Metal detector door metal classification identification and detection system

Abstract

This invention provides a metal classification and detection system for a metal detection gate, comprising: a passage status acquisition module, a near-zero differential magnetic field construction module, a magnetic field arrangement control module, an intelligent quantum magnetic field measurement module, a non-commutative residual extraction module, and a metal classification output module. The system generates a paired magnetic field arrangement trigger sequence based on the passage position and speed of the person being tested. In an empty gate state, it performs quantum magnetic field measurement using intelligent sensor nodes and constructs a near-zero differential magnetic field surface within the passage area using a compensation coil group. The system sequentially performs transverse pre-excitation and longitudinal near-zero field sweep, and longitudinal pre-excitation and transverse near-zero field sweep, extracting the first and second near-zero field fission trajectories. After aligning the two trajectories, it obtains the instantaneous non-commutative residual and the delayed fission closure residual, thereby determining the ferromagnetic material region, candidate internal region, and conductive metal category.

Description

Technical Field

[0001] This invention relates to the field of metal detection and classification technology, and in particular to a metal detection gate metal classification, identification and detection system. Background Technology

[0002] Metal detection gates typically use excitation and receiving coils located on both sides of the gate frame. When a person passes through the detection area, an alternating magnetic field is applied, and the gate determines whether the person is carrying metal based on the magnetic field disturbance caused by a metallic object. However, in areas where copper, silver, or other metal materials are processed, simply determining whether a person is carrying metal is insufficient for practical management needs.

[0003] Workers typically carry keys, belt buckles, steel-toed safety shoes, and steel tools, all of which generate a metallic response. If the system alarms for all metallic responses, it risks frequent false alarms; conversely, raising the alarm threshold might miss smaller copper pieces or particles. Furthermore, smuggled copper may be housed in metal boxes, steel tool casings, or steel-toed safety shoes, causing the strong response from the outer ferromagnetic material to mask the weaker response from the inner copper. Existing metal detection systems tend to mistake composite targets for ordinary iron objects, making it difficult to determine whether copper is present inside or near the ferromagnetic material.

[0004] Furthermore, the passage speed, carrying position, and orientation of metal objects change when people pass through metal detector gates. The magnetic field disturbances generated by the same metal object at different positions and orientations vary, and classification based solely on amplitude, phase difference, or fixed thresholds is easily affected by changes in spatial position and human movement. Even with highly sensitive magnetic field measuring devices, only a mixed signal formed by the superposition of multiple metal responses can be obtained, making it difficult to stably separate the responses of ferromagnetic materials from those of the internal conductive metals.

[0005] Therefore, this invention proposes a metal detection gate metal classification, identification, and detection system. The information disclosed in the background section is only for enhancing understanding of the background of this disclosure and may therefore contain prior art information that is not common knowledge to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by providing a metal detection gate metal classification, identification, and detection system, thereby solving the technical problems mentioned in the background section.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A metal detection gate metal classification, identification and detection system includes:

[0009] The passage status acquisition module is used to acquire the start time, end time, passage position and passage speed of the person to be tested by through-beam photoelectric sensors set on the entrance and exit sides of the metal detection gate and distance sensors set on the top crossbeam, generate passage status data, and generate a pair of magnetic field arrangement triggering sequence that includes the first detection sub-cycle, the protection interval and the second detection sub-cycle in sequence.

[0010] The near-zero differential magnetic field construction module is used to acquire background magnetic field data in the empty gate state, adjust the driving current of the compensation coil group according to the background magnetic field data, construct the initial near-zero differential magnetic field surface in the passage area, generate the sweep control sequence, and generate the first empty gate in-sequence reference magnetic field data and the second empty gate in-sequence reference magnetic field data respectively.

[0011] The magnetic field arrangement control module is used to sequentially execute the lateral pre-excitation, the first arrangement transition period, and the longitudinal near-zero field sweep during the first detection sub-cycle to form the first magnetic field arrangement. During the second detection sub-cycle, it sequentially executes the longitudinal pre-excitation, the second arrangement transition period, and the lateral near-zero field sweep to form the second magnetic field arrangement, and establishes a magnetic field arrangement measurement correspondence table.

[0012] The intelligent quantum magnetic field measurement module is used to control each intelligent sensor node to perform quantum magnetic field measurement under the first magnetic field arrangement and the second magnetic field arrangement according to the magnetic field arrangement measurement correspondence table. It subtracts the corresponding empty gate in-sequence reference magnetic field data to obtain the node target perturbation magnetic field value, and restores the near-zero differential magnetic field surface state according to the node target perturbation magnetic field value, and extracts the first near-zero field fission trajectory and the second near-zero field fission trajectory.

[0013] The non-commutative residual extraction module is used to align the first near-zero field fission trajectory and the second near-zero field fission trajectory based on the relative position coordinates of the personnel and the normalized sweep coordinates or normalized transition time coordinates, calculate the non-commutative fission trajectory residuals, and separate the instantaneous non-commutative residuals and delayed fission closure residuals.

[0014] The metal classification output module is used to determine ferromagnetic material regions and suspected composite target regions based on instantaneous non-exchange residuals, generate candidate internal regions, identify copper, silver or other non-ferromagnetic metals based on delayed fission closure residuals, and output metal classification detection results based on the spatial correlation between the conductive metal classification results and the candidate internal regions.

[0015] The passage status acquisition module is used to: determine the start time, end time, and passage direction of the person to be tested through through-beam photoelectric sensors on the entrance and exit sides; collect the passage distance sequence through the distance sensor on the top beam; and remove records of reverse passage, overlapping passage, and passage exceeding the time limit; perform continuity verification on the passage distance sequence and perform linear interpolation when the number of consecutive abnormal sampling points does not exceed the preset upper limit to generate passage status data; and divide the duration within the effective detection range into at least one paired detection cycle based on the passage status data. Each paired detection cycle includes a first detection sub-cycle, a protection interval, and a second detection sub-cycle in sequence, generating a paired magnetic field arrangement triggering sequence.

[0016] The near-zero differential magnetic field construction module is used to: control multiple smart sensor nodes to synchronously perform quantum magnetic field measurements during the empty gate period; correct the quantum magnetic field measurement results based on local temperature and reference magnetic field changes to generate background magnetic field data; adjust the driving current of the compensation coil group according to the background magnetic field data and the magnetic field influence coefficients corresponding to each compensation coil; connect the positions with magnetic field strengths that meet the preset near-zero magnetic field threshold and are spatially adjacent to each other to form the initial near-zero differential magnetic field surface; divide the transverse sweep path and the longitudinal sweep path into multiple sweep positions; set the driving current of each coil group in sequence; generate and verify the sweep control sequence; and generate the first empty gate in-sequence reference magnetic field data and the second empty gate in-sequence reference magnetic field data in the empty gate state.

[0017] The magnetic field arrangement control module is used to: sequentially execute lateral pre-excitation, the first arrangement transition period, and longitudinal near-zero field sweep during the first detection sub-cycle, generating the first magnetic field arrangement control record; sequentially execute longitudinal pre-excitation, the second arrangement transition period, and lateral near-zero field sweep during the second detection sub-cycle, generating the second magnetic field arrangement control record; ensure that the first and second magnetic field arrangements use the same pre-excitation field integral constraints, sweep position spacing, sweep position dwell time, quantum magnetic field measurement sampling period, allowable position error, and current error threshold, and filter out invalid cycles based on the displacement estimate of the person under test in the paired detection cycles; and associate the passage status data, sweep position, drive current of each coil group, and quantum magnetic field measurement trigger time based on a unified clock to generate a magnetic field arrangement measurement correspondence table.

