JP4376900B2 - Security scanner with capacitance and magnetic sensor array - Google Patents

Description

  The present invention generally relates to an improved apparatus and method for providing real-time 3D images of the contents of containers such as envelopes, parcels and baggage.

  Of course, the fight against crime and terrorism was a priority for all civilized societies. Since September 11, 2001, homeland security has been paying the greatest attention to airports, post offices and other areas dealing with mass entry. In recent years, in addition to the FBI and CIA in the United States, the “Department of Homeland Security (DHS)” has been established, indicating the priority of the US government to prevent further attacks. In the UK, airports often have fear of terrorist attacks.

  X-ray machines are commonly used to check airline passenger baggage and are effective in identifying objects of different densities, but have significant problems. (A) There are problems with radiation exposure, and in particular, members of security staff working with X-ray scanners will likely receive unacceptable radiation. (B) This machine is expensive to purchase and operate, and the operating cost is estimated to be about $ 1 million per X-ray machine each year, and is prohibited from use in situations where there is not much cash concentration. (C) Most X-ray machines can only provide 2D images, making it difficult for security staff members to identify dangerous objects. The generated video must be viewed or evaluated by an experienced / trained person. This is because there is currently no efficient automatic way to reliably trigger an alarm. (D) X-ray machines are bulky and non-portable, and security personnel cannot easily check a discarded bag using the X-ray machine. Therefore, there is a need for a separate and / or complementary security scanning tool.

  In the general aviation industry, such as small airports, non-standard operations, and sporting events or public theaters, daily use of X-ray machines is not practical. The most popular tool used for security checks is a metal detector. Metal detectors are very sensitive to ferrite materials, but cannot give information on shape and / or size, and can remove non-metallic objects (eg ceramic knives) and chemicals (eg explosives) It cannot be detected. These limitations of currently available detectors prevent the identification of dangerous objects and materials. Also, in many other cases, it is necessary to check bags and people for security reasons, for example, to manage entry into railway stations, public buildings and buses.

  Neither post offices nor parcel delivery services currently have tools to check the contents of envelopes, packages and boxes. Bombs, anthrax, guns and other illegal substances are sent through post offices and delivery services. This indicates the need to develop techniques and security scanning tools to address this issue. Various chemical sensors have been developed to detect explosives and chemicals, such as the Ionscan machine from Smith Detection (Elliot and Joets, 2003). However, if explosives or chemicals are sufficiently sealed, the chemical sensor will not detect them.

  In the US, Invision Technologies, Inc. has developed a CT scanner machine for security applications based on X-ray sensors (Invision, 2003). The sensing principle of such a machine is the same as a commonly used X-ray machine, but can provide 2D and 3D tomographic images. However, these machines have all the disadvantages of commonly used X-ray machines: radiation, expensive and bulky. X-ray tomography sensors must take measurements over a long period of time so that they can record a data set that can be used to generate an acceptable image. Recently, Invision announced a new machine based on quadrupole resonance analysis, the QScan QR500. In principle, this is similar to magnetic resonance imaging (MRI). Instead of using magnets, quadrupole resonance analysis uses high-frequency, low-intensity pulses in the probe that are carefully tuned to the molecular structure of the target material (Invision, 2003). However, this type of machine is limited to the detection of only certain liquid explosive materials, is large in size, is heavy and can span several tons.

  It has also been proposed to use penetrating non-ionizing radiation, eg millimeter waves, for security checks (Quinetic, 2001). Related organizations claim that this is non-hazardous radiation, but millimeter or microwave radiation is well established to be potentially harmful because of the high operating frequency of 35 or 95 GHz. (Quinetic, 2001). A similar scanner using millimeter-wave radiation has also been announced at the Pacific Northwest National Lab in the United States. The scanner is declared to allow airport security personnel to “see” the full spectrum of hidden weapons, including non-metallic threats such as plastics, explosives and ceramic knives. As a legitimate safety issue that has not yet been fully addressed, individuals are allowed to refuse to “receive microwaves”. This prevents the security agency from forcing anyone to scan.

  Tomography is widely used for diagnostic purposes in hospitals. X-ray CT scanners can provide 2D and 3D images of objects in the human body. In recent years, industrial process tomography (IPT) has been rapidly developed. The principle of IPT is similar to medical tomography, but there are some important differences. (A) IPT systems can be optimized to sense characteristics other than density, eg, contrast of dielectric properties, and thus distinguish different plastics, and (b) IPT systems can measure dozens of measurements per image. Can only get. This gives only images with relatively poor spatial resolution, but the temporal resolution is very good, on the order of milliseconds, and (c) the IPT system is small, portable, and relatively Low cost.

  Among the IPT techniques, dielectric constant, conductivity and permeability are measured and hence called electrical capacitance tomography (ECT), electrical resistance tomography (ERT) and electro-magnetic tomography (EMT) Electrical tomography provides a number of effects such as quick response, low cost, non-invasive and / or non-invasive and robust in a hostile environment. Electrical tomography includes gas / oil two-phase flow imaging in pipelines, gas / solid distribution in pneumatic conveyors and fluidized beds, combustion frames, liquid / liquid and solid / liquid mixing processes, and personal landmines. (York, 2001). The main disadvantage of electrical tomography is that it uses a relatively small number of sensing elements (usually 8, 12 or 16) due to the limited sensitivity of the electronic circuitry. As a result, the number of independent measurements is typically limited to, for example, <100, and thus current electrical tomography systems are only given a moderate spatial image resolution.

  While tomography has been used in the industrial and medical fields to form images inside opaque objects over the past 35 years, the characteristics of current tomography techniques have limited their usefulness. Problems inherent to the sensor itself are: (1) the size and weight of the sensor device using radiation is very large, (2) the safety of the living tissue of the object being scanned, and (3) the sensor signal Including low resolution. Other problems associated with the process of converting sensor signals into useful information are: (1) integration of object motion control with sensor signal capture cycle, (2) overall device cost, (3) Including the need for a long time between the generation of sensor signals and the generation of useful images, and (4) the inability to distinguish between different types of materials.

  Accordingly, there is a need for a new and improved security scanning device for non-invasively checking the contents of containers, particularly baggage, in transportation, shipping and postal facilities.

  Methods and apparatus for detecting hidden dangerous objects in baggage or other enclosures generally include scanning known objects with capacitive and inductive sensors. These sensors are preferably mounted in or on a transducer that may be portable or stationary. When each of a plurality of known objects is located near the sensor, the change in capacitance or inductance near the sensor is recorded in a computer database. A video and knowledge database is built that contains data about each known object. This database is indexed and the data is used for reverse transport, a form of supervised learning to image and identify hidden objects. Error data in the output from the sensor is used as feedback to the previous one, allowing the arrival weights to be updated.

  When capacitive and inductive sensors are used to scan baggage containing unknown objects, capacitance and inductance changes in the vicinity of the unknown object are compared to video and knowledge database data. Limited data regarding the nature, texture and shape of the object allows the computer to modify the excitation signal to facilitate the generation of an image of the unknown object and the determination of whether the object is a dangerous object. By providing feedback, the system can optimize the performance of video generation with different measurement protocols and different excitation frequencies delivered to the sensor. With video knowledge-based technology, real-time video analysis including the detection and identification of dangerous objects can be achieved. The operator can rotate the 3D image of the hidden object to quickly identify it.

  UK Patent GB 2329476, entitled “Image Creation in a Tomography System”, GB 2329476 discloses a method and apparatus for generating images from data obtained using electrical capacitance tomography. This patent discloses electrical capacitance tomography (ECT) that produces images of cross sections of conduits surrounding dielectric materials of different relative dielectric constants. This patent states that a series of capacitance measurements across the conduit can be obtained by exciting the electrodes located around the outer wall of the conduit. These capacitance measurements are used to construct an image of the section through the conduit that represents the relative proportion and location of the dielectric material within the conduit.

