EP2435955A1 - Systèmes d'inspection tomographique aux rayons x pour l'identification d'articles cibles spécifiques - Google Patents

Systèmes d'inspection tomographique aux rayons x pour l'identification d'articles cibles spécifiques

Info

Publication number
EP2435955A1
EP2435955A1 EP10781129A EP10781129A EP2435955A1 EP 2435955 A1 EP2435955 A1 EP 2435955A1 EP 10781129 A EP10781129 A EP 10781129A EP 10781129 A EP10781129 A EP 10781129A EP 2435955 A1 EP2435955 A1 EP 2435955A1
Authority
EP
European Patent Office
Prior art keywords
ray
image
scanning system
parameter
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
EP10781129A
Other languages
German (de)
English (en)
Other versions
EP2435955A4 (fr
Inventor
Edward James Morton
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Rapiscan Systems Inc
Original Assignee
Rapiscan Systems Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Rapiscan Systems Inc filed Critical Rapiscan Systems Inc
Priority to EP17187848.1A priority Critical patent/EP3267361A1/fr
Priority to EP17187847.3A priority patent/EP3282393A1/fr
Publication of EP2435955A1 publication Critical patent/EP2435955A1/fr
Publication of EP2435955A4 publication Critical patent/EP2435955A4/fr
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V5/00—Prospecting or detecting by the use of ionising radiation, e.g. of natural or induced radioactivity
    • G01V5/20—Detecting prohibited goods, e.g. weapons, explosives, hazardous substances, contraband or smuggled objects
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
    • G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
    • G01N23/046—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material using tomography, e.g. computed tomography [CT]
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V5/00—Prospecting or detecting by the use of ionising radiation, e.g. of natural or induced radioactivity
    • G01V5/20—Detecting prohibited goods, e.g. weapons, explosives, hazardous substances, contraband or smuggled objects
    • G01V5/22—Active interrogation, i.e. by irradiating objects or goods using external radiation sources, e.g. using gamma rays or cosmic rays
    • G01V5/226—Active interrogation, i.e. by irradiating objects or goods using external radiation sources, e.g. using gamma rays or cosmic rays using tomography
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/05—Recognition of patterns representing particular kinds of hidden objects, e.g. weapons, explosives, drugs