[0018] The intelligent quantum magnetic field measurement module is used to: control multiple intelligent sensor nodes to perform quantum magnetic field measurements under the first and second magnetic field arrangements, correct for temperature drift and subtract reference magnetic field changes from the quantum magnetic field measurement results, and subtract the first and second empty gate in-sequence reference magnetic field data respectively to obtain the node target perturbation magnetic field value; determine the near-zero magnetic field position based on the effective magnetic field value of adjacent intelligent sensor nodes, and connect spatially adjacent near-zero magnetic field positions to form a near-zero differential magnetic field surface state; and extract the first and second near-zero field fission trajectories by recording the spatial offset distance of the near-zero differential magnetic field surface, the number of near-zero magnetic field regions, the splitting time, the closing time, the splitting duration, the closing delay time, and the delay recovery time.

[0019] The non-exchangeable residual extraction module is used to: pair effective trajectory points in the first and second near-zero field fission trajectories based on the relative position coordinates of personnel and normalized sweep coordinates or normalized transition time coordinates to form aligned fission trajectory pairs; calculate the difference in spatial offset distance, number of near-zero magnetic field regions, node target disturbance magnetic field value, splitting duration, closure delay time and delayed recovery time in each aligned fission trajectory pair, and obtain the comprehensive non-exchangeable residual value after normalization; and collect the non-exchangeable fission trajectory residuals that are greater than the effective residual threshold and appear continuously in no less than 3 trajectory point pairs into instantaneous non-exchangeable residuals and delayed fission closure residuals according to the time period to which the data belongs.

[0020] The metal classification output module is used to: connect spatially adjacent and temporally continuous instantaneous non-commutative residuals into instantaneous residual regions; extract instantaneous residual feature vectors and compare them with the feature range of single iron instantaneous residuals to determine ferromagnetic material regions and suspected composite target regions; generate candidate internal regions based on ferromagnetic material regions and suspected composite target regions; extract delayed residual feature vectors from delayed fission closure residuals, calculate their category distances with copper, silver, and other non-ferromagnetic metals, and generate conductive metal classification results by combining them with the continuity consistency ratio; and output metal classification detection results corresponding to copper-clad targets, silver-clad targets, single ferromagnetic targets, bare copper targets, bare silver targets, or composite metal targets to be verified, based on the spatial correlation between the conductive metal classification results and the candidate internal regions.

[0021] The beneficial effects of this invention are as follows:

[0022] This invention verifies the continuity of the passage distance sequence and generates a pair of magnetic field triggering sequences based on the passage position and speed of the personnel being tested. This reduces the impact of changes in personnel passage speed, abnormal stops, and deviations in passage position on the detection results, thus improving the stability of metal classification and identification. By utilizing intelligent sensor nodes to perform quantum magnetic field measurements, combined with temperature correction, reference magnetic field change subtraction, and compensation coil group adjustment, a controllable near-zero differential magnetic field surface is constructed within the passage area. This suppresses background magnetic field interference caused by equipment start-up and shutdown, power transmission lines, and moving metal objects in the surrounding area.

[0023] This invention obtains the first and second near-zero field fission trajectories by sequentially performing transverse pre-excitation and longitudinal near-zero field sweep, and longitudinal pre-excitation and transverse near-zero field sweep, and extracts the non-exchange fission trajectory residuals. This enhances the difference between the mixed responses of ferromagnetic materials and internal conductive metals, and reduces the shielding effect of the outer ferromagnetic material on the response of the internal copper material. The ferromagnetic material region and suspected composite target region are determined based on the instantaneous non-exchange residuals, and copper, silver, or other non-ferromagnetic metals are identified based on the delayed fission closure residuals. This enables the differentiation of different metal categories, rather than simply outputting an alarm result indicating the presence of metal.

[0024] This invention, by analyzing the spatial correlation between conductive metal classification results and candidate internal regions, outputs targets for verification, including copper-clad targets, exposed copper targets, silver-clad targets, exposed silver targets, single ferromagnetic targets, or composite metal targets. This improves the ability to identify concealed smuggling activities in copper processing sites and reduces invalid alarms caused by mobile phones, keys, belt buckles, and steel tools. By verifying the classification results using in-sequence reference magnetic field data, effective residual thresholds, and continuous consistency ratios, isolated noise and transient fluctuations can be eliminated, improving the reliability of metal classification detection results. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the process of the metal classification, identification and detection system according to an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Example:

[0028] In this embodiment, the metal detection gate includes a left column, a right column, and a top beam. A passageway for personnel to pass through is formed between the left and right columns. The lateral direction refers to the direction from the left column to the right column, the longitudinal direction refers to the direction from the entrance side to the exit side of the metal detection gate, and the height direction of the gate frame refers to the direction from the bottom of the gate frame to the top beam.

[0029] Before the system was put into operation, calibration was completed using empty door state, single copper material sample, single silver material sample, single iron material sample, copper material sample with iron shell covering, copper material sample with steel tool shell inside, mobile phone, key, belt buckle and steel toe safety shoe, and calibration parameter library was established.

[0030] Each sample was repeatedly collected under different heights, placement orientations, and passage speeds. The calibration parameter library is used to store background magnetic field correction parameters, compensation coil drive parameters, sweep control parameters, empty gate in-sequence reference magnetic field data, residual normalization scale, residual weights, characteristic parameters of various metals, and classification thresholds.

[0031] like Figure 1 As shown, this embodiment provides a metal detection gate metal classification, identification, and detection system, which includes:

[0032] The passage status acquisition module is used to acquire the start time, end time, passage position, and passage speed of the person being tested through through-beam photoelectric sensors installed on the entrance and exit sides of the metal detection gate and a distance sensor installed on the top crossbeam. This generates passage status data and a paired magnetic field sequence triggering timing, which sequentially includes the first detection sub-cycle, the protection interval, and the second detection sub-cycle. Specifically, this includes:

[0033] S110. Two sets of through-beam photoelectric sensors are installed on the entrance and exit sides of the metal detection gate, spaced apart along the longitudinal direction. A distance sensor is installed on the top crossbeam, ensuring its detection range covers the central part of the passage area. When the through-beam photoelectric sensor on the entrance side generates an obstruction signal first, followed by the through-beam photoelectric sensor on the exit side, it is determined that the person to be detected has entered the passage area in the prescribed direction.

[0034] The start time is defined as the moment when the through-beam photoelectric sensor on the entrance side first generates an obstruction signal, and the end time is defined as the moment when the through-beam photoelectric sensor on the exit side releases the obstruction signal. The distance sensor collects human contour data according to a preset sampling period, selecting effective measurement points within the torso area from 0.8m to 1.5m above the ground. The median value of these effective measurement points is used as the travel distance at the corresponding sampling time, forming a travel distance sequence according to the sampling time order. Using the median value reduces the impact of arm swing, clothing swaying, and backpack edges on the travel distance. The sampling period of the distance sensor can be set from 10ms to 20ms.

[0035] When the through-beam photoelectric sensor on the exit side generates an obstruction signal first, the current record is marked as a reverse passage record; when the person being tested has not yet left the passage area and a new obstruction change occurs again on the entrance side, the current record is marked as an overlapping passage record; when the difference between the end time and the start time exceeds the preset maximum passage time, the current record is marked as a timeout passage record. The preset maximum passage time can be set to 5 seconds.

