Living animal intelligent detection and customs clearance management system based on millimeter wave radar

Abstract

The invention relates to the technical field of intelligent detection and customs clearance management, in particular to a living animal intelligent detection and customs clearance management system based on millimeter wave radar, which comprises a millimeter wave radar array module, a data processing and analysis module, a luggage sorting control module, a quarantine process management module, an alarm response module and a system management platform. The millimeter-wave radar array module acquires biological characteristics of a living body by adopting a multi-band fusion detection technology, and the data processing and analysis module identifies the living body and outputs a confidence score by applying algorithms such as an improved double-flow convolutional neural network; and the system management platform is linked with each module based on a detection result to realize luggage dynamic sorting, grading alarm, quarantine resource intelligent distribution and whole-process closed-loop control. According to the invention, hidden living animals can be effectively identified, biological invasion and epidemic disease transmission are prevented, and through automatic and intelligent management, the customs clearance efficiency is significantly improved, the manual intervention cost is reduced, and the traceability and safety of the clearance data are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent detection and customs clearance management, and specifically to a living animal intelligent detection and customs clearance management system based on millimeter wave radar. Background Art

[0002] Traditional customs baggage inspection mainly relies on manual inspection and single technical detection, which has many limitations.

[0003] Manual inspection relies on officers' experience and visual observation, making it inefficient and unable to cope with large-scale customs clearance. It is also prone to missing live animals hidden in luggage compartments and hidden compartments. Furthermore, frequent unpacking increases the risk of biological contamination, impacting the passenger's customs clearance experience. Existing detection technologies, such as X-ray imaging, can provide images of the luggage's internal structure, but struggle to effectively distinguish live animals from similarly shaped non-living items, resulting in a high rate of false positives. Infrared thermal imaging is also susceptible to interference from ambient temperature, making detection accuracy limited in complex scenarios.

[0004] As global customs business volume continues to grow, traditional detection methods can no longer meet the dual needs of efficient customs clearance and precise supervision. Therefore, to address the above problems, a millimeter-wave radar-based intelligent detection and customs clearance management system for live animals is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a system for intelligent detection and customs clearance management of living animals based on millimeter wave radar to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: A millimeter-wave radar-based intelligent detection and customs clearance management system for live animals, including a millimeter-wave radar array module, a data processing and analysis module, a baggage sorting control module, a quarantine process management module, an alarm response module, and a system management platform; Millimeter-wave radar array modules are deployed at key nodes of customs baggage conveyor belts to scan the biological characteristics of living people inside the luggage in real time; The data processing and analysis module uses millimeter wave feature extraction algorithms to identify the presence of living animals and outputs the detection results to the system management platform; The system management platform dynamically triggers the following collaborative processes based on the detection results: The baggage sorting control module adjusts the sorting path in real time to direct baggage containing live animals to the quarantine channel; Generate hierarchical alarm signals through the alarm response module and associate them with quarantine priorities; Quarantine resources are automatically allocated and electronic quarantine task orders are generated through the quarantine process management module.

[0007] As a preferred solution, the millimeter-wave radar array module adopts multi-band fusion detection technology, including: The low-frequency band is used to penetrate the luggage material to obtain the outline of the living body. The low-frequency band frequency is between 24-33GHz; The high-frequency band is used to capture micro-motion features, and the frequency of the high-frequency band is between 60-90GHz; A living biological feature model is established based on dual-band data fusion to distinguish living animals from static interference objects.

[0008] As a preferred solution, the data processing and analysis module includes: Millimeter wave signal preprocessing unit, which performs motion compensation and noise filtering on radar echoes; The living feature recognition unit extracts biological micro-motion spectrum features through an improved two-stream convolutional neural network model; The dynamic threshold judgment unit outputs a liveness confidence score based on an adaptive confidence calibration mechanism.

[0009] As a preferred solution, in the improved two-stream convolutional neural network model: The probability of the existence of a living body is obtained by calculating the weighted sum of the outputs of each convolutional layer using the Sigmoid function. The output of each convolution layer is generated by adding the low-frequency and high-frequency band data after being processed by the convolution kernel respectively, and then through the ReLU activation function; The dynamic frequency band weight coefficient is determined based on the exponential normalized value of the characteristic signal-to-noise ratio of each layer, where the signal-to-noise ratio weight is adjusted by the ambient noise energy value: The environmental interference adjustment factor approaches 1 as the environmental noise energy increases, and the material attenuation empirical constant ranges from 0.1 to 0.5.

[0010] As a preferred solution, in the adaptive confidence calibration mechanism: The final confidence score is the weighted sum of the probability of existence of living organisms and the Shannon entropy of micro-motion characteristics; The weighting coefficient is dynamically adjusted as the ambient temperature changes. For every 1 degree Celsius increase in temperature, the probability value weight increases by 0.05. The liveness threshold is determined by adding a compensation factor of the square of the material density to the base threshold of 0.6, and the material compensation factor is fixed at 0.02. When the final confidence score exceeds the dynamic threshold, it is determined that the baggage contains live animals.

[0011] As a preferred solution, the baggage sorting control module performs dynamic path optimization including: The sorting priority index is calculated by multiplying the confidence score by the negative exponential decay function of the quarantine channel load; When the real-time load of the channel reaches the benchmark load value, the priority exponential decays to 1 / e of the original value; By binding luggage and detection data through radio frequency identification technology, the sorting machine is controlled to transfer high-priority luggage to the quarantine area.

[0012] As a preferred solution, the alarm response module adopts a three-level collaborative disposal mechanism, including: Level 1 alarm: When the confidence score is greater than 0.9, an audible and visual alarm is triggered and the conveyor belt is frozen; Level 2 alarm: When the confidence score is 0.7 ≤ ≤ 0.9, a pending review label is generated and a manual inspection is notified; Level 3 alarm: When the confidence score is 0.6≤<0.7, only a low-risk alert will be sent to the quarantine management module; The alarm signal is pushed to the customs personnel’s mobile terminal in real time.

[0013] As a preferred solution, the quarantine process management module includes: A quarantine plan generation unit matches the preset disposal plan library according to the type of living organism; Resource scheduling unit, which allocates quarantine resources based on the luggage space coordinates and confidence scores; The electronic customs clearance form generation unit automatically generates electronic release instructions and updates the customs clearance system after quarantine is completed.

[0014] As a preferred solution, the resource scheduling unit executes an optimization algorithm including: The scheduling cost function minimizes the weighted response time of all bags to be quarantined; The weight factor is the ratio of the confidence score to the distance between the quarantine officer and the luggage; The response time is calculated based on the average time consumption of similar historical quarantines.

[0015] As a preferred solution, the system management platform implements closed-loop control of the entire process, including: Millimeter wave detection data is connected to the customs clearance system through an application programming interface; Build an electronic traceability chain from testing to release, and use blockchain technology to store data at each link; Generate a customs clearance efficiency heat map based on confidence scores and quarantine time to visually display process bottlenecks.