Definitions

  • the present invention is also a continuation-in-part of U.S. Patent Application Number 12/758,764, filed on April 12, 2010, which is a continuation of U.S. Patent Application No. 12/211,219, filed on September 16, 2008, and now issued U.S. Patent Number 7,724,868, which is a continuation of U.S. Patent No. 10/554,655, filed on October 25, 2005, and now issued U.S. Patent Number 7,440,543, which is a 371 national stage application of PCT/GB2004/001751, filed on April 23, 2004, and which, in turn, relies on Great Britain Patent Application Number 0309385.3, filed on April 25, 2003, for priority.
  • the present invention relates to X-ray scanning and, particularly to the security screening of baggage, packages and other suspicious objects, such as sharp objects, knives, nuclear materials, tobacco, currency, narcotics, and liquids.
  • X-ray computed tomography (CT) scanners have been used in security screening in airports for several years.
  • a conventional system comprises an X-ray tube that is rotated about an axis with an arcuate X-ray detector that is rotated at the same speed around the same axis.
  • the conveyor belt on which the baggage is carried is placed within a suitable aperture around the central axis of rotation, and moved along the axis as the tube is rotated.
  • a fan- beam of X-radiation passes from the source through the object to be inspected to the X-ray detector array.
  • the X-ray detector array records the intensity of X-rays passed through the object to be inspected at several locations along its length.
  • One set of projection data is recorded at each of a number of source angles. From these recorded X-ray intensities, it is possible to form a tomographic (cross-sectional) image, typically by means of a filtered back projection algorithm.
  • a tomographic image typically by means of a filtered back projection algorithm.
  • the X-ray source pass through every plane through the object. In the arrangement described above, this is achieved by the rotational scanning of the X-ray source, and the longitudinal motion of the conveyor on which the object is carried.
  • the rate at which X-ray tomographic scans can be collected is dependent on the speed of rotation of the gantry that holds the X-ray source and detector array.
  • the entire tube-detector assembly and gantry will complete two to four revolutions per second. This allows up to four or eight tomographic scans to be collected per second respectively.
  • the single ring of X-ray detectors has been replaced by multiple rings of detectors. This allows many slices (typically 8) to be scanned simultaneously and reconstructed using filtered back projection methods adapted from the single scan machines.
  • the source With a continuous movement of the conveyor through the imaging system, the source describes a helical scanning motion about the object. This allows a more sophisticated cone -beam image reconstruction method to be applied that can in principle offer a more accurate volume image reconstruction.
  • swept electron beam scanners have been demonstrated in medical applications whereby the mechanical scanning motion of the X-ray source and detectors is eliminated, being replaced by a continuous ring (or rings) of X-ray detectors that surround the object under inspection with a moving X-ray source being generated as a result of sweeping an electron beam around an arcuate anode.
  • This allows images to be obtained more rapidly than in conventional scanners.
  • the electron source lies on the axis of rotation, such swept beam scanners are not compatible with conveyor systems which themselves pass close, and parallel, to the axis of rotation.
  • the present invention provides an X-ray scanning system for inspecting items, the system comprising an X-ray source extending around a scanning volume, and defining a plurality of source points from which X-rays can be directed through the scanning volume, an X-ray detector array also extending around the scanning volume and arranged to detect X- rays from the source points which have passed through the scanning volume and produce output signals dependent on the detected X-rays, and a conveyor arranged to convey the items through the scanning volume.
  • the present invention further provides a networked inspection system comprising an X-ray scanning system, a workstation and connection means arranged to connect the scanning system to the workstation, the scanning system comprising an X-ray source extending around a scanning volume, and defining a plurality of source points from which X- rays can be directed through the scanning volume, an X-ray detector array also extending around the scanning volume and arranged to detect X-rays from the source points which have passed through the scanning volume and produce output signals dependent on the detected X- rays, and a conveyor arranged to convey the items through the scanning volume.
  • a networked inspection system comprising an X-ray scanning system, a workstation and connection means arranged to connect the scanning system to the workstation, the scanning system comprising an X-ray source extending around a scanning volume, and defining a plurality of source points from which X- rays can be directed through the scanning volume, an X-ray detector array also extending around the scanning volume and arranged to detect
  • the present invention further provides a sorting system for sorting items, the system comprising a tomographic scanner arranged to scan a plurality of scanning regions of each item thereby to produce a scanner output, analyzing means arranged to analyze the scanner output and allocate each item to one of a plurality of categories at least partly on the basis of the scanner output, and sorting means arranged to sort items at least partly on the basis of the categories to which they have been allocated.
  • the present invention further provides an X-ray scanning system comprising an X-ray source arranged to generate X-rays from a plurality of X-ray source positions around a scanning region, a first set of detectors arranged to detect X-rays transmitted through the scanning region, a second set of detectors arranged to detect X-rays scattered within the scanning region, and processing means arranged to process outputs from the first set of detectors to generate image data which defines an image of the scanning region, to analyze the image data to identify an object within the image, and to process the outputs from the second set of detectors to generate scattering data, and to associate parts of the scattering data with the object.
  • an X-ray scanning system comprising an X-ray source arranged to generate X-rays from a plurality of X-ray source positions around a scanning region, a first set of detectors arranged to detect X-rays transmitted through the scanning region, a second set of detectors arranged to detect X-rays scattered within the scanning region, and
  • the present invention further provides a data collecting system for collecting data from an X-ray scanner, the system comprising a memory having a plurality of areas each being associated with a respective area of an image, data input means arranged to receive input data from a plurality of X-ray detectors in a predetermined sequence, processing means arranged to generate from the input data X-ray transmission data and X-ray scattering data associated with each of the areas of the image, and to store the X-ray transmission data and the X-ray scattering data in the appropriate memory areas.
  • the present invention further provides an X-ray scanning system comprising a scanner arranged to scan an object to generate scanning data defining a tomographic X-ray image of the object, and processing means arranged to analyze the scanning data to extract at least one parameter of the image data and to allocate the object to one of a plurality of categories on the basis of the at least one parameter.
  • the present invention provides an X-ray scanning system comprising a scanner arranged to scan an object to generate scanning data defining a tomographic X-ray image of the object, and processing means arranged to analyse the scanning data to extract at least one parameter of the image data and to allocate the object to one of a plurality of categories on the basis of the at least one parameter.
  • the processing means comprise: one or more parameter extractors for identifying one or more predefined features in the tomographic X-ray image, the identified features being low level parameters of the X-ray image; each parameter extractor being arranged to perform a different processing operation to determine a different parameter; one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image; and a database searcher for mapping the X-ray image of the object as one of 'threat-causing' or 'clear' by using the constructed high level parameters of the X-ray image and predefined data stored in a database coupled with the database searcher.
  • the parameter extractors are designed to operate on one of 2-dimensional images, 3 -dimensional images and sinogram image data.
  • the present invention provides a method for detecting a predefined material by using an X-ray scanning system comprising a scanner arranged to scan an object to generate a tomographic X-ray image of the object, and processing means arranged to analyse the scanning data to extract at least one parameter of the image data and to allocate the object to one of a plurality of categories on the basis of the at least one parameter.
  • the method comprises the steps of: configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image, the identified features being low level parameters of the X-ray image; configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image; identifying the object as the predefined material by mapping the constructed high level parameters of the X-ray image with stored data defining the material.
  • the method further comprises the step of classifying the X-ray image of the object as one of 'threat-causing' or 'clear' by mapping the constructed high level parameters of the X-ray image with predefined stored data.
  • the present invention provides a method for detecting a liquid wherein the steps of configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image and configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image comprise the steps of: configuring a first parameter extractor to process the X-ray image to identify: the object being scanned; an outer envelope defining the object being scanned, and one or more flat surfaces within an outer envelope of the object being scanned; configuring a second parameter extractor to locate a contiguous volume of a material of uniform density extending from each identified flat surface in a vertical direction; and configuring a decision tree for: calculating the volume of the contiguous volume of material located; assigning the calculated volume to at least one predetermined shape corresponding to one of an oval bottle, a rectangular bottle and a triangular bottle; calculating a mean reconstructed density of the contiguous volume; and transferring the parameters corresponding to
  • the present invention provides a method for detecting a narcotic wherein, the steps of configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image and configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image comprise the steps of: configuring a first parameter extractor to process the X-ray image to identify contiguous volumes of low density material in at least sheet and bulk shapes; configuring a second parameter extractor to processes the identified contiguous volumes to determine statistical properties of the contiguous volumes; configuring a third parameter extractor for identifying one or more collections of randomly oriented parts of small volume; and configuring a decision tree for correlating the parameters corresponding to volume shape and statistical properties identified by a plurality of parameter extractors; and transferring the correlated data for identification of the narcotic material by mapping against a database.
  • the present invention provides a method for detecting a currency wherein, the steps of configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image and configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image comprise the steps of: configuring a first parameter extractor to identify one or more "bow-tie" shaped features in the X-ray image; configuring a second parameter extractor to identify one or more rectangular shapes in multiples of predefined physical dimension of predefined denominations of currency; configuring a third parameter extractor for identifying repeating patterns in the parameters identified by the first and the second parameter extractors; configuring a fourth parameter extractor for generating statistical properties of the patterns identified by the first, the second and the third parameter extractors; and configuring a decision tree for correlating the parameters identified by a plurality of parameter extractors; and transferring the correlated data for identification of the currency by mapping against a database.
  • the present invention provides a method for detecting cigarettes wherein, the steps of configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image and configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image comprise the steps of: configuring a first parameter extractor to identify repeating array structures with a length and width dimension consistent with predefined dimensions of cigarettes; configuring a second parameter extractor to identify rectangular volumes of predefined aspect ratio matching that of predefined cigarette packaging with a density that is consistent with predefined brands of cigarettes; and configuring a decision tree for correlating the parameters identified by a plurality of parameter extractors; and transferring the correlated data for identification of the cigarettes by mapping against a database.
  • the present invention provides a method for detecting a special nuclear material or a shielded radioactive source wherein, the steps of configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image and configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image comprise the steps of: configuring a first parameter extractor to identify highly attenuating regions in the X-ray image where the reconstructed pixel intensity is above a predefined threshold value; and configuring a decision tree for: determining whether the attenuating region is part of a larger structure; evaluating at least a shape, a location and a size of the attenuating region if it is determined that the attenuating region is not a part of a larger structure; and transferring the evaluated parameters corresponding to the attenuating region for identification of the special nuclear material or a shielded radioactive source by mapping against a database.
  • the present invention provides a method for detecting a pointed object or a knife wherein, the steps of configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image and configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image comprise the steps of: configuring a first parameter extractor to detect one or more protruding points in the X-ray image; configuring a second parameter extractor to identify one or more blades having a predefined length to width aspect ratio; configuring a third parameter extractor for identifying folded blades having a repeating structure of at least two air gaps and three material fills; and configuring a decision tree for correlating the parameters identified by a plurality of parameter extractors; and transferring the correlated data for identification of the pointed object or knife by mapping against a database.
  • the present invention provides a method for detecting a fire arm wherein, the steps of configuring a plurality of parameter extractors for identifying one or more predefined features in the tomographic X-ray image and configuring one or more decision trees for constructing high level parameters by analyzing the identified low level parameters of the X-ray image comprise the steps of: configuring a first parameter extractor to identify one or more cylindrical metal tubes in the X-ray image; configuring a second parameter extractor to identify one or more trigger mechanism and firing pin; configuring a third parameter extractor for identifying high density slugs and bullets with composition ranging from aluminium (density 2.7 g/cm3) to lead (density > 11 g/cm3); and configuring a decision tree for correlating the parameters identified by a plurality of parameter extractors and associated data; and transferring the correlated data for identification of the currency by mapping against a database.