[0036] The validity status of data for reverse passage records, overlapping passage records, and timeout passage records is set to invalid, and they do not proceed to S120; a passage event record is generated, which includes the test subject number, start time, end time, passage direction, passage distance sequence, and data validity status, and is sent to S120.

[0037] S120: Read the passage event records output by S110 that are valid, and perform continuity verification on the passage distance sequence. Record the k-th acquisition time as... The corresponding travel distance is recorded as When the following formula is satisfied, the k-th sampling point is marked as an abnormal sampling point:

[0038]

[0039] Where k represents the sampling point number; and Representing the k-th acquisition time and the k-th acquisition time respectively The travel distance corresponding to each data collection time; This indicates the speed limit that is allowed to be reached; This indicates the distance sensor measurement error and the position tolerance caused by local human body movements. The maximum allowable speed can be set to 2.0 m / s, and the position tolerance can be set to 0.05 m.

[0040] When the number of consecutive abnormal sampling points does not exceed 3, linear interpolation is performed according to the following formula:

[0041]

[0042] in, This indicates the corrected travel position corresponding to the k-th acquisition time. and These represent the travel distances (i.e., the original travel positions where no anomalies occurred) corresponding to the nearest valid sampling points before and after the abnormal sampling interval, respectively. and These represent the corresponding data collection times. When the number of consecutive abnormal sampling points exceeds three, the valid status of the corresponding passage event record will be updated to invalid.

[0043] After processing the abnormal sampling points, the passage speed is calculated according to the following formula:

[0044]

[0045] in, This represents the traffic speed at the k-th data collection time. This represents the corrected passage position corresponding to the (k-1)th acquisition time; for valid sampling points not marked as abnormal, their corrected passage positions are... Directly equal to the original travel distance at the corresponding moment .

[0046] The data of the person to be tested, the start time, the end time, each collection time, the corresponding passage position at each collection time, the passage speed at each collection time, and the data validity status are combined into passage status data and sent to S130, S330, and S510.

[0047] S130: Read the valid passage status data output from S120. Based on the start time, end time, passage position, and passage speed, determine the duration for which the person under test is within the valid detection range. Divide the duration into at least one paired detection cycle. Each paired detection cycle includes a first detection sub-cycle, a guard interval, and a second detection sub-cycle. The first detection sub-cycle is used to execute the first magnetic field arrangement, and the second detection sub-cycle is used to execute the second magnetic field arrangement. The two detection sub-cycles use the same duration, sampling period, and magnetic field strength constraints, only exchanging the pre-excitation direction and the near-zero field sweep direction.

[0048] The protection interval is used to wait for the residual magnetic field to decay after the first magnetic field arrangement is completed. At the end of the protection interval, the following should be met:

[0049]

[0050] in, This indicates the residual magnetic field strength within the passage area at the end of the protection interval; This indicates the peak magnetic field strength corresponding to the first magnetic field arrangement; This indicates the allowable residual percentage.

[0051] The allowable residual ratio can be set to 1% to 3%, with the specific value determined through gate calibration. If the duration of the person being tested within the effective detection range is insufficient to complete one paired detection cycle, the corresponding passage status data will be marked as insufficient detection time, a deceleration passage prompt will be output, and subsequent magnetic field arrangement will not be executed.

[0052] A pair of magnetic field arrangement triggering timing sequences are generated. The pair of magnetic field arrangement triggering timing sequences include the personnel number to be tested, the pair of detection cycle numbers, the start and end times of the first detection sub-cycle, the start and end times of the protection interval, the start and end times of the second detection sub-cycle, the corresponding passage position, and the corresponding passage speed, and are sent to S310, S320, S330, S410, and S420.

[0053] The near-zero differential magnetic field construction module includes a transverse excitation coil group, a longitudinal excitation coil group, a compensation coil group, and multiple intelligent sensor nodes. Each intelligent sensor node includes a quantum magnetic field measurement channel using an RF optically pumped atomic magnetometer, a reference magnetic field acquisition unit, a temperature detection unit, and an edge processing unit.

[0054] The near-zero differential magnetic field construction module is used to acquire background magnetic field data in the empty gate state. Based on the background magnetic field data, it adjusts the drive current of the compensation coil group to construct an initial near-zero differential magnetic field surface in the passage area, generates a sweep control sequence, and generates the first empty gate in-sequence reference magnetic field data and the second empty gate in-sequence reference magnetic field data, specifically including:

[0055] When neither the S210, entrance-side through-beam photoelectric sensor nor the exit-side through-beam photoelectric sensor detects an obstruction signal, and there are no pending passage events, the current period is designated as an empty door period. Once the empty door period lasts for the preset background data acquisition duration, the intelligent sensor nodes located on the left column, right column, and top beam synchronously perform quantum magnetic field measurements. The preset background data acquisition duration can be set from 300ms to 800ms.

[0056] Each smart sensor node has a unique node number and a fixed spatial installation location, and includes three mutually orthogonal quantum magnetic field measurement channels, a reference magnetic field acquisition unit, a temperature detection unit, and an edge processing unit. The three quantum magnetic field measurement channels are used to acquire magnetic field components in the lateral, longitudinal, and door frame height directions, respectively.

[0057] The quantum magnetic field measurement channel employs an RF optically pumped atomic magnetometer. When the RF optically pumped atomic magnetometer is a single-axis device, each smart sensor node is equipped with three RF optically pumped atomic magnetometers with mutually orthogonal sensing axes. The reference magnetic field acquisition unit is used to acquire changes in the reference magnetic field caused by the start-up and shutdown of equipment in the plant area, power transmission lines, and moving metal objects in the surrounding area. The temperature detection unit is used to acquire the local temperature.

[0058] The edge processing unit calculates the node background magnetic field value according to the following formula:

[0059]

[0060] Where i represents the smart sensor node number; r represents the direction of the magnetic field component; This represents the background magnetic field value of the i-th smart sensor node in direction r; This represents the measured value of the original quantum magnetic field in the corresponding direction; This represents the temperature correction factor; Indicates the current local temperature; Indicates the reference temperature; Indicates the reference magnetic field correction factor; This indicates the change in the reference magnetic field in the corresponding direction.

[0061] The temperature correction coefficient and the reference magnetic field correction coefficient are obtained through the empty gate calibration. The specific process is as follows: During the empty gate reference period when no person to be tested passes through and there are no moving metal objects in the vicinity, the system synchronously records the ambient temperature change curve and the reference magnetic field change curve of each smart sensor node; the least squares method is used to perform multiple linear regression fitting on the original quantum magnetic field measurement value, local temperature and reference magnetic field change, and the temperature bias slope corresponding to each smart sensor node in each direction is obtained as the temperature correction coefficient, and the reference magnetic field coupling ratio is obtained as the reference magnetic field correction coefficient.

[0062] If, within the preset background data acquisition period, the fluctuation range of the background magnetic field value of any smart sensor node exceeds the preset background stability threshold, the current background magnetic field data will be marked as invalid.

[0063] Output valid background magnetic field data, including node number, spatial installation location, acquisition time, node background magnetic field values ​​in three directions, local temperature, and background magnetic field stability status, and send it to S220.

[0064] S220: Read the background magnetic field data output from S210, ensuring the background magnetic field is stable and valid. Also, read the magnetic field influence coefficients of each compensation coil on each smart sensor node from the calibration parameter library. The magnetic field influence coefficient represents the change in the magnetic field value at the location of the corresponding smart sensor node when the driving current of a compensation coil changes by a unit current.

