WO2006067773A1

WO2006067773A1 - A detection system

Info

Publication number: WO2006067773A1
Application number: PCT/IE2005/000145
Authority: WO
Inventors: Rory Doyle; Anthony Mcenroe; Brendan Lyons
Original assignee: Farran Technology Limited
Priority date: 2004-12-22
Filing date: 2005-12-22
Publication date: 2006-06-29
Also published as: IE20050855A1; IES20050856A2

Abstract

A detection system for detecting concealed objects comprises sensors for receiving radiation in the frequency range of 1GHz to 10THz and a filter for partitioning at least two frequency bands in the received radiation. A processor generates an output corresponding to each frequency band, said outputs incorporating differences arising from differences in object interference and/or reflectivity for the different frequency bands. The processor controls a display of an image corresponding to each frequency band.

Description

"A Detection System"

Introduction

The invention relates to detection of concealed materials such as dielectric objects.

The use of millimetre waves to generate radiometric images of a scene has the advantageous property of being able to see through non-metallic material to image objects that are concealed, for example under clothing or in baggage. Materials such as metals and ceramics are readily detectable using this method.

Typical applications of millimetre wave imaging are portal screening to reveal weapons or contraband concealed under a persons clothing, poor weather visibility assistance and surveillance to remotely detect threat objects carried by people.

Materials such as metals that are electrically conductive are highly reflective to millimetre waves and can be easily distinguished and detected against a less conductive (and less reflective) background such as the human body. Lower conductivity materials are described as dielectric materials and can be characterised by their dielectric constant or relative permittivity value. Common dielectric materials such as glasses and plastics have typical permittivity values in the range 1 to 100 and in the lower end of this range we can categorise low dielectric materials as those with a value of dielectric constant below 30. Low dielectric constant materials such as plastics and explosives are more difficult to detect and image at millimetre-wave frequencies. This difficulty of detection is mainly due to their partial transparency to millimetre waves.

The invention is directed towards providing an improved detection system.

Summary of the Invention According to the invention, there is provided a detection system for detecting concealed objects, the system comprising:

sensor means for receiving radiation in the frequency range of IGHz to 10THz,

means for partitioning at least two frequency bands in the received radiation, and

a processor for processing signals from the sensor means to generate an output corresponding to each frequency band, said outputs incorporating differences arising from differences in object interference and/ or reflectivity for the different frequency bands.

In one embodiment, the processor controls a display of an image corresponding to each frequency band

In one embodiment, the sensor means comprises at least one sensor for receiving the radiation and filters for partitioning the frequency bands.

In one embodiment, the sensor means comprises an array of sensors, each for a spatial region of interest.

In one embodiment, the filter comprises a waveguide filter.

In one embodiment, the processor actively monitors the received radiation signal to generate the output indicating possibility of presence of a concealed object.

In one embodiment, the processor generates and stores pixel images, and analyses the pixel images to generate the output.

In one embodiment, the processor analyses patterns of pixel intensities In one embodiment, relative signal degradation or enhancement in different frequency bands is monitored by the processor.

In one embodiment, the sensor means comprises a plurality of sensors each of which detects at a specific frequency band.

In one embodiment, the sensor means comprises a detector whose operating frequency is swept over a range of frequencies.

In one embodiment, the frequency band bandwidth is in the range 0.1GHz to 40 GHz.

In one embodiment, the sensor means detects radiation at different frequency bands in ambient radiation coming from the scene.

In one embodiment, the system further comprises means for actively generating incident radiation for the scene.

Detailed Description of the Invention

The invention will be more clearly understood from the following description of some embodiments thereof, given by way of example only with reference to the accompanying drawings in which:-

Figs. l(a) and l(b) are representations of waves, and a reflection vs. frequency plot;

Fig. 2 is a diagrammatic representation of a detection system of the invention; Fig. 3 is a pair of plots showing theoretical and experimental reflectivity from polystyrene as a function of wavelength in air;

Figs. 4 and 5 are diagrams of different detectors of the invention;

Fig. 6 is a sample pair of displays of the same subject imaged a different frequencies; and

Fig. 7 is a plot of response characteristics of a single scene using different detection frequencies.