  • the present invention is an X-ray scanning system comprising: a non-rotating X-ray scanner that generates scanning data defining a tomographic X-ray image of the object; a processor executing programmatic instructions wherein said executing processor analyzes the scanning data to extract at least one parameter of the tomographic X-ray image and wherein said processor is configured to determine if said object is a bottle containing a liquid, a narcotic, tobacco, fire-arms, currency, sharp objects, or any other illicit object.
  • the processor executes programmatic instructions to allocate the object to one of a plurality of categories on the basis of the at least one parameter.
  • the programmatic instructions comprise at least one parameter extractor for identifying at least one predefined feature in the tomographic X-ray image wherein said predefined feature comprises a plurality of low level parameters of the X-ray image.
  • Such low level parameters can include basic dimension, density, size, and shape information.
  • the programmatic instructions comprise at least one decision tree for constructing high level parameters based upon the identified low level parameters of the X-ray image.
  • the high level parameters include determinations of the type of object, such as a bottle, repeating patterns, arrays of structures, among other defining variables.
  • the X-ray scanning system further comprises a database search tool for mapping the constructed high level parameters of the X-ray image to predefined data stored in a database.
  • the X-ray scanning system further comprises an alarm system for activating an alarm based on a result of said mapping, wherein said alarm defines an object as being a potential threat or not a potential threat.
  • the at least one parameter extractor is configured to operate on one 2 -dimensional images, 3 -dimensional images or sinogram image data.
  • the methods of the present invention described above are provided as a computer readable medium tangibly embodying a program of machine -readable instructions executable by a processor.
  • FIG. 1 is a longitudinal section of a real time tomography security scanning system according to a first embodiment of the invention
  • FIG. Ia is a perspective view of an X-ray source of the system of FIG. 1;
  • FIG. 2 is a plan view of the system of FIG. 1;
  • FIG. 3 is a schematic side view of the system of FIG. 1;
  • FIG. 4 is a schematic diagram of a data acquisition system forming part of the system of FIG. 1;
  • FIG. 5 is a schematic diagram of a threat detection system forming part of the system of FIG. 1;
  • FIG. 6 is a schematic diagram of a baggage sorting system according to an embodiment of the invention including the scanning system of FIG. 1;
  • FIG. 7 is a schematic diagram of a baggage sorting system according to a further embodiment of the invention.
  • FIG. 8a is a schematic diagram of baggage sorting systems according to further embodiments of the invention.
  • FIG. 8b is another schematic diagram of baggage sorting systems according to further embodiments of the invention.
  • FIG. 8c is another schematic diagram of baggage sorting systems according to further embodiments of the invention.
  • FIG. 9 is a schematic diagram of a networked baggage sorting system according to a further embodiment of the invention.
  • FIG. 10 is a schematic plan view of a stand-alone scanning system according to a further embodiment of the invention.
  • FIG. 11 is a schematic side view of the system of FIG. 10;
  • FIG. 12 is a schematic side view of a modular scanning system according to a further embodiment of the invention.
  • FIG. 13 is a diagram of an X-ray scattering event
  • FIG. 14 is a longitudinal section through a security scanning system according to a further embodiment of the invention
  • FIG. 15 is a further longitudinal section through the system of FIG. 14 showing how different scatter events are detected
  • FIG. 16 is a transverse section through the system of FIG. 14;
  • FIG. 17 is a schematic diagram of a data acquisition system of the scanning system of FIG. 14;
  • FIG. 18 is a partial view of a dual energy scanner according to a further embodiment of the invention.
  • FIG. 19 is a further partial view of the scanner of FIG. 18;
  • FIG. 20 is a schematic view of a dual energy X-ray source of a further embodiment of the invention.
  • FIG. 21 is a schematic view of a detector array of a scanner according to a further embodiment of the invention.
  • FIG. 22 is a schematic view of a detector array of a scanner according to a further embodiment of the invention.
  • FIG. 23 is a circuit diagram of a data acquisition circuit of the embodiment of FIG.
  • FIG. 24 is a circuit diagram of a data acquisition circuit of a further embodiment of the invention.
  • a concourse baggage scanning system 6 comprises a scanning unit 8 comprising a multi-focus X-ray source 10 and X-ray detector array 12.
  • the source 10 comprises a large number of source points 14 in respective spaced locations on the source, and arranged in a full 360. degree circular array around the axis X-X of the system. It will be appreciated that arrays covering less than the full 360. degree, angle can also be used.
  • the X-ray source 10 is made up of a number of source units 11 which are spaced around the scanning region 16 in a substantially circular arrangement, in a plane perpendicular to the direction of movement of the conveyor.
  • Each source unit 11 comprises a conductive metal suppressor 13 having two sides and an emitter element 15 extending along between the suppressor sides.
  • a number of grid elements in the form of grid wires 17 are supported above the suppressor 13 perpendicular to the emitter element 15.
  • a number of focusing elements in the form of focusing wires 19 are supported in another plane on the opposite side of the grid wires to the emitter element.
  • the focusing wires 19 are parallel to the grid wires 17 and spaced apart from each other with the same spacing as the grid wires, each focusing wire 19 being aligned with a respective one of the grid wires 17.
  • the focusing wires 19 are supported on two support rails 21 which extend parallel to the emitter element 15, and are spaced from the suppressor 13.
  • the support rails 21 are electrically conducting so that all of the focusing wires 19 are electrically connected together.
  • One of the support rails 21 is connected to a connector 23 to provide an electrical connection for the focusing wires 19.
  • Each of the grid wires 17 extends down one side of the suppressor 12 and is connected to a respective electrical connector 25 which provide separate electrical connections for each of the grid wires 17.
  • An anode 27 is supported above the grid wires 17 and focusing wires 19.
  • the anode 27 is formed as a rod, typically of copper with tungsten or silver plating, and extends parallel to the emitter element 15.
  • the grid and focusing wires 17, 19 therefore extend between the emitter element 15 and the anode 27.
  • An electrical connector 29 provides an electrical connection to the anode 27.
  • the grid wires 17 are all connected to a negative potential, apart from two which are connected to a positive potential. These positive grid wires extract a beam of electrons from an area of the emitter element 15 and, with focusing by the focusing wires 19, direct the electron beam at a point on the anode 27, which forms the X-ray source point for that pair of grid wires.
  • the potential of the grid wires can therefore be switched to select which pair of grid wires is active at any one time, and therefore to select which point on the anode 27 is the active X-ray source point at any time.
  • the source 10 can therefore be controlled to produce X-rays from each of the source points 14 in each of the source units 11 individually and, referring back to FIG. 1, X-rays from each source point 14 are directed inwards through the scanning region 16 within the circular source 10.
  • the source 10 is controlled by a control unit 18 which controls the electrical potentials applied to the grid wires 17 and hence controls the emission of X-rays from each of the source points 14.
  • Other suitable X-ray source designs are described in WO 2004/097889.
  • the multi-focus X-ray source 10 allows the electronic control circuit 18 to be used to select which of the many individual X-ray source points 14 within the multi-focus X-ray source is active at any moment in time.
  • the electronic control circuit 18 controls the electronic control circuit 18 to select which of the many individual X-ray source points 14 within the multi-focus X-ray source is active at any moment in time.
  • the illusion of X-ray source motion is created with no mechanical parts physically moving.
  • the angular velocity of source rotation can be increased to levels that simply cannot be achieved when using conventional rotating X-ray tube assemblies. This rapid rotational scanning translates into an equivalently speeded up data acquisition process and subsequently fast image reconstruction.
  • the detector array 12 is also circular and arranged around the axis X-X in a position that is slightly offset in the axial direction from the source 10.
  • the source 10 is arranged to direct the X-rays it produces through the scanning region 16 towards the detector array 12 on the opposite side of the scanning region.
  • the paths 18 of the X-ray beams therefore pass through the scanning region 16 in a direction that is substantially, or almost, perpendicular to the scanner axis X-X, crossing each other near to the axis.
  • the volume of the scanning region that is scanned and imaged is therefore in the form of a thin slice perpendicular to the scanner axis.
  • the source is scanned so that each source point emits X-rays for a respective period, the emitting periods being arranged in a predetermined order.
  • the signals from the detectors 12, which are dependent on the intensity of the X-rays incident on the detector are produced, and the intensity data that the signals provide are recorded in memory.
  • the detector signals can be processed to form an image of the scanned volume.
  • a conveyor belt 20 moves through the imaging volume, from left to right, as seen in
  • X-ray scatter shields 22 are located around the conveyor belt 20 upstream and downstream of the main X-ray system to prevent operator dose due to scattered X-rays.
  • the X-ray scatter shields 22 include lead rubber strip curtains 24 at their open ends such that the item 26 under inspection is dragged through one curtain on entering, and one on leaving, the inspection region.
  • the main electronic control system 18, a processing system 30, a power supply 32 and cooling racks 34 are shown mounted underneath the conveyor 20.
  • the conveyor 20 is arranged to be operated normally with a continuous scanning movement at constant conveyor speed, and typically has a carbon-fiber frame assembly within the imaging volume. Referring to FIG.
  • the processing system 30 includes an electronic data acquisition system and real-time image reconstruction system.
  • the X-ray detector array 12 comprises banks of individual X-ray detectors 50 configured in a simple linear pattern (e.g. 1. times.16). Multiple ring patterns (e.g. 8.times.l6) are also possible. Each detector 50 outputs a signal dependent on the intensity of the X-rays it detects.
  • a multiplexing block 52 multiplexes the output data signals from each of the input X-ray detectors 50, performs data filtering, gain and offset corrections and formats the data into a high-speed serial stream.
  • a selection block 53 takes input from all of the multiplexing blocks 52 and selects just the part of the entire X- ray data that is required for the image reconstruction.
  • the selection block 53 also determines the un-attenuated X-ray beam intensity, Io, for the appropriate X-ray source point (which will vary for every X-ray source point within the multi-focus X-ray tube), processes the X-ray intensity data, Ix, from the multiplexing block 52 by forming the result log o (Ix/Io) and then convolves this with a suitable 1-D filter.
  • the resulting projection data is recorded as a sinogram, in which the data is arranged in an array with pixel number along one axis, in this case horizontally, and source angle along another axis, in this case vertically. Data is then passed from the selection block 53 in parallel to a set of back projection- summation processor elements 54.
  • the processor elements 54 are mapped into hardware, using look-up tables with pre-calculated coefficients to select the necessary convolved X-ray data and weighting factors for fast back projection and summation.
  • a formatting block 55 takes the data representing individual reconstructed image tiles from the multiple processor elements 54 and formats the final output image data to a form suitable for generating a suitably formatted three dimensional image on a display screen. This output can be generated fast enough for the images to be generated in real time, for viewing in real time or off-line, hence the system is termed a real time tomography (RTT) system.
  • RTT real time tomography
  • the multiplexing block 52 is coded in software, the selection block
  • each of the final output images for each baggage item is then processed by a threat detection processor 60 within the processing system 30 which is arranged to determine whether the imaged baggage item represents a threat.
  • input X-ray tomographic image data 62 is passed in to a set of low- level parameter extractors 63 (level 1).
  • the parameter extractors 63 identify features in the image such as areas of constant grey level, texture and statistics. Some of the extractors work on the data for individual 2 dimensional images or slices, some work on the 3 dimensional images, and some work on the sinogram data. Where possible, each extractor works in parallel on the same set of input data, and each extractor is arranged to perform a different processing operation and to determine a different parameter.
  • the parameters determined by the parameter extractors 63 are passed up to a set of decision trees 64 (level 2). Details of the parameters extracted are given below.
  • the decision trees 64 each take a number (typically all) of the low level parameters and construct respective higher level information, such as information regarding contiguous volumes, with associated statistics.
  • a database searcher 65 maps the higher level parameters produced at level 2 into a "red" probability Pr(threat) of there being a threat present and a "green" probability Pr(safe) of the item under inspection being safe. These probabilities are used by the processing system 30 to allocate the scanned item to an appropriate safety category, and to produce an automatic sorting control output.
  • This automatic sorting control output can be either a first "green” output indicating that the item is allocated to a clear category, a second "red” output indicating that the item is allocated to a "not clear” category, or a third “amber” output indicating that the automatic sorting cannot be carried out with sufficient reliability to allocated the item to the "clear" or the "not clear” category. Specifically if Pr(safe) is above a predetermined value, (or Pr(threat) is below a predetermined value) then the automatic sorting output will be produced having a first signal form, indicating that the item should be allocated to the green channel.
  • Pr(threat) is above a predetermined value, (or Pr(safe) is below a predetermined value) then the automatic sorting output will be produced having a second signal form, indicating that the item should be allocated to the red channel. If Pr(threat) (or Pr(safe)) is between the two predetermined values, then the automatic sorting output is produced having a third signal form, indicating that the item cannot be allocated to either the red or green channel.
  • the probabilities can also be output as further output signals.
  • the parameters that will be determined by the parameter extractors 63 generally relate to statistical analysis of pixels within separate regions of the 2-dimensional or 3-dimensional image.
  • a statistical edge detection method is used. This starts at a pixel and then checks whether adjacent pixels are part of the same region, moving outwards as the region grows.
  • an average intensity of the region is determined, by calculating the mean intensity of the pixels within the region, and the intensity of the next pixel adjacent to the region is compared to that mean value, to determine whether it is close enough to it for the pixel to be added to the region.