  Yang's UK Patent GB 2329476 obtains an image of a cross section of a conduit surrounding dielectric materials of different relative dielectric constants by using an array of electrical capacitance tomography sensors distributed around the section of the conduit . The sensor array comprises a series of electrodes and provides a measurement of the capacitance between different pairs of electrodes across the conduit section. Using these capacitance measurements, an image is generated using a technique known as linear backprojection (LBP).

This British patent states that the LBP algorithm for a mixture of a low dielectric constant material and a high dielectric constant material is represented by the formula disclosed therein, provided that
N is the number of measurement electrodes constituting the sensor array,
F (p) is a fragment of high dielectric constant material present at position p in the cross section of the conduit;
Sjj (p) is the sensitivity of the pair of electrodes to the change in dielectric constant at position p in the conduit cross section,
Cij ′ is a capacitance between a pair of electrodes when the conduit is filled with a material having a low dielectric constant,
Cijh is a capacitance between a pair of electrodes when the conduit is filled with a material having a high dielectric constant.
Cijm is the capacitance between the pair of electrodes being measured, and Rjj is the normalized change in capacitance.

  The variable Sjj (p) is determined for each position p in the conduit by measuring the effect of the change in dielectric constant for each electrode pair. The resulting set of values is commonly referred to as a sensitivity map, and a separate map is generated for each electrode pair. The LBP algorithm forms an image of the section through the conduit by linearly superimposing the sensitivity maps for each electrode pair using the capacitance measurement as a weighting factor.

  The LBP video reconstruction algorithm is shown schematically in FIG. 1 of said patent, where X represents a set of normalized capacitance changes measured from the electrodes, and S is a set of A sensitivity map, and F is an image representing a normalized dielectric constant distribution. The LBP reconstruction algorithm can be thought of as a multivariable open loop system.

  The LBP algorithm cannot produce accurate quantitative images, but provides useful qualitative images. The preferred embodiment of the present invention uses an image reconstructed by the LBP algorithm to provide an initial set of values for subsequent iteration processes. Since the LBP algorithm is not an exact mathematical formula related to the measured capacitance value and image, the image reconstructed by the LBP algorithm will differ to some extent from the actual permittivity distribution. As a result, assuming that the video accurately represents the distribution, the normalized capacitance change calculated from the reconstructed video will be different from the capacitance change measured from the sensor. The difference between the measured capacitance value and the calculated capacitance value is used as an input to the iterative process.

The present invention is related to the improvement disclosed in U.K. GB 2329476 by We Kang Yang entitled "Image Creation in a Tomography System", the disclosure of which is incorporated herein by reference for all purposes.
The main object of the present invention is a method and apparatus for improving video resolution with ECT and EMT, in order to obtain more independent measurements, more sensors, more sensing circuits, and flexible Provide a method and apparatus in which an optimal measurement protocol is used.

  Another object of the present invention is to provide a method and apparatus for performing data fusion to couple the complementary sensitivity of ECT and EMT, possibly together with x-rays, to different material properties.

  Another object of the present invention is for noise filtering, feature extraction, object, 2D and 3D composition, generation of their attributes, and generation of feedback signals to multiple video generation units that can use various imaging techniques. And to provide a method and apparatus that uses the most appropriate technology.

  Yet another object of the present invention is to provide a method and apparatus using video knowledge-based architecture and implementation, which can characterize objects for which video attributes are obtained from multiple sensors. It is.

  Yet another object of the present invention is to provide a method and apparatus for performing real-time operation with a large number of sensors, parallel measurement channels and parallel data processing.

  The present invention described herein provides a method and apparatus for solving these problems, which provides a method and apparatus that enhances the potential usefulness of tomography in various applications, four preferred embodiments thereof: (1) 3D capacitive sensor with parallel sensor array, (2) flat sensor capacitance array, (3) magnetic scanner with flat Hall effect sensor array, and (4) 3D imaging of metal objects and chemicals The multi-plane ECT for this will be described below.

  For a better understanding of the present invention, reference will now be made to the accompanying drawings illustrating four preferred embodiments of the present invention. Like parts are designated by like reference numerals throughout the various views.

  Referring to FIG. 1, reference numeral 15 generally indicates a security scanning device, which includes a computer 20, a data acquisition module 30, an interface 40, and a transducer 50, which includes: A sensor for detecting a change in the capacitive or inductive field in the vicinity thereof is mounted. The transducer 50 is preferably connected to the data acquisition module 30 by a cable 51 or a wireless transceiver (not shown).

  In a preferred embodiment, the security scanning device 15 is based on electrical tomography (more specifically, ECT and EMT) and knowledge-based video analysis, and each device includes a sensing head, sensing electronics, It should be understood that it comprises a video reconstruction and video analysis microprocessor (microcontroller, DSP, laptop or desktop PC), a display unit, and accompanying software.

  The computer 20 is preferably a general purpose machine that processes data based on a set of instructions stored temporarily or permanently. The computer 20 is preferably a personal computer, laptop or handheld device. However, the computer 20 can take any configuration and includes an input / output device such as a VGA display or monitor 21, a keyboard 22 and a mouse 23, a data storage device, and a CPU processor for calculating, comparing and copying data. Preferably with slots for various peripheral devices.

  The computer 20 uses a wide area network to update the system and has the communication capability to check for updated definitions, patches and new features, device drivers, and system updates obtained from the central server. Is preferred.

  As best shown in FIGS. 1, 2 and 11, the data acquisition module 30 includes a digital output device 32, a digital / analog converter (DAC) 34, and an analog / digital converter (DAC) 36. preferable.

  In the illustrated embodiment, the interface 40 between the computer 20 and the data acquisition module 30 is a USB interface as shown in FIGS. 9-18.

  The data acquisition card is designed based on the USB Interface Module (USB-IFM) MOD2 with FT8U245AM integrated circuit (IC) chip, which is available from Future Technology Devices International (FTDI) Ltd, Scotland [1 ]. USB-IFM was chosen because of the easy and cost-effective way to transfer data between the peripheral device and the PC 20 at a rate of up to 8 Mbits (1 Mbyte) / second and the simple FIFO structure. Because it gives. The accompanying software makes it easy for users unfamiliar with the USB protocol to control other devices via the I / O port. FIG. 9 is a block diagram of the USB-IFM.

  The USB-IFM communicates with the PC 20 via a USB link and communicates with peripheral devices via an 8-bit parallel data port (D0-D7). All low-level operations associated with transmitting data between the USB-IFM and the PC 20, including operations between serial data and parallel data, are handled internally by the USB-IFM. When the PC 20 transmits data to the USB-IFM, the data is stored in the FIFO receive buffer and can be read one byte at a time from the data port by the peripheral device. Each rising edge of the RD # signal transmitted to the USB-IFM causes a new byte to be transferred to the data port, as shown in FIG. 10a. The peripheral device transmits data to the USB-IFM for transmission to the PC 20 by writing one byte at a time to the data port. Each falling edge of the WR signal transmitted to the USB-IFM causes a byte to be transferred to the FIFO transmission buffer (see FIG. 10B).

  The RD # and WR signals must be generated by the peripheral device. Two signals RXF # and TXE # are automatically generated by the USB-IFM to control the data flow. When the TXE # flag is “1”, data cannot be written to the USB-IFM data port. Similarly, when the RXF # flag is “1”, data cannot be read from the USB-IFM data port.