[0016] It can be seen from the technical solutions provided by the present invention that the intelligent detection and customs clearance management system for living animals based on millimeter wave radar provided by the present invention has the following beneficial effects: 1. Accurate testing and strengthening biosafety prevention and control: The millimeter-wave radar array module utilizes multi-band fusion detection technology, combined with advanced algorithms in the data processing and analysis module. It can effectively penetrate various luggage materials, accurately capture the minute physiological characteristics of live animals, and achieve highly reliable identification of hidden live animals. This technological breakthrough significantly reduces the risk of misjudgment and missed detection with traditional detection methods, effectively intercepting illegally entering live animals. 2. Intelligent linkage to improve customs clearance efficiency: The system management platform coordinates various functional modules to achieve automated and intelligent management of the entire process, from detection, sorting, alarming, and quarantine. The baggage sorting control module dynamically plans routes based on detection results and channel load, the quarantine process management module intelligently allocates resources, and the alarm response module accurately and categorically handles incidents. All links work closely together and operate efficiently. Compared with traditional manual operations, this significantly shortens the processing time of baggage containing live animals, greatly improving overall customs clearance efficiency at the port, reducing passenger waiting time, and optimizing the customs clearance experience. 3. Make scientific decisions and enhance emergency management capabilities: The alarm response module's hierarchical handling mechanism, combined with the system management platform's intelligent analysis capabilities, enables rapid and accurate responses to abnormal situations. It also predicts potential risks through data analysis, providing customs management with a scientific basis for decision-making. This proactive management model enables customs to plan resource allocation in advance, respond calmly to various emergencies, and effectively enhance the scientific and proactive nature of emergency management. 4. Full traceability to ensure regulatory compliance and security: The electronic traceability chain built with blockchain technology ensures that data throughout the customs clearance process cannot be tampered with and is fully traceable. This strictly complies with customs regulations and audit requirements, provides strong data support for law enforcement, and reduces enforcement risks. At the same time, the system's comprehensive security protection and authority management system effectively prevents data leakage and illegal operations, ensuring the security of customs business information. 5. Optimize configuration and reduce system operating costs: Automated and intelligent process design reduces manual intervention and lowers labor costs. Intelligent algorithm-driven precise resource scheduling avoids idle equipment and waste of resources, improves the efficiency of equipment and personnel, optimizes customs operation resource allocation, and effectively reduces long-term operating costs. 6. Flexible adaptation to expand the scope of application scenarios: The millimeter-wave radar array module has environmental adaptability and can dynamically adjust detection parameters according to different luggage materials and complex environments; the data processing and analysis module adapts to the differences in various port environments through online learning; this makes the invention widely applicable to different types of customs ports, international logistics hubs and other scenarios, demonstrating strong versatility and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of the overall structure of the millimeter-wave radar-based intelligent detection and customs clearance management system for living animals of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0020] like Figure 1 As shown, the embodiment of the present invention provides a millimeter-wave radar-based intelligent detection and customs clearance management system for live animals, including a millimeter-wave radar array module, a data processing and analysis module, a baggage sorting control module, a quarantine process management module, an alarm response module, and a system management platform; Millimeter-wave radar array modules are deployed at key nodes of customs baggage conveyor belts to scan the biological characteristics of living people inside the luggage in real time; The data processing and analysis module uses millimeter wave feature extraction algorithms to identify the presence of living animals and outputs the detection results to the system management platform; The system management platform dynamically triggers the following collaborative processes based on the detection results: The baggage sorting control module adjusts the sorting path in real time to direct baggage containing live animals to the quarantine channel; Generate hierarchical alarm signals through the alarm response module and associate them with quarantine priorities; Quarantine resources are automatically allocated and electronic quarantine task orders are generated through the quarantine process management module.

[0021] In this embodiment, the millimeter-wave radar array module adopts multi-band fusion detection technology, which specifically includes: The low-frequency band is used to penetrate the luggage material to obtain the outline of the living body. The low-frequency band frequency is between 24-33GHz; The high-frequency band is used to capture micro-motion features, and the frequency of the high-frequency band is between 60-90GHz; Establish a living organism feature model based on dual-band data fusion to distinguish living animals from static interference objects; Furthermore, the millimeter-wave radar array module is a key component of the system's liveness detection system. Its performance directly affects detection accuracy and efficiency. The following is a detailed explanation of this module from the aspects of function, composition, and technical principles: 1. Overview of overall functions: Millimeter-wave radar array modules are deployed at key nodes on customs baggage conveyor belts. They transmit millimeter-wave signals and receive reflected echoes, enabling non-contact scanning of living animals inside luggage. Their core function is to capture the spatial contours and micro-motion characteristics (such as minute displacements caused by breathing and heartbeats) of objects inside the luggage in real time, transmitting the raw radar data to the data processing and analysis module. Furthermore, the module possesses environmental adaptability, dynamically adjusting detection parameters based on luggage material and environmental interference, ensuring stable acquisition of high-quality detection data in complex scenarios. 2. Submodule composition and functions: (1) Multi-band radar transmitting and receiving unit: Dual-band signal transmission: This system integrates millimeter-wave transmitters in both the low-frequency (24–33 GHz) and high-frequency (60–90 GHz) bands. The low-frequency transmitter utilizes its long wavelength to penetrate common luggage materials such as fabric and plastic, acquiring the object's outline and location information. The high-frequency transmitter leverages its high-resolution advantage to capture Doppler frequency shift signals generated by minute physiological movements of living animals (e.g., respiratory rates of approximately 0.5-2 Hz and heart rates of approximately 1-3 Hz). Array antenna design: Utilizing phased array antennas in linear or planar array layouts, electronic scanning technology achieves a ±45° field of view, ensuring no blind spots when baggage moves on the conveyor belt. Each antenna unit independently receives echo signals at different angles, providing multi-dimensional information for subsequent data fusion. Echo signal acquisition: Equipped with a high-sensitivity receiving circuit, it performs low-noise amplification, frequency mixing, and demodulation on the reflected echo, converting the analog signal into a digital baseband signal. At the same time, it uses pulse compression technology to improve distance resolution, capable of distinguishing different objects with a distance of less than 5 cm. (2) Data preprocessing and synchronization unit: Time synchronization module: Through the Global Positioning System (GPS) or high-precision clock chip, it achieves nanosecond-level time synchronization of multi-band radar transmission and reception, ensuring the consistency of high-frequency and low-frequency data in the time dimension, laying the foundation for subsequent dual-band data fusion; Clutter suppression and filtering: Adaptive filtering algorithms (such as Kalman filtering) are used to process raw echo data to remove environmental noise (such as electromagnetic interference and clutter generated by conveyor belt vibration). Motion compensation technology is also used to eliminate Doppler frequency shift interference caused by luggage movement, thereby extracting pure target micro-motion signals. Data formatting: Convert pre-processed radar data into a range-Doppler map or tensor format, package it in time series, and transmit it to the data processing and analysis module. The data format includes parameters such as the timestamp and amplitude of low-frequency profile information and high-frequency micro-motion characteristics. (3) Environmental Adaptive Adjustment Unit: Material Identification Module: This module establishes a baggage material classification model based on the attenuation characteristics and reflection coefficient of radar echoes. It uses machine learning algorithms (such as support vector machines) to analyze the spectral characteristics of echo signals, identifying materials such as metal, plastic, and fabric in real time, and automatically adjusting radar transmit power and frequency switching strategies. For example, when metal is detected, it reduces high-frequency signal strength to avoid signal saturation while enhancing the penetration of low-frequency signals. Interference detection and response: Continuously monitors the intensity of environmental electromagnetic interference. When an interference signal is detected (such as electromagnetic waves in the same frequency band generated by nearby communication equipment), the frequency band hopping mechanism is activated to dynamically switch sub-bands within the 24–33 GHz or 60–90 GHz range to ensure the stability of the detection signal. 3. Key technical principles: (1) Principle of multi-band fusion detection: Based on the physical properties of millimeter waves at different frequency bands, the low-frequency band (24–33 GHz) uses its penetrating properties to construct a three-dimensional contour model of the target, while the high-frequency band (60–90 GHz) uses the Doppler effect to capture the physiological movements of tiny animals. A data-level fusion strategy is used to fuse low-frequency contour data with high-frequency micro-motion features in the feature space to form a multidimensional dataset containing position, shape, and life characteristics, significantly improving the ability to distinguish between living animals and static objects (such as clothing and electronic products). (2) Phased array antenna scanning principle: By controlling the phase and amplitude of each element in the array antenna, rapid electronic scanning of the millimeter-wave beam is achieved. By utilizing the principle of wave interference, beams pointing in different directions are synthesized in space, covering the entire conveyor belt area without mechanical rotation. This technology enables the radar to scan at millisecond frame rates, ensuring that even fast-moving luggage can be fully detected. (3) Principle of environmental adaptive adjustment: Combining the material recognition algorithm with the interference detection mechanism, a closed-loop feedback control system is constructed. When environmental changes are detected (such as changes in material or interference intensity), the radar transmission parameters (frequency band, power, pulse repetition frequency) are automatically adjusted, and the data preprocessing filter coefficients are updated at the same time to ensure the stability of radar performance and detection accuracy under different working conditions.