A sensor means has an array of detectors each receiving radiation from a spatial region of interest in a scene being monitored. Filters downstream of the detectors partition the radiation into at least two frequency bands. A processor processes the radiation in each band to generate an output. In one embodiment the output is a only a display image corresponding to each frequency band. An operator views the display images and discerns any objects in the scene according to differences in the display images arising from the fact that they correspond to different frequency bands. In another embodiment, the processor automatically performs image processing to actively generate an output indicating if there is a particular object in the scene. In doing so it monitors patterns of pixel intensities. The differences in the outputs for the different frequency bands arise from the fact that there is different object interference and/ or reflectivity for the different frequency bands. Relative signal degradation or enhancement are also monitored.

With such active operation of the processor, the use of an operator to interpret images in not required. In this case configurations of electronics hardware and software algorithms are implemented to identify the combination of signal responses that correspond to the presence of dielectric material in the scene. This method can be used to set up a means of automatically and/or remotely detecting the presence of a concealed dielectric material without the need for an attendant operator to examine and interpret the output images. The sensors sense at frequencies in the range of IGHz to 10 THz and the bandwidth of the frequency bands is between 0.1GHz and 40GHz. The bandwidth will determine the coherence length of the system where the coherence length is defined as c/B where c is the speed of light and B is the bandwidth.

Dielectric materials are known to exhibit specific reflectance behaviour to electromagnetic radiation that can be used to identify such materials using millimetre waves. Fig. l(a) shows the typical reflectance behaviour of incident radiation on a partially reflecting material. Incident radiation (13) upon a dielectric material (14) is reflected (internally and externally) and transmitted (15). The resultant reflectivity is a function of the sensed wavelength/frequency and the material thickness. Fig. l(b) shows the characteristic reflectivity response for a given material thickness over a range of frequencies.

A laboratory experiment demonstrates the principle as shown in Fig. 3. In this case the transmission of a piece of thin sheet dielectric material - 10mm thick polystyrene - is measured against a range of wavelengths (or frequencies where frequency = speed of light/ wavelength). The theoretical and measured response are shown. The agreement between the two is clearly seen.

Using the fact that the reflectance of a given dielectric material of specified refractive index and thickness varies as a function of frequency, the presence of that material can be deduced by viewing the material at multiple frequencies and looking for relative variation in the response between the different frequency returns.

The equations that govern the reflection of radiation at a dielectric layer placed in front of an object such as a human body depend on the thickness and refractive index of the layer and the frequency of the radiation, and are described below (basic properties and behaviour is referenced in O. S. Heavens, Optical properties of thin solid films, Dover Publications, Inc. (New York), pp.261, 1965). Consider u dielectric layers in the x-y plane, each layer m with a particular refractive index n_m and define^ as

where φ_m = angle of incidence, with respect to the normal to the boundary, of wave in layer m being considered λ = wavelength of the incident radiation

The direction of propagation of the radiation is the positive z direction and the plane of incidence is the z-x plane.

The amplitude of the electric vector of the wave incident on the first layer is E₀ ⁺ and that of the reflected wave is E₀ ^". In each layer m the results of all of the positive going (transmitted) waves sum to E_m ⁺ and those negative going (reflected) to E_m .

For non-normal incidence on an isotropic layer it is necessary to distinguish the plane of polarisation: vectors parallel to the plane of incidence have the subscript p and vectors perpendicular have the subscript s. Throughout the following only the components tangential to the boundaries between layers (i.e. the x and y components of the vectors) are used, as the boundary condition is that the tangential components of both the electric and magnetic fields be continuous across the boundaries. Most of the results derived here are general, for either polarisation, so care must be taken when calculating the final answer to consider the correct polarisation when substituting in for the Fresnel coefficients which differentiate between/? and s.

The Fresnel coefficients for a layer m (i.e. for reflection r and transmission t between layers (m-1) and (m) are:

The electric field amplitudes of the first and last media are related in matrix notation by

The characteristic matrix C of the whole system is the resultant of the product, in sequence, of the individual characteristic matrices

and

where t_m is the Fresnel transmission coefficient for the m^th boundary.

The characteristic matrix, C_m, relates the fields in the two adjacent media (m-1) (refractive index n_m.i) and m (n_m).

where

and d_m is the thickness of layer m.

The reflection and transmission coefficients for the system may therefore be calculated from equations (6-8) noting that there is no negative-going wave in the (u+ϊ)"' medium and so E₁₁₊/ = 0.

The frequency dependent reflectance and transmittance of the u layers may be found by multiplying the above equations by their complex conjugates:

the extra factor in the transmittance arising from energy conservation. These equations relate to the reflectance and transmission when viewing multiple layers - including airgaps.