  A complete data acquisition system 30 is shown in FIG. 11, which comprises a data acquisition card, a signal generator card, and up to six capacitance measurement cards, one for each capacitance electrode, 12 Provides one capacitance measurement channel. Two direct digital synthesizer (DDS) IC chips (AD7008) are used to generate two synchronized 500 kHz sine signals at 18V peak-to-peak, one as the excitation source and the other Serves as reference signal for phase sensitive demodulation (PSD). An analog multiplexer (MUX) (ADG526) is used to sequentially select a DC signal from each capacitance measurement channel. The differential amplifier (INA 105) subtracts the appropriate voltage generated by the 12-bit digital / analog converter 34 to cancel the standing capacitance. The signal now represents the measured capacitance change, which is further amplified. To handle a large dynamic measurement range, a DC PGA with selectable gains of 1, 2, 4, 8, and 16 is required. The analog signal is finally converted to a 12-bit digital signal by an analog / digital converter (ADC) and then transmitted to the PC 20 in two bytes.

但し、Ｖ_refは、ＤＡＣ３４の基準電圧であり、そしてＤは、デジタル入力である。 The offset signal used to balance the standing capacitance comes from the DAC 34 on the data acquisition card. The offset signal can vary from 0 to 5V in 4096 steps and is expressed as:

Where V _ref is the reference voltage of the DAC 34 and D is a digital input.

但し、Ｆは、ＡＤＣの全測定範囲である。 The ADC is configured for offset bipolar operation and operates over a range of V _m = −1V to 4V. The digital reading is given by:

However, F is the whole measurement range of ADC.

System operation is controlled by the digital output port of the data acquisition card and provides the following functions:
(1) Control of CMOS switch for selecting excitation and detection electrodes;
(2) Control of amplitude and frequency of excitation and reference signals and phase difference between those signals;
(3) MUX control to sequentially select DC signals from the capacitance measurement circuit; and (4) PGA gain control to fully use the ADC measurement range.

  A basic logic circuit for interfacing the USB-IFM to the electronic unit of the data acquisition system is shown in FIG. This circuit allows the PC 20 to select any unit and send data to or receive data from that unit.

  The output RXF # of the USB-IFM is connected to the RD # input via an inverter. The USB-IFM 8-bit I / O ports (D0-D7) are fetched to latch (1) (74LS273), which is activated when the RD # signal goes high. The output of latch (1) has two parts: bits D0-D3 as a 4-bit read data bus for all electronic units, and bits D4-D7 as a control bus that is fed into the 4-16 decoder chip (74LS154). The 16 outputs are used as control lines for selecting individual units.

  The operation is as follows. Initially, the USB-IFM receive buffer is empty, so the output RXF # is “1” and RD # is “0”. The PC 20 transmits the data byte to the USB-IFM reception buffer, and RXF # is automatically set to “0”. After a short time, the RD # and the CLK of the latch (1) become “1” via the inverter. The data byte is latched into latch (1), which also sets RXF # to “1” to indicate that no more data to be read is available, which is then RD # after a short delay. Set to low level. The system is now ready to receive more data from the PC 20 or send data to the PC 20.

  The PC 20 receives data one byte at a time from the USB-IFM transmission buffer using the command FT_Read. Similarly, data is transmitted one byte at a time to the USB-IFM reception buffer using the command FT_Write. The data byte received by the USB-IFM is divided by the logic circuit into two parts: the four least significant bits (LSB) as the read data bus and the four most significant bits (MSB) as the control bus. .

The control bus is taken to a 4-16 decoder to select one of the 16 control lines, which activates 16 different ICs in the data acquisition and measurement circuit as shown in Table 1.












Table 1: Control line connection

  A complete logic circuit for interfacing the USB-IFM to the DAC 34 is shown in FIG. A 4-bit read data bus and three control lines of the basic USB logic interface circuit (see FIG. 12) are used. The main thing is to provide 12-bit data input from the 4-bit read data bus.

  The 4-bit read data bus is extended to provide a 12-bit input signal to the DAC 34 by two additional latches, latch (2) and latch (3). The 4-bit read data bus goes to input bus D8-D11 of DAC 34, latch (2) (its output goes to input bus D4-D7), and latch (3) (its output goes to input bus D0-D3). Directly connected. A monostable chip (74HC123) is inserted into the control line (3) to ensure that the control signal width is adjusted to the correct DAC operation. Control lines 1, 2, and 3 activate latches (2), (3), and DAC 34, respectively.

  The method for loading the 12-bit signal into the DAC 34 is as follows. The PC 20 transmits a byte containing the signal and bits D0-D3 of binary 2 (control line of latch (3)). Latch (3) is activated and bits D0-D3 appear at the corresponding inputs of DAC 34. The PC 20 transmits the second byte including the signal and bits D4-D7 of binary 1 (control line of latch (2)). Latch (2) is activated and bits D4-D7 appear at the corresponding inputs of DAC 34. The PC 20 transmits a third byte including the signal and bits D8-D11 of binary 3 (DAC control line). Bits D8-D11 of the signal appear at the corresponding inputs of the DAC, so here a complete 12-bit signal appears at the DAC input bus. The signal on the control line (3) now locks out the DAC input so that the analog output is held at the value transmitted by the PC 20.

The first FT_Write instruction sends data bits D0-D3 to latch (3), and the second instruction sends data bits D4-D7 to latch (2). The third instruction sends data bits D8-D11 and activates DAC 34 with complete data bits D0-D11. The final instruction is used to reset the active control line to “1” without activating the unit. This is used after every operation. The DAC operations are summarized in Table 2.
Table 2: DAC operation

  The overall logic circuit design for interfacing USB-IFM to ADC is shown in FIG. The 8-bit I / O data bus and control line (5) of the basic USB logic interface circuit (FIG. 12) is used. Eight ADC output data lines D0-D3 (or D8-D11) and D4-D7 are connected to the 8-bit I / O data bus. The control line (5) is taken into the ADC's / RD input via a D flip-flop (74LS74) and an AND gate.

A method for obtaining a 12-bit signal from the ADC is as follows. The PC 20 sends a byte containing binary 5 (ADC control line) to the USB-IFM to activate the ADC conversion and the / BUSY signal goes low. At the end of the conversion, / BUSY goes high, where new data DB0-DB7 appear on the ADC output line (because HBEN is low) and use 3 delay chips (74HC123) Starts a series of three delay timers. After the first delay, a WR signal is generated, which causes data DB0-DB7 to be written to the USB-IFM transmission buffer. After the second delay HBEN is set high, the data DB8-DB11 and the four zeros appear at the ADC data output. After the third delay, a second WR signal is generated, which causes this data DB8-DB11 to be written to the USB-IFM transmission buffer. Here, the PC 20 can read the ADC conversion result from the USB-IFM transmission buffer in 2 bytes. The FT_Write command activates the ADC, and transmits the result as 2 bytes to the USB-IFM transmission buffer via the logic circuit. The FT_Read instruction transmits these two bytes to the PC 20. They do not activate any units and do not send any data to the read data bus. The ADC operations are summarized in Table 3.
Table 3: ADC operation

  The overall logic circuit design for interfacing USB-IFM to MUX is shown in FIG. The 4-bit read data bus and two control lines of the basic USB logic interface circuit (FIG. 12) are used. The 4-bit read data bus is connected to latch (3) and its output leads to MUX input buses A0-A3. Control lines (2) and (4) activate latches (3) and MUX, respectively.

  The method of operating MUX is as follows. The PC 20 transmits a byte including binary 2 (control line of the latch (3)) to the USB-IFM. Latch (3) is activated and the data on the read data bus appears at the corresponding input of the MUX. The PC 20 transmits the second byte including the binary 4 (MUX control line) to the USB-IFM. The MUX is activated and its output selects a channel based on the data on the 4-bit read data bus.

  The overall logic circuit design that interfaces the USB-IFM to the DC PGA is shown in FIG. The 4-bit read data bus and two control lines of the basic USB logic interface circuit (FIG. 12) are used. The 4-bit read data bus is connected to latch (2) and its last three outputs lead to DC PGA input buses A1-A3. Control lines (1) and (4) activate latch (2) and DC PGA, respectively.

