090] Module 4: Feature Extraction Module

[0091] In the rapid identity verification and trajectory tracking system for customs entry and exit personnel of the present invention, the feature extraction module is one of the core components. It is located between the signal denoising module and the classification module and is responsible for extracting personalized features from the denoised EEG signal that are helpful for identity verification and behavior analysis. These features will serve as the basis for subsequent classification, anomaly detection, and behavior analysis. Therefore, its accuracy directly affects the overall performance of the system.

[0092] Generally, the EEG signal contains a large amount of individual information. The brain waves of each person have uniqueness in both static and dynamic states. The goal of the feature extraction module is to extract the most representative personalized features from these complex signals so that the system can accurately identify the identities of different individuals and recognize their behavior patterns. To achieve this goal, the feature extraction module combines two advanced mathematical methods of fractal geometry and information theory.

[0093] In this embodiment, the feature extraction module first uses a Hurst exponent calculation unit based on fractal geometry to analyze the EEG signal. The Hurst exponent is a method for measuring the self-similarity of a time series. It can reflect the dependence of the signal on the long-term trend and is often used to describe the complex structure of the EEG signal. The calculation of the Hurst exponent is based on the following formula:

[0094]

[0095] Among them, R(N) is the range of the signal when the window size is N, S(N) is the standard deviation of the signal, and N is the length of the time window. By calculating the Hurst index, the system can quantify the self-similarity of the EEG signal, thereby extracting the unique brain wave characteristics of each individual. These characteristics are highly personalized for the verification of the identity of people entering and leaving the country, and can provide effective differentiation capabilities even in complex environments. As an option, the fractal geometry method can further enhance the personalized identification of the signal by observing the multi-scale changes of the signal. By adopting fractal features such as fractal dimension and Hurst index, the system can accurately identify the regularity in the signal in low-frequency and high-frequency noise, and then effectively separate the information related to individual identity. Specifically, the second part of the feature extraction module is the Shannon entropy calculation unit based on information theory. Shannon entropy can quantify the complexity of the signal. The larger the entropy value, the higher the uncertainty of the signal, and vice versa, the stronger the regularity of the signal. The calculation formula of Shannon entropy is:

[0096]

[0097] Among them, p(x i ) is the state x in the EEG signal i The system is able to evaluate the complexity of the signal and extract characteristic information related to the behavior patterns of people entering and leaving the country by calculating the Shannon entropy of the EEG signal, where n is the probability distribution of the state. This method is particularly suitable for comparing static and dynamic behavior patterns and can identify differences in behavioral state changes among different individuals.

[0098] In some embodiments, in addition to the Hurst exponent and Shannon entropy, the system can also combine other information theory indicators, such as Kolmogorov-Sinai entropy, to further enhance the accuracy of feature extraction. These additional entropy values ​​can help the system to more carefully classify and analyze the brain wave activity of individuals at different periods and behavioral states.

[0099] In one possible implementation, the output of the feature extraction module includes not only static EEG features, but also dynamic behavior pattern features. In real-time monitoring, EEG signals will change with the individual's behavioral state (such as quiet, walking, concentrating, etc.). The feature extraction module can accurately capture changes in dynamic behavior patterns by continuously monitoring these changes. This feature is particularly important in high-density crowds or complex scenarios, and can help the system quickly distinguish between normal behavior and potential abnormal behavior.

[0100] In some extended embodiments, in order to enhance the system's adaptability to different environmental conditions, the feature extraction module can also perform feature fusion by combining multimodal data. By combining EEG signals with other biometric information (such as heart rate, body temperature, etc.), the system can further improve the accuracy of identity verification, especially in complex environments or when multiple noise sources coexist, enhancing the system's robustness.

[0101] Module 5: Classification Module

[0102] In the rapid identity verification and trajectory tracking system for customs entry and exit personnel of the present invention, the classification module, as one of the core parts, undertakes the task of classifying the features of EEG signals. The function of the classification module is to analyze the personalized features extracted from the EEG signals to determine the identity and behavior status of the entry and exit personnel. Its accuracy is crucial for the overall performance of the system. Especially when facing a large number of people, how to quickly and accurately perform identity verification and behavior recognition is the key.

[0103] Generally, EEG signals contain unique electroencephalogram activity patterns for each individual. However, directly classifying EEG signals is often difficult because the features in the signals are highly complex. To accurately perform identification and classification, the present invention adopts the support vector machine (SVM) algorithm. SVM is a powerful supervised learning method that can effectively handle non - linear classification problems. Its advantage lies in mapping the data to a higher - dimensional space through a kernel function to find the optimal separating hyperplane for classification.