Thus, for a single layer of dielectric over an object - for instance the human body:

From equations (11-12), the reflection and transmission coefficients are

and so the reflectance and transmittance of a dielectric layer over the body are given by

Referring to Fig. 2 a detection system 1 comprises a processing circuit 2 connected to input sensors 3, 4, and 5. It is used, as illustrated, for detecting presence of a dielectric object 10 hidden behind an obscurant 11. The dielectric object may or may not be placed in front of a background object 12 with which it may be in contact or separated by an air gap. A typical scenario would be a sheet or layers of sheets of explosive material placed on or near the body and concealed underneath clothing.

The sensors 3, 4, 5 image the scene at different frequencies - in the millimetre and/ or Terahertz ranges. In a typical embodiment the centre frequencies of the receivers are located at a fixed spectral separation. There may be a single group of sensors detecting at different frequencies or an array of these sensor groups. In a different embodiment a single or multiple sensors may be swept in frequency and the response of the sensors observed as the frequency of operation is changed. In another embodiment, the sensor (or array of sensors) may simultaneously detect a number of frequency bands where the sensor(s) is configured to detect partitioned bands in different spectral regions.

The sensors can operate as either active transceivers or as passive receivers. In the case of the active transceiver, the sensor transmits at the frequency band that it is monitoring - reflected returns and normal scene emissions and reflections are measured by the receiver circuit. The passive receiver version on the other hand observes the emissions and reflections from the scene. The scene may be illuminated by active sources of suitable wavelength radiation or by thermally generated broadband radiation or by naturally occurring background radiation.

The illumination radiation and the sensed radiation may be manipulated by polarisation, rotation or other optical processing as part of the detection method.

For the embodiment of passive or active radiometry, the circuit 2 executes the following logic to detect the obscured dielectric object 10. In the case where sensors of fixed but separated frequency are used, the received signal from each separate frequency sensor or band is monitored and compared for consistency of response. While there may be a difference in level of response when looking at the same scene, the relative scale of the signals can be expected to track each other. In the case where there is a dielectric material in the scene, there will be a relative variation in the response levels from the separate frequency channels corresponding to the interference and reflectivity difference of the dielectric material at the receiver frequency. By discriminating between the variations in received signal, the presence of the dielectric can be detected. Similarly for the case of a swept frequency receiver, there will be variation in signal corresponding to the changes in dielectric reflectivity over frequency. These variations are identified by the processing circuit 2.

By processing the signals from the sensors 3 - 5 in this manner, the circuit can detect the presence of concealed materials, such as objects concealed beneath clothing on a person or in baggage. This is highly advantageous. Also, the system may be used to detect objects such as metals and ceramics with millimetre-wave radiation, which have been more conventionally detected in the past.

Referring to Fig. 4, in this case a detection channel is configured to receive the incoming radiation at an antenna, 20. This signal is amplified in one or more amplification stages, 21. The amplified signal is then passed to a power splitter, 22, where the signal is split into two or more segments. The split segments are then further processed through amplification and filtering components, 23, 24. The order, number and sequence of the amplifiers and filters can be varied. The filter components 24 are configured to filter the signal in different portions of the frequency spectrum. These portions of the frequency spectrum can be distinct and separate or may overlap each other. The output of each of the filtered signals is passed to a detector device that converts the signal power to a voltage level before passing to the processing electronics (2, Fig 2). Depending on the content of the scene being viewed by the sensor, the level of the output signal can vary from one channel segment to the next because they are operating in different regions of the frequency spectrum, 26. In particular, if a dielectric sheet is present in the scene, this will provide a different response depending on which region of the frequency spectrum it is being viewed in. The processing electronics 2 are configured to identify and highlight these response differences.

Another embodiment is referred to in Fig. 5 where a given point in the scene is observed over a range of frequencies. In this embodiment, the received signal is collected in a front end antenna 30. This signal is then amplified in one or more amplification stages 31. The amplified signal is passed to a frequency sweeping component 32. This sweeps a window of given bandwidth over some or all of the operating frequency range of the receiver. The receiver works by sweeping through the operating frequency range of the channel while observing a single point in the scene. This scene point may be observed by a static measurement or by using a scanning system to direct the receiver to that scene point. In the case of a scanned observation, there may be some movement in the observed point during the process of sweeping through the channel bandwidth. The output of the frequency sweeper may be further processed - for example by an amplifier stage 33. The signal is finally passed to a detecting device that converts the signal power to a voltage level before passing to the processing electronics 2 (Fig. 2). Depending on what the receiver is viewing during the process of sweeping through the channel bandwidth, the level of the received signal may vary. Variations in output level can be due to specific components in the scene such as a dielectric sheet. The processing electronics are configured to identify and highlight such variations in received signal level as a scene point is observed over a range of frequencies.