[0022] In this embodiment, the data processing and analysis module includes: Millimeter wave signal preprocessing unit, which performs motion compensation and noise filtering on radar echoes; The living feature recognition unit extracts biological micro-motion spectrum features through an improved two-stream convolutional neural network model; A dynamic threshold determination unit that outputs a liveness confidence score based on an adaptive confidence calibration mechanism; Among them, in the improved two-stream convolutional neural network model: The probability of the existence of a living body is obtained by calculating the weighted sum of the outputs of each convolutional layer using the Sigmoid function. The output of each convolution layer is generated by adding the low-frequency and high-frequency band data after being processed by the convolution kernel respectively, and then through the ReLU activation function; The dynamic frequency band weight coefficient is determined based on the exponential normalized value of the characteristic signal-to-noise ratio of each layer, where the signal-to-noise ratio weight is adjusted by the ambient noise energy value: The environmental interference adjustment factor approaches 1 as the environmental noise energy increases, and the material attenuation empirical constant ranges from 0.1 to 0.5; Among them, in the adaptive confidence calibration mechanism: The final confidence score is the weighted sum of the probability of existence of living organisms and the Shannon entropy of micro-motion characteristics; The weighting coefficient is dynamically adjusted as the ambient temperature changes. For every 1 degree Celsius increase in temperature, the probability value weight increases by 0.05. The liveness threshold is determined by adding a compensation factor of the square of the material density to the base threshold of 0.6, and the material compensation factor is fixed at 0.02. When the final confidence score exceeds the dynamic threshold, it is determined that the baggage contains live animals; Furthermore, the data processing and analysis module acts as the "brain" of the system, responsible for converting the raw data collected by the millimeter-wave radar array module into accurate detection results. The following provides a comprehensive and in-depth explanation of this module from multiple aspects, including its function, structure, principles, and processes. 1. Overview of overall functions: The data processing and analysis module is primarily responsible for receiving raw radar data transmitted by the millimeter-wave radar array module. After a series of complex data processing and intelligent analysis operations, it accurately determines whether there are live animals in the luggage and promptly outputs the detection results to the system management platform. Its core functions include pre-processing the millimeter-wave radar echo signal, extracting biological micro-motion characteristics using advanced deep learning algorithms, and assigning a confidence score for the presence of living organisms using an adaptive threshold mechanism. Furthermore, the module continuously optimizes algorithms and calibration parameters to continuously improve detection accuracy and stability, providing solid data support and reliable decision-making for efficient customs clearance management. 2. Submodule composition and functions: (1) Millimeter wave signal preprocessing unit: Motion Compensation Module: When luggage moves on a conveyor belt, it generates Doppler frequency shift interference, affecting detection accuracy. The Motion Compensation Module deeply analyzes the speed and direction of the luggage to construct an accurate motion model. Based on this model, it precisely translates and stretches the time axis of the original radar echo signal, effectively eliminating the false Doppler component caused by object motion. This makes micro-motion features such as breathing and heartbeat clearer and more prominent, laying a good foundation for subsequent feature extraction and analysis. Noise filtering module: To remove noise from the signal, this module combines median filtering and wavelet transform techniques. Median filtering can specifically remove salt and pepper noise while effectively protecting the signal's edge information and avoiding edge blurring. Wavelet transform decomposes the signal into different frequency scales and, by setting appropriate thresholds, effectively removes high-frequency environmental noise, such as electromagnetic interference and noise generated by conveyor belt mechanical vibration. It also fully preserves the 0.1-5Hz characteristic frequency band of the micro-motion signal, ensuring the purity of the input data. Data Normalization Module: Because raw radar data varies in amplitude and frequency range, this can affect the performance of subsequent feature extraction algorithms. The data normalization module normalizes preprocessed radar data, mapping data of different scales to a unified interval, such as [0, 1]. This not only eliminates the adverse effects of data scale differences but also significantly improves the convergence speed and stability of the algorithm, making subsequent feature extraction and analysis more efficient and accurate. (2) Living feature recognition unit: Improved two-stream convolutional neural network model: This network uses a dual-input structure to process low-frequency and high-frequency data from millimeter-wave radar, fusing multi-band features through dynamic weighting to achieve accurate identification of living animals; Additive frequency band weight fusion: (in, For the The fusion weight coefficient of low-frequency and high-frequency features in the layer convolution, ; For the The signal-to-noise ratio of the layer feature, calculated as the ratio of signal power to noise power; is the environmental disturbance regulation factor, ; is the material attenuation empirical constant, and its value range is , unit is c / g is used to quantify the attenuation characteristics of different materials to millimeter waves; the ambient noise energy value is the energy integral of the ambient noise spectrum collected in real time, calculated by Fourier transform); Forward propagation calculation: (in, Is the probability output value of living body, the value range is ; Sigmoid activation function maps linear output to probability distribution; is the dynamic frequency band weight coefficient, calculated by the above formula; 、 They are the low-frequency and high-frequency band feature convolution kernel parameter matrices, respectively, with dimensions of , 、 are the number of input and output channels respectively; is a convolution operator, which uses a two-dimensional convolution operation with a step size of 1; 、 They are respectively the preprocessed low-frequency and high-frequency band radar data tensors, with dimensions of , 、 is the feature map height and width; For the Layer bias term, used to adjust the threshold of the activation function; ReLU is the rectified linear unit activation function, the expression is , used to introduce nonlinear characteristics; is the total number of network convolutional layers. ; Network structure optimization: Using batch normalization layer: (in, For input data; 、 are the mean and variance of the mini-batch data respectively; To prevent the denominator from being zero, take the value ; 、 are learnable scaling and translation parameters; is the normalized data); The Dropout layer uses the probability Randomly drop neurons to prevent overfitting; Transfer learning optimization: Based on a pre-trained 3DCNN model (such as ResNet3D), fine-tune the parameters using the following formula: (in, are the model parameters after fine-tuning; are the pre-training model parameters; is the gradient calculated based on the customs dataset; is the learning rate, with an initial value of 0.001 and an exponential decay strategy, which decays to 0.9 times the original value every 10 training cycles); (3) Dynamic threshold determination unit: Adaptive confidence calibration mechanism: Confidence score calculation: using the formula (in, is the final confidence score, ranging from , a higher value indicates a stronger reliability in detecting living animals; is the probability-characteristic entropy balance coefficient, ; Indicates the ambient temperature value in °C; is the probability of living body existence output by the two-stream network; is the micro-motion eigenvector, ; is the micromotion eigenvector The Shannon entropy is calculated as , used to measure the uncertainty of micromotion characteristics; The final confidence score is calculated based on the energy distribution probability of the characteristic frequency domain In this way, the model output probability and feature stability are balanced, effectively avoiding the misjudgment that may be caused by judging based on only a single indicator; Dynamic threshold setting: through formula (in, is the threshold for determining liveness; is the basic threshold, with a value of 0.6; is the material compensation coefficient, the value is 0.02; is the baggage material density obtained by radar inversion, in units of ) Calculate the liveness determination threshold This formula enables the judgment threshold to be dynamically adjusted according to the different baggage materials, thus better adapting to various complex detection scenarios. Result output and feedback: When the calculated confidence score When a live animal is detected in the baggage, it is determined that the baggage contains a live animal and the detailed test results, including the confidence score and the location of the suspected live animal, are promptly sent to the system management platform to trigger the corresponding baggage sorting, alarm response and other operations. At the same time, the unit will also compare and analyze historical test data with actual quarantine results, and feed the comparison results back to the algorithm for continuous optimization of algorithm parameters and continuous improvement of detection accuracy and reliability. (1) Principle of dual-stream convolutional neural network fusion: Based on the multi-band characteristics of millimeter-wave radar, the two branches of the dual-stream convolutional neural network extract the object's contour, position and other spatial features from the low-frequency data, and capture the micro-motion spectrum features from the high-frequency data; through dynamic weighting , achieving adaptive fusion at the feature level; this innovative design breaks the limitations of traditional single-stream networks, fully combines the advantages of data from different frequency bands, significantly improves the accuracy of identifying living animals, and can more accurately detect living animals from complex radar data; (2) Adaptive confidence calibration principle: Considering that factors such as ambient temperature and luggage material may affect the detection results, this module introduces a temperature adjustment coefficient and material compensation threshold , to achieve dynamic adjustment of confidence scores and judgment criteria; at the same time, using feature entropy Quantify the stability of micro-motion features, effectively avoiding the model's over-reliance on low-quality data and resulting in erroneous judgments, ensuring that the detection results are highly reliable and accurate in various complex environments; (3) Principles of real-time online learning: This module establishes an efficient closed-loop feedback mechanism, using the actual quarantine results returned by the system management platform, such as live or non-live labels confirmed by manual re-inspection, as supervisory signals. Based on this feedback, the weight parameters and confidence calibration coefficients of the dual-stream network are updated online. Through this real-time online learning approach, the algorithm can automatically adapt to differences in inspection environments and changes in baggage types at different ports, continuously improving its detection performance and maintaining optimal working conditions. 4. Module workflow: (1) Data reception and preprocessing stage: The data processing and analysis module receives the raw detection data transmitted by the millimeter-wave radar array module. This data contains radar echo signals in the low-frequency and high-frequency bands and serves as the basis for subsequent analysis. After the data enters the millimeter wave signal preprocessing unit, it undergoes motion compensation, noise filtering, and data normalization. This series of preprocessing steps removes interference and noise from the data, unifies the data scale, and generates standardized data, preparing for subsequent live feature extraction. (2) Liveness feature extraction stage: The preprocessed and standardized data are fed into the low-frequency and high-frequency branches of the improved two-stream convolutional neural network, respectively. Within the network, spatial features and micro-motion features are extracted from the data through multiple layers of convolution and pooling operations. Using dynamic band weight coefficients The extracted high-frequency and low-frequency features are fused so that the network can fully utilize the information of data in different frequency bands; finally, the probability of the existence of living organisms is output through the Sigmoid activation function. , complete the extraction and preliminary judgment of living features; (3) Confidence Assessment and Decision-Making Stage: The dynamic threshold judgment unit is based on the current ambient temperature and luggage material density , calculate the probability-characteristic entropy balance coefficient and the threshold for determining liveness The calculation of these parameters fully considers the influence of the actual testing environment to ensure the rationality of the judgment criteria; Combining the probability of the two-stream network output and Entropy , calculate the final confidence score ,Will and the judgment threshold Perform a comparison and output a clear determination of whether there are living animals in the luggage based on the comparison results; (IV) Result output and feedback stage: Once the detection result is confirmed, the module will send detailed information, including the confidence score and the location of the suspected live animal, to the system management platform. Based on this information, the system management platform will promptly trigger subsequent processes such as baggage sorting and alarm response, ensuring efficient handling of baggage containing live animals. The module collects actual quarantine results and compares and analyzes them with historical detection data. Based on the analysis results, it iteratively optimizes the parameters and confidence calibration mechanism of the dual-stream network to continuously improve the module's detection performance and accuracy, making it better suited to actual application needs. (V) Ending stage: When the system stops running or receives a stop command, the data processing and analysis module will save important information such as the current model parameters, calibration coefficients, and historical detection records, then release the occupied computing resources and enter standby mode, waiting for the next task to start.