[0065] The controller uses a preset starting position within the passage area as the target area and constructs a Jacobian sensitivity matrix based on the magnetic field influence coefficients of each compensation coil on each smart sensor node. Based on the deviation of the current magnetic field strength of each node from the preset near-zero magnetic field, and combined with the Jacobian sensitivity matrix, the controller uses a damped least squares algorithm to calculate the update step size of the drive current for each compensation coil. The controller then performs multi-variable collaborative iterative adjustment of the drive current of the compensation coil group to bring the magnetic field strength within the target area into the preset near-zero magnetic field threshold range. The preset near-zero magnetic field threshold is determined according to the following formula:

[0066]

[0067] in, This indicates the preset near-zero magnetic field threshold; This represents the average value of the absolute value of the magnetic field residual under the empty gate compensation state; The standard deviation represents the absolute value of the magnetic field residual; This represents the threshold coefficient, which can be set to 3 to 5.

[0068] After each adjustment, the smart sensor node re-executes the quantum magnetic field measurement. When the absolute value of the magnetic field strength corresponding to each smart sensor node within the target area is not greater than the preset near-zero magnetic field threshold, the adjustment stops, and the current driving current of each compensation coil is determined as the compensation current parameter. Near-zero magnetic field positions that meet the threshold requirement and are spatially adjacent are combined according to their connection relationship to form an initial near-zero differential magnetic field surface. The initial near-zero differential magnetic field surface refers to the low magnetic field cross-section formed by connecting spatial positions where the absolute value of the magnetic field strength is not greater than the preset near-zero magnetic field threshold under the empty-gate compensation state.

[0069] If the threshold requirement is still not met after reaching the preset maximum number of iterations, or if the driving current of any compensation coil exceeds the rated safety range, the current construction process is marked as invalid, and the process returns to S210 to re-acquire background magnetic field data. The compensation current parameters, the initial near-zero differential magnetic field surface and its effective state are output and sent to S230, S310 and S320.

[0070] S230: When the initial near-zero differential magnetic field surface is in an effective state, the compensation current parameters determined in S220 are invoked to divide the preset sweep path into sweep positions arranged in the execution order. The transverse sweep path is set along the transverse direction, and the longitudinal sweep path is set along the longitudinal direction. The distance between adjacent sweep positions is determined based on the installation spacing of the smart sensor nodes and the minimum size of the copper material to be identified, and can be set to 0.03m to 0.08m.

[0071] The controller reads the correspondence between each sweep position and the drive current of each coil group from the calibration parameter library, and sequentially sets the drive current of the transverse excitation coil group, the longitudinal excitation coil group, and the compensation coil group, so that the near-zero differential magnetic field surface moves continuously from the starting position to the ending position. When the difference in drive current between adjacent sweep positions exceeds the preset upper limit of change, a transition current value is inserted between the two sweep positions to avoid the sudden change in drive current affecting the quantum magnetic field measurement.

[0072] Each sweep position is maintained for at least one complete sampling cycle. The intelligent sensor node determines the actual near-zero differential magnetic field surface position based on the quantum magnetic field measurement results; when the distance between the actual position and the target sweep position does not exceed the preset sweep position error, the corresponding sweep position is marked as valid. The preset sweep position error can be set to 0.01m to 0.03m.

[0073] After completing the sweep position verification, a complete sweep is performed in the empty gate state according to the first magnetic field arrangement and the second magnetic field arrangement respectively. The effective magnetic field value of each smart sensor node at each sweep position and at each quantum magnetic field measurement trigger time is recorded to form the first empty gate in-sequence reference magnetic field data and the second empty gate in-sequence reference magnetic field data, which are then stored in the calibration parameter library.

[0074] Output the sweep control sequence, the first empty gate in-sequence reference magnetic field data, and the second empty gate in-sequence reference magnetic field data. The sweep control sequence includes the sweep path type, sweep position number, spatial coordinates, execution time, dwell time, drive current of each coil group, and sweep position validity status, and is sent to S310, S320, S330, and S430; the first empty gate in-sequence reference magnetic field data is sent to S410, and the second empty gate in-sequence reference magnetic field data is sent to S420.

[0075] The magnetic field arrangement control module is used to sequentially execute the lateral pre-excitation, the first arrangement transition period, and the longitudinal near-zero field sweep during the first detection sub-cycle to form the first magnetic field arrangement; and sequentially execute the longitudinal pre-excitation, the second arrangement transition period, and the lateral near-zero field sweep during the second detection sub-cycle to form the second magnetic field arrangement, and establish a magnetic field arrangement measurement correspondence table, specifically including:

[0076] S310 reads the paired magnetic field arrangement trigger timing output from S130, the compensation current parameters output from S220, and the sweep control sequence output from S230. At the beginning of each first detection sub-cycle, the transverse excitation coil group is controlled to output a transverse pre-excitation pulse, causing the metal target under test to form a transverse pre-excitation response state. After the transverse pre-excitation ends, the first arrangement transition period begins.

[0077] The first transition period refers to the time interval between the end of the transverse pre-excitation and the start of the longitudinal near-zero field sweep, used to attenuate the transient magnetic field generated by the coil switching to within a preset transient threshold. After the first transition period ends, the compensation coil group continues to operate according to the compensation current parameters, and the longitudinal excitation coil group is controlled according to the sweep control sequence to sequentially call the drive current corresponding to each sweep position, so that the near-zero differential magnetic field surface moves along the longitudinal sweep path.

[0078] To ensure comparability between lateral and longitudinal pre-excitation, the pre-excitation field integral is calculated during the open-gate calibration phase:

[0079]

[0080] in, Indicates direction The corresponding pre-excitation field integral; =H indicates the horizontal direction. =L indicates the vertical direction; and These represent the start and end times of the pre-excitation, respectively. This represents the magnetic field strength at the center point of the passage area at time t.

[0081] The difference between the pre-excitation field integrals of the transverse pre-excitation and the longitudinal pre-excitation shall not exceed 5%. The drive circuits of each coil group synchronously acquire the actual drive current; when the actual drive current deviates from the target drive current by more than 2% to 5% of the absolute value of the target drive current, the current first detection sub-cycle is marked as invalid.

[0082] Output the first magnetic field arrangement control record, including the test subject number, the paired detection cycle number, the start and end times of the transverse pre-excitation, the first arrangement transition period, the longitudinal sweep position number, the target drive current of each coil group, the actual drive current of each coil group, and the effective status of the first magnetic field arrangement, and send it to S330 and S410.

[0083] S320 reads the paired magnetic field arrangement triggering timing output from S130, the compensation current parameters output from S220, and the sweep control sequence output from S230. At the beginning of the second detection sub-cycle corresponding to the first detection sub-cycle, the longitudinal excitation coil group is controlled to output a longitudinal pre-excitation pulse, causing the metal target under test to form a longitudinal pre-excitation response state. After the longitudinal pre-excitation ends, the second arrangement transition period begins.

[0084] When the transition period of the second arrangement ends and the transient magnetic field decays to within the preset transient threshold, the compensation coil group continues to operate according to the compensation current parameters, and the transverse excitation coil group is controlled according to the sweep control sequence to sequentially call the driving current corresponding to each sweep position, so that the near-zero differential magnetic field surface moves along the transverse sweep path. The second magnetic field arrangement uses the same pre-excitation field integral constraint, sweep position spacing, sweep position dwell time, quantum magnetic field measurement sampling period, allowable position error, and current error threshold as the first magnetic field arrangement, only exchanging the pre-excitation direction and sweep direction.