  • the standard deviation of the pixel intensity within the region is determined, and if the intensity of the new pixel is within the standard deviation, then it is added to the region. If it is not, then it is not added to the region, and this defines the edge of the region as being the boundary between pixels in the region and pixels that have been checked and not added to the region.
  • parameters of the region can be measured.
  • One such parameter is a measure of the variance of the pixel intensity within the region. If this is high this might be indicative of a lumpy material, which might for example be found in a home-made bomb, while if the variance is low this would be indicative of a uniform material such as a liquid.
  • Another parameter that is measured is the skewedness of the distribution of pixel value within the region, which is determined by measuring the skewedness of a histogram of pixel values.
  • a Gaussian, i.e. non-skewed, distribution indicates that the material within the region is uniform, whereas a more highly skewed distribution indicates non-uniformities in the region.
  • these low-level parameters are passed up to the decision trees 64, where higher level information is constructed and higher level parameters are determined.
  • One such higher level parameter is the ratio of the surface area to the volume of the identified region.
  • Another is a measure of similarity, in this case cross-correlation, between the shape of the region and template shapes stored in the system.
  • the template shapes are arranged to correspond to the shape of items that pose a security threat, such as guns or detonators.
  • an in-line real time tomography baggage sorting system comprises the scanning system 6 of FIG. 1 with the conveyor 20 passing through it. Downstream of the scanning system 6 a sorting device 40 is arranged to receive articles of baggage from the conveyor 20 and move them onto either a clear or "green" channel conveyor 42 or a not clear or “red” channel conveyor 44.
  • the sorting device 40 is controlled by the automatic sorting output signals via a control line 46 from the processing system 30, which are indicative of the decision of the processing system 30 as to whether the item is clear or not, and also by signals from a workstation 48 to which it is connected via line 45.
  • the images from the scanning system 6 and signals from the processing system 30, indicative of the red and green probabilities and the nominal decision of the processing system 30, are also fed to the workstation 48.
  • the workstation is arranged to display the images on a screen 47 so that they can be viewed by a human operator, and also to provide a display indicative of the green and red probabilities and the nominal automatic sorting decision.
  • the user at the workstation can review the images and the probabilities, and the automatic sorting output, and decide whether to accept or override the decision of the scanning system, if that was to allocate the item to the red or green category, or to input a decision if the scanning system decision was to allocate the item to the "amber" category.
  • the workstation 48 has a user input 49 that enables the user to send a signal to the sorting device 40 which can be identified by the sorting device as over-riding the decision of the scanning system. If the over-riding signal is received by the sorting device, then the sorting device does over-ride the decision of the scanning system. If no over-ride signal is received, or indeed if a confirming signal is received from the workstation confirming the decision of the scanning system, then the sorting device sorts the item on the basis of the decision of the scanning system. If the sorting system receives an "amber" signal from the scanning system relating to an item, then it initially allocates that item to the "red" category to be put into the red channel. However, if it receives an input signal from the workstation before it sorts the item indicating that it should be in the "green” category, then it sorts the item to the green channel.
  • the sorting can be fully automatic, with the processing system giving one of just two sorting outputs, "clear” and “not clear", allocating the item to either the green or the red channel. It would also be possible for the processing system to determine just one probability Pr(threat) with one threshold value and allocate the item to one of the two categories depending on whether the probability is above or below the threshold. In this case the allocation would still be provisional and the operator would still have the option of overriding the automatic sorting. In a further modification the automatic category allocation of the scanning system is used as the final allocation, with no user input at all. This provides a fully automated sorting system.
  • the scan speed is matched to the conveyor velocity, so that the baggage can be moved at a constant velocity from a loading area where it is loaded onto the conveyor 20, through the scanning system 6, and on to the sorting device 40.
  • the conveyor 20 extends for a distance L, between the exit of the scanning system 6 and the sorting device 40.
  • an operator can view the image data of the item under inspection, and the initial category allocation determined by the scanning system, and confirm or reject the automated decision of the RTT system.
  • the baggage would then be either accepted into the Clear channel and passed forward ready for transportation or rejected into the Not Cleared channel for further investigation.
  • the RTT scanning unit 8 is able to operate at full baggage belt speed, and hence no baggage queuing or other divert mechanism is required for optimal system operation.
  • the limited throughput capability of conventional rotating source systems is a significant constraint. Often this means placing multiple conventional CT machines in parallel, and using sophisticated baggage handling systems to switch the item for inspection to the next available machine. This complexity can be avoided with the arrangement of FIG. 6.
  • a second embodiment of the invention comprises a redundant system in which two RTT scanning systems 70, 72 are located in series on the same conveyor 74 such that if one system were to be taken out of service, then the other could continue to scan baggage. In either case, the conveyor belt 74 would continue to run through both scanning systems 70, 72 at standard operating belt speed.
  • FIG. 8a in a third embodiment there is provided a more complex redundant system in which two RTT systems 82, 84 are operated in parallel.
  • a first main incoming conveyor 86 brings all items to be sorted to a first sorting device 88 which can transfer items onto either one of two further conveyors 90, 92.
  • Each of these two conveyors 90, 92 passes through a respective one of the scanning systems 82, 84, which will scan the items and enable a decision to be made as to whether to clear the item or not.
  • a further sorting device 94, 96 is provided on each of the two conveyors 90, 92 which is arranged to sort the baggage onto a common "green channel” conveyor 98 for onward transportation, or a "red channel” conveyor 100 if it is not cleared, where it can undergo further investigation.
  • the main incoming conveyor 86 and the common "green channel" conveyor move at a speed of 1 m/s whereas the scanning conveyors 82, 84 travel at half that speed, i.e. 0.5 m/s.
  • the system can be expanded with more parallel RTT systems, with the ratio of the speed of the main incoming conveyor to that of the scanner conveyors being equal to, or substantially equal to, the number of parallel scanners, although the sorting devices may become unreliable at more than about 1 m/s main conveyor speed.
  • a baggage sorting system comprises a number of RTT scanners 81b, 82b, 83b, typically up to about 60 in one system, each associated with a respective check-in desk.
  • a sorting device 84b, 85b, 86b is associated with each RTT scanner, and baggage is conveyed on a conveyor from each RTT scanner to its associated sorting device.
  • Each sorting device 84b, 85b, 86b sorts the baggage, in response to signals from its scanner, onto either a common clear channel conveyor 88b, or a common reject channel conveyor 87b.
  • a further backup RTT scanner 89b is provided on the reject channel conveyor 87b, with an associated sorting device 90b, that can leave baggage on the reject channel conveyor 87b, or transfer it to the clear channel conveyor 88b.
  • each of the primary scanners 81b, 82b, 83b sorts the baggage, and the backup or redundant scanner 89b simply provides a further check on items sorted into the reject channel. If that scanner determines that an item of baggage represents no, or a sufficiently low threat, then it transfers it to the clear channel. If one of the primary scanners is not functioning or has a fault, then its associated sorting device is arranged to sort all baggage from that scanner to the reject channel. Then, the back-up scanner 89b scans all of that baggage and controls sorting of it between the clear and reject channels. This enables all the check-in desks to continue to function while the faulty scanner is repaired or replaced. Referring to FIG.
  • baggage from each of the check-in desks is transferred via a plurality of separate conveyors onto a central loop or carousel 81c, on which it circulates continuously.
  • a number of sorting devices 82c, 83c, 84c are each arranged to transfer items of baggage from the loop 81c to a respective conveyor leading to a respective RTT scanner 85c, 86c, 87c.
  • the sorting devices 82c, 83c, 84c are controlled by the scanners to control the rate at which baggage items are fed to each of the scanners.
  • conveyors transfer all of the baggage items to a common exit conveyor 88c leading to a further sorting device 89c. This is controlled by all of the scanners to sort each of the baggage items between a clear channel 90c and a reject channel 91c.
  • each item is given a 6-digit
  • the scanners can therefore identify which item of baggage is being scanned at any one time, and associate the scanning results with the appropriate item.
  • the sorting devices can therefore also identify the individual baggage items and sort them based on their scanning results.
  • the number of scanners and the speeds of the conveyors in this system are arranged such that, if one of the scanners is not functioning, the remaining scanners can process all of the baggage that is being fed onto the loop 81c from the check-in desks.
  • the sorting devices 82c, 83c, 84c that select which items are transferred to each scanner are not controlled by the scanners, but are each arranged to select items from the loop 81c so as to feed them to the respective scanner at a predetermined rate.
  • a networked system comprises three scanning systems 108 similar to that of FIG. 6, and four operator workstations 148.
  • the video image outputs from the three RTT scanning systems 108 are connected via respective high bandwidth point-to-point video links to real time disk arrays 109 which providing transient storage for the raw image data, to a redundant video switch 110.
  • the disk arrays 109 are in turn connected to each of the workstations 148.
  • the video switch 110 is therefore able to transmit the raw video image output from each of the scanning systems 108 from its temporary storage, to any one of the workstations 148, where it can be used to create 3- dimensional video images which can be viewed off-line.
  • a networked system comprises a single scanning system 108 connected to a server and a workstation 148.
  • the video image output from the scanning system 108 is connected to a real time disk array 109, which provides transient storage for the raw image data.
  • the disk array 109 is in turn connected to the workstation 148.
  • the probability signal and allocation signal outputs are sent to the workstation 148 together with the associated video image output to be monitored by an operator.
  • the networked single scanning system may be part of a networked system with multiple scanning systems.
  • an in-line scanner has a conveyor belt 160 just as long as the main scatter shields 162.
  • the item for inspection is placed onto the conveyor belt 160 and the item loaded into the system.
  • the item is then scanned through the scanner machine 164 and images are generated.
  • the item is pre-screened with a simple transmission X-ray system to identify likely threat areas prior to computed tomography screening of selected planes in the object.
  • Such applications are simple for a real-time multi- focus system to cope with. Here, no pre-screening would be used and a true three- dimensional image of the complete item would be obtained.
  • the locus of source points in the multi-focus X-ray source will extend in an arc over an angular range of only 180 degrees plus the fan beam angle (typically in the range 40 to 90 degrees).
  • the number of discrete source points is advantageously selected to satisfy the Nyquist sampling theorem.
  • a complete 360 degree ring of source points is used. In this case, the dwell-time per source point is increased over a 180+ fan beam configuration for a given scan rate and this is advantageous in improving reconstructed image signal-to-noise ratio.
  • the scanner system of FIG. 1 is an integrated scanner system, in that the control, processing, power supply, and cooling units 18, 30, 32, 34 are housed in a unit with the scanning system 8 and the screening 22.
  • FIG. 12 in a further embodiment there is provided a modular system in which some, or all, of the control, processing, power supply, and cooling racks 218, 230, 232, 234 are located remotely from the scanning unit 208 comprising multi-focus X-ray source and sensor array. It is advantageous to use a modular design to facilitate easy installation, particularly in baggage handling hall environments, where systems may be suspended from the ceiling or in regions with restricted access. Alternatively, a complete system can be configured as an integrated unit with the sub- assembly units co-located within a single housing.
  • a single X-ray detector ring is used. This is inexpensive to construct and provides adequate signal-to-noise performance even at high image scanning rates with a simple fan-beam image reconstruction algorithm.
  • the use of a multi-ring sensor increases dwell-time per source point resulting in larger integrated signal size and consequent improvement in signal-to-noise ratio in the reconstructed image.
  • a multi-focus X-ray source based computed tomography system for baggage inspection provides a good ratio of angular source rotational speed to linear conveyor speed for the purposes of high probability detection of threat materials and objects in the item under inspection.
  • the conveyor speed is 0.5 m/s as is common in airport systems.
  • the source can achieve 240 source rotations about the conveyor per second, so the object under inspection will move a distance of 2.08 mm through the imaging slice during the scan. In a conventional system with source rotation of 4 revolutions per second, the object under inspection will move a distance of 62.5 mm through the imaging slice during the scan at the same belt speed.
  • the primary goal of an inspection system for detection of threat materials is to detect accurately the presence of threat materials and to pass as not suspect all other materials.
  • RTT real-time tomography
  • the X-ray focus will be switched to maximize the distance of the current anode irradiation position from all previous irradiation positions in order to minimize the instantaneous thermal load on the anode.
  • Such instantaneous spreading of the X-ray emission point is advantageous in minimizing partial volume effect due to conveyor movement so further improving reconstructed pixel accuracy.
  • RTT systems allow a high level of accuracy to be achieved in automated threat detection. With this high level of accuracy, RTT systems can be operated in unattended mode, producing a simple two-state output indication, with one state corresponding to a green or clear allocation and the other to a red or not clear allocation. Green bags are cleared for onward transport. Red bags represent a high level of threat and should be reconciled with the passenger and the passenger barred from traveling.
  • n is an integer
  • ⁇ is the wavelength of the X-ray
  • d is the inter-atomic distance in the object. Therefore the amount of Bragg scattering gives information about the atomic structure of the object. However, it does not vary smoothly with atomic number.
  • a security scanning system comprises a multi-focus X-ray source 410 which is the same as that of FIG. 1, and a circular detector array 412 and conveyor 420 that are also the same as those of FIG. 1.