[0023] In this embodiment, the baggage sorting control module performs dynamic path optimization including: The sorting priority index is calculated by multiplying the confidence score by the negative exponential decay function of the quarantine channel load; When the real-time load of the channel reaches the benchmark load value, the priority exponential decays to 1 / e of the original value; By binding baggage with inspection data through radio frequency identification technology, the sorting machine is controlled to transfer high-priority baggage to the quarantine area; Furthermore, the baggage sorting control module, as the core execution unit for the system's intelligent customs clearance management, undertakes the critical task of converting detection results into precise sorting actions. The following will elaborate on the technical solutions in the claims from the perspectives of functional architecture, core algorithms, and workflow: 1. Overview of overall functions: The baggage sorting control module receives the live animal detection results (confidence score) output by the data processing and analysis module , combining the real-time load status of the quarantine channel to dynamically optimize the path of baggage on the conveyor belt. Its core functions include calculating sorting priorities based on confidence and channel load, binding baggage and inspection data through radio frequency identification (RFID) technology, accurately controlling the sorting machine to direct baggage containing live animals to the quarantine channel, and simultaneously updating baggage status information on the system management platform to ensure the efficient and orderly operation of the customs clearance process. 2. Submodule composition and functions: (1) Priority calculation unit: Data access and integration: Real-time acquisition of baggage liveness detection confidence scores output by the data processing and analysis module (Value range ), and the real-time load of each quarantine channel fed back by the quarantine process management module (Unit: piece, representing the number of bags waiting for inspection in the current channel), channel benchmark load value (Preset normal channel carrying capacity); Priority algorithm execution: using the formula priority index Calculate the sorting priority of each bag; the formula uses an exponential function to convert the channel load Perform attenuation processing to ensure that high-confidence bags (close to the presence of live animals) or bags in low-load channels receive higher priority; for example, when a bag , where the channel , When the priority index is ; Priority sorting and output: Arrange the calculation results in descending order to generate a real-time sorting queue, bind the priority information with the baggage ID (obtained through RFID), and transmit it to the sorting execution unit; (2) RFID data management unit: Tag read and write control: Before the luggage enters the inspection area, a fixed RFID reader is used to assign a unique electronic tag to each piece of luggage and write basic information about the luggage (such as flight number, passenger ID); after the inspection is completed, additional inspection data (confidence score) is written. , detection time); Data association and tracking: Establish a mapping relationship between baggage ID, inspection results, and sorting routes. By updating tag data in real time, dynamic tracking of bags from inspection to sorting is achieved. For example, when a bag's route needs to be changed due to priority adjustment, the system automatically updates the target channel information in the tag. Anti-collision processing: A time division multiple access (TDMA) algorithm is used to resolve signal conflicts when multiple tags are read and written simultaneously, ensuring data reading and writing accuracy in scenarios with dense baggage transmission. (3) Sorting execution unit: Path planning engine: Based on the real-time status of priority queues and quarantine channels, the Dijkstra shortest path algorithm is used to generate the optimal sorting path. For example, when multiple quarantine channels are available, the channel closest to the current baggage location is prioritized. Equipment collaborative control: Control instructions are sent to sorters (such as cross-belt sorters and swing-arm sorters) via industrial Ethernet (such as the Profinet protocol) to accurately adjust sorter action parameters (such as sorting angle and trigger time). At the same time, the conveyor belt speed controller is linked to ensure that baggage enters the target channel smoothly. Abnormal fault tolerance mechanism: When a mechanical failure or path blockage of the sorting machine is detected, an emergency plan is automatically triggered: the affected baggage is temporarily transferred to the backup buffer, maintenance personnel are notified through the alarm response module, and the fault log of the system management platform is updated at the same time; 3. Key technical principles: (1) Dynamic priority calculation principle: Based on Bayesian decision theory, the confidence score As a basis for the probability of the existence of living organisms, combined with channel load To assess the impact on sorting timeliness, a priority model is constructed using an exponential decay function. This model can dynamically balance quarantine urgency and channel resource utilization. For example, when confidence is high (close to the presence of live animals) but the channel is busy, the priority can be appropriately lowered to avoid excessive channel congestion. (2) RFID full-process tracking principle: Leveraging RFID's contactless identification capabilities, luggage is uniquely linked to inspection data. Real-time updates of tag data allow for the establishment of an electronic traceability chain for the entire process, from inspection and sorting to quarantine. This ensures data traceability and operational auditability, meeting customs regulatory compliance requirements. (3) Sorting path optimization principle: Heuristic search algorithms (such as the Dijkstra algorithm) are combined with real-time path status (channel occupancy, equipment operating status) to dynamically plan the shortest sorting path. By introducing path weight coefficients (such as distance weight and time weight), the baggage transmission efficiency in complex conveyor belt networks is optimized, reducing sorting delays. 4. Module workflow: (1) Initialization phase: Start the RFID reader, sorter and other hardware equipment, and complete equipment self-test and parameter calibration (such as sorter angle calibration and conveyor belt speed calibration); Loads preset configuration parameters, including quarantine channel baseline load values , sorter action parameter thresholds (such as maximum sorting angle, minimum trigger interval), and establish communication connection with the system management platform; (2) Data collection and calculation stage: Receive the detection results of the data processing and analysis module in real time (confidence score and channel load data of the quarantine process management module ( 、 ); The priority calculation unit is based on the formula priority index Calculate the sorting priority of each piece of luggage and generate a sorting queue; (III) Path planning and execution phase: The path planning engine calculates the optimal sorting path based on the priority queue and channel status, and generates control instructions including the target channel ID and sorting trigger time; The sorting execution unit obtains the baggage electronic tag information through the RFID data management unit, verifies the data integrity, and sends the control instruction to the sorting machine to drive the baggage into the corresponding quarantine channel; (IV) Status feedback and update stage: After sorting is completed, the arrival of the baggage is confirmed by the RFID reader installed at the entrance of the channel, and the actual sorting result (success / failure) is fed back to the system management platform; Update the status of the baggage electronic tag (e.g., mark it as "sorted") and record the sorting time and path information in the system database for subsequent tracing and statistical analysis; (V) Ending stage: When the system stops running or receives a stop command, the baggage sorting control module stops data processing and equipment control operations, saves data such as the current priority queue and sorting log, turns off the power of the RFID reader and sorter, and releases system resources.