[0085] The displacement of the person under test during the paired testing period shall be verified according to the following formula:

[0086]

[0087] in, This represents the estimated displacement of the person being measured within the current paired detection cycle; This indicates the average passage speed within the current paired detection cycle; Indicates the duration of the current paired detection cycle; This indicates the upper limit of the allowable periodic displacement. The upper limit of the allowable periodic displacement is determined based on the installation spacing of the smart sensor nodes and the minimum size of the copper material to be identified, and can be set to 0.05m to 0.10m. If the upper limit of the allowable periodic displacement is exceeded, the current paired detection cycle will be marked as invalid.

[0088] Output the second magnetic field arrangement control record, including the test subject number, the paired detection cycle number, the start and end time of the longitudinal pre-excitation, the second arrangement transition period, the lateral sweep position number, the target drive current of each coil group, the actual drive current of each coil group, and the effective status of the second magnetic field arrangement, and send it to S330 and S420.

[0089] The drive circuits of S330, the passage status acquisition module, the magnetic field arrangement control module, the intelligent sensor node, and each coil group all use the unified clock output by the system controller to record the time. The system reads the passage status data output by S120, the paired magnetic field arrangement trigger timing sequence output by S130, the sweep control sequence output by S230, the first magnetic field arrangement control record output by S310, and the second magnetic field arrangement control record output by S320.

[0090] For any sweep moment When it is located at two adjacent acquisition times and During this period, calculate the location of the person to be tested using the following formula:

[0091]

[0092] in, Indicates the moment of sweeping The corresponding location of the person being tested; and These represent the two adjacent corrected passage positions output by S120.

[0093] Only records are retained for cases where both the first and second magnetic field arrangement states are valid, the estimated displacement of the paired detection cycle does not exceed the upper limit of the allowable cycle displacement, the sweep position is valid, and both detection sub-cycles have complete quantum magnetic field measurement sampling periods. The magnetic field arrangement measurement correspondence table also records the quantum magnetic field measurement trigger times within the first and second arrangement transition periods for subsequent extraction of instantaneous non-commutative residuals.

[0094] Output the magnetic field arrangement measurement correspondence table, including the personnel number under test, the paired detection cycle number, the magnetic field arrangement type, the detection sub-cycle type, the pre-excitation direction, the sweep direction, the sweep position number, the sweep position spatial coordinates, the sweep time, the personnel under test's passage position, the personnel under test's passage speed, the target driving current of each coil group, the actual driving current of each coil group, the quantum magnetic field measurement trigger time during the arrangement transition period, the quantum magnetic field measurement trigger time during the sweep, and the valid status of the paired detection cycle, and send it to S410, S420, S430, and S510.

[0095] The intelligent quantum magnetic field measurement module is used to control each intelligent sensor node to perform quantum magnetic field measurements under the first and second magnetic field arrangements according to the magnetic field arrangement measurement correspondence table. It subtracts the corresponding empty-gate in-sequence reference magnetic field data to obtain the node target perturbation magnetic field value, and recovers the near-zero differential magnetic field surface state based on the node target perturbation magnetic field value. It also extracts the first and second near-zero field fission trajectories, specifically including:

[0096] S410 reads the first magnetic field arrangement control record output from S310 (showing the first magnetic field arrangement as valid), the magnetic field arrangement measurement correspondence table output from S330, and the first empty gate sequence reference magnetic field data output from S230. During the first arrangement transition period and the longitudinal near-zero field sweep, the intelligent sensor node synchronously samples according to the quantum magnetic field measurement trigger time in the magnetic field arrangement measurement correspondence table, and subtracts temperature drift and reference magnetic field changes according to the correction rule in S210 to obtain the node's effective magnetic field value.

[0097] For the m-th magnetic field arrangement, the node target disturbance magnetic field value of the i-th smart sensor node in the direction r corresponding to time t is calculated according to the following formula:

[0098]

[0099] Where m represents the magnetic field arrangement type. The time indicates the first magnetic field arrangement. The time indicates the second magnetic field arrangement; i indicates the smart sensor node number; r indicates the direction of the magnetic field component; This represents the effective magnetic field value at the node after deducting temperature drift and reference magnetic field changes. Indicates the reference magnetic field value of the empty gate sequence; This represents the magnetic field value of the nodal target disturbance caused by the metal target under test.

[0100] At least 5 consecutive sampling points are retained at each sweep position. If the deviation between the actual sampling time and the quantum magnetic field measurement trigger time exceeds 0.2ms to 0.5ms, or if the effective magnetic field value of the node exceeds the range, the corresponding sampling point is marked as invalid.

[0101] Output the first quantum magnetic field measurement data, including node number, data time period, sweep position number, acquisition time, effective magnetic field values ​​of the node in three directions, target disturbance magnetic field values ​​of the node in three directions, and effective node status, and send it to S430.

[0102] S420 reads the second magnetic field arrangement control record output from S320 (showing the second magnetic field arrangement as valid), the magnetic field arrangement measurement correspondence table output from S330, and the second empty gate sequence reference magnetic field data output from S230. During the second arrangement transition period and the lateral near-zero field sweep, the intelligent sensor node uses the same sampling period, correction rules, and node valid state determination rules as S410 to obtain the node valid magnetic field value and node target disturbance magnetic field value corresponding to the second magnetic field arrangement.

[0103] If the number of invalid smart sensor nodes exceeds 10% of the number of smart sensor nodes participating in the measurement of the current sweep location at the same sweep location, the sweep location will be marked as invalid.

[0104] Output the second quantum magnetic field measurement data, including node number, data time period, sweep position number, acquisition time, effective magnetic field values ​​of the node in three directions, target disturbance magnetic field values ​​of the node in three directions, and effective node status, and send it to S430.

[0105] The system reads the sweep control sequence output from S230, the magnetic field arrangement measurement correspondence table output from S330, the first quantum magnetic field measurement data output from S410, and the second quantum magnetic field measurement data output from S420. The system establishes node adjacency relationships based on the fixed spatial installation positions of each smart sensor node. These adjacency relationships indicate that position interpolation is allowed between spatially adjacent smart sensor nodes. For each valid acquisition moment, the magnetic field component corresponding to the current sweep direction is used as the analysis component.

[0106] When the effective magnetic field values ​​of two adjacent smart sensor nodes have opposite signs, the near-zero magnetic field position between the two nodes is determined according to the following formula:

[0107]

[0108] in, Indicates a position near zero magnetic field; and These represent the spatial installation locations of two adjacent smart sensor nodes; and The abbreviated symbols represent the effective magnetic field values ​​of these two adjacent smart sensor nodes at the current acquisition time and corresponding analysis components (i.e., the aforementioned...). ).

[0109] If the absolute value of the effective magnetic field of a node is not greater than the preset near-zero magnetic field threshold determined by S220, the spatial installation position of the node is directly determined as the near-zero magnetic field position. Near-zero magnetic field positions obtained at the same acquisition time and that are spatially adjacent are connected to form a near-zero differential magnetic field surface state; isolated areas formed by fewer than 3 consecutive near-zero magnetic field positions are discarded as noise areas.

[0110] The current near-zero differential magnetic field surface state is compared with the state of the empty-gate in-sequence reference magnetic field surface recovered from the empty-gate in-sequence reference magnetic field data. The spatial offset distance, the number of near-zero magnetic field regions, the splitting time, and the closing time are recorded. When the originally continuous near-zero differential magnetic field surface forms two or more unconnected near-zero magnetic field regions, it is determined to be in a split state; when the split near-zero magnetic field regions are reconnected to form a continuous region, it is determined to be in a closed state.