The method for operating the DC PGA is as follows. The PC 20 transmits a byte including binary 1 (the control line of the latch (2)) to the USB-IFM. Latch (2) is activated and the data on the read data bus appears at the corresponding input of the DC PGA. The PC 20 transmits the second byte including the binary 4 (DC PGA control line) to the USB-IFM. The DC PGA is activated and its output selects the gain based on the data on the last 3-bit read data bus. Note that control line (4) activates both MUX and DC PGA. The operation of UX and DC PGA is summarized in Table 4.
Table 4: MUX and DC PGA operations

  The first instruction sends data bits D0-D3 via latch (3) to the four input channel select bits of the MUX. The second instruction sends data bits D1-D3 via latch (2) to the three gain select bits of the DC PGA. The third instruction activates both the MUX and DC PGA.

  The DDS chip AD7008 is a complex device and its control is more complex than other units in the data acquisition system. A circuit diagram of a DDS signal generator board controlled by a USB data acquisition card is shown in FIG.

  The frequency, amplitude, and phase of the DDS output sine wave are set by three internal registers that must be loaded externally via the data inputs (D0 and D7) and a parallel assembly register. The data inputs (D0-D7) of the DDS chip are supplied via the DDS latch. The inputs to these latches come from the data lines of the USB and data acquisition cards (4-bit read data bus and latch (3)). The four LSBs of the DDS latch output are also connected to the DDS transfer logic inputs TC0-TC3. When data is written to the parallel assembly register, these bits can be loaded into the appropriate internal registers based on TC0-TC3.

  The control lines (9) and (10) from the USB data acquisition card are used to set the DDS latch, while the control signals for the two DDS chips (ie RESET, / WR and ROAD) are It is taken from the data line of the data acquisition card to latch (4) (bits 8-10).

  The operation to set up two DDS chips sends data to each chip via individual latches and sets three DDS control inputs via latches (4) as shown in Tables 5 and 6 ( On a USB data acquisition card).

  All control signals to the AD7008 are connected to the USB data acquisition card latch (4) via 64 connectors. When the AD7008 is activated, the control bus is set to "0" and the read data bus is set to activate the appropriate control input.

テーブル６：ＡＤ７００８のデータラインオペレーション

The data line of the AD7008 is controlled by the USB data acquisition card via two latches DDS latch (1) and DDS latch (2). The first FT_Write instruction sends data bits D0-D3 to the latch (3) of the USB data acquisition card. The second instruction sends data bits D4-D7 and activates one of the DDS latches with complete data bits D0-D7.
Table 5: AD7008 control line operation

Table 6: AD7008 data line operations

The USB data acquisition card software is written in Visual C ++ 6.0, which is useful for interfacing peripheral devices and displays and designing user interfaces. FIG. 18 shows a Windows user interface. This has the following functions.
(1) Control of the signal generator board;
(2) System parameter menu and setting management, and system initialization;
(3) Control of capacitance measurement board and acquisition of capacitance data;
(4) DAC and ADC control for automatic DC offset and gain functions; and (5) Real-time data acquisition.

  From the foregoing, it will be readily apparent that this method and apparatus provides a high speed USB data acquisition card for an ECT system. Electronic circuits are designed and PCBs are manufactured to meet Eurocase and Eurocard standards. The card can support up to 12 capacitance measurement channels and has been successfully incorporated into the system. The software provides a user-familiar Windows (registered trademark) with automatic calibration to obtain DC offset and DC amplifier gain settings to cancel standing capacitance, and collection of measurement data for video reconstruction and analysis. ) Equipped with interface.

  As described above, the computer 20 can take various configurations based on a specific application. For example, portable laptop computers can be used for some applications such as checking baggage left unattended in an airport lobby. For other applications, the personal computer 20 may be used as a stand-alone system, workstation, and file server in a local area network or in a wide area network via the Internet.

  The transducer 50 shown in FIGS. 1 and 3 comprises a plastic cylinder or pipe 52 with eight electrodes 54 positioned around the cylinder. A similar electrical capacitance tomography (ECT) transducer is disclosed in UK Patent No. GB 2329476, entitled “Image Creation in a Tomography System”, incorporated herein by reference. This patent describes the use of a device to retrieve a series of capacitance measurements across the conduit obtained by exciting electrodes 54 located around the outer wall of the conduit 52. These capacitance measurements are used to construct an image of the section through the conduit 52 that represents the relative percentage and location of the dielectric material within the conduit 52. The above patent combines video representing internal characteristics of a region with video correction data derived from repeated feedback algorithms that are quickly executed on a “pentium computer” from measured electrical parameters around the region. Describes the method for generating.

  A number of patents issued to Kathleen Hennessy disclose video database knowledge databases that have been used to detect printed circuit board defects. These patents are U.S. Pat. Nos. 5,515,453, 5,553,168, 5,703,969, 6,014,461, 6,091,846, 205,239, 6,246,787, and 6,246,788, the disclosures of which are incorporated herein by reference.

  As described in detail below, using backpropagation, a supervised learning to build a video and knowledge-based database is provided, and error data is backpropagated to the previous one, Allow incoming weights to be updated. Video indexes and searches based on video primitives can be used to generate 3D real-time video of hidden items.

  It is important to note that an image of an object need not be formed in order to detect a dangerous object. The transducer signal can be used not only to detect the gradient characteristics of the edges of the object, but also to detect and characterize the distinction between adjacent objects in the material. Video generation is only required to provide visual information to the human operator after the initial tomographic examination. Objects and their properties can be represented in symbolic superspace beyond the dimensions available for visual display methods.

  In further development of the method disclosed in US Pat. No. 5,515,453, signal measurements obtained from the panel of sensors / transducers are used to delineate the edges of the object and Characterize the external shape and texture of the material. A selected sample set of dangerous objects is used in the same way that semiconductor defects are characterized in US Pat. Nos. 6,014,461, 6,091,846 and 6,205,239. Thus, a dangerous object knowledge base is generated.

  The characteristics of each sample of objects to be used to generate rules for dangerous objects in the dangerous object knowledge base are obtained by the following steps 1-7.

  Step 1: Transducer signal representing either (1) a discontinuity in the intensity characteristics of the edge segment of the object, or (2) a range of values of adjacent transducer signals characterizing one or more textures of the interior material of the object. Generate a set of primitives from the values of.

  Step 2: Associate a large number of primitives with each other based on the spatial accessibility and similarity of the primitive values, and connect the edge segment of the object and the internal material to their attribute values, eg, relative position, sharpness Representing the length, width, curvature, and other characteristics derived from the transducer signal produces a set of high level primitives.

  Step 3: Define the object by joining the high level primitives based on shared properties such as edge and internal material similarity.

Step 4: The set of object descriptor values such as object size, curvature and sharpness of its edges, metallic texture, unique chemical texture of nitrate, etc., as shown in Table 7 below: Occurs from the shared characteristics.
Table 7: Characterization of dangerous objects 2D 3D cubes of sharp objects Sharp edges Inside edges of objects Internal ID Cross-section Body volume * Eccentricity * Curvature * Texture Texture
0 = yen coefficient * variation *
100 = Worm knife 1 104 223 93 64 16 -48 ¹ 17
Knife 2 108 231 87 71 19 -41 ¹ 12
Mine 1 52 304 21 6 91 53 22
Mine 2 48 291 24 8 93 49 18
Razor 1 60 312 33 23 18 51 28
Razor 2 64 303 37 28 22 15 16
* Normalization, 0-100
¹ Metal exhibits negative capacitance.
² Razor 1 contains explosives and razor 2 does not.

  Step 5: For other similar dangerous objects whose characteristic similarity is shown in the graph of FIG. 24a as opposed to different objects whose divergent characteristics are shown in FIG. Repeat, usually 5 times or less.