[0024] In this embodiment, the quarantine process management module includes: A quarantine plan generation unit matches the preset disposal plan library according to the type of living organism; Resource scheduling unit, which allocates quarantine resources based on the luggage space coordinates and confidence scores; Electronic customs clearance form generation unit, which automatically generates electronic release instructions and updates the customs clearance system after quarantine is completed; Furthermore, the quarantine process management module is the core hub of the customs clearance management system to ensure the scientific and efficient implementation of quarantine work. Driven by test results, it realizes the intelligent allocation of quarantine resources and closed-loop management of processes. The following is a detailed explanation of its functional positioning, module composition, technical principles, and workflow: 1. Overview of overall functions: The quarantine process management module is based on the baggage liveness detection results (confidence score) output by the data processing and analysis module. ) and baggage location information, automatically completing quarantine plan matching, dynamic resource scheduling, and electronic customs clearance form generation. Its core functions include: intelligently matching quarantine disposal plans according to the type of live animals, rationally allocating quarantine personnel, equipment and other resources based on optimization algorithms, and automatically generating electronic release instructions and updating the customs clearance system after quarantine is completed, ensuring the traceability and efficiency of the entire quarantine process, providing solid support for customs' prevention and control of biosafety risks. 2. Submodule composition and functions: (1) Quarantine plan generation unit: Identification and classification of live animal types: This module receives characteristic information of suspected live animals from the data processing and analysis module, combines it with historical quarantine data and a species database (including physiological characteristics and risk levels of commonly smuggled live animals), and uses pattern recognition algorithms to preliminarily determine the type of live animal (e.g., mammal, reptile, bird, etc.) and risk level (high risk, medium risk, low risk). Intelligent quarantine plan matching: A quarantine plan library is established, covering standardized handling procedures (such as isolation and observation, sampling and testing, and disinfection) for different types of living organisms and risk levels. Based on the identification results, the optimal quarantine plan is automatically retrieved and matched, and an electronic task sheet is generated with operational steps and a list of required resources. For example, if a high-risk alien species is detected, a strict quarantine and professional laboratory testing plan is automatically matched. Dynamic plan adjustment mechanism: supports manual intervention and plan updates. Quarantine personnel can manually adjust quarantine plans based on actual conditions (such as abnormal health status of live animals). At the same time, the system regularly optimizes the plan library based on the latest quarantine standards and case data to ensure the scientific nature and compliance of the quarantine process. (2) Resource Scheduling Unit: Real-time monitoring of resource status: Connect to various resource management systems at the quarantine site to obtain real-time resource data such as quarantine personnel's work status (busy, idle, on standby), equipment availability (testing instruments, sampling tools, disinfection equipment, etc.), site occupancy information (quarantine rooms, isolation areas), etc., to build a dynamic resource pool; Optimize the execution of the scheduling algorithm: use the formula to schedule the cost (in, The number of luggage to be quarantined; For the Baggage confidence score; Current location of quarantine personnel to luggage The distance in meters (m); The optimal scheduling solution is calculated by taking into account the baggage risk level (confidence score) ), resource accessibility (distance ) and disposal efficiency (time ), prioritize allocating high-confidence luggage to quarantine personnel who are close by and have more experience, to maximize resource utilization efficiency; Dispatching instruction issuance and feedback: Generated resource dispatch instructions (such as assigning quarantine personnel to a specific baggage location or deploying specific equipment to a quarantine area) are sent to relevant execution terminals (such as quarantine personnel's handheld terminals and equipment management systems), and the execution status of the instructions is monitored in real time. If a resource conflict occurs (such as equipment failure or personnel emergency), a re-dispatching mechanism is automatically triggered to ensure uninterrupted quarantine work. (3) Electronic customs clearance form generation unit: Quarantine result data integration: Receive quarantine process records (sampling data, test reports, treatment measures, etc.), on-site monitoring video clips, and relevant data from other modules of the system (such as test time and sorting path) uploaded by quarantine personnel through handheld terminals to form a complete quarantine file; Automated customs clearance decision-making: Based on a pre-set customs clearance rule base (including laws and regulations, health standards, risk thresholds, etc.), quarantine results are automatically reviewed. If quarantine passes, an electronic customs clearance form is automatically generated, including release instructions, quarantine conclusion, validity period, and other information. If quarantine fails, a return or destruction notice is generated, and the baggage status is simultaneously marked as "pending"; Data synchronization and system integration: Electronic customs clearance form data is pushed to the customs clearance system in real time through the application programming interface (API) to update the baggage clearance status. Quarantine files are also encrypted and stored on the blockchain to ensure that the data cannot be tampered with and is fully traceable, meeting customs supervision and audit requirements. 3. Key technical principles: (1) Principle of intelligent matching of quarantine plans: Based on knowledge graph technology, a live animal-quarantine plan association network is constructed, structurally associating knowledge nodes such as species characteristics, risk levels, and quarantine standards. Natural language processing (NLP) technology is used to analyze feature descriptions in test results, quickly searching for matching paths in the knowledge graph to achieve accurate recommendations for quarantine plans, improving the accuracy and efficiency of plan matching. (2) Resource optimization scheduling principle: Using multi-objective optimization theory, a mathematical model is constructed with the goal of minimizing scheduling costs (balancing risk management timeliness and resource consumption). This model is solved using heuristic algorithms (such as genetic algorithms and simulated annealing algorithms) to rapidly search for the global optimal solution in complex resource combinations and task allocation scenarios, ensuring that quarantine resources remain efficiently allocated in a dynamically changing environment. (3) Principles of electronic customs clearance automation: Utilizing smart contract technology, customs clearance rules are converted into automatically executable code logic. When quarantine results meet pre-set conditions, the smart contract automatically triggers the generation of electronic customs clearance forms and system data updates, enabling unmanned operations from quarantine to customs clearance, reducing errors and delays caused by human intervention. 4. Module workflow: (1) Initialization phase: After the quarantine process management module is started, it loads the latest quarantine plan library, customs clearance rule library, species database and other basic data, and completes the system parameter configuration (such as risk level threshold and scheduling algorithm weight coefficient); Establish communication connections with other system modules (data processing and analysis module, baggage sorting control module, and alarm response module) to obtain real-time information on baggage to be quarantined and system status data; (2) Task acceptance and analysis stage: Receive information about luggage containing live animals sent by the baggage sorting control module, including luggage ID and location coordinates , confidence score and preliminary results of the living organism type determination; The quarantine plan generation unit quickly matches the quarantine plan based on the received information and generates an electronic task order; the resource scheduling unit simultaneously analyzes the number, distribution and resource status of the luggage to be quarantined and initiates resource scheduling calculations; (3) Resource dispatch and quarantine implementation phase: The resource scheduling unit issues task instructions to quarantine personnel and equipment management systems based on the scheduling plan determined by the optimization algorithm. Quarantine personnel receive tasks through handheld terminals and go to designated locations to perform quarantine operations. During the quarantine process, quarantine personnel upload sampling data, test results, and other information to the system in real time; the resource scheduling unit continuously monitors resource usage and quarantine progress, and dynamically adjusts scheduling strategies (such as replenishing consumed test reagents and deploying reinforcements); (IV) Customs clearance form generation and feedback stage: The electronic customs clearance form generation unit collects complete quarantine result data, conducts automated review based on customs clearance rules, generates an electronic customs clearance form or processing notice, and pushes it to the customs clearance system and relevant personnel terminals; The entire quarantine process data (including plan execution records, resource usage details, and customs clearance results) is encrypted and stored on the blockchain. At the same time, the baggage status dashboard on the system management platform is updated to show customs management personnel the customs clearance progress. (V) Ending stage: When all the baggage to be quarantined has been processed or the system receives a stop command, the quarantine process management module stops task scheduling and data processing, backs up quarantine files and system logs, releases occupied computing resources, and enters standby mode.