[0111] Based on the spatial coordinates of the personnel's passage position and sweep position output by S330, it is determined whether the relative position of the personnel where the metal target is located has left the effective detection neighborhood corresponding to the current sweep position; the moment when the target first leaves the effective detection neighborhood is determined as the moment when the metal target leaves the corresponding sweep position.

[0112] The duration of the split is calculated using the following formula:

[0113]

[0114] in, Indicates the duration of the split; Indicates the moment of splitting; Indicates the closing time.

[0115] The closing delay time is calculated using the following formula:

[0116]

[0117] in, Indicates the closing delay time; This indicates the time when the metal target being tested leaves the corresponding sweep position.

[0118] After the metal target leaves the effective detection area, when the spatial offset distance does not exceed the preset recovery error threshold for 3 to 5 consecutive sampling points, the first sampling time that meets the condition is determined as the recovery time.

[0119] The delayed recovery time is calculated using the following formula:

[0120]

[0121] in, Indicates the duration of delayed recovery; Indicates the time of recovery.

[0122] When no splitting occurs, the splitting duration and closure delay time are both set to 0; if the splitting is not completed before the end of the observation window, the end time of the observation window is taken as the temporary closure time, and the corresponding trajectory point is marked as the truncated trajectory point. S430 forms the first near-zero field fission trajectory and the second near-zero field fission trajectory, respectively.

[0123] Each trajectory point includes the personnel number to be tested, the paired detection cycle number, the magnetic field arrangement type, the time period to which the data belongs, the sweep position number, the personnel's passage position, the acquisition time, the spatial offset distance, the number of near-zero magnetic field regions, the splitting state, the splitting time, the closing time, the time when the metal target to be tested leaves the corresponding sweep position, the splitting duration, the closing delay time, the recovery time, the delay recovery time, and the trajectory point's valid status. S430 sends the first near-zero field fission trajectory and the second near-zero field fission trajectory to S510.

[0124] The non-commutative residual extraction module is used to align the first near-zero field fission trajectory and the second near-zero field fission trajectory based on the relative position coordinates of personnel and normalized sweep coordinates or normalized transition time coordinates, calculate the non-commutative fission trajectory residuals, and separate the instantaneous non-commutative residuals and delayed fission closure residuals, specifically including:

[0125] S510 reads the passage status data output by S120, the magnetic field arrangement measurement correspondence table output by S330, and the first and second near-zero field fission trajectories output by S430. For the trajectory points formed during the sweep, establish the relative position coordinates of the personnel and the normalized sweep coordinates. The relative position coordinates of the personnel refer to the passage position of the personnel to be measured relative to the reference plane on the entrance side of the door frame.

[0126] Normalized sweep coordinates are used to represent the relative progress of the near-zero difference magnetic field surface from the starting position to the ending position in the current sweep path, and are calculated according to the following formula:

[0127]

[0128] Where u represents the normalized sweep coordinate; This indicates the position coordinates of the current sweep location in the corresponding sweep direction; Indicates the coordinates of the starting position of the sweep path; This represents the coordinates of the end position of the sweep path. Normalized sweep coordinates range from 0 to 1.

[0129] For the trajectory points formed during the arrangement of transition periods, the normalized transition period coordinates are calculated according to the following formula:

[0130]

[0131] in, represents the coordinates of the normalized transition period; t represents the acquisition time within the staging transition period. and These represent the start and end times of the transition period, respectively.

[0132] For trajectory points formed during the sweeping phase, in the second near-zero field fission trajectory, trajectory points belonging to the same person under test, in the same paired detection cycle, with a relative position coordinate difference of no more than 0.03m to 0.06m, a normalized sweeping coordinate difference of no more than 0.03 to 0.08, and all trajectory points in valid states are selected to form sweeping trajectory point pairs. For trajectory points formed during the arrangement of transition periods, transition trajectory point pairs are formed according to the following conditions: the same person under test, in the same paired detection cycle, a normalized transition period coordinate difference of no more than the preset transition coordinate error, and all trajectory points in valid states.

[0133] Output aligned fission trajectory pairs, including trajectory point pair number, data time period, relative position coordinates of personnel, normalized sweep coordinates or normalized transition time period coordinates, trajectory feature values ​​of the first and second trajectory points respectively, and the validity status of the trajectory point pair, and send them to S520.

[0134] S520 reads the output of S510 and confirms that the trajectory point pair is valid and is an aligned fission trajectory pair. For transition trajectory point pairs, the spatial offset distance residual, the number of near-zero magnetic field regions residual, and the residual of the nodal target disturbance magnetic field value are calculated respectively. For sweep trajectory point pairs, the spatial offset distance residual, the number of near-zero magnetic field regions residual, the splitting duration residual, the closure delay time residual, and the delayed recovery time residual are calculated respectively. Since each residual component corresponds to distance, quantity, magnetic field strength, and time respectively, a normalized scale is used to eliminate dimensional differences.

[0135] The combined non-commutative residual value of the q-th trajectory point pair is calculated according to the following formula:

[0136]

[0137] in, This represents the integrated non-commutative residual value corresponding to the q-th trajectory point pair; A represents the number of residual components involved in the current calculation; a represents the residual component number. This represents the weight of the a-th residual component, and the sum of all weights is 1; and These represent the values ​​of the a-th residual component in the first and second trajectory points, respectively. This represents the normalized scale corresponding to the a-th residual component.

[0138] The normalization scale and weights were obtained through calibration using samples of empty gate state, single copper material, single silver material, single iron material, and coated copper material. The specific calibration rules are as follows: Extract the residual component sets of each type of calibration sample under different passage positions and speeds; set the difference between the maximum and minimum values ​​of the a-th residual component in the calibration sample set as the normalization scale corresponding to that residual feature; calculate the ratio of inter-class variance to intra-class variance for each residual component when distinguishing between ferromagnetic materials and conductive metal samples; normalize the ratios of each residual component and use them as the weight of the a-th residual component, so that residual components with higher classification discrimination automatically receive greater weights. The effective residual threshold is determined according to the following formula:

[0139]

[0140] in, Indicates the effective residual threshold; This represents the average value of the composite non-exchangeable residuals under the empty gate state; This represents the standard deviation of the aggregated non-commutative residuals under an empty gate state; This represents the residual threshold coefficient, which can be set to 3 to 5. When the aggregated non-commutative residual value is greater than the effective residual threshold, the corresponding residual is marked as effective.

[0141] Output the non-exchangeable fission trajectory residuals, including the personnel number, paired detection cycle number, trajectory point pair number, data time period, personnel relative position coordinates, normalized sweep coordinates or normalized transition time period coordinates, each residual component, comprehensive non-exchangeable residual value and residual validity status, and send them to S530.

[0142] S530 reads the output of S520, and the valid residual status is the non-exchangeable fission trajectory residual. The valid non-exchangeable fission trajectory residuals formed during the first and second staging transition periods are aggregated into instantaneous non-exchangeable residuals according to the relative position coordinates of the personnel and the normalized transition period coordinates. The instantaneous non-exchangeable residuals are used to characterize the difference in directional correlation response generated by the measured metal target after being subjected to lateral pre-excitation and longitudinal pre-excitation.

[0143] The effective non-commutative fission trajectory residuals formed between the moment the tested metal target leaves its corresponding sweep position and the recovery time are aggregated into delayed fission closure residuals according to the personnel's relative position coordinates, normalized sweep coordinates, and delay duration. These delayed fission closure residuals characterize the differences in the recovery process that still exist after the tested metal target leaves its corresponding sweep position. When three or more consecutive trajectory point pairs at the same personnel's relative position coordinates all have effective residuals, they are connected to form a residual region; isolated effective residuals formed by fewer than three consecutive trajectory point pairs are discarded as discrete noise.