Step 6: Assign a weight to the descriptor and give a large weight to descriptors with similar and limited range values, as shown in Table 8 below, with different or widely varying range values The descriptor is given a small weight or no weight.
Table 8: Weighted object specified in the characteristics of dangerous object 2D 3D cubic Edge inside object edge Internal internal ID Cross section Body volume Sharpness Eccentricity Curvature Texture Texture
Coefficient Floating knife 11 14 94 61 16 11 18
Mine 32 28 6 44 42 91 23
Razor W 6 14 23 14 18 88 11

Step 7: The set of descriptor values for each group of dangerous objects is reduced to a single set of descriptor value midpoints and weights, and each such set is dangerous as shown in Table 9 below. Store as object knowledge base rules.






Table 9: Hazardous Object Knowledge Base: Attribute Value Intermediate Point and Weight
2D 3D Cubic Edge Inside Object Edge Internal Rule Cross Section Volume Volume Sharpness Eccentricity Curvature Texture Texture
Coefficient variation
MPt Wt MPt Wt MPt Wt MPt Wt MPt Wt MPt Wt MPt Wt
Knife 106 11 227 14 93 94 64 61 16 16 -48 11 17 18
Mine 50 32 298 28 22 6 7 44 92 42 51 91 20 23
Razor W 63 6 307 14 35 23 26 14 20 18 38 88 22 11

  The distribution, use, maintenance and special application improvement of the dangerous object knowledge base will be described in the following steps 8-12.

  Step 8: Once the dangerous object knowledge base is populated with rules for dangerous objects that reflect the accumulated relative information about a wide variety of objects, properties and their relative importance, this is used to By following steps 1-4, potentially dangerous objects in the enclosed area, such as in parcels and baggage, can be detected and classified.

  Step 9: A dangerous object knowledge base with a normal size of 15000 bytes or less is distributed electronically to a security unit for use in a particular embodiment of the scanning device.

  Step 10: Modification of the dangerous object knowledge base is accomplished by adding, deleting or replacing one or more instances of the object, and the set of rules in the dangerous object knowledge base is reformulated, i.e. usually only takes a few minutes. And a process that can be easily reversed.

  Step 11: Generate a dangerous object knowledge base master set, modify it on a similar device at a location remote from the inspection site, and electronically send it to any location to keep up-to-date on dangerous objects throughout the security network Give rules.

  Step 12: Generate and maintain multiple versions of a dangerous object knowledge base compiled to detect explosives and accelerators for different security tasks such as cargo and package checks, Generate another version of the dangerous objects knowledge base for in-flight items with added rules for detecting sharp metal objects.

  Numerous embodiments of transducers that can be used separately or in combination are shown in FIGS. 4, 5, 6 and 19-23.

  Four security scanning devices will be described below as the next embodiment. (1) a 3D capacitive scanner with a parallel sensor array; (2) a shoe scanner with a flat capacitance sensor array; (3) a magnetic scanner with a flat Hall effect sensor array; and (4) metal objects and chemicals. Multi-planar ECT for 3D image formation.

Embodiment 1 3D Capacitive Scanner with Parallel Sensor Array A conventional ECT sensor is composed of a number of electrodes surrounding a cross section. Another design is to use two parallel plates, one for excitation and the other for detection, with a capacitance sensor array, with each plate having, for example, 256 (or other) electrodes. As shown in FIG. 19, when one electrode 19e is energized in the upper plate 19a, for example, 25 capacitance measurements can be obtained from the lower plate 19b. The top plate excitation electrodes are sequentially energized, for example, more than 1000 capacitance measurements can be collected from the bottom plate sensing electrodes. To obtain a security distribution (Yang et al., 1999), a limited difference method can be used, which is more convenient than a limited element method (FEM) because of the shape of the sensing space. Each electrode can be divided into, for example, 3 × 3 = 9 (or other number) pixels. The space between the excitation plate and the detection plate can be divided into multiple layers by software so that the thickness of each layer is similar to the dimensions of the pixels. In this way, the sensing space can be divided into a large number of small cubes. This enables true 3D video reconstruction and can achieve excellent spatial resolution.

  Both capacitance and loss conductance components can be measured at different frequencies and different phases. The measurement circuit may be based on an AC-based ECT system design that uses sinusoidal excitation and phase-sensitive demodulation (Yan and York, 1999). To measure the capacitance component, a high frequency excitation signal (eg, 1 MHz) can be applied to one or more excitation electrodes for an optimal sensing field, and a quadrature signal is phase sensed from the detection electrode. It can be measured by demodulation. To measure the loss conductance component, a low frequency signal (eg, 10 kHz) can be applied to measure the in-phase signal. The measurement protocol can be optimized to obtain a data set with attributes that are optimal for video resolution (Polyride and McCann, 2002). Parallel data acquisition channels can be used to achieve real-time imaging.

  Many dangerous objects such as guns and knives (metal or ceramic) can be uniquely characterized only by their shape. Others such as explosives can be identified by the unique electrical properties of the material. On the other hand, certain harmless objects may trigger false positive alarms because they have a suspicious shape, such as a set of wires, or because of the suspicious nature of the material, such as a hair spray. To employ automatic inspection or scanning techniques, it is important to eliminate false positive alarms. The ability of knowledge-based video analysis to characterize objects not only by shape, but also by electrical properties of materials, substantially reduces false positive alarms, as its application in semiconductor wafer inspection shows. (Lin 2001, Fan 2002).

  In order to make the best use of the information obtained from tomographic sensors, capacitance and loss conductance data can be processed or extracted as a set of features, which can be used to capture this information in terms of material image, shape, texture and properties. Can be represented as seamlessly integrated with other information such as, and then stored in a database. This allows data from the sensor to be assigned to the relevant properties of the object, and the system can extract, formulate and use rules derived from the relevant data. Capacitance and loss conductance data can now be part of a dynamic knowledge base of objects and their properties in relation to the similarity of capacitance and loss conductance profiles.

  Development of this device involves acquiring raw video data, filtering noise, and extracting features. This requires software and hardware facilities for knowledge base functions. Information can also be provided to evaluate the performance of each video generation technique with respect to shape description, object structure and feedback capabilities. Development of this device can also include acquiring knowledge-based characteristics of all 3D images of dangerous objects and performing extensive field tests to ensure acceptable system performance in terms of use. The knowledge base provides feedback to the ECT system, like an autofocus facility, so that the system can optimize the performance of video generation with different measurement protocols and different excitation frequencies. With video knowledge-based technology, real-time video analysis including detection and identification of dangerous objects can be achieved.

  This device can be used to check (but is not limited to) baggage at airports and envelopes and / or parcels at post offices and parcel delivery services. An object can be placed between two plates. If there is a metal or ceramic knife or explosive material inside, the imaging system can show the contents as a 3D image and identify them. This device is the first of its kind to prevent criminals from sending bombs, anthrax and other dangerous materials through airports, post offices and parcel delivery services. The device may also be used as a portable device for checking abandoned document bags, which is a major problem at airports.

Embodiment 2: Shoe Scanner with a Flat Capacitance Sensor Array In the past, terrorists have relied on hiding explosives in shoes, which has been referred to as “shoe bombs”. In particular, at many airports in the United States, all passengers must take off their shoes for security personnel to check. There is a clear problem with this method. (A) Unable to find explosives hidden in soles or heels; (b) Time consuming, additional staff required and long queues formed; (c) This experience has been Uncomfortable for both security staff.

  Experiments have shown that using a capacitance sensor array can generate 2D images showing plastic or metal objects buried in sand. The 256 sensor array has been shown to visualize the shape of the foot and determine its size.