[0025] In this embodiment, the alarm response module adopts a three-level collaborative handling mechanism, including: Level 1 alarm: When the confidence score is greater than 0.9, an audible and visual alarm is triggered and the conveyor belt is frozen; Level 2 alarm: When the confidence score is 0.7 ≤ ≤ 0.9, a pending review label is generated and a manual inspection is notified; Level 3 alarm: When the confidence score is 0.6≤<0.7, only a low-risk alert will be sent to the quarantine management module; The alarm signal is pushed to the customs personnel’s mobile terminal in real time; Furthermore, the alarm response module serves as the security sentinel within the customs clearance management system, shouldering the key responsibility of real-time early warning and rapid handling of abnormal situations. This module triggers a hierarchical alarm mechanism through multi-dimensional risk assessment. Combined with intelligent scheduling and visual command, it establishes an efficient and accurate emergency response system to ensure that customs personnel can respond to various emergencies in a timely manner. The following details the functional architecture, technical implementation, and workflow. 1. Overview of overall functions: The alarm response module is based on the detection results of the data processing and analysis module (confidence score ) and system operating status data to monitor abnormal situations during the customs clearance process in real time; its core functions include: determining the alarm level through a risk threshold algorithm, triggering audible and visual alarms and multi-channel notifications (SMS, email, and app push), linking with the on-site monitoring system to lock the target location, generating a task work order with a disposal plan, and recording the entire alarm process data for subsequent analysis and system optimization; 2. Submodule composition and functions: (1) Risk Assessment Unit: Multi-dimensional threshold detection: Liveness detection risk: based on confidence score Set the three-level threshold ( 、 、 ),when When the warning is triggered, It is upgraded to a medium alarm. Activate advanced alarm when Equipment abnormality monitoring: This system collects operating parameters (such as temperature, current, and vibration frequency) of equipment such as sorters and RFID readers in real time, and uses machine learning algorithms to build equipment health models. When parameters deviate from the normal range and exceed preset thresholds (such as temperature exceeding 75°C or vibration frequency fluctuation exceeding ±15%), it determines that the equipment has a potential failure risk. System operation status: monitors indicators such as system throughput, response time, and data transmission success rate. When the data packet loss rate exceeds 3% or the processing delay exceeds 5 seconds, an abnormal system performance alarm is triggered; Comprehensive calculation of risk level: using the formula risk index (in, Score the severity of the equipment anomaly, ranging from 0 to 10; Score the system operation status, ranging from 0 to 10; 、 、 is the weight coefficient, and , with default values of 0.6, 0.25, and 0.15 respectively) to comprehensively calculate the risk index; this formula uses dynamic weight allocation to ensure that the actual risk level is accurately reflected in different scenarios; (2) Alarm trigger unit: Hierarchical alarm mechanism: Low-level alarm (risk index <6): triggers a yellow warning signal, and the alarm information is only displayed on the system management platform to notify the on-duty personnel to pay attention; Intermediate alarm (6 ≤ risk index < 8): Activate the audible and visual alarm devices (such as flashing lights and sounding buzzers), and send a text message and app push notification to designated personnel, with brief alarm details (such as the location of the suspected living person and the risk level); Advanced alarm (risk index ≥ 8): Based on the intermediate alarm, the system automatically calls the responsible person and sends an email containing a response plan. The on-site monitoring system is linked to focus the camera on the alarm location and start the recording function. Alarm filtering and suppression: A time window mechanism (e.g., within 10 minutes) is used to consolidate repeated alarms of the same type and location to avoid wasting resources. Known temporary anomalies (e.g., parameter fluctuations during equipment maintenance) are filtered through a preset whitelist to reduce false alarm interference. (3) Emergency response unit: Disposal plan generation: Build an emergency response knowledge base and pre-set standardized response processes for different types of alarms (such as live animal alarms and equipment failure alarms). When an alarm occurs, the corresponding response plan is automatically matched and a task work order is generated, including the operation steps, responsible personnel, and required resources. For example, for high-risk live animal alarms, a three-level response process is automatically generated: "Isolate luggage immediately - notify quarantine experts - initiate biosafety protection measures." Resource Scheduling and Collaboration: This system interacts with the quarantine process management module and the baggage sorting control module in real time to adjust resource allocation when an alarm is triggered. For example, when a live animal alarm is detected, quarantine personnel and testing equipment are prioritized, and baggage sorting processes in the relevant areas are suspended to ensure uninterrupted handling. Handling process tracking: QR code technology is used to generate a unique identifier for each alarm task. Handling personnel scan the code to confirm when performing each step of the operation. The system updates the task status (such as received, processing, completed) in real time and records the operation time and executor, forming a complete handling closed loop; 3. Key technical principles: (1) Intelligent threshold adaptive mechanism: Based on historical alarm data and disposal results, the risk threshold is dynamically adjusted using reinforcement learning algorithms (such as 、 、 When the system finds that the false alarm rate of a certain type of alarm is too high, it automatically increases the corresponding threshold; if missed alarms occur frequently, it appropriately lowers the threshold and optimizes the accuracy of alarm triggering through continuous learning; (2) Multimodal alarm fusion technology: Adopting DS evidence theory to integrate alarm information from different sources (such as liveness detection, equipment monitoring, and system logs) for decision-making; by calculating the trust and likelihood of each piece of evidence, a comprehensive judgment is made on whether to trigger an alarm and the final risk level, effectively reducing the potential for misjudgment from a single data source; (3) Emergency Response Knowledge Map: A knowledge graph centered around "alarm type-handling plan-resource requirements" is constructed, associating and storing structured knowledge such as customs emergency response standards, historical cases, and expert experience. When an alarm occurs, the optimal handling path is quickly inferred through a graph neural network, and a handling plan that complies with the latest regulatory requirements is dynamically generated. 4. Module workflow: (1) Initialization phase: After the alarm response module is started, the preset risk threshold parameters (such as 、 、 , normal range of equipment parameters), alarm classification rules and emergency response knowledge base, and complete system configuration initialization; Establish real-time data channels with the data processing and analysis module, baggage sorting control module, and quarantine process management module, and subscribe to key monitoring indicators (such as confidence score and equipment operating status); (2) Monitoring and evaluation stage: The risk assessment unit continuously receives data transmitted by each module, performs real-time analysis on liveness detection results, equipment operating parameters, and system performance indicators, and calculates the risk index; When the risk index exceeds the preset threshold, an alarm event is generated, recording key information such as alarm time, location, type, and risk level, and the event is pushed to the alarm trigger unit; (III) Alarm and disposal stage: The alarm trigger unit activates the corresponding alarm mechanism (sound and light alarm, multi-channel notification) according to the risk level, and sends the alarm details and emergency response plan to the designated personnel terminal; The emergency response unit generates a task order, dispatches relevant resources (such as quarantine personnel and testing equipment) to implement the response plan, and tracks the response progress in real time through QR code scanning. During the response process, if the risk level changes, the alarm level and response measures are automatically adjusted. (IV) Feedback and Archiving Stage: When the disposal is completed, the system automatically collects the disposal results (e.g., live animals have been properly handled, equipment failures have been repaired), updates the alarm status to "resolved," and stores the disposal process data (including operation records, on-site photos, and video clips) in the blockchain for evidence storage; Regularly conduct statistical analysis on alarm data to evaluate the rationality of alarm thresholds and the effectiveness of disposal plans, providing a basis for system parameter optimization and knowledge base updates; (V) Ending stage: When the system stops running or receives a stop command, the alarm response module stops data collection and analysis, saves the current alarm status and system configuration, turns off the sound and light alarm device and communication channel, and releases system resources.