[0144] Output the instantaneous non-commutative residual and the delayed fission closure residual. The instantaneous non-commutative residual includes the personnel number, the paired detection cycle number, the personnel's relative position coordinates, the normalized transition time coordinates, the instantaneous residual start time, the instantaneous residual end time, the instantaneous residual duration, each residual component, the comprehensive non-commutative residual value, and the effective status of the instantaneous residual region, and sends them to S610.

[0145] The delayed fission closure residual includes the personnel number to be tested, the paired detection cycle number, the personnel's relative position coordinates, the normalized sweep coordinates, the time when the metal target to be tested leaves the corresponding sweep position, the recovery time, the duration of the delayed residual, each residual component, the comprehensive non-exchange residual value, and the effective status of the delayed residual region, and is sent to S620.

[0146] The metal classification output module is used to determine ferromagnetic material regions and suspected composite target regions based on instantaneous non-commutative residuals, generate candidate internal regions, identify copper, silver, or other non-ferromagnetic metals based on delayed fission closure residuals, and output metal classification detection results based on the spatial correlation between the conductive metal classification results and the candidate internal regions. Specifically, this includes:

[0147] S610 reads the output of S530 and identifies valid real-time non-commutative residuals in the real-time residual region. Based on the relative position coordinates of personnel and the normalized transition time coordinates, spatially adjacent and temporally continuous real-time non-commutative residuals are connected to form real-time residual regions. Each real-time residual region includes at least three consecutive trajectory point pairs. For each real-time residual region, the average spatial offset distance residual, the average number of near-zero magnetic field region residuals, the average residual of the node target disturbance magnetic field value, the real-time residual duration, and the average comprehensive non-commutative residual value are calculated and combined into a real-time residual feature vector.

[0148] The instantaneous residual feature vector is compared with the instantaneous residual feature range of a single iron material in the calibration parameter library. If it falls within the range, the corresponding instantaneous residual region is marked as a ferromagnetic material region. If at least one feature value exceeds the upper limit of the instantaneous residual feature range of a single iron material, and this state persists for at least two adjacent paired detection cycles, the corresponding instantaneous residual region is marked as a suspected composite target region. Based on the sweep trajectory points that belong to the same paired detection cycle as the instantaneous residual region and are adjacent to the relative position coordinates of the personnel, the normalized sweep coordinate ranges corresponding to the ferromagnetic material region and the suspected composite target region are determined.

[0149] Candidate internal regions are generated according to the following formula:

[0150]

[0151] in, represents the candidate internal region; x represents the relative position coordinates of the personnel, in meters; u represents the normalized sweep coordinates. Indicates a region of ferromagnetic material; Indicates a region suspected of being a composite target area; and These represent the relative position coordinates and normalized sweep coordinates of personnel within the ferromagnetic material region or the suspected composite target region, respectively; This indicates the extended distance of the relative position coordinates of personnel, and can be set to 0.10m to 0.20m; This represents the normalized sweep coordinate expansion, which can be set from 0.05 to 0.15. It outputs the ferromagnetic material region, the suspected composite target region, and the candidate internal region, and sends them to S620 and S630.

[0152] S620 reads all delayed fission closure residuals output from S530 that are valid in the delayed residual region, and the candidate internal regions output from S610. For each delayed residual region, the average values ​​of the spatial offset distance residual, splitting duration residual, closure delay time residual, delayed recovery time residual, and comprehensive non-commutative residual value are calculated and combined into a delayed residual feature vector.

[0153] For the For metals, the category distance is calculated using the following formula:

[0154]

[0155] in, Indicates the region of delayed residual to be identified and the first Category distance between metals; Metalloids include copper, silver, and other non-ferromagnetic metals; C represents the number of features involved in the classification; b represents the classification feature number; This represents the weight of the b-th classification feature, and the sum of all weights is 1. This represents the b-th eigenvalue in the delayed residual eigenvector; and They represent the first The calibrated mean and calibrated standard deviation are for the b-th classification feature corresponding to the metalloids. When the calibrated standard deviation is lower than the measurement resolution, the measurement resolution is used as the lower limit for calculation.

[0156] The continuity consistency ratio is calculated using the following formula:

[0157]

[0158] in, Indicates the first The proportion of continuous consistency corresponding to metalloids; This indicates that within the same delayed residual region, the category with the smallest class distance and falling into the [missing category] is [missing category]. The number of valid trajectory point pairs within the permissible distance range for metal-like structures; This represents the total number of valid trajectory point pairs within the delay residual region.

[0159] A fixed value is selected from 0.70 to 0.85 for the preset consistency threshold. When the continuous consistency ratio is not less than the preset consistency threshold, the corresponding conductive metal category is output. If two or more categories simultaneously meet the condition, or if all categories fail to meet the condition, the corresponding delay residual region is marked as a region to be reviewed. If the conductive metal classification result is located within the candidate's internal region, it is marked as an associated classification result; if it is located outside the candidate's internal region, it is marked as a non-associated classification result.

[0160] Output the conductive metal classification results, including the personnel number, paired detection cycle number, delay residual region number, personnel relative position coordinate range, normalized sweep coordinate range, delay residual feature vector, category distance corresponding to each metal category, continuity consistency ratio, conductive metal category, spatial association type, and conductive metal classification validity status, and send them to S630.

[0161] S630 reads the ferromagnetic material region, suspected composite target region, and candidate internal region output by S610, and the conductive metal classification result output by S620. When the associated classification result is copper, the corresponding target is marked as a copper-clad target; when the associated classification result is silver, the corresponding target is marked as a silver-clad target; when a ferromagnetic material region exists but no stable associated classification result is obtained, the corresponding target is marked as a single ferromagnetic target; when the non-associated classification result is copper, the corresponding target is marked as a bare copper target; when the non-associated classification result is silver, the corresponding target is marked as a bare silver target; when the conductive metal classification result is a region to be verified, the corresponding target is marked as a composite metal target to be verified.

[0162] The copper-coated target refers to the classification and detection tag generated by the system when the copper area is located in the candidate internal area corresponding to the ferromagnetic material area or the suspected composite target area; the tag is used to indicate that the copper and the ferromagnetic material have a spatial relationship, but is not used to limit the copper to be completely enclosed by the ferromagnetic material.

[0163] For copper-coated targets, exposed copper targets, and composite metal targets awaiting verification, an interception prompt will be output. For single ferromagnetic targets, silver-coated targets, exposed silver targets, and other non-ferromagnetic metal targets, a manual verification prompt, registration prompt, or release prompt will be output according to the plant's preset management rules. When the same person to be tested corresponds to multiple spatially separated target areas, separate metal target records will be generated.

[0164] Output the metal classification and detection results, including the person number to be tested, access event record, metal target record number, relative position coordinate range of the person, ferromagnetic material judgment result, conductive metal category, spatial association type, system classification label, continuity consistency ratio, prompt type and data generation time, to complete the metal classification identification and detection process for the current person to be tested.

[0165] All the above formulas are performed using dimensionless numerical calculations; the relevant formulas are based on empirical models that approximate the real situation, obtained through extensive data collection and software simulation fitting. The preset parameters and thresholds involved in the formulas can be conventionally set and adjusted by those skilled in the art according to the physical constraints of the actual application scenario.

