Personnel screening with threat detection and identification

Abstract

A method includes: receiving data representing a sample obtained by the plurality of magnetic field receivers including magnetic sensors, a sample of a combination of a first magnetic field and a second magnetic field generated by the interaction of the first magnetic field with an object; determining a polarizability index of the object using the received data, the polarizability index representing a magnetic polarizability property of the object; using the determined polarizability index, classifying the object as a threat or a non-threat; and providing the classification. Related apparatus, systems, techniques, and articles are also described.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of and priority under 35 U.S.C. §119 to U.S. Provisional Patent Application No. 62 / 751,490, filed on October 26, 2018, the entire contents of which are hereby expressly incorporated herein by reference. Technical Field

[0003] The subject matter described herein relates to a personnel screening system that, in some exemplary embodiments, is capable of performing threat detection and identification without removing personal items. Background Art

[0004] Airport security personnel attempt to prevent any threatening or potentially dangerous situations from occurring or entering a country. Some existing radio frequency (RF) imaging systems (e.g., those used by airport security personnel for passenger screening) are bulky, expensive, and require the individual to remain stationary while an antenna rotates around the stationary individual to capture an image. In addition, these existing RF imaging systems may require the individual being screened to remove personal items such as cell phones, keys, wallets, etc. This removal requirement can reduce the throughput and availability of the imaging system.

[0005] Some existing inspection systems (e.g., walk-through metal detectors) may include coils for generating and measuring changes in a magnetic field caused by magnetic or conductive materials (e.g., metal) passing through the magnetic field. These existing inspection systems are able to measure metal objects that pass a threshold, but are unable to distinguish personal items (e.g., cell phones, laptops, keys, belt buckles, etc.) from threats (e.g., firearms or improvised explosive devices). Therefore, these exemplary existing inspection systems require the removal of personal items, thereby limiting their throughput and usability. Summary of the invention

[0006] In one aspect, a method includes receiving data representing samples obtained by the plurality of magnetic field receivers including magnetic sensors, a combination of a first magnetic field and a second magnetic field generated by an interaction of the first magnetic field with an object; determining a polarizability index of the object using the received data, the polarizability index representing a magnetic polarizability property of the object; classifying the object as a threat or a non-threat using the determined polarizability index; and providing the classification.

[0007] One or more of the following features may be included in any feasible combination. For example, the polarizability index of the object may characterize at least the shape, magnetic permeability, and electrical conductivity of the object. The polarizability index of the object may include a complex tensor, the complex tensor including at least six elements characterizing the directional polarizability component of the object at one or more frequencies used by the transmission system. Determining the polarizability index may include solving a set of trial solutions by a pre-calculated pseudo-inverse, determining the residual of each trial solution, and selecting the trial solution that produces the minimum residual. Determining the polarizability index may include defining a set of trial solutions, each trial solution including position, velocity, and time-shift; calculating the associated polarizability index and the associated residual for each trial solution; and selecting a final trial solution, the final trial solution including the trial solution in the set of trial solutions associated with the minimum residual. The method may further include positioning the object within the volume being inspected, the positioning including determining the object velocity, the object position, and the object time-offset to a predetermined plane. The method may further include generating one or more signals for driving the magnetic field transmitter at a frequency less than 1000 Hertz (Hz).

[0008] The plurality of magnetic field receivers may include fluxgate sensors that measure the amplitude and phase of a combination of a first magnetic field and a second magnetic field on at least three axes at the plurality of magnetic field receiver positions. The samples obtained by the plurality of magnetic field receivers may characterize the amplitude and phase. Classifying an object as a threat or a non-threat using a polarizability index may include comparing the amplitude, phase, and polarizability index with a predetermined threat signature library. The first magnetic field may be generated by at least a first magnetic field transmitter and a second magnetic field transmitter. The first magnetic field transmitter and the second magnetic field transmitter may be oriented in a first direction. The first magnetic field transmitter and the second magnetic field transmitter may be transverse to a predefined direction of motion and transverse to the first direction and offset in space.

[0009] In another aspect, a system may include: a magnetic field transmitter configured to generate a first magnetic field; a plurality of magnetic field receivers including magnetic sensors, the plurality of magnetic field receivers configured to sample a combination of the first magnetic field and a second magnetic field generated by an interaction of the first magnetic field with an object; and at least one data processor configured to at least: receive data characterizing samples obtained by the plurality of magnetic field receivers; determine a polarizability index of the object using the received data, the polarizability index characterizing a magnetic polarizability property of the object; classify the object as a threat or a non-threat using the polarizability index; and provide the classification.

[0010] One or more of the following features may be included. For example, a transmit driver may be coupled to the magnetic field transmitter and configured to generate one or more signals for driving the magnetic field transmitter at a frequency less than 1000 Hertz (Hz). The transmit driver may be configured to generate one or more signals for driving the magnetic field transmitter at a first frequency and a second frequency. The magnetic field transmitter may be driven at a first frequency and a second frequency. The first frequency may be between 5 Hz and 40 Hz, and the second frequency may be between 100 Hz and 1000 Hz. A second magnetic field transmitter may be included. The transmit driver may include: a first drive electronics device for driving the magnetic field transmitter; and a second drive electronics device for driving the second magnetic field transmitter. Samples may be obtained by a plurality of magnetic field receivers and the samples may characterize amplitude and phase. Classifying an object as a threat or non-threat using a polarizability index may include comparing the amplitude, phase, and polarizability index with a predetermined threat signature library.

[0011] The plurality of magnetic field receivers may include fluxgate sensors that measure the amplitude and phase of a combination of a first magnetic field and a second magnetic field on at least three axes at a plurality of magnetic field receiver locations. A data acquisition base station may be included that is coupled to the plurality of magnetic field receivers and configured to filter, demodulate, and digitize samples obtained by the plurality of magnetic field receivers. The data acquisition base station may be further configured to determine in-phase and quadrature data characterizing the second magnetic field. Determining the polarizability index may include determining a polarizability tensor. Determining the polarizability index may include solving a set of trial solutions by a pre-calculated pseudo-inverse, determining the residual of each trial solution, and selecting the trial solution that produces the smallest residual.

[0012] The at least one data processor may be further configured to: locate the object within the inspected volume, the locating comprising determining the object velocity, the object position, and the object time offset to a predetermined plane. Determining the polarizability index may include: defining a set of trial solutions, each trial solution comprising a position, a velocity, and a time offset; calculating an associated polarizability index and an associated residual for each trial solution; and selecting a final trial solution, the final trial solution comprising a trial solution in the set of trial solutions associated with a minimum residual. The at least one data processor may be further configured to determine a measure of distortion of the second magnetic field based on an amount of metal present at a location where the system is deployed.

[0013] The system may also include a base plate configured to couple the first column and the second column. The magnetic field transmitter may be coupled to the first column, and the plurality of magnetic field receivers may be coupled to the second column. The base plate may include a modular mounting system configured to couple the base plate to the surface on which the base plate is located. The modular mounting system may include at least one of a plurality of suction cups, a plurality of clamping mechanisms, a plurality of perforating mechanisms, and a plurality of auger mechanisms. The base plate may include a base plate frame hidden within the base plate. The base plate frame may be configured to permanently mount the base plate at a location where the system is deployed. The proximal end of the first column and the proximal end of the second column may be coupled to the base plate. The distal end of the first column and the distal end of the second column are not connected to an arch that couples the distal end of the first column to the distal end of the second column. At least one of an accelerometer or an inclinometer may be coupled to the first column and / or the second column.

[0014] The system may also include at least one camera coupled to the first column or the second column. The at least one camera may be configured to capture images and / or video data of an object. The camera may be selected from a group including a depth camera, a rear-facing camera, a fisheye camera, an infrared camera, a thermal imaging camera, a surface image camera, an optoelectronic camera, and a stereo camera. The system may include a first camera coupled to the first column and a second camera coupled to the second column. The first camera may be configured to capture an image at a first perspective, and the second camera may be configured to capture image data at a second perspective. The second perspective is different from the first perspective. The image data captured by the first camera and the second camera are registered in a coordinate system of the system. In response to classifying the object as a threat, the processor may be further configured to generate a graphical indication identifying the location of the object and superimpose the graphical indication on the image data in the display.

[0015] The details of one or more variations of the subject matter described herein are set forth in the accompanying drawings and the description below. Other features and advantages of the subject matter described herein will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a system block diagram of an exemplary personnel screening system capable of performing threat detection and identification without removing personal belongings;

[0017] Figure 2 is a schematic diagram showing four example diagrams of the spatial arrangement of emitters;

[0018] Figure 3 is a schematic diagram illustrating an arrangement of an exemplary personnel screening system according to some embodiments;

[0019] Figure 4is a process block diagram illustrating an exemplary process of an exemplary inspection system according to some aspects of the current subject matter;

[0020] Figure 5 is a process block diagram illustrating an exemplary process for determining a polarizability index of an object in an exemplary inspection system in accordance with some aspects of the current subject matter;

[0021] Figure 6 is a process block diagram illustrating an exemplary process for detecting an object on an individual in a large group of people in an exemplary inspection system in accordance with some aspects of the current subject matter;

[0022] Figure 7 is a schematic diagram illustrating an exemplary embodiment of a personnel inspection system;

[0023] Figure 8 is a schematic diagram showing an exemplary configuration of a personnel checking system including a plurality of sensors and transmitters located at different locations; and

[0024] Fig. 9 is a schematic diagram showing an exemplary configuration of a personnel checking system including a plurality of cameras.