  The sensing pad as the second embodiment can be configured by, for example, 256 (or any number) of electrodes that are sequentially energized. When one electrode is energized, the interelectrode capacitance can be measured, for example, from 24 (or other) surrounding electrodes. Considering that the excitation electrodes near the edge do not have 24 surrounding electrodes, there are only measurements less than 256 × 24 = 6144. Similar to the first embodiment, the sensitivity distribution can be found by the limited difference simulation. The video reconstruction algorithm can be implemented for 2D and 3D displays. With the measurement protocol and prior knowledge about the sensitivity distribution, a high video resolution can be expected. Similar to embodiment 1, each electrode can be divided into 9 (or other) pixels, giving 2304 pixels in 2D video and more cubes in 3D video, for example. Can do. Similar to the solution proposed for Embodiment 1, knowledge-based video analysis can be applied in view of metals, ceramics, explosives and other hazardous materials. In addition to being used as a shoe scanner, handheld versions are also contemplated for other portable applications, for example as a video tool to replace conventional metal detectors and plastic landmine detection.

Embodiment 3: Magnetic Scanner with Flat Hall Effect Sensor Array While current metal detectors can only detect the presence of metal, the magnetic scanner of the present invention relates to the size and / or shape of the metal object (s) Information can be given. As shown in FIG. 20, this sensor includes, for example, nine (or other number) coils 20c for generating an AC magnetic field and, for example, twenty-eight (or other number) Hall effect sensors 20d as detectors. It consists of. The coil 20c can be sequentially energized. When one coil 20c is energized, for example, 28 (or other numbers) measurement values can be obtained from the 28 (or other numbers) Hall effect sensors 20d. In total, for example, 252 measurements are obtained from the sensor. In order to reconstruct an image from the 252 measurement values, a 3D sensitivity map is required.

  As shown in FIG. 2, the sensor panel includes, for example, nine (or other number) coils for generating a magnetic field of 10 kHz, for example, and 28 Hall effect sensors as detectors. For example, sensors made of strained indium-gallium-arsenide (InGaAs) heterostructures, or other Hall effect sensors, are small in size (eg, 1.5 × 3 mm) and large dynamic range (eg, up to 1 Tesla). Gives a high resolution (eg 0.1 μ Tesla) (Hand and Missor, 2003). The coils can be sequentially energized. When one coil is energized, 28 measurements can be taken from the Hall effect sensor. In total, for example, 252 measurements can be taken from the sensor panel. The data acquisition system may be similar to that used to measure capacitance, except for the transducer circuit.

  Reconstructing an image from the 252 measurements requires a 3D sensitivity distribution, which can be calculated using, for example, a limited element software package from Unsoft Inc. (USA). . After the object has been interrogated from a number of different angles, the measured data can be used to reconstruct a 2D and / or 3D image of the metal object (s). The device can be designed using a PC 20 or other type of desktop computer for data acquisition, video reconstruction, video processing and video display. The handheld tool 60 shown in FIGS. 21-23 uses a microcontroller, DSP, or other type of microprocessor with an LCD display panel and is battery operated. This can be used by security personnel to scan people and small bags. The device of the present invention can also be used to detect metal landmines. Again, the advantage of this device over conventional metal detectors is that it provides an image showing the shape and / or size of the metal object and can reduce false alarms. Note that the success rate of landmine detection using conventional metal detectors is on the order of 0.1% (Mr. Allosop, 2000). Similar to the first and second embodiments, knowledge-based video analysis can be applied to the third embodiment. In the future, this device can be combined with a flat capacitance-based imaging device (Embodiment 2) to obtain improved video by fusing data and video.

As shown in FIGS. 21-23, the handheld device of the present invention preferably has a plurality of layers including:
Layer 1: LCD display and keypad Layer 2: Sensing electronics and DSP
Layer 3: Magnetic shield Layer 4: Excitation coil Layer 5: Preamplifier and MUX
Layer 6: Hall Effect Sensor The device of the present invention can be combined with a flat capacitance imaging sensor array to obtain an improved image.

Embodiment 4: Multiplanar ECT for 3D imaging of metal objects and chemicals
In the past, ECT techniques have been developed for imaging dielectric materials of different dielectric constants such as gas (τ _r = 1.0) and oil (τ _r = 2.1). The metal was observed to generate a very strong signal that represents a negative capacitance change. Experiments have demonstrated that metal objects can be seen in existing ECT systems by simply inverting the capacitance data during video reconstruction. This shows that the ECT system can be used as a security scanner to detect, position and form an image of a metal object in a cross section. An important issue is improving the resolution of the video and reliably identifying metal objects. In Embodiment 4, the sensor for 3D imaging is composed of, for example, four (or other number) of capacitance electrodes, and each ring includes, for example, 16 (or other number) electrodes. . These electrodes can surround a round or rectangular tunnel. Similar to Embodiments 1 and 2, both capacitance and loss conductance components can be measured from the sensor array to obtain more independent measurements to more reliably identify dangerous objects or materials.

  Since metal objects produce capacitance changes that are completely different from dielectric materials, it is necessary to find a sensitivity distribution specific to the metal object in order to reconstruct the image of the metal object. Loss conductance measurements can also be used to reconstruct a suitable image. In both cases, it is necessary to calculate the 3D sensitivity distribution using, for example, Ansoft software or other FEM or limited difference software. The two images can be integrated together by image fusion to generate a more accurate display of the metal object.

A 2D image can be reconstructed from a set of measurement data from each electrode ring, and then, for example, 4 images from all electrode rings can be stacked together to give a 3D image. For use at an airport, for example, a tunnel with a diameter of 60 cm is designed so that each ring has, for example, 16 (or other) electrodes, for example 10 cm in length, into which a handbag can be placed. can do. When operating the system for 2 _^1/2 D image formation, it can provide the 16 electrodes, for example, 120 pieces of capacitance measurements and for example, 120 pieces of Los conductance measurements within each ring. For example, four videos can be reconstructed from capacitance data using Landweber's iterative algorithm or other video reconstruction algorithms (Yang et al., 1999, Yang and Penn, 2003) and loss -Four images can be reconstructed from conductance data. Then, two virtual 3D images can be constructed from the capacitance image and the loss conductance image, respectively. Finally, 3D images can be generated by fusing the two images together. In order to obtain a true 3D image, the measurement protocol can be much more complex than conventional solutions. In this case, once an electrode or group of electrodes is energized, capacitance and loss conductance measurements can be made from all other electrodes of all rings. In addition, more independent measurement data can be obtained from the sensor and improved video can be extracted. Similar to Embodiments 1, 2, and 3, knowledge-based video analysis can be applied to Embodiment 4.

From the above explanation, the following will become clear.
1. A security scanning device based on electrical tomography (more specifically ECT and EMT) and knowledge-based image analysis, comprising a sensing head, sensing electronics, image reconstruction and image analysis microprocessor (microcontroller, DSP) , Either a laptop or a desktop PC), a device each having a display unit and accompanying software, provides a significant advantage over conventional devices.

  2. The security scanning device can improve the video resolution associated with ECT and EMT. More sensors, more sensing circuits, and flexible / optimal measurement protocols can be used to obtain more independent measurements.

  3. The security scanning device can perform data fusion to combine the complementary sensitivity of ECT and EMT (possibly with x-rays) to different material properties.

  4). The security scanning device is optimized for noise filtering, feature extraction, object structure, generation of their attributes, and generation of feedback signals to a number of video generation units that can use a variety of imaging techniques. Can be determined.

  5). The security scanning device provides an architecture and means for implementing a video knowledge base that can characterize objects for which video attributes are obtained from multiple sensors.

  6). The security scanning device can operate in real time with a large number of sensors requiring parallel measurement channels and parallel data processing.