[0166] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0167] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0168] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A metal detection door metal classification identification and detection system, characterized in that, include: The passage status acquisition module is used to acquire the start time, end time, passage position and passage speed of the person to be tested through a through-beam photoelectric sensor and a distance sensor, generate passage status data, and generate a pair of magnetic field arrangement triggering timing sequence that includes the first detection sub-cycle, the protection interval and the second detection sub-cycle in sequence. The near-zero differential magnetic field construction module is used to acquire background magnetic field data in the empty gate state, adjust the drive current of the compensation coil group according to the background magnetic field data, construct the initial near-zero differential magnetic field surface in the passage area, and generate a sweep control sequence. The magnetic field arrangement control module is used to sequentially execute the lateral pre-excitation, the first arrangement transition period, and the longitudinal near-zero field sweep during the first detection sub-cycle to form the first magnetic field arrangement. During the second detection sub-cycle, it sequentially executes the longitudinal pre-excitation, the second arrangement transition period, and the lateral near-zero field sweep to form the second magnetic field arrangement, and establishes a magnetic field arrangement measurement correspondence table. The intelligent quantum magnetic field measurement module is used to control each intelligent sensor node to perform quantum magnetic field measurement under the first and second magnetic field arrangements according to the magnetic field arrangement measurement correspondence table. It subtracts the corresponding empty gate in-sequence reference magnetic field data to obtain the node target perturbation magnetic field value, and restores the near-zero differential magnetic field surface state according to the node target perturbation magnetic field value, and extracts the first near-zero field fission trajectory and the second near-zero field fission trajectory.

2. The metal classification and detection system of claim 1, wherein, Also includes: The non-commutative residual extraction module is used to align the first near-zero field fission trajectory and the second near-zero field fission trajectory based on the relative position coordinates of the personnel and the normalized sweep coordinates or normalized transition time coordinates, calculate the non-commutative fission trajectory residuals, and separate the instantaneous non-commutative residuals and delayed fission closure residuals.

3. The metal classification and detection system of claim 2, wherein, Also includes: The metal classification output module is used to determine ferromagnetic material regions and suspected composite target regions based on instantaneous non-exchange residuals, generate candidate internal regions, identify copper, silver or other non-ferromagnetic metals based on delayed fission closure residuals, and output metal classification detection results based on the spatial correlation between the conductive metal classification results and the candidate internal regions.

4. The metal classification and detection system of claim 1, wherein, The passage status acquisition module is used for: The start time, end time and direction of travel of the person being tested are determined by through-beam photoelectric sensors on the entrance and exit sides, and the travel distance sequence is collected by distance sensors on the top crossbeam to eliminate records of reverse travel, overlapping travel and overtime travel. The continuity of the passage distance sequence is checked, and linear interpolation is performed when the number of consecutive abnormal sampling points does not exceed a preset upper limit to generate passage status data; Based on the traffic status data, the duration within the effective detection range is divided into at least one paired detection cycle. Each paired detection cycle includes a first detection sub-cycle, a protection interval, and a second detection sub-cycle in sequence, generating a paired magnetic field arrangement triggering sequence.

5. The metal classification and detection system of claim 1, wherein, The near-zero differential magnetic field construction module is used for: During the empty gate period, multiple intelligent sensor nodes are controlled to synchronously perform quantum magnetic field measurements. The quantum magnetic field measurement results are corrected according to the local temperature and reference magnetic field changes to generate background magnetic field data. Based on the background magnetic field data and the magnetic field influence coefficients corresponding to each compensation coil, the driving current of the compensation coil group is adjusted, and the positions where the magnetic field strength meets the preset near-zero magnetic field threshold and are spatially adjacent are connected to form the initial near-zero differential magnetic field surface. The transverse and longitudinal sweep paths are divided into multiple sweep positions. The drive current of each coil group is set sequentially, the sweep control sequence is generated and verified, and the first empty gate in-sequence reference magnetic field data and the second empty gate in-sequence reference magnetic field data are generated in the empty gate state.

6. The metal classification, identification, and detection system for a metal detector gate according to claim 1, characterized in that, The magnetic field arrangement control module is used for: Within the first detection sub-cycle, the transverse pre-excitation, the first staging transition period, and the longitudinal near-zero field sweep are executed sequentially to generate the first magnetic field staging control record; During the second detection sub-cycle, longitudinal pre-excitation, the second staging transition period, and the lateral near-zero field sweep are executed sequentially to generate the second magnetic field staging control record; The first and second magnetic field arrangements adopt the same pre-excitation field integral constraints, sweep position spacing, sweep position dwell time, quantum magnetic field measurement sampling period, allowable position error and current error threshold, and invalid cycles are screened out based on the displacement estimate of the person to be measured in the paired detection cycle. Based on a unified clock, the passage status data, sweep position, driving current of each coil group and quantum magnetic field measurement trigger time are correlated to generate a magnetic field arrangement measurement correspondence table.

7. The metal classification and detection system of claim 1, wherein, The intelligent quantum magnetic field measurement module is used for: Multiple intelligent sensor nodes are controlled to perform quantum magnetic field measurements under the first and second magnetic field arrangements respectively. Temperature drift correction and reference magnetic field change subtraction are performed on the quantum magnetic field measurement results. The first and second empty gate in-sequence reference magnetic field data are subtracted respectively to obtain the node target disturbance magnetic field value. The near-zero magnetic field position is determined based on the effective magnetic field value of adjacent smart sensor nodes, and the spatially adjacent near-zero magnetic field positions are connected to form a near-zero differential magnetic field surface state. By recording the spatial offset distance of the near-zero differential magnetic field surface, the number of near-zero magnetic field regions, the splitting time, the closing time, the splitting duration, the closing delay time, and the delay recovery time, the first near-zero field fission trajectory and the second near-zero field fission trajectory are extracted.

8. The metal classification and detection system of claim 2, wherein, The non-commutative residual extraction module is used for: Based on the relative position coordinates of the personnel and the normalized sweep coordinates or normalized transition time coordinates, the effective trajectory points in the first near-zero field fission trajectory and the second near-zero field fission trajectory are paired to form aligned fission trajectory pairs. The spatial offset distance, number of near-zero magnetic field regions, nodal target disturbance magnetic field value, splitting duration, closure delay time and delay recovery time in each aligned fission trajectory pair are calculated by difference, and the comprehensive non-commutative residual value is obtained after normalization. Non-exchangeable fission trajectory residuals that are greater than the effective residual threshold and appear consecutively in no less than 3 trajectory point pairs are grouped into instantaneous non-exchangeable residuals and delayed fission closure residuals according to the time period to which the data belongs.

9. The metal classification and detection system of claim 3, wherein, The metal classification output module is used for: Connect spatially adjacent and temporally continuous instantaneous noncommutative residuals into instantaneous residual regions, extract instantaneous residual feature vectors, and compare them with the feature range of instantaneous residuals of a single iron material to determine ferromagnetic material regions and suspected composite target regions. Candidate internal regions are generated based on ferromagnetic material regions and suspected composite target regions; The delayed residual feature vector is extracted from the delayed fission closure residual. The category distance between the feature vector and copper, silver and other non-ferromagnetic metals is calculated. The classification result of conductive metals is generated by combining the continuity consistency ratio.

10. The metal detection door metal classification identification and detection system according to claim 9, characterized in that, Also includes: Based on the spatial correlation between the conductive metal classification results and the candidate internal regions, the metal classification detection results corresponding to copper-coated targets, silver-coated targets, single ferromagnetic targets, bare copper targets, bare silver targets, or composite metal targets to be verified are output.

Images