[0025] Like reference numerals in the various drawings indicate like elements. DETAILED DESCRIPTION

[0026] Personnel screening systems are used to detect threats that may be introduced into a specific area that the screening system seeks to protect. For example, common personnel screening systems may include metal detectors placed at the entrance of a courthouse or stadium, or body scanners at an airport. These screening systems are configured to generate data that can be processed to determine if a threat is present. A threat can be considered any person, object, or element that passes through the system and, if allowed to enter the protected area, could cause harm, cause a security concern, and / or disrupt an event or activity occurring within the protected area. For example, a firearm is a threat that a personnel screening system attempts to detect at an airport or stadium. Typically, personnel screening systems are configured to detect threats that include metal objects.

[0027] Traditional personnel screening systems have many drawbacks. Personnel screening systems are often configured to scan or evaluate data associated with a single individual at a time (e.g., a queue of individuals in a secure area of ​​an airport). Each individual must be scanned or processed before another can be scanned or processed for threat detection, which can result in delays and long waits to enter a protected area. In traditional personnel screening systems, individuals often experience higher levels of anxiety and mistrust when being evaluated for potential threats.

[0028] Conventional personnel screening systems also require individuals to remove all potential clutter objects, such as any metal objects, before entering an area where the screening system is deployed. In this way, conventional screening systems may broadly identify potential threats as any detected metal object that may be left on the person passing through the screening system, without discerning the size, type, or composition of the object. Using such a broad binary discrimination threshold results in a large false alarm rate and requires individuals to undergo subsequent screening processing to remove objects that were not accurately identified as threat objects. For example, conventional screening systems generally fail to uniquely discern the object and material characteristics of a belt buckle from the object and material characteristics of a firearm. In conventional screening systems, both objects are detected and characterized as potential threats equally, however, belt buckles pose much less threat, or even no threat, than firearms. Individuals often have some metal objects on their person that may be incorrectly identified as threats in conventional screening systems. Shoe or boot grommets, belt buckles, eyeglasses, as well as cell phones, laptops, hearing aids, and pacemakers all include metal that conventional screening systems may incorrectly identify as detected threats. Therefore, personnel screening systems require that individuals passing through the screening system remove or unload all potentially threatening objects.

[0029] The current subject matter can include an improved person screening system that, in some exemplary embodiments, is capable of performing threat detection and discrimination in highly cluttered environments where individuals may be carrying personal items (e.g., cell phones and laptops) without the need to remove personal items. In some embodiments, the person screening system can perform threat detection and discrimination at high throughput, allowing individuals to pass through the detector at normal walking speeds, thereby not requiring the individual to slow down for screening, and in some embodiments, the screening threshold can allow multiple people to pass through the threshold side by side (e.g., two or more side by side).

[0030] Advantages of the improved personnel screening system may include a higher throughput of individuals undergoing assessments, a reduced incidence of false positives due to more accurate discrimination of metal objects as threats or non-threats, and reduced stress levels and improved emotional responses of individuals undergoing assessments utilizing the improved personnel screening system. Additionally, the improved personnel screening system may more accurately discriminate metal objects on the person of an individual passing through the improved screening system as threats or non-threats without requiring the individual to remove the metal objects from the person. The data collected, processed, and generated by the improved screening system may also be used in the context of other security-focused operations, such as notifying system operators of individuals in possession of detected threat objects, training screening system operators or supervisors, and overall process improvements to security procedures that may occur before or after individuals are screened or assessed using the improved screening system.

[0031] Some exemplary embodiments of the improved inspection systems disclosed herein may include a highly sensitive continuous wave magnetic detection system capable of detecting disturbances of up to 1 part in 10,000 in its transmitted field. To increase this sensitivity, the system may be configured to transmit a steady magnetic field and measure the transmitted magnetic field using low noise methods, since magnetic disturbances caused by unintentional system noise may be difficult to distinguish from magnetic disturbances caused by metal objects. In this case, system noise may include many signal disturbances, including traditional electronic noise, amplitude variations in the transmitted magnetic field, and / or digital errors that may be caused by harmonic mismatches between intentional signals and sampling rates associated with analog-to-digital conversion.

[0032] Some exemplary embodiments may include an active magnetic system that can acquire a series of magnetic field measurements of an observation domain; determine in-phase and quadrature components of the magnetic field measurements; determine measurements of polarizability (e.g., polarizability tensor, polarizability index) of objects in the observation domain; locate objects, including determining the velocity, position, and time offset of the objects; and perform threat detection and / or discrimination of objects in the presence of clutter using magnetic field measurements, polarizability, and / or location information. In certain embodiments, the system can be configured to detect firearms and / or improvised explosive devices (IEDs).

[0033] In some embodiments, the system can determine the polarizability of the object being inspected, and the polarizability of the object can be used to perform threat detection and discrimination (e.g., classification). By determining and utilizing the polarizability of the object, certain threats (e.g., firearms and improvised explosive devices) can be more accurately detected, thereby improving personnel inspection systems.

[0034] In a personnel screening system configured with magnetic field sensing to detect illegal or threatening objects (e.g., firearms), the frequency band used to optimize the signal-to-noise ratio (SNR) is different from the frequency band used to optimize the discrimination between threatening and non-threatening objects. In addition, one or both of the optimized frequency bands may be too low to be effectively detected by the system's receiver. Challenges faced by personnel screening systems configured with magnetic field sensing can include determining how to best measure detected threatening objects in the appropriate frequency band while maintaining the benefits provided by both optimized frequency bands.

[0035] The improved inspection system described herein can be configured to interrogate an object in such a manner that it obtains information in a first frequency band having good SNR characteristics and in a second frequency band having good discrimination characteristics, the first frequency band being different from the second frequency band. The use of a fluxgate in addition to or in place of an induction coil receiver can allow the system disclosed herein to measure magnetic fields at very low frequencies (e.g., below 1 kHz). Certain characteristics of the object (e.g., its position, velocity, direction, etc.) are determined by using frequency band measurements from the first frequency band having good SNR characteristics before recovering additional characteristics of the object via the second frequency band having good discrimination characteristics. Some exemplary embodiments of the magnetic sensing algorithms and systems described herein can exploit the benefits of both frequency bands simultaneously.

[0036] For example, many consumer electronics contain non-ferrous metals (such as aluminum and copper), while many firearms contain ferrous metals such as steel. Since eddy currents vary in proportion to frequency, maximum discrimination between ferrous and non-ferrous metals can be achieved by exciting the object with a magnetic field at a very low frequency (less than 1kHz). Such low frequencies can be effectively measured using fluxgate receivers compared to induction coils, and it is simpler and easier to establish induction coils at higher frequencies. However, the noise characteristics of fluxgate receivers (and their digitizing electronics) are generally better at higher frequencies. The systems and magnetic sensing algorithms described herein utilize frequency measurements at low frequencies (e.g., ~30Hz) and high frequencies (e.g., ~230Hz). For example, the systems and magnetic sensing algorithms described herein can utilize frequency measurements between about 5Hz and 100Hz and frequency measurements between about 100Hz and 1000Hz to simultaneously utilize the enhanced discrimination characteristics of low frequencies and the superior SNR of high frequencies, thereby improving the detection and discrimination of metal objects. Thus, in some embodiments, the system can operate at frequencies below 1 kilohertz (Hz) to improve firearm detection and discrimination in the presence of common personal items (e.g., cell phones). At relatively low frequencies (e.g., below 1 kHz), the magnetic contribution to the polarizability magnitude can dominate the conductive contribution to the polarizability magnitude. Thus, the signal amplitude driven by the total metal content of the threat may be less than the signal amplitude driven by materials that are unique to the characteristics of many threats and not found in typical consumer electronics.

[0037] The above description of the processing performed by the inspection system and magnetic sensing algorithm described herein may be further considered with respect to a single object passing through the inspection system. A single object will have certain properties that vary with frequency (e.g., material properties and / or magnetic dipole moments and / or polarizability tensor elements), and have some properties that are common across a frequency band, such as position, direction, and velocity. The inspection system and magnetic sensing algorithm described herein provide enhanced detection capabilities by utilizing measurements for determining frequency-invariant properties of an object, and utilizing good discrimination properties primarily for classification.

[0038] In some embodiments, the magnetic sensing algorithms described herein are configured to initially perform a search operation on a frequency band with favorable SNR characteristics to solve for position, direction, and velocity, and subsequently retrieve object characteristics, such as material properties, magnetic dipole moment, and / or the object's polarizability tensor or index on another frequency band. The information determined from the first search operation can be used to constrain the second search and improve the overall accuracy of the detection.

[0039] In other embodiments of the magnetic sensing methods described herein, the magnetic sensing algorithm can retrieve all object characteristics in a single step by using a weighted cost function that more closely evaluates higher frequencies for frequency-invariant characteristics. In both embodiments, the characteristics of objects in more discriminating frequency bands will be recovered with higher fidelity than when recovered independently. Subsequently, these characteristics can be used in a classification step to determine whether an object belongs to a particular category (e.g., a firearm or a consumer electronic device).

[0040] Figure 1 is a system block diagram of an exemplary inspection system 100 capable of performing threat detection and identification without removing personal belongings.