[0026] In this embodiment, the system management platform implements full-process closed-loop control, including: Millimeter wave detection data is connected to the customs clearance system through an application programming interface; Build an electronic traceability chain from testing to release, and use blockchain technology to store data at each link; Generate a heat map of customs clearance efficiency based on confidence scores and quarantine time, visualizing process bottlenecks; Furthermore, the system management platform, serving as the "central nervous system" of the millimeter-wave radar-based intelligent detection and customs clearance management system for live animals, undertakes the core functions of cross-module collaboration, full-process control, and data-intelligent decision-making. By integrating the data and operations of each functional module, it achieves closed-loop management from detection to release, while also utilizing information and intelligent technologies to enhance the transparency and efficiency of customs supervision. The following will elaborate on the functional architecture, core modules, technical principles, and workflow. 1. Overview of overall functions: The data-driven system management platform connects the millimeter-wave radar array module, data processing and analysis module, baggage sorting control module, quarantine process management module, and alarm response module, enabling information sharing and collaborative operations among multiple modules. Its core functions include: real-time monitoring of system operating status and dynamic scheduling of tasks among modules; building an electronic traceability chain to ensure data traceability throughout the customs clearance process; generating visual analysis reports based on test results and quarantine time consumption to provide decision-making support for customs management; and ensuring system data security and legal and compliant operation through permission management and security protection mechanisms. 2. Submodule composition and functions: (1) Data integration and interaction module: Multi-source data access: Establish communication with each functional module through standardized interfaces (such as RESTful API, WebSocket), and receive the raw data collected by the millimeter wave radar and the detection results (confidence score) output by the data processing module in real time. ), multi-source heterogeneous data such as baggage sorting status, quarantine process records and alarm information; Data cleaning and storage: Received data is formatted, missing values are filled, and outliers are filtered. Structured data (such as test results and device status) is stored in a relational database (such as MySQL), and unstructured data (such as radar images and surveillance videos) is stored in a distributed file system (such as Ceph). Indexes are created to support fast queries. Data sharing and distribution: Based on the needs of each module, the integrated data is pushed to the corresponding module in real time. For example, the confidence score and location information of luggage containing live animals are sent to the baggage sorting control module, and the quarantine task list and related resource requirements are passed to the quarantine process management module. (2) Process collaborative control module: Task Scheduling Engine: Dynamically triggers collaborative operations among modules based on detection results. When the data processing and analysis module determines that the baggage contains live animals, it calls the baggage sorting control module to adjust the sorting path, the alarm response module to trigger hierarchical alarms, and the quarantine process management module to allocate quarantine resources, forming an automated "detection-sorting-alarm-quarantine" workflow. Conflict detection and resolution: Real-time monitoring of the execution status of tasks in each module. By establishing a task dependency graph (e.g., quarantine tasks must be initiated after baggage sorting is completed), resource conflicts during concurrent execution of multiple tasks are avoided. If conflicts arise (e.g., multiple quarantine tasks competing for the same equipment), task priorities are automatically adjusted or resources reallocated. Exception handling mechanism: When a module fails (such as a sorting machine freeze or quarantine equipment failure), the relevant tasks are immediately suspended, an exception message is sent to the alarm response module, and an emergency plan is initiated (such as activating an alternative sorting path or deploying backup equipment) to ensure that the overall operation of the system is not seriously affected; (III) Electronic traceability and supervision module: Blockchain Evidence Storage System: Utilizing consortium blockchain technology, data from the entire process from millimeter-wave radar detection to final release (test records, sorting logs, quarantine reports, customs clearance forms, etc.) is encrypted and stored on-chain. Each data node contains information such as timestamp, operator, and data hash value, ensuring that the data cannot be tampered with and is traceable. Traceability Query Service: This service provides customs supervisors with a visual traceability interface, enabling quick query of the customs clearance history of specific luggage by baggage ID, time range, operator, and other criteria. For example, the service can trace the inspection time, sorting route, quarantine results, and handling personnel of a particular piece of luggage, providing a basis for law enforcement audits and accountability tracing. Compliance Check: A built-in library of customs regulations and business rules (such as the Law of the People's Republic of China on Import and Export Animal and Plant Quarantine) automatically compares customs clearance process data with regulatory requirements. If any violations are found (such as direct release without quarantine as required), an alert is immediately triggered and a rectification work order is generated. (IV) Data analysis and decision-making module: Visual monitoring screen: Through visualization tools such as ECharts and D3.js, the system operation status (such as the current number of detected luggage, the load of each quarantine channel, alarm statistics), customs clearance efficiency heat map (based on confidence score) is displayed in real time. and quarantine time) and abnormal event distribution, helping managers quickly grasp the overall situation; Intelligent analysis model: Uses machine learning algorithms (such as time series analysis and cluster analysis) to mine historical data and predict future business trends (such as changes in live animal detection rates during peak hours) and equipment failure probabilities (based on abnormal trends in operating parameters), providing decision support for resource allocation and preventive maintenance. Report generation and export: Regularly generate various statistical reports (such as monthly live animal detection reports and comparison tables of customs clearance efficiency at various ports), and support export in Excel, PDF and other formats to meet the needs of customs business statistics and reporting to higher authorities; (V) System security and rights management module: Identity authentication and authorization: Multi-factor authentication (username / password + dynamic token) is used to ensure the legitimacy of user identities. The role-based access control (RBAC) model assigns differentiated permissions to different users (such as inspectors, quarantine officers, and administrators) (for example, inspectors can only view test data, while administrators can modify system configurations). Data encryption and protection: SSL / TLS encryption protocols are used for transmitted data, and AES-256 encryption algorithms are used for stored data. Intrusion detection systems (IDS) and firewalls are deployed to monitor network attacks in real time to prevent data leakage or malicious tampering. Log auditing and backup: Record all user operation logs (login time, data access, configuration modification), and conduct regular security audits; establish an off-site disaster recovery backup mechanism, perform daily incremental backups of core data, and weekly full backups to ensure data security and system reliability; 3. Key technical principles: (1) Microservice architecture collaboration principle: The system management platform adopts a microservices architecture, splitting each functional module into independently running services (such as data integration services and process control services). Efficient interaction between services is achieved through lightweight communication protocols (such as gRPC). This architecture improves the scalability and fault tolerance of the system. A service failure will not affect the overall operation, and it facilitates the addition of new functional modules or the upgrade of existing services. (2) Principle of blockchain electronic traceability: Based on the blockchain's distributed ledger and consensus mechanism, customs clearance data is packaged into blocks in chronological order and linked into chains using a hash algorithm. Each block contains the hash value of the previous block, ensuring data integrity and immutability. The consortium chain model limits data read and write access to authorized nodes (such as customs departments), meeting regulatory data privacy and compliance requirements. (3) Principles of intelligent decision-making analysis: Utilizing big data analysis technology, massive amounts of historical data are cleansed, aggregated, and modeled. By training predictive models (such as LSTM-based equipment failure prediction models and random forest-based live animal risk assessment models) and combining them with real-time data input, intelligent analysis of system operating status and trend prediction are achieved, assisting customs management personnel in making scientific decisions. 4. Module workflow: (1) Initialization phase: After the system management platform is started, it loads the configuration file (database connection parameters, service interface address, permission policy, etc.) and completes the initialization of each sub-module; Establish connections with external modules such as the millimeter-wave radar array module and the data processing and analysis module, and subscribe to key data topics (such as test results and device status changes); (2) Data processing and interaction stage: The data integration and interaction module continuously receives data sent by each functional module, cleans and converts it, stores it in the corresponding database, and distributes it to other modules as needed; If a test result containing live animals is received, the process collaborative control module immediately triggers baggage sorting, alarming, and quarantine tasks, generates task scheduling instructions, and sends them to relevant modules; (III) Process monitoring and management stage: Monitor the execution status of tasks in each module in real time, ensure tasks are executed in sequence through the process coordination control module, and handle task conflicts and abnormal situations; The electronic traceability and supervision module stores data from all links on the chain in real time and updates baggage traceability information. At the same time, it checks process compliance against the regulatory database and issues are promptly reported. (IV) Data analysis and decision-making stage: The data analysis and decision-making module regularly extracts data from the database and generates visual reports and analysis reports to display system operation status and business trends; Mining historical data based on intelligent analytical models to predict potential risks (such as equipment failures and decreased customs clearance efficiency), and delivering early warning information and optimization suggestions to managers; (V) Security and maintenance stage: The system security and permission management module continuously monitors user operations and network security, records logs and conducts audits; regularly backs up data to ensure safe and stable system operation; Adjust the configuration, update functions, or expand services of each sub-module based on business needs or system upgrade requirements, such as adding new data interfaces and optimizing task scheduling algorithms; (6) Ending stage: When the system stops running or receives a stop command, the system management platform closes the connection with the external module, saves the current running status and configuration parameters, releases resources, and stops all services.