  • the system comprises a further cylindrical array of detectors 422 which also extends around the conveyor at the same radius as the circular detector array 412 but on the other side axially of the source 410.
  • the cylindrical detector array 422 is arranged to detect X-rays scattered in the object.
  • the scatter detector array 422 is made up of a number of circular arrays or rings 422a, 422b of detectors, and the detectors in each ring are equally spaced around the conveyor so that they are arranged in a number of straight rows extending in the axial direction of the scanner.
  • the detectors in the scatter detector array 422 are energy resolving detectors such that individual X-ray interactions with each detector produce a detector output that is indicative of the energy of the X-ray.
  • Such detectors can be fabricated from wide bandgap III-V or II-IV semiconductor materials such as GaAs, HgI, CdZnTe or CdTe, a narrow gap semiconductor such as Ge, or a composite scintillation detector such as NaI(Ti) with photomultiplier tube readout.
  • a collimator 428 is provided in front of the scattering detectors 422.
  • the collimator 428 provides a barrier that prevents X-rays from reaching each detector unless it comes from a particular receiving direction.
  • the receiving direction passes through the central longitudinal axis X-X of the scanner, as can be seen in FIG. 16.
  • the receiving direction is not perpendicular to the axis X-X, but is inclined at about 5. degree, to the plane of the detector rings 422a, 422b in the direction towards the source 410, as can be seen in FIG. 15.
  • X-rays incident on any one of the detectors of the array 422 must have been scattered from a respective small sub -volume within the thin imaged volume that lies both in the path of the X-ray beam and in the line of the receiving direction from the detector 422.
  • the axial position of the detector that detects it will be determined by the distance from the active X- ray source point at which the scattering occurred. Detectors nearest the source 410 in the axial direction will detect X-rays scattered furthest from the active X-ray source point.
  • X-rays scattered from the point x which is nearest the active X-ray source point 410a
  • the axial position of the detector which detects the scattered X-ray can be used to determine the position of the scattering along the X-ray beam direction.
  • X-rays from an active source point 410a that undergo coherent scattering will only be detected by the row of detectors 422a that is on the opposite side of the scanner axis to the active source point, and possibly one or more of the rows close to it on either side depending on how narrowly focused the collimator is. If X-rays are confined to a straight narrow "pencil" beam, then any X-rays that are scattered incoherently through larger angles will not be detected at all as they will be cut off by the collimator 428. An example of such an X-ray is shown by arrow "a" in FIG. 16.
  • X-rays directed further away from the scanner axis can undergo incoherent scattering and reach detectors to either side of the row 422a opposite the active source point. Examples of such X-rays are shown by the arrows b and c. It will be noted that, to reach any detector 422b, the scattering event must take place in the plane passing through the scanner axis and that detector 422b.
  • the position of the scattering event of a detected X-ray can be identified as being in the plane passing through the scanner axis and that detector. If the exact position of the scattering event is to be determined then other information is needed. For example if information regarding the position of objects within the imaging volume is available, for example from tomographic imaging data, then the scattering can be associated with the most likely object as will be described in more detail below.
  • the combination of the X-ray energy and the scatter angle can be used to determine the inter-atomic distance d of the material in which the scattering event took place.
  • the scatter angle can be assumed to be constant, and the energy used to distinguish between different materials.
  • the level of scattering from each volume of the scanning volume gives an indication of the density of the material in that volume.
  • the ratio of Compton to coherent scatter can also be determined and used as a further parameter to characterize the material of the imaged object.
  • each sub-volume Due to the short dwell time for each X-ray source point, the number of detected scattered X-rays for each source point will always be very low, typically less than five. In order to form a reasonable coherent scatter signal it is necessary to collect scatter data for all source points within a tomographic scan and then accumulate the results for each sub-volume of the imaging volume. For a scanner with 500 source points, and an average of one coherent diffraction scatter result per sub-volume per scan, then following accumulation of the set of data, each sub-volume will have 500 results associated with it, corresponding to 500 scattering events within that sub-volume.
  • a typical sub-volume occupies an area within the imaging plane of a few square centimeters, with a volume thickness of a few millimeters.
  • the data acquisition system arranged to accumulate data from the scatter detector array 422 of the scanner of FIGS. 14 to 16 comprises a multi-channel analyzer 500 associated with each of the detectors 422.
  • Each MCA 500 is arranged to receive the output signals from the detector, and allocate each X-ray detected to one of a number of X-ray energy ranges or channels, and output a signal indicative of the energy range in which the detected X-ray falls.
  • a multiplexer 502 is arranged to receive the outputs from each of the MCAs 500.
  • a look-up table 504 is also provided which has entries in it that, for a given source point and detector, identify the sub-volume within the imaging volume in which the X-ray was scattered.
  • the system further comprises an image memory 506 which includes a number of memory areas 508, each of which is associated with a respective sub-volume within the scanner imaging plane.
  • the look up table is loaded with coefficients prior to scanning that map each combination of detector 422 and MCA 500 to a respective image location 508, one look up table entry per X-ray source position.
  • Those pixels, i.e. detectors 422, that are in the forward direction, i.e. substantially in the direction that the photon is traveling from the source prior to any interaction, are assumed to record coherent scatter photons at small beam angles of about 4-6 degrees.
  • Those pixels 422 that are not in the forward direction are assumed to record incoherent scattered photons due to the Compton scattering effect.
  • the image memory 506 is actually "three dimensional" —two dimensions represent location in the image while the third dimension holds scattered energy spectra for both coherent (Io 8 -bits) and incoherent scattering (hi 8 bits).
  • the look up table 504 will also instruct the multiplexer 502 as to the type of data that is being collected for each MCA 500 at each projection so that the appropriate memory space is filled.
  • the data is transferred to and synchronized, by a projection sequencer 510, with the main RTT data acquisition system 512, which is described above with reference to FIG. 4.
  • the reconstructed image data and scatter data are passed through simultaneously to the threat detection system, which can use it to determine suitable parameters for analysis.
  • the tomographic image data from the transmission detectors 412 produces data relating to the X-ray attenuation for each pixel of the image, which in turn corresponds to a respective sub-volume of the tomographic imaging volume. This is obtained as described above with reference to FIG. 4.
  • the data from the scatter detectors 422 provides, as described above, data relating to the amount of coherent scattering within each sub- volume, and data relating to the amount of incoherent scattering within each sub-volume. This data can therefore be analyzed in a threat detection processor similar to that of FIG. 5.
  • the parameters of the data which are extracted can relate to the image data or the scatter data or combinations of two or more types of data.
  • parameters that are extracted from the data are the ratio of coherent to incoherent scatter, material types as determined from coherent scatter data, material density as determined from incoherent scatter data, correlation of CT image pixel values with scatter data. Also parameters for the scatter data corresponding to those described above for the transmission data can also be determined.
  • the transmission detectors 512 that are used to generate the tomographic image data are arranged to measure the X-ray transmission over different energy ranges. This is achieved by having two sets of detectors 512a, 512b, each forming a ring around the conveyor. The two sets are at different axial locations along the direction of travel of the conveyor, in this case being adjacent to each other in the axial direction.
  • the first set 512a has no filter in front of it, but the second set 512b has a metal filter 513 placed between it and the X-ray source 510.
  • the first set of detectors 512a therefore detects transmitted X-rays over a broad energy range, and the second set 512b detects X-rays only in a narrower part of that range at the high energy end.
  • each thin volume or slice of it can be scanned once using the first set of detectors 512a and then scanned again using the second set 512b.
  • the same source 510 is used to scan two adjacent volumes simultaneously, with data for each of them being collected by a respective one of the detector sets 512a, 512b.
  • two sets of image data can be formed using the two different X-ray energy ranges, each image including transmission (and hence attenuation) data for each pixel of the image.
  • the two sets of image data can be combined by subtracting that for the second detector set 512a from that of the first 512b, resulting in corresponding image data for the low energy X-ray component.
  • the X-ray transmission data for each individual energy range, and the difference between the data for two different ranges, such as the high energy and low energy, can be recorded for each pixel of the image.
  • the data can then be used to improve the accuracy of the CT images. It can also be used as a further parameter in the threat detection algorithm.
  • two sets of detectors are used that are sensitive to different energy X-rays.
  • stacked detectors comprising a thin front detector that is sensitive to low energy X-rays but allows higher energy X-rays to pass through it, and a thick back detector sensitive to the high energy X-rays that pass through the front detector.
  • the attenuation data for the different energy ranges can be used to provide energy specific image data.
  • two scans are taken of each slice of the object with two different X-ray beam energies, achieved by using different tube voltages in the X-ray source, for example 160 kV and 100 kV.
  • the different energies result in X-ray energy spectra that are shifted relative to each other. As the spectra are relatively flat over part of the energy range, the spectra will be similar over much of the range. However, part of the spectrum will change significantly. Therefore comparing images for the two tube voltages can be used to identify parts of the object where the attenuation changes significantly between the two images. This therefore identifies areas of the image that have high attenuation in the narrow part of the spectrum that changes between the images. This is therefore an alternative way of obtaining energy specific attenuation data for each of the sub -volumes within the scanned volume.
  • two different X-ray energy spectra are produced by providing an anode 600 in the X-ray tube that has target areas 602, 604 of two different materials.
  • the anode comprises a copper base 606 with one target area 602 of tungsten and one 604 of uranium.
  • the electron source 610 has a number of source points 612 that can be activated individually.
  • a pair of electrodes 612, 614 is provided on opposite sides of the path of the electron beam 616 which can be controlled to switch an electric field on and off to control the path of the electron beam so that it strikes either one or the other of the target areas 602, 604.
  • the energy spectrum of the X-rays produced at the anode will vary depending on which of the target areas is struck by the electron beam 616.
  • This embodiment uses an X-ray source similar to that of FIG. Ia, with the different target areas formed as parallel strips extending along the anode 27.
  • the source can be arranged to switch between the two target areas for each electron source point while it is active.
  • the scan along the anode 27 can be performed twice, once for one target material and once for the other. In either case further electron beam focusing wires may be needed to ensure that only one or the other of the target materials is irradiated by the electron beam at one time.
  • the beams from the two target areas 602, 604 can in some cases be arranged to pass though the same imaging volume and be detected by a common detector array. Alternatively they may be arranged to pass through adjacent slices of the imaging volume and detected by separate detector arrays. In this case the parts of the imaged item can be scanned twice as the item passes along the conveyor in a similar manner to the arrangement of FIG. 18.
  • two detector arrays are provided in a single scanner, adjacent to each other in the axial direction, one 710 corresponding to that of FIG. 1 and being arranged to form a RTT image, and the other, 712, being of a higher resolution, and being arranged to produce a high resolution projection image of the scanned object.
  • the high resolution detector array 712 comprises two parallel linear arrays 714, 716 each arranged to detect X-rays at a different energy, so that a dual energy projection image can be produced.
  • the high resolution array 812 comprises two stacked arrays, a thin array on top arranged to detect lower energy X-rays but transparent to higher energy X-rays, and a thicker array beneath arranged to detect higher energy X-rays. In both cases, the two detector arrays are arranged close enough together in the axial direction to be able to detect X-rays from a single linear array of source points.
  • each detector 718, 818 in the high resolution array is connected to an integrator 750.
  • the integrator comprises an amplifier 752 in parallel with a capacitor 754.
  • An input switch 756 is provided between the detector 718 and the amplifier 752, a reset switch 758 is provided across the input terminals of the amplifier, and a further reset switch 759 connected across the capacitor 754, and a multiplexing switch 760 is provided between the integrator and an analogue to digital converter ADC.
  • This provides an analogue signal to the ADC determined by the level of charge on the capacitor 754, and therefore indicative of the number of X-rays that have been detected by the detector 718 during the period for which it was connected to the integrator.
  • the ADC then converts this analogue signal to a digital signal for input to the data acquisition system.
  • all of the high resolution detectors are used to collect data at the same time, when one of the X-ray source points is active.
  • each detector 718 is connected to two integrators 750a, 750b in parallel, each of which is identical to that of FIG. 23.
  • the outputs from the two integrators are connected via their output switches 760a, 760b to an ADC.
  • each integrator to be arranged to integrate the signal from the detector 718 at a different point in the scan of the X-ray source, and therefore to collect data for a separate image, the two images being from different angles with different X-ray source points.
  • this can be used to produce projection images from orthogonal directions which can be used to build up a high resolution 3 -dimensional image, from which the position of features in the imaged package can be determined in three dimensions.
  • the high resolution image can be useful when combined with the RTT image, as it can help identify items for which higher resolution is needed, such as fine wires.
  • the threat detection processor which executes programmatic instructions that process input image data 62 and code the decision trees 64 and parameter extractions 63 shown in Figure 5, is tuned for the detection of liquids.