[0041] like Figure 1As shown, the system 100 includes a magnetic receiver 105 connected to a data acquisition base station 115. The data acquisition base station 115 can be configured to filter, demodulate and digitize the magnetic field measurement data received from the receiver 105. The transmitter 106 and the magnetic receiver 105 can be arranged to detect an observation domain (OD) 107, sometimes referred to as a "scene", for example, a threshold or other defined area. The OD 107 can be considered to include voxels that define a volume. The OD 107 can be a single continuous area or a plurality of separate areas. The system 100 also includes the transmitter 106 connected to the transmission driver 160. The transmission driver 160 can be configured to generate a signal to drive the transmitter 106. The system 100 also includes a processing system 120, which is configured to analyze the received magnetic field measurements. The processing system 120 includes a data acquisition module 130, a calibration module 135, a reconstruction module 140, an automatic threat identification module 145, a presentation module 150 and a memory 155. The system 100 may also include a display 165 for providing output and sensors 125 for providing additional input to the system 100 .

[0042] In some embodiments, the system can be configured to operate as a distributed lock-in amplifier to accurately extract in-phase (I) and quadrature (Q) information from a specific transmit frequency of the system using synchronous homodyne digital bi-phase demodulation techniques. Demodulation can be achieved by digitally mixing or multiplying the desired signal with a reference signal and then filtering the result using a low pass filter. The reference signal can be a direct measurement signal related to or derived from the drive signal used in the transmitter, or it can be a synthesized analog signal. By using two versions of the reference signal, one phase shifted or delayed from the other, the amplitude, phase and / or I and Q of the measured signal can be reconstructed.

[0043] In some embodiments, the transmitting driver 160 may include a set of combined digitally controlled high-precision direct digital synthesis (DDS) waveform generators, digitally controlled summation programmable gain amplifier (PGA) circuits, and closed-loop class D power amplifiers with enhanced power supply rejection ratio (PSRR). Such a system provides flexibility in the frequency and amplitude of the transmission waveform while achieving high stability in the transmission magnetic field to meet the necessary signal-to-noise ratio in the measured data. The system 100 can digitally control the amplitude and frequency of the transmission magnetic field. Digital control of the amplitude and frequency of the transmission magnetic field can be performed dynamically and in an arbitrary or specific manner. In some embodiments, the system may include a closed-loop microcontroller-based feedback system, which is configured to measure and dynamically adjust the amplitude of each frequency of the transmission field, thereby improving the stability and predictability of the system.

[0044] The transmitter 106 may include at least two wire loop transmitters capable of generating a magnetic field in response to a drive signal having an operating (e.g., characteristic) frequency (e.g., a modulation frequency). For example, the transmitter 106 may operate at 30 Hz and 130 Hz. In general, the wire loop may be considered to be located within a principal plane. In some embodiments, the system 100 may include transmitters arranged to transmit a field with sufficient diversity to detect all cardinal directions (e.g., Cartesian coordinates) of the entire OD 107. In a static system where the inspected object is stationary, at least three orthogonally oriented (e.g., the principal planes of each of the three transmitters may be oriented orthogonally to each other) or spatially offset transmitters may be included. If the object is moving in a particular direction, such as when the object passes through the inspection system 100, two transmitters may be used if the directions of the two transmitters are orthogonal to the direction of motion or transverse to the direction of motion and its shared direction. This configuration represents a reasonable constraint on the object's motion (e.g., in one direction) and may further represent the minimum number of transmitter coils that can achieve sufficient field diversity to fully detect a given object.

[0045] like Figure 1 As shown, the emitter driver 160 can generate one or more signals for driving the emitters 106. In some embodiments, the emitters 106 can be driven by looping through the emitters in time, one after another, until all desired measurements are captured. Advantages of this approach can include that the drive electronics can be shared between all emitters 106. However, this approach can impose a duty cycle on each emitter 106, thereby reducing its signal-to-noise ratio. In such a configuration, the emitters 106 may not measure at the same time, which may introduce motion-induced artifacts if the object is in motion.

[0046] In some embodiments, the transmitters 106 can be driven simultaneously, but at frequencies that are slightly (e.g., 10 Hertz (Hz)) offset. The frequencies can be offset enough so that they can be clearly demodulated in post-processing, which can be set by the bandwidth required to resolve the object's motion, which can be approximately 5-10 Hz for an object moving at a typical walking speed of 1.3 meters per second (m / s). At the same time, the frequencies can be selected to be similar enough that the dispersion in the polarizability is negligible. In some embodiments, the offset can be 10 Hz, which can be considered a negligible difference other than the vanishing frequency. In this exemplary frequency multiplexing method, the transmit driver 160 can include separate drive electronics to drive each transmitter individually, which can achieve an improved signal-to-noise ratio without (and / or reduced) risk of motion blur.

[0047] In some embodiments, the transmit driver 160 can generate a drive signal that can be distributed to the transmitter 106, which can establish a fully phase-coherent measurement system across all receive-transmit pairs. In addition, the drive signal can be provided as a reference signal that is routed from the transmitter driver 160 to the data acquisition base station 115, which can be used for demodulation, as described more fully below.

[0048] The magnetic receiver 105 may include a fluxgate sensor, which can directly measure the magnetic field (e.g., amplitude and phase) compared to the coil that measures the rate of change of the magnetic field. In some embodiments, one or more of the receivers 105 may include a three-axis fluxgate magnetometer. In some embodiments, one or more of the receivers 105 may include a 2-axis fluxgate magnetometer. The advantage of the fluxgate magnetometer is that it can operate with high sensitivity, high linearity, and low local noise compared to the coil receiver. The receiver 105 can provide accurate magnetic measurements at frequencies that are too low for traditional methods.

[0049] A fluxgate sensor can measure the magnetic field amplitude in three axes (e.g., x, y, and z) at the location of the fluxgate sensor. A fluxgate sensor can include a sensing coil surrounding an internal drive coil that is tightly wound around a high magnetic permeability core material (e.g., mu metal). An alternating current can be applied to the drive winding, which can drive the core in a continuous repeating cycle of saturation and desaturation. In the presence of an external magnetic field, the core is in a high magnetic permeability state, and such a magnetic field is locally attracted or gated through the sensing winding. This continuous gating of the external field into and out of the sensing winding induces a signal in the sensing winding, whose main frequency is twice the drive frequency, and whose strength and phase direction vary directly with the external field amplitude and polarity.

[0050] In some embodiments, fluxgate sensors can be used at operating frequencies below 1 kHz (e.g., 130 Hz and 30 Hz). At these relatively low operating frequencies, fluxgate sensors can operate with an increased noise floor, for example, some fluxgates can achieve volt-to-field ratios on the order of 20 microvolts / nano-Tesla.

[0051] The data acquisition base station 115 can demodulate, filter, and digitize the data received from the receiver 105. The data acquisition base station 115 can aggregate the received data, determine in-phase and quadrature data (I and Q data, respectively) based on the received and aggregated digitized data, and send the aggregated data as in-phase and quadrature data to the processing system 120. The filtering and amplification of the raw magnetometer signals provided to the data acquisition module 130 allows the system to achieve a high dynamic range of frequencies of interest, for example, frequencies below 1 kHz (such as 130 Hz and 30 Hz) by rejecting large ambient direct current (DC) magnetic signals. The bandwidth and design of the filters used in the hardware and / or software of the system 100 can be selected to reject unwanted signals in the environment (e.g., 50 Hz and 60 Hz signals generated by alternating current (AC) lines) while maintaining sufficient bandwidth in the demodulated signal to recover the motion of the object.

[0052] The sensor 125 may include an infrared (IR) camera, a thermal imaging camera, an ultrasonic distance sensor, a video camera, an electro-optical (EO) camera, and / or a surface / depth map camera. The sensor 125 creates an additional information image or video (e.g., an optical image) of at least the OD 107. In some embodiments, the sensor 125 sends the image or video to the processing system 120 for further analysis. The system 100 may include multiple sensors 125. The sensor 125 may also be used to detect the presence of an object in the OD 107. Detection of the presence of an object in the OD 107 may be used to trigger a scan of the system 100. In some embodiments, the sensor 125 may include a radio frequency identification (RFID) reader.

[0053] The system can also present an image to the operator via display 165, in which the visible portions of the visitor and / or their belongings that are most likely to contain an object are segmented, highlighted, or otherwise provide notification to the operator and help the operator respond. In addition, aspects of the object can be determined based on the image obtained from the depth camera. The obtained aspects can be associated with the classification of the object. For example, if the object is in a plan view, the magnetic sensing algorithm can determine the object category, for example, whether the object is a laptop or an umbrella. If the object is hidden, the magnetic sensing algorithm can determine the part of the body of the person or the location on the person where the object is hidden, such as the person's clothing pocket, the person's ankle or wrist, or a bag that the person may be carrying. The data associated with these locations can be combined with information from the magnetic field data in a classification step, which utilizes all available information to achieve higher prediction accuracy during threat detection.

[0054] The processing system 120 includes multiple modules for processing magnetic field data and additional information images from the sensor 125 of the OD 107, including a data acquisition module 130, a calibration module 135, a reconstruction module 140, an automatic threat identification module 145, a presentation module 150 and a memory 155.

[0055] The data acquisition module 130 acquires a time series of voltage measurements representing magnetic field measurements from the DAS base station 115 and acquires additional information images from the sensor 125. In some embodiments, the sampling rate of the data acquisition module 130 is derived from the same master clock used to generate the transmit field via the transmitter 106. For each receiver 105, the data acquisition module 130 derives I and Q data from the time series via demodulation using a companion reference signal in post-processing. The timing of the I and Q data can be synchronized between the receivers 105, and the data acquisition module 130 can publish the synchronized data as frames (e.g., time slices) for further analysis by the system 100.