[0104] In this embodiment, the classification module classifies the extracted EEG signal features through the SVM algorithm. First, the system passes the feature vectors of the EEG signals into the SVM classifier. The SVM constructs a decision boundary by learning the distribution of these feature vectors to classify each entry and exit personnel. To handle the non - linear features of the EEG signals, the SVM adopts a Gaussian kernel function to map the input data. Specifically, the form of the Gaussian kernel function is as follows:

[0105]

[0106] where x and y are two data points to be compared, and σ is the parameter of the Gaussian kernel, which determines the smoothness of the data mapping. Through this mapping, the SVM can find an optimal hyperplane in the high - dimensional space to separate data of different classes. Specifically, the Gaussian kernel can effectively handle those EEG features with complex non - linear relationships, enabling the system to better identify the identities of entry and exit personnel in the high - dimensional space.

[0107] As an option, the SVM classification module can also be optimized by combining other kernel functions, such as polynomial kernels, radial basis functions (RBF), etc. By selecting different kernel functions, the system can adapt to different types of data characteristics and further improve the classification accuracy. For example, in an environment with more noise, the RBF kernel may perform more robustly, while in the case of relatively stable signals, the polynomial kernel may show better classification results.

[0108] Specifically, the SVM classifier finds the optimal classification hyperplane by maximizing the classification margin. The goal of the optimization problem is to minimize the following objective function:

[0109]

[0110] where w is the normal vector of the hyperplane, b is the bias term, y i is the class label of the i-th sample, and x i is the feature vector of the i-th sample, and n is the total number of samples. By solving this optimization problem, SVM can find the optimal hyperplane to separate data points of different classes to the greatest extent, thus achieving efficient classification.

[0111] In some extended embodiments, the classification module can also combine multi-classification methods to identify multiple identities of inbound and outbound personnel. For example, in addition to verifying identities, it can also infer the intentions of personnel by analyzing behavior patterns, such as whether there are potential abnormal behaviors. In this case, the system not only classifies the identities of personnel but also makes comprehensive judgments by combining behavior characteristics, further improving the intelligence level of the system.

[0112] In one possible implementation, the SVM classification module can also perform cross-validation to avoid overfitting. Through cross-validation, the system can automatically adjust the parameters of SVM, such as the penalty factor C and the kernel function parameter σ, during the training process, so that the model can maintain good generalization ability on different data sets. This method is particularly important when dealing with unknown data and can effectively improve the stability of the system in practical applications.

[0113] In some extended embodiments, the classification module can also combine multi-classification methods to identify multiple identities of inbound and outbound personnel. For example, in addition to verifying identities, it can also infer the intentions of personnel by analyzing behavior patterns, such as whether there are potential abnormal behaviors. In this case, the system not only classifies the identities of personnel but also makes comprehensive judgments by combining behavior characteristics, further improving the intelligence level of the system.

[0114] Module 6: Anomaly Detection Module

[0115] In the rapid identity verification and trajectory tracking system for customs entry and exit personnel of the present invention, the anomaly detection module plays a crucial role. This module is responsible for real-time monitoring of anomaly patterns during the behavior process of entry and exit personnel, and can identify potential abnormal behaviors or identity disguises, thereby ensuring the security and accuracy of the system in practical applications. The anomaly detection module is closely combined with the aforementioned classification module and feature extraction module. Using the personalized features extracted from EEG signals and combining with the dynamic time warping (DTW) algorithm, it ensures that the changes in personnel behavior can be effectively detected and early warnings can be issued in a timely manner.

[0116] Generally, in complex environments, especially in high-security scenarios such as customs and border inspections, the behavior patterns of personnel may vary. For example, entry and exit personnel may exhibit EEG characteristics different from normal behaviors due to factors such as nervousness and anxiety. Therefore, it is crucial to identify these abnormal behaviors. To ensure the efficiency of the system, the anomaly detection module adopts the dynamic time warping algorithm (DTW), which can calculate the similarity between different time series, and then helps the system identify behaviors that do not conform to the normal pattern.

[0117] In this embodiment, the DTW algorithm determines whether there is an abnormal behavior by comparing the time series differences between the real-time EEG signal and the preset normal behavior pattern. When the similarity between the EEG signal of the entry and exit personnel within a specific time period and the normal pattern is lower than the preset threshold, the system will determine it as an abnormal behavior. For example, a certain entry and exit personnel may show extremely high anxiety during the customs clearance process, resulting in significant changes in their brain wave activities. The DTW algorithm can accurately detect this abnormal change and issue an alarm immediately.

[0118] The basic principle of the DTW algorithm is to measure the similarity in time between two time series by calculating the distance between them. The calculation formula of DTW is as follows:

[0119]

[0120] Among them, D(i,j) represents the matching cost at the i-th and j-th time points, and d(x i ,y j ) represents the distance between two time points, usually calculated using the Euclidean distance. Through this algorithm, the system can calculate the matching differences between each pair of signal points, thereby obtaining the overall similarity of the signals. If the distance between two signal sequences is greater than the preset threshold, it is determined as an abnormal behavior.