  • a three-dimensional segmentation calculation is used for performing liquid detection. Variations of segmentation calculations, commonly known to persons of ordinary skill in the art and based around a surface detector (for example using a dilate-erode algorithm in all three dimensions) and a volume filling algorithm that grows out from the detected surface, can be implemented.
  • a volume filling algorithm from a seed point which extends out until a surface is reached is used.
  • the surface is determined by analyzing each new pixel in terms of its probability of being part of the volume or otherwise.
  • a pixel value within, for example, one sigma of the mean value of the volume may be considered to be part of the volume while a pixel lying more than two sigma from the mean value may be considered to be not part of the volume.
  • Seed points may be determined by analyzing the image to find regions of similar pixel density and placing a seed point at the centroid location of each such region. In various embodiments, a suitable method is determined based on the statistical properties of the image, the type and severity of image reconstruction artefacts in the image and the intrinsic spatial resolution of the image.
  • a first parameter extraction block 63 is configured to process the input image data 62 to identify a) an object under inspection, b) the outer envelope defining the object under inspection, and c) any flat surfaces within the outer envelope of the object under inspection which lie parallel to the surface of the conveyor in the horizontal plane. Such parallel surfaces can be reliably used to uniquely identify the existence of a liquid contained within a vessel.
  • a second parameter extraction block is configured to locate a contiguous volume of material of uniform density that extends from each such parallel surface element in a vertical direction downwards towards the surface of the conveyor. This method uses a simplified version of the segmentation methods outlined above since the only direction of importance here is in the vertical plane rather than vertical and horizontal planes.
  • the information obtained by the parameter extraction blocks is passed from the parameter extraction blocks 63 to an associated decision tree 64, where the actual volume of the contiguous volume is calculated and assigned a fit to a predetermined shape matching one or more of oval bottle, rectangular bottle and triangular bottle.
  • the mean reconstructed intensity of the contiguous volume is calculated and this information (volume, shape, density) is passed to the database searcher 65.
  • a 1.5 liter bottle of sugar laden beverage will have a common exterior dimension of approximately 400mm (L) x 100mm (diameter) within a thin plastic container with a neck drawing in to a cap and a base with rounded or moulded features.
  • the density of such beverages is typically just greater that that of water and the volume of the liquid will be no greater than 1.5 liters within a container of volume just greater than 1.5 liters.
  • an alcoholic beverage is more often in a glass bottle with thicker wall thickness with smaller volume such as 750 ml (bottle of wine) or 330 ml (spirit). Alcohol based bottles will generally be full (i.e.
  • volume of liquid is close to volume of container) and will include characteristic features such as a cork or screw top. Density of the liquid will also be in well known bands. Known threat materials tend to have density just below that of water and so stand out quite clearly. In the case that a clear match is made against a benign material, then no further action is taken. In the case that a threat material is detected, an alarm shall be raised on the operator workstation and the threat object highlighted for further inspection. In one embodiment, density calculations are obtained and then compared, by the processor, against liquid densities stored in a database.
  • Using a high resolution three dimensional image allows a very accurate estimate of liquid volume to be made which helps substantially in separating a full vessel from a partially filled bottle. It is noted that virtually all liquids are sold in standard sized quantities (e.g. 1500 ml, 1000 ml, 750 ml, 500 ml, 330 ml, 250 ml) with a high level of filling accuracy. Thus, volume estimation can play a key role in detecting vessels which have been tampered with.
  • the high speed of transit of the object through the data acquisition system of the present invention causes the liquid to have "waves" on its surface.
  • the high speed of data acquisition enables reconstruction of the shape of these waves which are themselves characteristic of the viscosity of the fluid within the container. This additional information is then used to enable differentiation in detection of one liquid from another. It shall be evident to one skilled in the art that the architecture described here is capable of being implemented in a parallel architecture and that therefore multiple algorithms can be constructed, each looking for either the same type of feature or quite different types of feature (such as liquids and narcotics) at the same time.
  • the threat detection processor which executes programmatic instructions that process input image data 62 and code the decision trees 64 and parameter extractions 63 shown in Figure 5, is tuned for the detection of narcotics.
  • the threat detection processor is tuned by selecting algorithms used in the parameter extraction blocks and by weighting of the results from the parameter extraction blocks 63 as they propagate through the decision trees 64.
  • the tuning information is stored in the form of parameters which can be updated easily without having to re-program the underlying algorithms and methods.
  • a first parameter block 63 is tuned to process the input image data 62 to look for contiguous volumes of low density material in both sheet and bulk shapes.
  • volumes in the range 1 g/cm3 to 3 g/cm3 are identified.
  • a second parameter block processes these same volumes to determine basic statistical properties of the contiguous volumes to include mean value, standard deviation and skew.
  • a third parameter block searches for collections of randomly oriented parts of small volume which may be determined to be pharmaceuticals. Such randomly oriented particles include tablets in a jar or bag while a structured arrangement of tablets may be observed in pop -out packaging materials. All of this data (volume, shape and statistical properties) is passed to the decision tree which correlates the data from the multiple parameter blocks. The data is then passed to the database searcher 65 where the data is compared against values stored in a database of known threat materials.
  • raw narcotics such as heroin or cocaine in powder form
  • raw narcotics tend to be packaged in relatively ordered shapes with very thin, generally polythene, wrapping.
  • a package of volume 5 cm3 upwards to 100 cm3 with almost undetectable wrapping with density in the range Ig/cm3 to 3 g/cm3 is likely to be a suspect bulk narcotic material.
  • an alarm shall be raised on the operator workstation and the threat items highlighted for further inspection.
  • the threat detection processor which executes programmatic instructions that process input image data 62 and code the decision trees 64 and parameter extractions 63 shown in Figure 5, is tuned for the detection of currency . It shall be evident to a person skilled in the art that a parallel algorithm architecture may be implemented which allows simultaneous detection of currency and other potential threat items.
  • a first parameter block 63 is tuned to look for "bow-tie" shaped features in the object under inspection. Bundles of currency are typically bound towards their centre, such that the centre of a bundle is marginally thinner than the ends.
  • a second parameter block is tuned to search for rectangular shapes in multiples of the physical dimension of common denominations of currency.
  • a third parameter block is tuned to look for repeating patterns in the blocks found by the first two parameter blocks. These patterns are characteristic of the stacks of individual bundles that tend to be used to make up a collection of currency.
  • a fourth parameter block is tuned to generate statistical properties of the bundles identified by the other three parameter blocks.
  • a stack of US dollar bills will have a well defined area (approx 150 mm x 65 mm) with a thickness which is dependent on the amount of currency involved. The thicker the stack, the less the bow-tie effect.
  • each type of currency has certain security features embedded into it (such as a metal strip) or a variation in material type or printing density. This arrangement of subtle information is amplified when many notes are stacked together and can be registered by a repeating block algorithm.
  • This collection of information is passed to the decision tree 64 which correlates the information from the currency specific parameter blocks with the data from all the other parameter blocks and the complete setoff data is passed to the database searcher 65. When a clear match is found between an item in the cargo under inspection and a know type of currency, an alarm is raised on the operator workstation and the relevant currency is highlighted for further inspection.
  • the threat detection processor which executes programmatic instructions that process input image data 62 and code the decision trees 64 and parameter extractions 63 shown in Figure 5, is tuned for the detection of tobacco, and cigarettes in particular.
  • Cigarettes are characterized by a very repeatable structure which is first determined by the length and diameter of individual cigarettes. Secondly, groups of 10 and/or 20 cigarettes are packed together into a box of known and repeatable size. Thirdly, groups of boxes of cigarettes are packed into arrays of dozens of boxes which occupy a significant volume. Therefore, a first parameter block 63 is tuned to look for repeating array structures with a length and width dimension consistent with known dimensions of cigarettes.
  • a second parameter block is tuned to detect rectangular volumes of predefined aspect ratio matching that of common cigarette packaging with a density that is consistent with typical brands of cigarettes.
  • the results from these two parameter blocks are passed through to a decision tree which correlates the data with other known data from the object under inspection.
  • an exemplary brand of cigarette has an overall length of 90mm with a filter length of 15mm and a diameter of 8mm
  • a pack of 20 cigarettes will therefore have an external dimension of 92 mm (L) x 82 mm (W) x 22m (D).
  • a stack of 48 packets arranged in a 6 x 8 configuration will have an overall dimension of 132 mm (H) x 656mm (W) x 92mm (D).
  • Within thus structure will be a set of rectangular planes which are established by the higher density cardboard boxes and it is all of this information which is picked up by the automated detection algorithm
  • This information is then passed to a database search which compares the information with known tobacco products. When a clear match is found between an item in the cargo under inspection and a known type of tobacco, and alarm is raised on the operator workstation and the relevant tobacco product is highlighted for inspection. As would be apparent to a person skilled in the art, similar algorithms may be implemented for the detection of cigars and related products.
  • the threat detection processor which executes programmatic instructions that process input image data 62 and code the decision trees 64 and parameter extractions 63 shown in Figure 5, is tuned for the detection of special nuclear materials, such as Uranium and Plutonium, and for other shielded radioactive sources such as cobalt-60 and cesium- 137.
  • a first parameter extractor 63 is tuned to look for highly attenuating regions in the image where the reconstructed pixel intensity is above a threshold value. This data is passed to the decision tree which seeks to determine whether the attenuating object is part of a larger structure (such as metallic rails which are commonly used to support the framework of a piece of baggage).
  • the decision tree will evaluate shape, location (for example, is the attenuation object at the surface of the cargo item in which case it may be a fastener) and size prior to passing the new data on to the database search tool. If a match is found between the dark object and a set of characteristics that are common to radioactive sources (typically no re-entrant volume of lower intensity), then an alarm is raised on the operator workstation and the relevant nuclear material is highlighted for inspection.
  • high density shielding typically has a density of 11 g/cm3 (lead) and above.
  • Standard engineering materials e.g. steel
  • This specific range enables the detection of radioactive source shielding relatively easily.
  • the threat detection processor which executes programmatic instructions that process input image data 62 and code the decision trees 64 and parameter extractions 63 shown in Figure 5, is tuned for the detection of sharp objects and knives.
  • Sharp objects are characterized by the presence of a sharp point which is typically distinguished from points that might occur at the non-threatening corners of rectangular objects such as books and electrical goods.
  • Knives are further characterised by the presence of a blade shape which is typically long in one dimension compared to another. In many cases, blades are folded into a supporting housing, and, in this case, the blade is characterized by a repeating structure of high density (case or blade) and low density (surrounding air).
  • a first parameter extractor is tuned for the detection of protruding points and not simply corners.
  • a second parameter extractor is tuned for the detection of blades in which there is a significant length to width aspect ratio.
  • a third parameter extractor is tuned for the detection of folded blades in which there is a repeating structure of at least two air gaps and three material fills.
  • a three inch blade generally has an aspect ratio (length:width) of at least
  • the width of a blade is generally in proportion to its length, with a width to length ratio on the order of 1 :60, and typically less than 1 :100 but more than 1 :20.
  • Data from these parameter extractors is passed to a decision tree which correlates the data with associated information such as the presence of a handle on a knife.
  • the filtered information is then passed to a database searcher which analyses the data against known threat items. If a reasonable match is determined, an alarm is raised on the operator workstation and the relevant point or blade is highlighted for further inspection.
  • the threat detection processor which executes programmatic instructions that process input image data 62 and code the decision trees 64 and parameter extractions 63 shown in Figure 5, is tuned for the detection of firearms, including gun barrels, steel tubes of a certain diameter (or range of diameters), rifles; and ammunition (a high density slug such as lead, copper, tungsten) next to and connected to a brass cylinder loaded with powder of a known range of densities.
  • firearms including gun barrels, steel tubes of a certain diameter (or range of diameters), rifles; and ammunition (a high density slug such as lead, copper, tungsten) next to and connected to a brass cylinder loaded with powder of a known range of densities.
  • Fire-arms are characterized by the presence of metallic tubes of a certain diameter. Accordingly, a first parameter extractor is tuned for the detection of cylindrical metal tubes. A second parameter extractor is tuned for detection of a trigger mechanism and firing pin. A third parameter extractor is tuned for detection of high density slugs and bullets with composition ranging from aluminium (density 2.7 g/cm3) to lead (density > 11 g/cm3). Generally, bullets will have a lower density core filled with gun powder (density typically 1 g/cm3).
  • Data from these parameter extractors is passed to a decision tree which correlates the data with associated information such as the presence of a gun barrel.
  • the filtered information is then passed to a database searcher which analyses the data against known threat items. If a reasonable match is determined, an alarm is raised on the operator workstation and the relevant cylinder is highlighted for further inspection.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Geophysics (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Radiology & Medical Imaging (AREA)
  • Analytical Chemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Biochemistry (AREA)
  • Pulmonology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Multimedia (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