[0056] In some embodiments, the master clock of the system 100 can be distributed over several meters of space in the system using a low jitter low skew clock fanout and an internal network of low voltage differential signaling (LVDS) converters. This configuration allows the sampling rate to be an integer harmonic of each transmit frequency, thereby eliminating digitization errors that would otherwise damage the sensitivity of the system. By configuring each device in the data acquisition module 130 on the same clock domain, it can be correctly assumed that the receivers 105, which can be several meters apart, receive samples at the same time interval without drift due to frequency mismatch. Therefore, for a given frame, the data acquisition module 130 publishes a set of data for each receiver 105 and sensor 125. In some embodiments, data can be acquired and frames published at a rate sufficient to resolve the carrier frequency.

[0057] In some embodiments, the data acquisition module 130 removes static background signals (eg, primary fields). In some embodiments, the data acquisition base station 115 can remove static background signals (eg, primary fields) so that the I and Q data represent the secondary fields instead of the primary fields.

[0058] The calibration module 135 applies calibration corrections to the published data. The calibration corrections may include compensating the published data for serial time sampling. In addition, the calibration module 135 may compare the measured primary field to one or more field model predictions and compensate for any differences. In some embodiments, the calibration may account for amplitude and phase changes of the transmitter that occur due to normal wear, manufacturing variations, or temperature changes.

[0059] The reconstruction module 140 converts the calibration data into an image and / or feature map. An image may be created for each receiver 105 and / or based on a composite of measurements obtained by multiple receivers 105. The reconstruction module 140 may include determining a polarizability metric (e.g., tensor) and a location of an object.

[0060] The polarizability can be characterized as a proportionality constant relating the far-field response of an object to the primary field that caused it. The polarizability can have units of volume and can depend on the shape of the object, the magnetic permeability and electrical conductivity, and the frequency of the applied field. To determine the polarizability, in some embodiments, a best fit algorithm can be used to implement a minimum residual matched filter.

[0061] The transmitter field can be calculated based on the model of a rectangular coil. The receiver field can be calculated based on the dipole field along the specific axis of the sensor, so that the three-axis receiver node is considered as 3 independent and orthogonal dipoles.

[0062] In some embodiments, image data from sensor 125 can be used to further enforce sparsity constraints beyond those provided by a priori knowledge of items or subjects that may occupy OD 107. Specifically, images of OD 107 acquired by sensor 125 can be used to determine the spatial location of an object (e.g., voxels of OD 107 where the object is located and empty voxels of OD 107). Empty voxels do not contain any objects and therefore can be considered zero for compressed sensing (e.g., enabling better and / or faster estimation of solutions to underdetermined linear systems).

[0063] Additionally, an appropriately sized OD 107 can produce a scene that is sparse enough for compressed sensing. For example, if OD 107 is a volume of 2 meters by 1 meter by 0.5 meters and is divided into 8,000,000 5-mm voxels, a typical human being located within the OD 107 will only occupy about 10% of the voxels (e.g., about 800,000 voxels) at any given moment. The set of polarizable objects retrieved from sensor 125 can be used to determine the three-dimensional surface within the volume of OD 107, thereby determining the voxel in which the person is located. When retrieving the set of polarizable objects, empty voxels can be forced to zero, while non-zero voxels can be changed during reconstruction (e.g., can be treated as variables to find an optimal reconstruction solution for an underdetermined linear system).

[0064] The reconstruction module 140 can reconstruct the magnetic retrieval set of one or more polarizable objects. In addition, the reconstruction module 140 can create an aggregate retrieval set of polarizable objects by combining the retrieval sets of multiple independent polarizable objects. In some embodiments, the reconstruction module 140 can treat all receivers 105 as a large sparse aperture and reconstruct a single retrieval set of polarizable objects using information obtained from all receivers 105 in a single aperture.

[0065] The reconstruction module 140 can perform localization of the object using multiple time slices. This method can use a single model fitting method to solve for the position (e.g., x, y, and t intersection), velocity, and polarizability tensor of the object. An exemplary localization method will be described more fully below.

[0066] The reconstruction module 140 can generate a feature map based on the reconstructed image. The feature map can include characteristics or features of the magnetic measurements. Statistical analysis can be performed across multiple images. Some exemplary features include field amplitude, field phase, and polarizability tensor properties (discussed further below). Other features are also possible.

[0067] The automatic threat identification module 145 analyzes the image and / or feature map to determine the presence of a threat object. Threat objects may include dangerous items that a person may conceal on their person, such as firearms and explosives. The automatic threat identification module 145 may use, for example, a classifier that evaluates the feature map generated by the reconstruction module 140 to identify threats. The classifier may be trained on known threat signatures. In some embodiments, the threat identification process may compare the determined image to a library of predetermined polarizability signatures.

[0068] In some embodiments, features (eg, categorical variables) may include field amplitude, phase, and polarizability tensor properties at one or more operating frequencies.

[0069] The presentation module 150 generates or presents an image representing the results of the threat identification analysis performed by the threat identification module 145. The image can be presented on the display 165. For example, the presentation module 150 can show an avatar of the scanned person and any identified threats. The presentation module 150 can show that the automatic threat identification module 145 did not detect any features of the threat.

[0070] Figure 2 205, the three transmitters are arranged so that they are oriented orthogonally to each other, so that the configuration of 205 is capable of measuring all dimensions of static (e.g., stationary) objects. In diagram 210, the four transmitters are arranged so that they are positioned in the same plane, but offset in space, so that all dimensions of static objects can be measured. In diagram 215, the two transmitters are arranged orthogonally to each other, so that all dimensions of objects in motion can be measured. Similarly, in diagram 220, the two transmitters are arranged so that they are positioned in the same plane, but offset in space, so that all dimensions of objects in motion can be measured. As discussed in more detail below, in some embodiments, the system 100 may include transmitters 106 arranged according to the configuration shown in diagram 220. In some embodiments, such a configuration can provide a desirable form factor and reduced cost.

[0071] Figure 3 is a schematic diagram showing the arrangement of an exemplary personnel inspection system 300 according to some embodiments. Two transmitters 106 are arranged so that they are positioned in the same plane, but offset in space, so that all dimensions of objects in motion can be measured. In this example, the transmitter 106 is a coil capable of transmitting between 20 Hz and 1000 Hz, and can be configured to transmit multiple signals with frequency offsets so as to work simultaneously. For example, the transmitter 106 can operate at 30 Hz and 130 Hz. Other offset frequencies are possible, for example, 5-10 Hz. In some embodiments, the transmitter 106 can operate (e.g., be driven) at a first frequency between 20 Hz and 40 Hz and a second frequency between 120 Hz and 140 Hz. Considerations for selecting a frequency may include: lower frequencies increase the relative amplitude of ferrous metals compared to non-ferrous metals, higher frequencies may have better background noise for certain fluxgate sensors, and certain frequencies (e.g., 50 Hz and 60 Hz, harmonics) may be avoided entirely due to interference in typical environments. The receivers 105 are arranged vertically on posts (eg, vertical poles) on either side of the transmitter to provide dual channels to allow individuals being inspected to pass through the system.

[0072] In order to recover the magnetic signature of an object being detected independently of its position and orientation, it may be desirable to collect measurements of the object while it is exposed to a magnetic field radiated toward it from orthogonal directions. Figure 3 The configuration of transmitter 106 shown may reduce the total number and overall complexity of coils of an inspection system, which may reduce overall operating and maintenance costs as well as the size of the physical footprint of the inspection system.

[0073] Figure 3 The configuration of the transmitter 106 shown in can enable some exemplary systems disclosed herein to collect measurements of an object when the object is interrogated by magnetic fields in substantially orthogonal directions, so that magnetic characteristics can be determined substantially independently of the position and orientation of the object. At the same time, in order to reduce cost and physical footprint, the total number of coils and the complexity of the system can be minimized. By exploiting the movement of the object past the transmitter, field diversity can be achieved in a system with a minimal transmitter configuration (e.g., two transmitters). In addition, system complexity can be further reduced by simultaneously powering two transmitters with orthogonal time-varying modes (e.g., two sine waves with slightly offset frequencies). The frequencies can be similar enough that dispersion in the polarizability index can be negligible.

[0074] Figure 4 is a process block diagram illustrating an exemplary process 400 of an exemplary inspection system in accordance with some aspects of the current subject matter.

[0075] At 405, the processing system 120 may receive data representing samples obtained by the plurality of magnetic field receivers 105. For example, magnetic field samples may be obtained by the receivers 105, and images (e.g., video) may be obtained by the cameras or other sensors 125. An event may be identified (e.g., identifying that a person is approaching and / or entering the observation domain). For example, an event may be identified based on an event data signal received from a photocell test, occupancy of the system, or motion in the field of view of a camera feed. The received event data may cause the system to initiate object detection, or alternatively, the system may be configured to search for an object in the absence of event data.

[0076] At 410, the data acquisition module 130 aggregates a large amount of magnetic field data for processing by subsequent components. For example, the data acquisition module 130 may send a copy of a circular buffer containing magnetic field time samples from multiple receivers in the system. The calibration module 135 may be configured to account for deviations in magnetic field data compared to a pre-prepared model including amplitude, phase, or environmental characteristics. The reconstruction module 140 may be configured to determine the best fit object or an object that may be defined by a set of attributes including position, velocity, time offset, and polarizability index. The best fit may be determined by a cost function that measures the difference (e.g., residual) between the actual measured value and the model predicted value, and may also include other attributes that may indicate the rationality of the solution, such as the isotropy of the polarizability index. The polarizability index may include a complex tensor that includes 1-6 unique elements that characterize the directional polarizability components of the object at one or more frequencies emitted by the transmitter 106. These attributes (including the polarizability index) may be determined simultaneously or continuously using an optimization routine (e.g., a gradient descent algorithm or a nested parameter search). For example, the model can be an isotropic or anisotropic model, with uniform or non-uniform motion through the scan area. In some embodiments, features derived from the polarizability index of the object can characterize the shape, magnetic permeability, and electrical conductivity of the object in units of volume.