[0121] As an option, the DTW algorithm can be combined with other distance measurement methods, such as Manhattan distance, cosine similarity, etc., and can be flexibly selected according to different application scenarios and signal characteristics. In some complex environments, Manhattan distance or cosine similarity may be more suitable for comparing behavior patterns.

[0122] Specifically, the DTW algorithm is suitable for processing dynamic and time-series data. By comparing real-time EEG signals with historical behavior patterns, DTW can detect anomalies in a timely manner when people's behavior changes. The system automatically marks abnormal behaviors based on the calculated time series differences and the set thresholds. For example, when a person entering or leaving the country shows brain wave activity that is significantly different from his or her normal behavior pattern when passing through customs inspection, the DTW algorithm will detect the difference and issue an alarm, reminding staff to conduct additional inspections on the person.

[0123] In some embodiments, the anomaly detection module not only compares static behavior patterns through the DTW algorithm, but also tracks changes in personnel behavior in real time by comparing historical behavior patterns multiple times. This real-time monitoring capability enables the system to capture subtle behavioral anomalies, further improving the sensitivity of the system.

[0124] In one possible implementation, the system will combine historical behavioral data, entry and exit records, and EEG signals of personnel to form a multi-dimensional behavioral feature library. When abnormal behavior is identified, the DTW algorithm not only compares the current behavior pattern with the normal pattern, but also uses historical data for comprehensive judgment. This enhanced feature can improve the accuracy of anomaly detection, especially when dealing with long-term behavior patterns.

[0125] In some extended embodiments, in order to further optimize the detection accuracy, the anomaly detection module can be combined with other types of machine learning algorithms for auxiliary judgment. For example, clustering-based algorithms such as K-means or DBSCAN can be used to classify behavior patterns and identify those behavior patterns that show similar anomalies. Combined with these methods, the results of the DTW algorithm will be enhanced, and the potential abnormal behavior of people entering and leaving the country can be detected more accurately.

[0126] Module 7: Trajectory Tracking Module

[0127] In the customs entry and exit personnel rapid identity verification and trajectory tracking system of the present invention, the trajectory tracking module is one of the key parts of the system, responsible for real-time tracking of the behavior trajectory of entry and exit personnel. The main task of this module is to provide comprehensive and accurate personnel location tracking information by integrating RFID tags, GPS positioning technology and EEG signals, and closely cooperate with the aforementioned modules (such as feature extraction module and anomaly detection module). The trajectory tracking module can not only monitor the actual location of personnel, but also analyze whether they stay in unauthorized areas based on their behavior patterns. In this way, the system can further improve security efficiency and real-time monitoring of personnel behavior while ensuring identity verification.

[0128] In general, the tracking of the trajectories of inbound and outbound personnel is the basis for identity verification and behavior monitoring. In practical applications, the identity verification of personnel may not only rely on static information (such as ID cards or passports), but also requires further verification of the legality of their behavior through real-time trajectory monitoring and behavior analysis. To ensure that personnel do not cross the specified area or engage in illegal activities, the trajectory tracking module relies on the combination of advanced positioning technologies and biometric technologies.

[0129] In this embodiment, the trajectory tracking module uses RFID tags and the GPS positioning system to track the location information of inbound and outbound personnel in real time. The RFID tags can be positioned by being embedded in the items of inbound and outbound personnel (such as passports, luggage, and smart devices). Whenever a person passes through the inspection area of the system or enters a specific monitoring area, the RFID tag will communicate with the reader in the system, thereby recording the person's location information. In cooperation with the RFID system, the GPS system is used to obtain the specific location of the person in real time over a large area. Especially in the case of cross-regional or cross-border situations, GPS can provide more accurate location data.

[0130] As an option, the combination of RFID tags and GPS technology can provide more accurate positioning services. Especially in large-scale personnel management scenarios, the high-frequency identification ability of RFID enables the system to quickly tag and locate personnel, while GPS provides high-precision location information in long-distance and outdoor environments. The combination of the two can ensure the efficiency of the system in different environments and scenarios.

[0131] Specifically, the trajectory tracking module will also analyze the behavior patterns of personnel by combining EEG signals. For example, by analyzing the EEG signals of inbound and outbound personnel, the system can identify whether they have abnormal behaviors, such as staying in a certain area for a long time or frequently changing positions. This function is particularly important for high-security places such as customs inspections, which can avoid potential security threats.