[0027] While 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 these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. Millimeter-wave radar-based intelligent detection and customs clearance management system for living animals, characterized by: It includes millimeter-wave radar array module, data processing and analysis module, baggage sorting control module, quarantine process management module, alarm response module and system management platform; The millimeter-wave radar array module is deployed at key nodes of the customs baggage conveyor belt to scan the biological characteristics of living people inside the luggage in real time; The data processing and analysis module identifies the presence of living animals through a millimeter wave feature extraction algorithm and outputs the detection results to the system management platform; The system management platform dynamically triggers the following collaborative processes based on the detection results: The baggage sorting control module adjusts the sorting path in real time to direct baggage containing live animals to the quarantine channel; Generate hierarchical alarm signals through the alarm response module and associate them with quarantine priorities; Quarantine resources are automatically allocated and electronic quarantine task orders are generated through the quarantine process management module.

2. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 1 is characterized by: The millimeter-wave radar array module adopts multi-band fusion detection technology, specifically including: The low-frequency band is used to penetrate the luggage material to obtain the outline of the living body. The low-frequency band frequency is between 24-33GHz; The high-frequency band is used to capture micro-motion features, and the frequency of the high-frequency band is between 60-90GHz; A living biological feature model is established based on dual-band data fusion to distinguish living animals from static interference objects.

3. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 1 is characterized in that: The data processing and analysis module includes: Millimeter wave signal preprocessing unit, which performs motion compensation and noise filtering on radar echoes; The living feature recognition unit extracts biological micro-motion spectrum features through an improved two-stream convolutional neural network model; The dynamic threshold judgment unit outputs a liveness confidence score based on an adaptive confidence calibration mechanism.

4. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 3 is characterized by: In the improved two-stream convolutional neural network model: The probability of the existence of a living body is obtained by calculating the weighted sum of the outputs of each convolutional layer using the Sigmoid function. The output of each convolution layer is generated by adding the low-frequency and high-frequency band data after being processed by the convolution kernel respectively, and then through the ReLU activation function; The dynamic frequency band weight coefficient is determined based on the exponential normalized value of the characteristic signal-to-noise ratio of each layer, where the signal-to-noise ratio weight is adjusted by the ambient noise energy value: The environmental interference adjustment factor approaches 1 as the environmental noise energy increases, and the material attenuation empirical constant ranges from 0.1 to 0.

5.

5. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 3 is characterized by: In the adaptive confidence calibration mechanism: The final confidence score is the weighted sum of the probability of existence of living organisms and the Shannon entropy of micro-motion characteristics; The weighting coefficient is dynamically adjusted as the ambient temperature changes. For every 1 degree Celsius increase in temperature, the probability value weight increases by 0.

05. The liveness threshold is determined by adding a compensation factor of the square of the material density to the base threshold of 0.6, and the material compensation factor is fixed at 0.

02. When the final confidence score exceeds the dynamic threshold, it is determined that the baggage contains live animals.

6. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 1, characterized in that: The baggage sorting control module performs dynamic path optimization including: The sorting priority index is calculated by multiplying the confidence score by the negative exponential decay function of the quarantine channel load; When the real-time load of the channel reaches the benchmark load value, the priority exponential decays to 1 / e of the original value; By binding luggage and detection data through radio frequency identification technology, the sorting machine is controlled to transfer high-priority luggage to the quarantine area.

7. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 1, characterized in that: The alarm response module adopts a three-level collaborative disposal mechanism, including: Level 1 alarm: When the confidence score is greater than 0.9, an audible and visual alarm is triggered and the conveyor belt is frozen; Level 2 alarm: When the confidence score is 0.7 ≤ ≤ 0.9, a pending review label is generated and a manual inspection is notified; Level 3 alarm: When the confidence score is 0.6≤<0.7, only a low-risk alert will be sent to the quarantine management module; The alarm signal is pushed to the customs personnel’s mobile terminal in real time.

8. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 1, characterized in that: The quarantine process management module includes: A quarantine plan generation unit matches the preset disposal plan library according to the type of living organism; Resource scheduling unit, which allocates quarantine resources based on the luggage space coordinates and confidence scores; The electronic customs clearance form generation unit automatically generates electronic release instructions and updates the customs clearance system after quarantine is completed.

9. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 8, characterized in that: The resource scheduling unit executes an optimization algorithm, including: The scheduling cost function minimizes the weighted response time of all bags to be quarantined; The weight factor is the ratio of the confidence score to the distance between the quarantine officer and the luggage; The response time is calculated based on the average time consumption of similar historical quarantines.

10. The millimeter-wave radar-based intelligent detection and customs clearance management system for living animals according to claim 1, characterized in that: The system management platform implements full-process closed-loop control, including: Millimeter wave detection data is connected to the customs clearance system through an application programming interface; Build an electronic traceability chain from testing to release, and use blockchain technology to store data at each link; Generate a customs clearance efficiency heat map based on confidence scores and quarantine time to visually display process bottlenecks.

Images