[0077] At 420, the automatic threat identification module 145 may classify the object as a threat or a non-threat. The threat / non-threat decision may be based on physical properties, such as matching the polarizability index or a subset of its components to known polarizability index examples of previously determined threat objects (e.g., a gun barrel or a knife). Alternatively, the threat / non-threat decision may use a classifier trained through a machine learning process, where many labeled examples of threats and non-threats are used to train a magnetic sensing algorithm to determine whether a newly detected object should be characterized as a threat or a non-threat. Classification may include using frequency and shape information to determine threat / non-threat.

[0078] Finding multiple objects in the same domain (e.g., spatially or temporally) can indicate a detection event associated with a visitor passing through the inspection system disclosed herein. Compared to a classifier that operates independently on each object, the magnetic sensing algorithm can combine these objects for classification through an aggregate classifier that can additionally use the combined information. For example, the classifier can add together some attributes of adjacent objects as if they constitute an underlying large and / or distributed object that is best considered as a whole.

[0079] At 425, a classification may be provided. The classification may be provided, for example, via a display such as display 165. In some embodiments, the classification may be provided to a backend security management system that may coordinate multiple assets (e.g., screening or inspection equipment). In some embodiments, the classification may be stored in memory 155 of processing system 120. In some embodiments, the classification may be stored in a database configured within the inspection system to store polarizability indices associated with determined threats and determined non-threats.

[0080] At 430 , system 100 may repeat the steps of process 400 in an iterative manner.

[0081] Figure 5 is a process block diagram illustrating an exemplary process 500 for determining a polarizability index of an object in an exemplary system 100 according to some aspects of the current subject matter. In some embodiments, a set of magnetic field samples can be provided as an input to process 500. The set of magnetic field samples can be collected simultaneously while the object is detected using a set of transmission fields. Process 500 can include introducing a set of trial solutions that can be solved independently by pseudo-inverse, from which the trial solution with the smallest residual can be selected. Process 500 can be performed to determine isotropic and anisotropic polarizability indices. Process 500 can be extended to include samples collected over time while the object is in motion relative to system 100.

[0082] At step 505, the automatic threat identification module 145 may define a set of trial solutions. The set of trial solutions may be indexed to include magnetic field samples associated with possible locations where an object may be present. For each location, the automatic threat identification module 145 may determine a corresponding polarizability index based on calculating a transfer matrix and pseudo-inverse for each magnetic field sample to determine a set of trial solutions. In some embodiments, the transfer matrix and pseudo-inverse may be predetermined and stored in the memory 155.

[0083] At step 510, the automatic threat identification module 145 may calculate the associated polarizability index and the associated residual for each trial solution. For example, the isotropic polarizability index may be determined based on selecting the trial solution that best fits the magnetic field sample. When determining the anisotropic polarizability index, the polarizability index may be considered as a complex symmetric tensor defined by 6 unique polarizability elements. A pseudo-inverse may be applied, and the polarizability index associated with each of the 6 unique polarizability elements may be calculated.

[0084] At step 515, the automatic threat identification module 145 may select a final trial solution. The final trial solution may be selected as the trial solution with the smallest or lowest residual. During operation, the automatic threat identification module 145 may capture magnetic field samples in real time. The pseudo-inverse may be applied by matrix multiplication. Subsequently, a transfer matrix may be applied to calculate the residual. The trial solution with the lowest or smallest residual may then be selected.

[0085] At 520 , automatic threat identification module 145 may repeat the steps of process 500 in an iterative manner.

[0086] Figure 6 600 is a process block diagram illustrating an exemplary process for detecting objects on an individual in a large group of people using an exemplary inspection system in accordance with some aspects of the current subject matter. The inspection system described herein is also capable of accurately detecting metal objects on a group of people walking together, and can be configured to detect metal objects on a large number of people at once, compared to conventional systems that are limited to scanning one person at a time through the inspection system. Because a large number of people gathered in an unorganized manner can make it difficult for the inspection system to determine when to start and / or stop scanning, the magnetic sensing algorithm used in the system described herein can dynamically adapt to situations where multiple objects can be sensed simultaneously and where objects pass through the system in a temporally overlapping manner (which can result in situations where there is no clear start or stop of scanning).

[0087] At step 605, the processing system 120 may process the sensor data received from the sensor 125 in a streaming manner by the magnetic sensing algorithm, such that each time the one or more objects pass through the system, the magnetic sensing algorithm is able to discover the one or more objects and is able to identify when insufficient data has been collected about the objects. The processing system 120 may determine that a sufficient amount of data has been collected based on the goodness of fit of the model to the real data, or by the proximity of the discovered objects to the end of the data buffer (e.g., the current time). If the processing system 120 determines that insufficient data has been discovered, nothing is done, and the magnetic sensing algorithm simply executes again at the next available moment or after an allotted amount of additional data has been collected.

[0088] At step 610, the processing system 120 can then store and account for the found objects to avoid repeatedly detecting the same objects. For example, the modeled signals associated with the found objects can be recalculated and subtracted from the measurement data before searching for the next object. The magnetic sensing algorithm can also be biased to search for objects far from previously found objects, for example, by modifying the cost function in the reconstruction module 140 to penalize objects found near already stored objects.

[0089] The processing system 120 and magnetic sensing algorithm can be configured to execute at regular intervals so that the amount of time that elapses from when enough data has been collected on an object to when the magnetic sensing algorithm is executed to find the object can be minimized. The magnetic sensing algorithm can detect an object to be found in a time period that extends slightly into the future (e.g., beyond the point in time at which object data has been collected and processed so far (if this is the best fit to the available data)), and use that scenario as a criterion for knowing when to wait for more data.

[0090] At step 615, the magnetic sensing algorithm of the processing system 120 can determine "quiet times" by opportunistically processing the retained buffer to re-estimate the background noise in subsequent executions, which "quiet times" can be defined as time periods when the processed signal changes little relative to adjacent time periods. The "quiet time" data can be used as a proxy for situations where there are no objects near the sensor. Considering the situation where one or more objects may be near some sensors but not near other sensors at different points in time, a different set of optimal "quiet time" samples can be found for each sensor. When an object is found, it can be stored in a buffer or memory 155 of the processing system 120. This buffered data can be processed in the next iterative execution of the magnetic sensing algorithm, which subtracts all previously discovered objects from the available data to avoid contamination / distortion in the current object search.

[0091] Figure 7 is a schematic diagram illustrating an exemplary embodiment of a personnel inspection system 700 according to the subject matter disclosed herein. The system 700 includes Figures 1 to 3 The components described in the discussion are similar components that perform similar functions.

[0092] Conventional personnel screening systems often include an arch through which an individual must pass in order to undergo threat assessment and detection. In these conventional screening systems, the arch or some other elevated member may transmit wires between one or more components of the screening system to carry power and data signals, ensure alignment of sensors disposed on or within the components of the screening system, and increase the stability of the overall structure of the screening system with respect to environmental conditions (e.g., wind or uneven mounting surfaces). However, conventional screening systems that include arches or other elevated members are often unsightly and may induce feelings of distrust, anxiety, and claustrophobia in individuals passing through the arch.

[0093] about Figure 7 The improved inspection system described provides the advantages of transmitting power and data signals to various inspection system components, ensuring alignment of the inspection system components, reducing emotional response of individuals, and not requiring individuals to pass through arches or overhead elements of the inspection system.

[0094] Another problem addressed by the improved personnel detection system disclosed herein is the non-uniformity of the operating environment in which the system operates. The operating environment may include hard but somewhat smooth surfaces (e.g., tile floors, asphalt surfaces, or concrete floors), as well as softer surfaces that may be non-uniform (e.g., a sand-covered surface at a beach entrance or a grass surface at an outdoor music festival entrance).

[0095] The system 700 disclosed herein can be deployed in various different locations indoors and outdoors, and can be configured based on different types of surfaces where the system can be positioned. For example, the base plate 710 can be configured as a universal base plate, and various modular mounting systems can be attached to the universal base plate according to the site where the system is located and / or the surface where the system 700 is located. The modular mounting system can achieve positioning, leveling, and connection or adhesion to the surface of the position where the system is deployed for operation. For example, the base plate 710 can be configured with a suction cup on the bottom side (the side of the surface facing the position where the system is deployed) to fix the base plate 710 to a hard surface, such as a tile or marble. In some embodiments, the base plate 710 may include a clamping or perforating mechanism that can fix the base plate 710 to a soft surface (e.g., a carpet). In some embodiments, the base plate 710 may include a screw or a screw drill type mechanism to connect the base plate 710 to a soil surface or the ground. In some embodiments, the base plate 710 may be configured with a base plate frame hidden in the base plate 710 and providing a permanent installation of the system.

[0096] Traditional security systems require visitors to pass through a portal for detection, where an arch is used to transmit data signals and power between each component of the system, ensure proper alignment of the various sensors, and add stability to the entire structure. Improper alignment may create unwanted deviations in system performance and arch structural movement, for example, in environments that may include high winds. Improper alignment may result in the creation of unwanted distortions that are difficult to separate from the desired signal. However, the arch lacks visual beauty and can create a feeling of distrust and unpleasantness for visitors passing through the arch.