  7. Security scanning devices provide tomographic or non-tomographic images of dangerous objects (eg guns and knives, metals or ceramics) and chemicals (eg explosives) in 2D or 3D. Supported by analyzing and understanding knowledge-based video for identification.

  8). Security scanning devices can be used in airports, post offices, and other areas that handle public access, such as public theaters, railway stations, public buildings, sporting events, and other unprotected venues.

  Terms such as “left”, “right”, “clockwise”, “counterclockwise”, “horizontal”, “vertical”, “up”, “down”, etc., when used with reference to the drawings, It refers to the orientation of the component in the illustrated embodiment, not necessarily the orientation in use. It should be understood that these terms are only meant to refer to relative positions and / or orientations for convenience and are not meant to be limiting in any way.

  While various embodiments of the invention have been illustrated and described, it will be apparent that changes can be made without departing from the inventive concept. Thus, the present invention is not limited to the examples or embodiments described above.

It is a perspective view which shows a security scanning device. It is a block diagram which shows a security scanning apparatus. FIG. 3 is a cross-sectional view substantially along the line 3-3 in FIG. 1. It is a circuit diagram which shows the switch of a sensor. 1 is a schematic diagram of a Hall transducer with multiple magnetic sensing elements. FIG. 1 is a schematic diagram of a Hall sensing device having multiple coils and magnetic sensing elements. (A) is a graph showing a hierarchical distribution, (b) is a graph showing a core distribution, and (c) is a graph showing a circular distribution. (A) is a hierarchical distribution image, (b) is a core distribution image, and (c) is a circular distribution image. It is an internal book figure of USB-IFM. It is a USB-IFM timing diagram. It is a figure of the data acquisition system which shows a USB interface card. It is a figure which shows a USB-IFM logic interface circuit. It is a figure which interfaces USB-IFM to DAC. It is a figure which interfaces USB-IFM to ADC. It is a figure which interfaces USB-IFM to MUX. It is a figure which interfaces USB-IFM to DC PGA. It is a figure which shows a DDS signal generator board. It is a figure which shows a Windows user interface. FIG. 3 is a schematic diagram showing two parallel plates, one for excitation and the other for detection, and a capacitance sensor array on each plate. 1 is a schematic diagram of a magnetic scanner having an excitation coil and a Hall effect sensor. FIG. It is an orthographic view of a handheld scanner. 1 is an enlarged side view of a handheld scanner with multiple layers. FIG. It is a top view of a handheld scanner. It is a graph which shows the alignment state of the value of the descriptor of the same object. It is a graph which shows the divergence state of the value of the descriptor between different objects.

Claims (35)