[0132] In some embodiments, to improve the adaptability of the system, the trajectory tracking module can also be linked with the video surveillance system to provide multi-dimensional data support. The video surveillance system can identify the facial features and behavior patterns of personnel in real time, so as to compare with the RFID and GPS data, further improving the accuracy of trajectory tracking. For example, when the RFID and GPS systems detect that a certain person has stayed in the designated area for too long, the video surveillance system can further verify their identity and behavior to ensure that the system issues an alarm in a timely manner.

[0133] In one possible implementation, the trajectory tracking module can record and display the location and path of personnel in real time, and issue warnings for abnormal activities based on the results of behavioral analysis. For example, if a person entering or leaving the country stays outside the area where he or she should stay for more than a preset time, the system will automatically identify the behavior based on RFID and GPS data and further analyze it through the anomaly detection module. This automated behavioral analysis can greatly improve the efficiency of personnel monitoring and reduce human intervention.

[0134] In some extended embodiments, in order to further improve the accuracy and robustness of the system, the trajectory tracking module can also adopt an integrated positioning algorithm to reduce the errors that may be caused by a single sensor by fusing multi-source data from RFID, GPS and EEG signals. For example, if the GPS signal fails in certain areas (such as underground parking lots or closed areas), the system can calculate the movement trajectory of the person through the combination of RFID and EEG signals. Through multi-sensor data fusion, the system can provide more accurate and reliable positioning information.

[0135] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A system for rapid identity verification and trajectory tracking of customs entry and exit personnel, characterized in that: include: Wearable device module, used to collect brain wave signals of people entering and leaving the country; Wireless transmission module, used to receive and transmit brain wave signals to the central processing system in real time; A signal denoising module is used to denoise the received brain wave signal; Feature extraction module, used to extract personalized features from EEG signals; A classification module, used to classify the extracted EEG signal features; Anomaly detection module, used to detect abnormal behavior of people entering and leaving the country; The trajectory tracking module is used to track the trajectories of people entering and leaving the country.

2. A customs entry and exit personnel rapid identity verification and trajectory tracking system according to claim 1, characterized in that: The system includes a data analysis module for summarizing and analyzing the data of each module and generating a decision report; The signal transmission module is used to ensure the real-time transmission and processing of EEG signals and behavioral data, using low-latency, high-bandwidth communication technology.

3. A system for rapid identity verification and trajectory tracking of customs entry and exit personnel according to claim 1, characterized in that: The denoising process of the signal denoising module includes performing multi-resolution time-frequency analysis on the signal using wavelet transform to remove high-frequency noise, and performing smoothing processing on the signal through heat conduction equation to eliminate low-frequency noise.

4. A system for rapid identity verification and trajectory tracking of customs entry and exit personnel according to claim 1, characterized in that: The feature extraction module includes a Hurst index calculation unit based on fractal geometry. The Hurst index evaluates the self-similarity of the signal by calculating the ratio of the signal's range to the standard deviation, thereby extracting the personalized EEG features of the inbound and outbound personnel. The feature extraction module includes a Shannon entropy calculation unit based on information theory. Shannon entropy is used to quantify the complexity of the EEG signal, thereby identifying the individual behavior patterns of the inbound and outbound personnel.

5. According to claim 1, a system for rapid identity verification and trajectory tracking of customs entry and exit personnel, characterized in that: The classification module classifies EEG signal features through a support vector machine algorithm. SVM uses a Gaussian kernel function to perform nonlinear mapping on the EEG signal and classifies the data according to the maximum margin principle.

6. A customs entry and exit personnel rapid identity verification and trajectory tracking system according to claim 1, characterized in that: The anomaly detection module compares the time series similarity between the real-time EEG signal and the preset normal behavior pattern through a dynamic time warping algorithm. If the similarity is lower than a preset threshold, it is determined to be an abnormal behavior.

7. A system for rapid identity verification and trajectory tracking of customs entry and exit personnel according to claim 1, characterized in that: The trajectory tracking module tracks the location of inbound and outbound personnel in real time through RFID tags and GPS positioning systems, and further analyzes behavior patterns through EEG signals to determine whether personnel stay in unauthorized areas.

8. A system for rapid identity verification and trajectory tracking of customs entry and exit personnel according to claim 2, characterized in that: The data analysis module processes and comprehensively analyzes signal, behavior and trajectory data in real time based on the big data analysis platform, generates decision reports, and helps customs management personnel make real-time responses.

9. A system for rapid identity verification and trajectory tracking of customs entry and exit personnel according to claim 1, characterized in that: The wavelet transform in the signal denoising module performs noise suppression through multi-scale analysis and adaptive threshold selection.

10. A system for rapid identity verification and trajectory tracking of customs entry and exit personnel according to claim 1, characterized in that: The trajectory tracking module integrates RFID, GPS and EEG signals to update the trajectory of personnel in real time and provide early warning of abnormal activities in combination with behavioral pattern analysis.

Images