[0097] like Figure 7 As shown, the system 700 disclosed herein can be configured to provide a column 705 (which can include Figure 1 and Figure 3 705) and provide correct orientation / leveling of the column 705. In some embodiments, the base plate 710 can be configured to accommodate the column 705 and ensure proper alignment of the column at the appropriate and predetermined position. In some embodiments, Figure 7 The system 700 shown may include an inclinometer configured to guide an operator during the leveling process. In some embodiments, the inclinometer may be used by the magnetic sensing algorithm to compensate for known misalignments. In some embodiments, the system 700 may include an accelerometer to compensate for structural instability in the system and track the movement of system components. Accelerometer data may be provided to the magnetic sensing algorithm. In some embodiments, the signals generated by the inclinometer and / or accelerometer and the power supplied to the inclinometer and / or accelerometer may be routed through the base plate 710. In some embodiments, the signals generated by the inclinometer and / or accelerometer may be transmitted wirelessly to the system 100. In some embodiments, the base plate 710 may include a plurality of slots for accommodating a column 705 or other structure suitable for mounting one or more sensors 125.

[0098] The base plate 710 may be a semi-rigid low pad configured to accurately locate the position and orientation of the base of the Rx and Tx columns 705. The inclinometer 415 may determine the relative tilt angle between the Rx and Tx columns 705 when installed. When the system 700 determines that the relative tilt angle is above a certain threshold, the operator may be alerted and asked to improve or remedy the leveling of the column 705.

[0099] For example, in some embodiments, the tilt angle may be used to modify the position and orientation of the sensor. The position and orientation of the sensor may be further used to create a sensor model that the reconstruction module 140 relies on.

[0100] In some embodiments, the accelerometer 720 can be configured to dynamically track tilt angles, which enables the system to track the motion of various sensors in the system. For example, when combined with known field gradients that can be determined with respect to the motion of the various sensors, the calibration module 135 can predict the expected distortion of the measured motion. This prediction can then be removed from the measured signal to recover a more accurate representation of the measured signal as if there were no motion. In this example, the field to be modeled and removed can be the measured motion, e.g., the displacement or tilt of the sensor multiplied by the field gradient for each direction / type of motion.

[0101] A magnetic field-based personnel screening system may detect or receive magnetic fields that have been transmitted by the system (e.g., by transceiver 106), as well as secondary fields of objects that may be affected by the presence of metal in the environment in which the screening system is deployed. Ambient magnetic fields (e.g., those that may be reflected by metal near the screening system, such as metal in a metal floor in which the screening system is located) may negatively impact the screening system's ability to accurately discern and detect threat objects. Determining and characterizing the amount of metal in every environment in which the screening system is located may be difficult, expensive, and often impossible.

[0102] like Figure 7As further shown in , the system 700 can be configured to account for physical, metallic structures that may be present in the location where the system is deployed, such as a metal floor. The receiver 105 can detect the transmission field 730 generated by the transmitter 106, as well as the secondary field 735 of the detected object. The secondary field can be affected by the presence of metal in the environment near the system (for example, in the floor 725 below the system). The presence of metal in the floor can have a negative impact on the detection performance and accuracy of the system. It is difficult and expensive to locate and characterize metal in each environment. In order to account for the presence of metal in the environment, the system 700 can fit an appropriate parameterized model of the metal present in the environment based on determining how the magnetic field 735 emitted from the metal floor 725 is distorted relative to a known model. The system 700 can include an image model, whereby the metal in the floor 725 is interpreted as a composite weighted mirror image of the entire system including the transmitter 106 and the receiver 105. For example, the image model can simulate a perfect electrical conductor, which is known to change the magnetic field in a very predictable and analytically described manner. The image model can then be parameterized by an overall composite weight of depth relative to the sensor coordinate system and an amplitude less than or equal to 1. The measurements of the emission field 730 can be used to perform an optimization routine by fitting the composite weights of the mirrors and the plane depths of the mirrors. Such an optimization routine can search (e.g., by trying various predefined solutions or by gradient descent) for model parameters that best fit the measured data. For example, the optimization routine can perform a search by trying previously predefined solutions or using gradient descent optimization. These model parameters can then be used for system operation during threat / non-threat detection. As a result, the system 700 can more accurately automatically detect objects in a deployed environment that may or may not include metal near the system.

[0103] Figure 8 is a schematic diagram showing an exemplary configuration of a personnel inspection system 800 as disclosed herein, which includes sensors and transmitters 106 located at different locations, according to some embodiments. In the example shown, the system 800 includes two transmitters 106, two vertical columns 805A and 805B, each of which includes a plurality of three-axis fluxgate receivers 810. Each of the three-axis fluxgate receivers is shown as an "x" and is configured on or within the columns 805A and 805B. The system 800 also includes a vertically oriented column 805C located near the two transmitters 106. The column 805C includes a plurality of dual-axis fluxgate receivers 815. The number of each type of component (e.g., transmitter 106 and / or fluxgate receiver 105) can be determined by balancing the cost and complexity of the additional component with the marginal gain that the additional component adds to the fidelity of the magnetic sensing algorithm.

[0104] The system 800 may also include a camera 820 integrated with the system 800. In some embodiments, the camera 820 may be configured on or within one or more of the poles 805A, 805B, and 805C, as well as on or within the transmitter 106. The camera 820 may generate images and videos in a streaming or recorded manner. The integrated camera 820 may be configured to have an appropriate viewing angle. For example, in some embodiments, the camera 820 may be a rear-facing camera.

[0105] When an alarm is generated in response to a threat being detected by a personnel screening system, the alarm must be resolved by the screening system operator or other security team member. Resolving the alarm requires the operator or security team member to interact with the individual and follow security protocols to search for the individual. The operator or security team member may not always follow security protocols correctly. Security team management or the screening system operator may not determine if proper security and search protocols were followed. About Figure 8 The described improved personnel screening system solves these problems.

[0106] For example, the camera 820 can provide images or video with sufficient field of view so that the supervisor of the inspection system operator can evaluate the inspection system operator's response to an alarm or detected object. In this way, the camera 820 integrated within the system 800 can enable the supervisor or management of the inspection system operator to evaluate the operator's compliance with screening procedures or policies, provide training feedback, provide images or video for evidentiary purposes, and record the alarm resolution actions taken by the operator. In some embodiments, the images or video from the camera 820 can be combined with images of the detected object generated by the system and used to refute or support allegations of improper handling by the inspection system operator.

[0107] Personnel inspection systems that use magnetic sensing are limited to sensing metal on the person passing through the inspection system and are unable to detect the body or physical characteristics of the person passing through the inspection system. As such, the personnel inspection system lacks a concept of the person passing through the inspection system and can therefore generate limited information about the individual to the inspection system operator.

[0108] This article describes and Figure 8The improved inspection system shown in the figure may include a camera 820 for detecting a person passing through the inspection system 800. In some embodiments, the camera 820 may be a depth camera. The depth camera may include a camera configured to use stereo vision to calculate the depth in the image acquired by the depth camera. The depth camera may include a depth sensor and an infrared projector. In some embodiments, the depth camera may include a color sensor configured to detect light in a red, green, blue (RGB) scale, also known as an RGB sensor. The output of the depth camera can be used to determine the position, direction, or configuration of the detected object, as well as the speed or gait of the object passing through the inspection system. The output of the depth camera can allow the inspection system to count the number of objects or individuals passing through the inspection system and track one or more individuals passing through the inspection system. The inspection system disclosed herein (when configured to include a depth camera) may provide many advantages over an inspection system that may not include a depth camera. These advantages may include more rapid identification of threatening objects or individuals, and providing more robust notification or alarm data to identify threatening objects or individuals.

[0109] In such an embodiment, the depth camera 820 can register its coordinate system to the Figure 1 An automatic threat identification subsystem 145 is shown configured within the inspection system 100 .

[0110] In this way, the system can understand the location / configuration of the visitor and any metal objects on their person at the same time in a common coordinate system. Such an embodiment enables the magnetic sensing algorithm to point to those voxels occupied by the visitor and / or specific areas on the visitor (e.g., pockets, bags, and ankles). For example, the voxels of OD 107 can be compared with the pixels obtained via the depth camera to determine which voxels are occupied (e.g., there are pixels with coordinates close enough to the voxel), unoccupied (e.g., there are no pixels with coordinates close enough to the voxel and there is no chance of occlusion), or unknown (e.g., it has been found that the pixel may obscure OD107, regardless of whether the object exists in the voxel). The magnetic sensing algorithm can then be restricted to searching only for objects in occupied and / or unknown voxels.

[0111] Additionally, utilizing knowledge of the visitor's speed and gait in the magnetic sensing algorithm can improve the accuracy of the algorithm, as can counting and tracking visitors entering and exiting the system for statistical reporting. The accuracy of the optimization routine is generally improved when the number of variables that the optimization routine must solve for can be reduced or constrained. Knowing the visitor's speed and position will allow these properties to be constrained or imposed on the objects that the optimization routine is solving for.

[0112] In such an embodiment, the depth camera 820 (or any sensor 125 or combination of sensors capable of providing RGB and depth values ​​at various pixels, such as a structured light camera or a stereo camera) can be integrated directly into the Figure 1 In this way, the image data obtained by the depth camera can be used in the magnetic sensing algorithm. Given the knowledge of the position, orientation and lens characteristics of the depth camera, the pixels obtained by the depth camera can be registered to Figure 1 The coordinate system of the automatic threat identification subsystem 145 is converted into the three-dimensional coordinate system of the magnetic subsystem through a series of transformations so that each pixel of a given returned depth value can be converted into the three-dimensional coordinate system of the magnetic subsystem through a series of transformations.