In a security scanning device for scanning, identifying and imaging the contents of a device in a sensing space,
A plurality of arrays of sensor means adjacent to the sensing space;
Responsive to an input signal applied to the excitation electrode of the sensor means, a plurality of electric fields are generated between at least one excitation electrode of the sensor means and each of the plurality of detection electrodes of the second detection electrode array. Means for
Software based on different sensor signals to divide the sensing space into multi-layered small cubes;
Means for delivering a signal from each detection electrode of the second detection electrode array of the sensor means to a data acquisition system;
Means for analyzing the signal received by the data acquisition system to determine the nature of the device in the sensing space and to image a cross section of the device;
A knowledge database associated with the means for analyzing a signal received by the data acquisition system, the database indexed to use the data in backpropagation to identify hidden objects;
Security scanning device comprising: The sensing space of claim 1 , wherein the plurality of arrays of sensing means adjacent to the sensing space is configured to perform data fusion to couple the complementary sensitivity of ECT and EMT to different material properties. A security scanning device for scanning, identifying and imaging the contents of devices within. The means for delivering a signal from each detection electrode of the second detection electrode array of the sensor means to a data acquisition system includes noise filtering, feature extraction, object structure, generation of their attributes, and various images The device in a sensing space according to claim 1 , wherein the contents of the device in the sensing space are configured to be able to determine an optimum technique for generating feedback signals to a number of image generating units utilizing forming techniques. And a security scanning device for image formation. Wherein analyzing the signal received by the data acquisition system, and determines the nature of the device in the sensing space, said means for the cross-section of the apparatus the image formation, the then large number of sensors image attributes security for but with the architecture and means for carrying out the movies image knowledge base characterizing the object to be acquired, to scan the contents of the device in the sensing space according to claim 1, to identify and imaging Scanning device. A security scan for scanning, identifying and imaging the contents of a device in a sensing space according to claim 1 , wherein said plurality of arrays of sensor means adjacent to the sensing space includes capacitive and inductive sensors. apparatus. Wherein the plurality of arrays of sensor means adjacent to the sensing space includes a capacitive sensor, scanning the contents of the device in the sensing space according to claim 1, identify and security scanning device for imaging. Wherein the plurality of arrays of sensor means adjacent to the sensing space includes an inductive sensor, scanning the contents of the device in the sensing space according to claim 1, identify and security scanning device for imaging.   In a security scanning device for scanning, identifying and imaging the contents of a device in a sensing space,
  A plurality of arrays of sensor means adjacent to the sensing space;
  Responsive to an input signal applied to the excitation electrode of the sensor means, a plurality of electric fields are generated between at least one excitation electrode of the sensor means and each of the plurality of detection electrodes of the second detection electrode array. Means for
  Means for dividing the sensing space into multi-layered small cubes;
  Means for delivering a signal from each detection electrode of the second detection electrode array of the sensor means to a data acquisition system;
  Means for analyzing the signal received by the data acquisition system to determine the nature of the device in the sensing space and to image a cross section of the device;
  A knowledge database associated with the means for analyzing a signal received by the data acquisition system, the database indexed to use the data in backpropagation to identify hidden objects;
With
  The means for analyzing the signal received by the data acquisition system to determine the nature of the device in the sensing space and to image the cross-section of the device includes: A security scanning device comprising an architecture and means for implementing a video knowledge base characterizing an object to be acquired.   9. The sensing space of claim 8, wherein the plurality of arrays of sensing means adjacent to the sensing space is configured to perform data fusion to couple complementary ECT and EMT sensitivities to different material properties. A security scanning device for scanning, identifying and imaging the contents of devices within.   The means for delivering a signal from each detection electrode of the second detection electrode array of the sensor means to a data acquisition system includes noise filtering, feature extraction, object structure, generation of their attributes, and various images 9. Scan and identify the contents of the device in the sensing space of claim 8 configured to determine an optimal technique for generating feedback signals to a number of image generation units utilizing forming techniques. And a security scanning device for image formation.   The database associated with the means for analyzing a signal received by the data acquisition system is configured to characterize an object whose video attributes are acquired from multiple sensors;
  The video and knowledge database is constructed according to the following steps to contain data about each known object, said steps comprising:
  1) From the value of the sensor signal, either (1) the intensity discontinuity that characterizes the edge segment of the object, or (2) the range of adjacent sensor signal values that characterize one or more textures of the internal material of the object Generating a set of primitives that represent
  2) Generating a set of high-level primitives by associating multiple primitives with each other based on the spatial accessibility and similarity of the primitive values, such as their attribute values, eg, relative position, sharpness, Representing the edge segment and internal material of the object by width, curvature, and other characteristics derived from the sensor signal;
  3) joining the high level primitives based on shared properties, eg, edge and internal material similarity, to define the object;
  4) Generating a set of descriptor values for the object from the shared properties, eg, object size, curvature and sharpness of its edges, metal texture, unique chemical texture, eg nitrate texture Stages,
  5) repeating steps 1-4 with at least three other similar dangerous objects;
  6) The descriptors with similar and limited range values are given high weight, and the descriptors with different or significantly changing range values are given little or no weight. Assigning weights to the descriptors;
  7) Decreasing the set of descriptor values for each group of dangerous objects to a single set of descriptor values, ranges and weights, each to be stored as a rule in the dangerous object knowledge database. When,
A security scanning device for scanning, identifying and imaging the contents of a device in a sensing space according to claim 8.   9. A security scan for scanning, identifying, and imaging the contents of a device in a sensing space according to claim 8, wherein the plurality of arrays of sensor means adjacent to the sensing space includes capacitive and inductive sensors. apparatus.   The database associated with the means for analyzing a signal received by the data acquisition system is configured to characterize an object whose video attributes are acquired from multiple sensors;
  The video and knowledge database is constructed according to the following steps to contain data about each known object, said steps comprising:
  Positioning a known object relative to the capacitive and inductive sensors;
  Recording changes in capacitance and inductance near the capacitive and inductive sensors when each of a plurality of known objects is located nearby;
  Building a video and knowledge database containing data about each said known object;
  Moving baggage including an unknown object relative to the capacitive and inductive sensors; comparing capacitance and inductance changes near the unknown object with data in the video and knowledge database;
  Determining whether the object is a dangerous object;
A security scanning device for scanning, identifying and imaging the contents of a device in a sensing space according to claim 12.   Means for implementing the video knowledge base characterizing an object from which a plurality of sensors have their video attributes obtained;
  Means for providing tomographic and non-tomographic images of dangerous objects and chemicals in 2D or 3D, supported by knowledge-based video analysis and understanding to identify dangerous objects and chemicals 12. A security scanning device for scanning, identifying and imaging the contents of a device in a sensing space according to claim 11 comprising means.   9. A security scanning device for scanning, identifying and imaging the contents of a device in a sensing space according to claim 8, wherein the plurality of arrays of sensor means adjacent to the sensing space includes a capacitive sensor.   9. The means for generating a plurality of electric fields between at least one excitation electrode and each of a plurality of sensing electrodes includes means for delivering a high frequency sinusoidal excitation signal to the excitation electrode. A security scanning device for scanning, identifying and imaging the contents of a device within the described sensing space.   9. Scanning the contents of a device in a sensing space according to claim 8, wherein the means for analyzing a signal received by the data acquisition system comprises phase sensing demodulation for electrical tomography and knowledge based imaging. Security scanning device for identifying, identifying and imaging.   The plurality of arrays of sensor means adjacent to the sensing space includes a capacitive scanner with a parallel sensor array for scanning, identifying, and imaging the contents of the device within the sensing space. Security scanning device.   The array of sensor means adjacent to the sensing space includes a plurality of electrodes surrounding the sensing space for scanning, identifying and imaging the contents of the device within the sensing space. Security scanning device.   9. A device in a sensing space according to claim 8, wherein the plurality of arrays of sensor means adjacent to the sensing space comprises a plurality of capacitance sensor arrays with a plurality of excitation electrodes and a plurality of detection electrodes mounted on a plate. Security scanning device for scanning, identifying and imaging the contents of a machine.   An electrical capacitance tomography (ECT) system for scanning, identifying and imaging the contents of a device within a sensing space, comprising:
  An array of excitation and detection electrodes adjacent to the sensing space;
  An excitation signal generator configured to apply a signal to at least one of the excitation electrodes to allow measurement of a change in capacitance between the excitation electrode and each of the plurality of detection electrodes;
  Means for delivering capacitance signals from each sensing electrode to the data acquisition system;
  Means for analyzing a capacitance signal received by the data acquisition system to determine the nature of the device in the sensing space and to image a cross-section of the device;
  The capacitance signal received by the data acquisition system is positive when the device in the sensing space is a dielectric material, negative when the device in the sensing space is metal and negative of the capacitance. And means for analyzing the signal received by the data acquisition system inverts capacitance data during reconstruction of the image to allow imaging of metal devices using the capacitance signal. An electrical capacitance tomography (ECT) system, characterized in that: In a method for scanning, identifying and imaging a device in a sensing space,
Positioning a plurality of arrays of sensor means adjacent to the sensing space;
Positioning the device to be scanned, identified and imaged in the sensing space;
Responsive to an input signal applied to the first excitation electrode array of the sensor means between at least one excitation electrode of the first excitation electrode array of the sensor means and each detection electrode of the second detection electrode array. Generating a plurality of electric fields;
Dividing the sensing space into multi-layered small cubes;
Delivering a signal from each detection electrode of the second detection electrode array of the sensor means to a data acquisition system;
Analyzing the signal received by the data acquisition system to determine the nature of the device in the sensing space and the identity of the device;
With a method. Analyzing the signal received by the data acquisition system comprises:
Obtaining raw video data from the sensor means;
Processing the raw video data;
Seamlessly integrating information with other information, including material images, shapes, textures and properties, and then extracting the data as a set of features represented to be stored in a database;
23. A method of scanning, identifying and imaging a device in a sensing space according to claim 22 . Analyzing the signal received by the data acquisition system comprises:
Generating a video database from a collection of videos related to a particular subject;
Generating a knowledge database with other data related to the particular subject, wherein the knowledge database cross-references the video database;
24. A method of scanning, identifying and imaging a device in a sensing space according to claim 23 . 23. A method for scanning, identifying and imaging a device in a sensing space according to claim 22 , wherein the sensor means comprises an electrical capacitance sensor. 23. A method for scanning, identifying and imaging a device in a sensing space according to claim 22 , wherein the input signal applied to the first excitation electrode array of the sensor means comprises a high frequency sinusoidal excitation signal. Analyzing the signals received by the data acquisition system to detect and identify devices in the sensing space includes phase sensing demodulation for electrical tomography and knowledge base imaging; 23. A method for scanning, identifying and imaging a device in a sensing space according to claim 22 . 23. A method for scanning, identifying and imaging a device in a sensing space according to claim 22 , wherein the sensor means comprises a capacitive scanner with a parallel sensor array. The method of scanning, identifying and imaging a device in a sensing space according to claim 22 , wherein the sensor means comprises a plurality of electrodes surrounding the sensing space. 23. Scanning, identifying and imaging a device in a sensing space according to claim 22 , wherein said sensor means comprises a plurality of capacitance sensor arrays with a plurality of excitation electrodes and a plurality of detection electrodes mounted on a plate. how to. 23. Sensing according to claim 22 , wherein the sensor means comprises two parallel plates, one for excitation and the other for detection, and the capacitance sensor array comprising a plurality of excitation electrodes on each plate. A method of scanning, identifying and imaging a device in space. 23. Sensing according to claim 22 , wherein the sensor means comprises two parallel plates, one for excitation and the other for detection, and a capacitance sensor array with a plurality of detection electrodes on each plate. A method of scanning, identifying and imaging a device in space. The sensor means includes two parallel plates, one for excitation and the other for detection, and the capacitance sensor array comprises a plurality of excitation electrodes on each plate and a plurality of detections on each plate. 23. A method for scanning, identifying and imaging a device in a sensing space according to claim 22 comprising electrodes. In response to an input signal applied to the first excitation electrode array of the sensor means, a plurality of excitation electrodes between the first excitation electrode array of the sensor means and each detection electrode of the second detection electrode array Said step of generating an electric field comprises:
Energizing a single excitation electrode of the first excitation electrode array and delivering a signal from each of a plurality of detection electrodes of the second detection electrode array of the sensor means to a data acquisition system. A method for scanning, identifying and imaging a device in a sensing space according to claim 22 . In response to an input signal applied to the first excitation electrode array of the sensor means, a plurality of excitation electrodes between the first excitation electrode array of the sensor means and each detection electrode of the second detection electrode array Said step of generating an electric field comprises:
Sequentially energizing a single excitation electrode of the first excitation electrode array, and a plurality of detections of the second detection electrode array of the sensor means when each of the plurality of excitation electrodes is energized Delivering a signal from each of the electrodes to the data acquisition system;
23. A method of scanning, identifying and imaging a device in a sensing space according to claim 22 .

Images