L'invention concerne un système de balayage aux rayons X avec un scanner à rayons X non rotatif qui génère des données de balayage définissant une image tomographique aux rayons X de l'objet, et un processeur exécutant des instructions de programme. Le processeur d'exécution analyse les données de balayage pour extraire au moins un paramètre de l'image tomographique aux rayons X, ce processeur étant configuré pour déterminer si l'objet contient un liquide, un objet pointu, des substances narcotiques, des devises, des matières nucléaires, des cigarettes ou des armes à feu.
EP10781129.1A 2009-05-26 2010-05-26 Systèmes d'inspection tomographique aux rayons x pour l'identification d'articles cibles spécifiques Ceased EP2435955A4 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP17187848.1A EP3267361A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques
EP17187847.3A EP3282393A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US18106809P 2009-05-26 2009-05-26
PCT/US2010/036183 WO2010138574A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique aux rayons x pour l'identification d'articles cibles spécifiques

Related Child Applications (2)

Application Number Title Priority Date Filing Date
EP17187847.3A Division EP3282393A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques
EP17187848.1A Division EP3267361A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques

Publications (2)

Publication Number Publication Date
EP2435955A1 true EP2435955A1 (fr) 2012-04-04
EP2435955A4 EP2435955A4 (fr) 2016-12-21

Family

ID=43223043

Family Applications (3)

Application Number Title Priority Date Filing Date
EP17187847.3A Withdrawn EP3282393A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques
EP17187848.1A Ceased EP3267361A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques
EP10781129.1A Ceased EP2435955A4 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique aux rayons x pour l'identification d'articles cibles spécifiques

Family Applications Before (2)

Application Number Title Priority Date Filing Date
EP17187847.3A Withdrawn EP3282393A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques
EP17187848.1A Ceased EP3267361A1 (fr) 2009-05-26 2010-05-26 Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques

Country Status (6)

Country Link
EP (3) EP3282393A1 (fr)
JP (2) JP2012528397A (fr)
CN (1) CN102483803B (fr)
GB (7) GB2501024B (fr)
HK (1) HK1248873A1 (fr)
WO (1) WO2010138574A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10901112B2 (en) 2003-04-25 2021-01-26 Rapiscan Systems, Inc. X-ray scanning system with stationary x-ray sources
US10976271B2 (en) 2005-12-16 2021-04-13 Rapiscan Systems, Inc. Stationary tomographic X-ray imaging systems for automatically sorting objects based on generated tomographic images