[0113] In this way, the depth camera 820 can identify which voxels in the scanning area of ​​the magnetic subsystem are occupied and / or unoccupied at a given moment, and the magnetic sensing algorithm can be guided accordingly. The speed and gait of the subject can also be estimated and used in the magnetic sensing algorithm to more accurately identify any hidden objects. Objects can be discovered by the magnetic sensing algorithm using specific three-dimensional coordinates. Objects can then be associated with a subset of pixels in the depth camera image based on a simple distance threshold.

[0114] Additionally, the subset of pixels can be associated with a larger neighboring object identified and segmented by the depth camera 820 by its depth value or RGB value, or both. For example, the system can identify the outline of a person who is most likely carrying the discovered object. The neighborhood around the object or the attributes of the person identified as holding the object can be used for its classification. For example, the face of the person can be compared to a watch list by a facial recognition algorithm, and the past history or knowledge of this person can be used to classify the object. At the same time, the threat overlay image formed by applying the threat overlay on top of the RGB image can be enhanced by highlighting the visible container of the object or the person holding the object, or both, to assist the operator in identification. This helps improve the operator's reaction time and accuracy in resolving alarms triggered by the system.

[0115] Fig. 9is a schematic diagram showing an exemplary configuration of a personnel inspection system 900 as disclosed herein, which, according to some embodiments, includes multiple cameras. By configuring the inspection system 900 with multiple cameras, the inspection system can better determine which specific individual to search for and where to search for the specific individual. The personnel inspection system lacking threat localization utilizes multiple cameras to generate multiple perspectives, so that a large number of unorganized people can be screened without forming a personal screening queue, and potential threats located in three dimensions can be identified. In addition, when an alarm related to a detected threat is generated, the use of multiple cameras that generate multiple perspectives enables the system to visually identify individuals associated with the detected threat using captured images, so as to notify the inspection system operator or other security personnel where to search for the identified individual for the detected threat. Alternative solutions may use lights or separate still images to identify individuals associated with the detected threat, which may only provide an obstructed perspective of a single camera. A single perspective may not be sufficient to safely and effectively identify detected threats on individuals in a crowd passing through the inspection system.

[0116] like Fig. 9 As shown, system 900 includes poles 905A and 905B. Pole 905A and 905B include a plurality of three-axis fluxgate receivers 910. Pole 905C includes a plurality of two-axis fluxgate receivers 915. Pole 905C also includes a transmitter 106.

[0117] In some embodiments, multiple individuals may pass between posts 905A, 905B, and 905C at any one time. For example, two people may pass posts 905A and 905B at the same time, which may cause a single sensor (e.g., Figure 1 The field of view of the sensor 125 described above is obscured. As a result, in embodiments including a single sensor, the field of view of the sensor may be obstructed. In some embodiments, multiple sensors may be included in the inspection system to avoid obstruction of the field of view at a single sensor and to provide multiple viewpoints.

[0118] like Fig. 9As shown, multiple cameras, for example, cameras 905A and 905B, can be configured in the personnel inspection system 900. Cameras 905A and 905B can be registered to the system coordinate system representing the coordinates of OD 107 to perform threat positioning from multiple viewpoints. Since objects are tracked over time, each camera can determine the location of the threat across multiple time slices or in multiple time slices. Therefore, even if the view from a single camera 905A is blocked at a certain moment, the possibility that the view from the second camera 905B is also blocked at all available moments is very low. The system can provide images and / or videos of all view angles to the inspection system operator, and can use images and / or videos from one or more cameras to determine the actual threat location. In some embodiments, the presentation module 150 can also be configured to automatically select the best view angle and / or time slice for presentation to help the operator identify and respond to detected threats. For example, an image with the largest number of occupied pixels near the found object is likely to give a reasonable and unobstructed view of the object.

[0119] For example, Fig. 9 As shown, system 900 includes two fisheye cameras 905A and 905B. Each camera is mounted on opposing poles 905A and 905C to capture the scan area from two very different perspectives. Each pixel in the images and / or videos generated by the two cameras is mapped to the coordinate system of system 900 on multiple planes along the direction of the object or person passing through the channel formed between poles 905A and 905C. System 900 can be configured to generate an alarm positioned in a specific viewing plane associated with camera 905A or 905B. Visual representations or threat indicators can be superimposed on the images and / or videos generated by each camera and centered on the appropriate registered pixels corresponding to the detected threat. System 900 can be configured to repeat these operations for multiple viewing planes as an object or person passes through the system to generate a threat overlay image or video from each camera.

[0120] Video and / or images from one or all of the cameras may be displayed to an operator via a computing device (e.g., a laptop, tablet, or other mobile computing device configured with a display) communicatively coupled to the system 900. In some embodiments, the system may utilize a threshold criterion to determine whether the view of one of the cameras is obstructed. For example, the system 900 may determine occlusion by evaluating motion in the video stream, or by incorporating a depth camera to determine a depth estimate for the pixel in question. If the system 900 determines that the depth estimate is inconsistent with the 3D coordinates returned by the system, the view should be considered obstructed, and the system may determine that the image is not useful for guiding the operator's response.

[0121] In some embodiments, the inspection system 900 can process additional sensor data (e.g., video or images) and can relay images, videos, or video frames of the subject of the classification results together with or superimposed on the classification results. For example, the inspection system 900 can superimpose a graphical indicator on the image, and the graphical indicator can identify the detected threat or object. In some embodiments, additional metadata about individuals, detected objects, or system parameters can be provided to the image superimposed with the graphical indicator. In some embodiments, the image superimposed with the graphical indicator can be provided to an individual (e.g., an inspection system operator or security guard), who can be located further downstream in the sequence of objects or individuals inspected by the inspection system. In this way, the detection system can provide the inspection system operator or security guard with an image superimposed with a graphical indicator for additional monitoring and / or interception of detected objects, threats, or individuals.

[0122] Although some variations have been described in detail above, other modifications or additions may be made. For example, the number of receivers is not limited, and some embodiments may include any number of receivers. The transmitter is not limited to a specific frequency, for example, coils with different characteristics (operating frequency, location, etc.) may be used. Different reconstruction algorithms may be used, and different features may be used for threat detection.

[0123] Without limiting in any way the scope, interpretation, or application of the claims appearing below, the technical effects of one or more exemplary embodiments disclosed herein may include one or more of the following effects, for example: Some exemplary embodiments of the current subject matter may perform threat detection and identification in a high-clutter environment where individuals may carry personal items (e.g., cell phones and laptops) without removing personal items. In some embodiments, the personnel inspection system may perform threat detection and identification at a high throughput, allowing individuals to pass through metal detectors at normal walking speeds, so that individuals are not required to slow down for inspection, and in some embodiments, the inspection threshold may allow multiple individuals to pass through the threshold side by side (e.g., two or more side by side). In some configurations, multiple individuals walking in close proximity may be screened, thereby eliminating the need for the screened individuals to remain stationary during the screening process.

[0124] One or more aspects or features of the subject matter described herein can be implemented in digital electronic circuits, integrated circuits, specially designed application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) computer hardware, firmware, software, and / or combinations thereof. These different aspects or features can include embodiments in one or more computer programs that are executable and / or interpretable on a programmable system, the programmable system including at least one programmable processor, a storage system, at least one input device, and at least one output device, the programmable processor can be dedicated or general, it can be connected to the storage system to receive data and instructions from the storage system, and send data and instructions to the storage system. A programmable system or computing system can include a client and a server. The client and the server are usually far away from each other, and usually interact through a communication network. The relationship between the client and the server is generated by a computer program that runs on each computer and has a client-server relationship with each other.

[0125] These computer programs may also be referred to as programs, software, software applications, applications, components or codes, which include machine instructions for programmable processors and may be implemented in high-level programming languages, object-oriented programming languages, functional programming languages, logical programming languages ​​and / or assembly / machine languages. As used herein, the term "machine-readable medium" refers to any computer program product, device and / or equipment (e.g., disks, optical disks, memories and programmable logic devices (PLDs)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor. A machine-readable medium may store such machine instructions non-temporarily, for example, a non-temporary solid-state memory or a magnetic hard drive or any equivalent storage medium. A machine-readable medium may store such machine instructions alternatively or additionally in a transient manner, for example, a processor cache or other random access memory associated with one or more physical processor cores.

[0126] To provide interaction with a user, one or more aspects or features of the subject matter described herein may be implemented on a computer having a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) or a light emitting diode (LED) monitor for displaying information to a user) and a keyboard and pointing device (e.g., a mouse or trackball) through which a user can provide input to the computer. Other types of devices may also be used to provide interaction with a user. For example, the feedback provided to the user may be any form of sensory feedback, such as visual feedback, auditory feedback, or tactile feedback; and input from the user may be received in any form, including, but not limited to, acoustic, voice, or tactile input. Other possible input devices include, but are not limited to, a touch screen or other touch-sensitive device (e.g., a single-point or multi-point resistive or capacitive tracking pad), voice recognition hardware and software, an optical scanner, an optical pointer, a digital image capture device and related interpretation software, and the like.