The invention is not limited to the embodiments described but may be varied in construction and detail. For example, the system may include an opto-mechanical or electronically steered scanning mechanism. The sensors may detect at any frequency within the range of IGHz to 10 THz, with a chosen frequency separation between the sensors. There may be receiver or transceiver sensors.

An example of the method is shown in the image in Fig. 6. In this case two receivers were used to view a target made up of three thin sheet polymer materials. The observed materials in this instance were polyvinylchloride (PVC) and High Density Polyethylene (HDPE) sheets. These sheets varied in thickness from 3 millimetres to 6 millimetres. The target materials were imaged with two receivers operating at different frequencies. In this case the receivers operated at 94GHz with lOGhz bandwidth and 77GHz with 3GHz bandwidth. It can be seen that the materials present a different response depending on which frequency they are viewed at. In the case of the 77GHz image the 3mm HDPE appears bright in the image whereas it is dark when viewed at 94GHz. Similarly, differences in response level can also be seen for the other two materials.

The basis for an automated detection of the presence of dielectric material is presented in the response plots shown in Fig. 7. In this case the output from two different frequency bands is overlaid as a line scan of the scene is taken. The centre frequencies used in the detectors in this instance were 84GHz and 94GHz - each with a bandwidth of 12HGz. The scan contains a number of objects as indicated in the diagram including a sheet of dielectric material. It can be seen that the responses of the different frequencies track each other to a close correspondence along all of the line scan except where the dielectric material is being imaged. Where there is a sheet of dielectric in the scene, the responses track at levels that are different to each other in a manner that is clearly dissimilar to the close tracking throughout the rest of the scan. This divergence of response is identified by processing the output signals using computer hardware and software to action a means of automatically flagging the presence of a dielectric material in the scene.

Similar testing has been conducted with narrower band detectors where the operating bandwidth is reduced to 2GHz. In this case the dielectric detection effect is more pronounced although the overall detector sensitivity is reduced.

The invention is not limited to the embodiments described but may be varied in construction and detail. For example, the file=ter for discriminating the frequency bands may comprises a waveguide, planar, cavity based or software algorithm filter.

Claims

1. A detection system for detecting concealed objects, the system comprising:

sensor means for receiving radiation in the frequency range of IGHz to

10THz,

a processor for processing signals from the sensor means to generate an output corresponding to each frequency band, said outputs incorporating differences arising from differences in object interference and/or reflectivity for the different frequency bands.

2. A detection system as claimed in claim 1, wherein the processor controls a display of an image corresponding to each frequency band

3. A detection system as claimed in claims 1 or 2, wherein the sensor means comprises at least one sensor for receiving the radiation and filters for partitioning the frequency bands.

4. A detection system as claimed in claim 3, wherein the sensor means comprises an array of sensors, each for a spatial region of interest.

5. A detection system as claimed in claims 3 or 4, wherein the filter comprises a waveguide filter.

6. A detection system as claimed in any preceding claim, wherein the processor actively monitors the received radiation signal to generate the output indicating possibility of presence of a concealed object.

7. A detection system as claimed in claim 6, wherein the processor generates and stores pixel images, and analyses the pixel images to generate the output.

8. A detection system as claimed in claim 7, wherein the processor analyses patterns of pixel intensities

9. A detection system as claimed in any of claims 6 to 8, wherein relative signal degradation or enhancement in different frequency bands is monitored by the processor.

10. A detection system as claimed in claims 1 or 2, wherein the sensor means comprises a plurality of sensors each of which detects at a specific frequency band.

11. A detection system as claimed in any of claims 3 to 10, wherein the sensor means comprises a detector whose operating frequency is swept over a range of frequencies.

12. A detection system as claimed in any preceding claim where the frequency band bandwidth is in the range 0.1 GHz to 40 GHz.

13. A detection system as claimed in any preceding claim, wherein the sensor means detects radiation at different frequency bands in ambient radiation coming from the scene.

14. A detection system as claimed in any of claims 1 to 12 further comprising means for actively generating incident radiation for the scene.