[0127] In the above description and claims, phrases such as "at least one" or "one or more" may appear before a conjunction list of elements or features. The term "and / or" may also appear in a list of two or more elements or features. Unless otherwise implied or explicitly contradictory in the context in which the phrase is used, the phrase means any element or feature listed separately, or a combination of any listed element or feature with any other listed element or feature. For example, the phrases "at least one of A and B"; "one or more of A and B" and "A and / or B" mean "A alone, B alone, or A and B together", respectively. A similar interpretation also applies to lists including three or more items. For example, the phrases "at least one of A, B, and C"; "one or more of A, B, and C" and "A, B, and / or C" mean "A alone, B alone, C alone, A and B together, A and C together, B and C together, or A and B and C together", respectively. In addition, the term "based on" used in the above description and in the claims means "based at least in part on", thereby also allowing unlisted features or elements.

[0128] The subject matter described herein can be implemented in systems, devices, methods and / or articles according to the desired configuration. The embodiments set forth in the above description do not represent all embodiments consistent with the subject matter described herein. On the contrary, they are only examples consistent with the aspects related to the subject matter. Although some variants have been described in detail above, other modifications or additions can be made. In particular, in addition to the features and / or variants described herein, other features and / or variants can also be provided. For example, the above embodiments can be directed to various combinations and sub-combinations of the disclosed features and / or combinations and sub-combinations of multiple other features disclosed above. In addition, the logic flows depicted in the drawings and / or described herein do not necessarily require the specific order or sequence shown to achieve the desired results. Other embodiments may fall within the scope of the appended claims.

Claims

1. A method for performing a personnel inspection, comprising: receiving data representative of samples obtained by the plurality of magnetic field receivers, a combination of a first magnetic field and a second magnetic field resulting from an interaction of the first magnetic field with the object, from a plurality of magnetic field receivers including magnetic sensors, the first magnetic field including at least two frequency components, the first frequency component being between 100 Hz and 1000 Hz and the second frequency component being between 5 Hz and 100 Hz; Determine a polarizability index of the object using the received data, wherein the polarizability index represents a magnetic polarizability characteristic of the object, wherein the polarizability index includes a first polarizability index component determined based on at least the first frequency component and a second polarizability index component determined based on at least the second frequency component; Using the determined polarizability index, classifying the object as a threat or a non-threat, wherein the classification includes determining a first property of the object based on at least a first polarizability index component associated with the first frequency component and determining at least one material property of the object based on at least a second polarizability index component associated with the second frequency component; and Provides classification.

2. The method according to claim 1, wherein: The polarizability index of an object characterizes at least the shape, magnetic permeability and electrical conductivity of the object.

3. The method according to claim 1, wherein: The polarizability index of the object includes a complex tensor including at least six elements characterizing the directional polarizability components of the object at one or more frequencies employed by the transmission system.

4. The method according to claim 1, wherein: Determining the polarizability index includes solving a set of trial solutions via a pre-computed pseudo-inverse, determining the residual for each trial solution, and selecting the trial solution that produces the smallest residual.

5. The method according to claim 1, wherein: Determining the polarizability index involves: A set of trial solutions is defined, each trial solution including position, velocity and time displacement; Calculate the associated polarizability index and the associated residual for each trial solution; and A final trial solution is selected, the final trial solution comprising the trial solution in the set of trial solutions associated with the minimum residual.

6. The method according to claim 1, further comprising: The object is located within the inspected volume, including determining the object velocity, the object position, and the object time offset to a predetermined plane.

7. The method according to claim 1, further comprising: One or more signals for driving the magnetic field transmitter are generated at a first frequency and a second frequency.

8. The method according to claim 1, wherein: The plurality of magnetic field receivers include a fluxgate sensor that measures the amplitude and phase of a combination of the first magnetic field and the second magnetic field on at least three axes at the locations of the plurality of magnetic field receivers.

9. The method according to claim 1, wherein: The samples obtained by the plurality of magnetic field receivers characterize amplitude and phase; wherein classifying the object as a threat or non-threat using the polarizability index comprises comparing the amplitude, phase and polarizability index with a predetermined library of threat signatures.

10. The method according to claim 1, wherein: The first magnetic field is generated by at least a first magnetic field transmitter and a second magnetic field transmitter, which are oriented in a first direction and are spatially offset transversely to a predefined direction of movement and transversely to the first direction.

11. A personnel inspection system, comprising: A magnetic field transmitter configured to generate a first magnetic field, wherein the first magnetic field includes at least two frequency components, the first frequency component is between 100 Hz and 1000 Hz, and the second frequency component is between 5 Hz and 100 Hz; a plurality of magnetic field receivers including magnetic sensors configured to sample a combination of a first magnetic field and a second magnetic field resulting from an interaction of the first magnetic field with the object; as well as At least one data processor configured to at least: receiving data characterizing a sample acquired by a plurality of magnetic field receivers; Determine a polarizability index of the object using the received data, wherein the polarizability index represents a magnetic polarizability characteristic of the object, wherein the polarizability index includes a first polarizability index component determined based on at least the first frequency component and a second polarizability index component determined based on at least the second frequency component; Using the polarizability index, classifying the object as a threat or a non-threat, the classification comprising determining a first property of the object based on at least a first polarizability index component associated with the first frequency component and determining at least one material property of the object based on at least a second polarizability index component associated with the second frequency component; as well as Provides classification.

12. The personnel checking system according to claim 11, further comprising: A transmit driver is coupled to the magnetic field transmitter and is configured to generate one or more signals for driving the magnetic field transmitter at a frequency less than 1000 Hertz (Hz).

13. The personnel checking system according to claim 12, wherein: The transmit driver is configured to generate one or more signals for driving a magnetic field transmitter at a first frequency and a second frequency, the magnetic field transmitter being driven at the first frequency and the second frequency.

14. The personnel inspection system according to claim 12, further comprising a second magnetic field transmitter, wherein: The transmitting driver further includes: first drive electronics that drives the magnetic field transmitter; and Second drive electronics drives the second magnetic field transmitter.

15. The personnel checking system according to claim 11, wherein: The samples obtained by the plurality of magnetic field receivers characterize amplitude and phase; wherein classifying the object as a threat or non-threat using the polarizability index comprises comparing the amplitude, phase and polarizability index with a predetermined library of threat signatures.

16. The personnel checking system according to claim 11, wherein: The plurality of magnetic field receivers include a fluxgate sensor that measures the amplitude and phase of a combination of the first magnetic field and the second magnetic field on at least three axes at the locations of the plurality of magnetic field receivers.

17. The personnel checking system according to claim 11, further comprising: a data acquisition base station coupled to the plurality of magnetic field receivers and configured to filter, demodulate and digitize samples acquired by the plurality of magnetic field receivers; The data acquisition base station is further configured to determine in-phase and quadrature data characterizing the second magnetic field.

18. The personnel checking system according to claim 11, wherein: Determining the polarizability index includes determining the polarizability tensor.

19. The personnel checking system according to claim 18, wherein: Determining the polarizability index includes solving a set of trial solutions via a pre-computed pseudo-inverse, determining the residual for each trial solution, and selecting the trial solution that produces the smallest residual.

20. The personnel checking system according to claim 11, wherein: The at least one data processor is further configured to: The object is located within the inspected volume, including determining the object velocity, the object position, and the object time offset to a predetermined plane.

21. The personnel checking system according to claim 11, wherein: Determining the polarizability index involves: A set of trial solutions is defined, each trial solution including position, velocity and time displacement; Calculate the relevant polarizability index and the relevant residual for each trial solution; and A final trial solution is selected, the final trial solution comprising the trial solution in the set of trial solutions associated with the minimum residual.

22. The personnel screening system of claim 11, further comprising a base plate configured to be coupled to the first column and the second column, wherein: The magnetic field transmitter is coupled to the first pole and a plurality of magnetic field receivers are coupled to the second pole.

23. The personnel checking system according to claim 22, wherein: The base plate includes a modular mounting system, which is configured to connect the base plate to the surface on which the base plate is located, and the modular mounting system includes at least one connecting device, which is selected from the group consisting of multiple suction cups, multiple clamping mechanisms, multiple perforating mechanisms and multiple auger mechanisms.

24. The personnel checking system according to claim 22, wherein: The chassis includes a chassis frame concealed within the chassis, the chassis frame being configured to permanently mount the chassis at a location where the system is deployed.

25. The personnel checking system according to claim 22, wherein: The proximal end of the first post and the proximal end of the second post are coupled to the base plate, and the distal end of the first post and the distal end of the second post are not connected to an arch that couples the distal end of the first post to the distal end of the second post.

26. The personnel checking system according to claim 22, wherein: At least one of an accelerometer or an inclinometer is coupled to the first column and / or the second column.

27. The personnel checking system according to claim 11, wherein: The at least one data processor is further configured to determine a measure of distortion of the second magnetic field based on an amount of metal present at a location where the system is deployed.

28. The personnel screening system of claim 11, further comprising a first pole, a second pole, and at least one camera coupled to the first pole or the second pole, the at least one camera configured to capture images and / or video data of the object, wherein: The camera is selected from the group consisting of a depth camera, a rear-facing camera, a fisheye camera, an infrared camera, a thermal imaging camera, a surface image camera, a photoelectric camera, and a stereo camera.

29. The personnel inspection system of claim 28, further comprising a first camera connected to the first pole and a second camera connected to the second pole, wherein the first camera is configured to capture image data at a first perspective, and the second camera is configured to capture image data at a second perspective, the second perspective being different from the first perspective.

30. The personnel checking system according to claim 29, wherein: The image data captured by the first camera and the second camera are registered in the coordinate system of the system.

31. The personnel checking system according to claim 30, wherein: In response to classifying the object as a threat, the processor is further configured to generate a graphical indication identifying the location of the object and superimpose the graphical indication on the image data in the display.

Images