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103308535B (zh) 2012-03-09 2016-04-13 同方威视技术股份有限公司 用于射线扫描成像的设备和方法
CN104636707B (zh) * 2013-11-07 2018-03-23 同方威视技术股份有限公司 自动检测香烟的方法
US9542525B2 (en) * 2014-06-24 2017-01-10 Siemens Product Lifecycle Management Software Inc. Additive smoothing of sharp concave edges on designed 3D printable polygonal mesh models
CA2962618C (fr) 2014-09-26 2024-05-28 Battelle Memorial Institute Article de test de qualite d'image
US10656304B2 (en) * 2015-09-10 2020-05-19 American Science And Engineering, Inc. Backscatter characterization using interlinearly adaptive electromagnetic X-ray scanning
CN106932414A (zh) * 2015-12-29 2017-07-07 同方威视技术股份有限公司 检验检疫用检查系统及其方法
GB2595986A (en) * 2016-02-22 2021-12-15 Rapiscan Systems Inc Systems and methods for detecting threats and contraband in cargo
JP6764709B2 (ja) * 2016-06-30 2020-10-07 株式会社日立製作所 Ｘ線自動判定装置、ｘ線自動判定方法
EP3309589B1 (fr) * 2016-10-13 2021-01-20 VEGA Grieshaber KG Appareil de mesure radiométrique pour fonctionnement à deux conducteurs avec une conduite de dérivation
EP3309590B1 (fr) * 2016-10-13 2021-02-24 VEGA Grieshaber KG Appareil de mesure radiométrique pour la collecte d'un spectre d'amplitudes en fonctionnement à deux fils
KR101884575B1 (ko) * 2017-12-06 2018-08-02 한국공항공사 보안 검색 시스템
CA3092996C (fr) 2018-03-14 2024-05-28 Monsanto Technology Llc Systemes et methodes pour l'imagerie de graines au moyen de plusieurs cameras
MX2020013129A (es) * 2018-06-11 2021-02-18 Monsanto Technology Llc Seleccion de semillas.
CN110941973B (zh) * 2018-09-21 2023-09-15 驭势科技（北京）有限公司 一种障碍物检测方法、装置、车载设备及存储介质
KR101990123B1 (ko) * 2018-10-31 2019-06-18 (주)제이엘케이인스펙션 영상 분석 장치 및 방법
JP7177721B2 (ja) * 2019-02-15 2022-11-24 日本信号株式会社 検査システム
JP7199996B2 (ja) * 2019-02-19 2023-01-06 アンリツ株式会社 Ｘ線検査装置
JP7200001B2 (ja) * 2019-02-22 2023-01-06 アンリツ株式会社 Ｘ線検査装置
DE102019111463A1 (de) * 2019-05-03 2020-11-05 Wipotec Gmbh Röntgenstrahlungsdetektorvorrichtung und Vorrichtung zur Röntgeninspektion von Produkten, insbesondere von Lebensmitteln
JP7249300B2 (ja) * 2020-02-21 2023-03-30 株式会社日立製作所 アラート出力装置、アラート出力方法、及びアラート出力プログラム

Family Cites Families (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2840611B2 (ja) * 1989-05-24 1998-12-24 明治乳業株式会社 包装体内容物の非破壊検査方法及び検査装置
US5367552A (en) * 1991-10-03 1994-11-22 In Vision Technologies, Inc. Automatic concealed object detection system having a pre-scan stage
US5600700A (en) * 1995-09-25 1997-02-04 Vivid Technologies, Inc. Detecting explosives or other contraband by employing transmitted and scattered X-rays
WO1997018462A1 (fr) * 1995-11-13 1997-05-22 The United States Of America As Represented By The Dispositif et procede pour la reconnaissance automatique d'objets caches par tomographie assistee par ordinateur a energies multiples
US6317509B1 (en) * 1998-02-11 2001-11-13 Analogic Corporation Computed tomography apparatus and method for classifying objects
US6347132B1 (en) * 1998-05-26 2002-02-12 Annistech, Inc. High energy X-ray inspection system for detecting nuclear weapons materials
US6195444B1 (en) * 1999-01-12 2001-02-27 Analogic Corporation Apparatus and method for detecting concealed objects in computed tomography data
JP2002257751A (ja) * 2001-03-01 2002-09-11 Kawasaki Heavy Ind Ltd 手荷物検査方法および手荷物検査システム
US6661867B2 (en) * 2001-10-19 2003-12-09 Control Screening, Llc Tomographic scanning X-ray inspection system using transmitted and compton scattered radiation
CA2365045A1 (fr) * 2001-12-14 2003-06-14 Cedara Software Corp. Methode de detection d'armes a feu et de munitions par balayage a rayons x de contenants a des fins de securite
US7103137B2 (en) * 2002-07-24 2006-09-05 Varian Medical Systems Technology, Inc. Radiation scanning of objects for contraband
GB0309379D0 (en) 2003-04-25 2003-06-04 Cxr Ltd X-ray scanning
GB0525593D0 (en) * 2005-12-16 2006-01-25 Cxr Ltd X-ray tomography inspection systems
GB0309383D0 (en) 2003-04-25 2003-06-04 Cxr Ltd X-ray tube electron sources
GB0309374D0 (en) 2003-04-25 2003-06-04 Cxr Ltd X-ray sources
GB0309387D0 (en) 2003-04-25 2003-06-04 Cxr Ltd X-Ray scanning
GB0309371D0 (en) 2003-04-25 2003-06-04 Cxr Ltd X-Ray tubes
GB0309385D0 (en) 2003-04-25 2003-06-04 Cxr Ltd X-ray monitoring
US7856081B2 (en) * 2003-09-15 2010-12-21 Rapiscan Systems, Inc. Methods and systems for rapid detection of concealed objects using fluorescence
IL159828A0 (en) * 2004-01-12 2005-11-20 Elbit Systems Ltd System and method for identifying a threat associated person among a crowd
EP1719000A2 (fr) * 2004-02-11 2006-11-08 Reveal Imaging Technologies, Inc. Systemes et procedes pour la detection de contrebande
US7277577B2 (en) * 2004-04-26 2007-10-02 Analogic Corporation Method and system for detecting threat objects using computed tomography images
US7356174B2 (en) * 2004-05-07 2008-04-08 General Electric Company Contraband detection system and method using variance data
US7302083B2 (en) * 2004-07-01 2007-11-27 Analogic Corporation Method of and system for sharp object detection using computed tomography images
US7231017B2 (en) * 2005-07-27 2007-06-12 Physical Optics Corporation Lobster eye X-ray imaging system and method of fabrication thereof
US20070189597A1 (en) * 2005-08-23 2007-08-16 Limer Daniel J Machine vision counting system apparatus and method
US7548606B2 (en) * 2006-08-31 2009-06-16 Ge Homeland Protection, Inc. System and method for integrating explosive detection systems
CA2676913C (fr) * 2006-09-18 2010-11-30 Optosecurity Inc. Procede et appareil d'evaluation de caracteristiques de liquides
US7539283B2 (en) * 2007-01-17 2009-05-26 Ge Homeland Protection, Inc. Combined computed tomography and nuclear resonance fluorescence cargo inspection system and method
GB0716045D0 (en) * 2007-08-17 2007-09-26 Durham Scient Crystals Ltd Method and apparatus for inspection of materials
WO2009006044A2 (fr) * 2007-06-21 2009-01-08 Rapiscan Security Products, Inc. Systèmes et procédés pour améliorer le criblage dirigé de gens
US7869566B2 (en) * 2007-06-29 2011-01-11 Morpho Detection, Inc. Integrated multi-sensor systems for and methods of explosives detection
US8331675B2 (en) * 2007-08-02 2012-12-11 L-3 Communications Security And Detection Systems, Inc. Reducing latency in a detection system
JP4853444B2 (ja) * 2007-09-28 2012-01-11 株式会社デンソー 移動物体検出装置
CN101403711B (zh) * 2007-10-05 2013-06-19 清华大学 液态物品检查方法和设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See references of WO2010138574A1 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10901112B2 (en) 2003-04-25 2021-01-26 Rapiscan Systems, Inc. X-ray scanning system with stationary x-ray sources
US11796711B2 (en) 2003-04-25 2023-10-24 Rapiscan Systems, Inc. Modular CT scanning system
US10976271B2 (en) 2005-12-16 2021-04-13 Rapiscan Systems, Inc. Stationary tomographic X-ray imaging systems for automatically sorting objects based on generated tomographic images

Also Published As

Publication number Publication date
CN102483803B (zh) 2015-12-16
JP2012528397A (ja) 2012-11-12
GB2482819B (en) 2014-02-12
HK1248873A1 (zh) 2018-10-19
GB201312004D0 (en) 2013-08-21
EP3282393A1 (fr) 2018-02-14
GB2501024A (en) 2013-10-09
GB2501022B (en) 2014-02-12
JP2015099147A (ja) 2015-05-28
GB2482819A (en) 2012-02-15
GB2501022A (en) 2013-10-09
GB201120242D0 (en) 2012-01-04
GB2501023A (en) 2013-10-09
CN102483803A (zh) 2012-05-30
GB2501024B (en) 2014-02-12
GB201312007D0 (en) 2013-08-21
WO2010138574A1 (fr) 2010-12-02
EP3267361A1 (fr) 2018-01-10
GB2501026B (en) 2014-02-12
GB201312003D0 (en) 2013-08-21
GB201312005D0 (en) 2013-08-21
GB201312008D0 (en) 2013-08-21
GB201312006D0 (en) 2013-08-21
EP2435955A4 (fr) 2016-12-21
JP5968403B2 (ja) 2016-08-10
GB2501023B (en) 2014-02-12
GB2501025B (en) 2014-02-12
GB2503358B (en) 2014-02-12
GB2501025A (en) 2013-10-09
GB2503358A (en) 2013-12-25
GB2501026A (en) 2013-10-09

Similar Documents

Publication Publication Date Title
US20200200690A1 (en) X-Ray Tomographic Inspection Systems for the Identification of Specific Target Items
EP3282393A1 (fr) Systèmes d'inspection tomographique à rayons x pour le identification d'articles cibles spécifiques
US10976271B2 (en) Stationary tomographic X-ray imaging systems for automatically sorting objects based on generated tomographic images
US8885794B2 (en) X-ray tomographic inspection system for the identification of specific target items
EP3686901A1 (fr) Procédé d'inspection tomographique par rayons x
CN102269826B (zh) X射线断层摄影检查系统

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20111123

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK SM TR

DAX Request for extension of the european patent (deleted)
RA4 Supplementary search report drawn up and despatched (corrected)

Effective date: 20161123

RIC1 Information provided on ipc code assigned before grant

Ipc: G06K 9/00 20060101AFI20161117BHEP

Ipc: G01V 5/00 20060101ALI20161117BHEP

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20171114

REG Reference to a national code

Ref country code: DE

Ref legal event code: R003

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN REFUSED

18R Application refused

Effective date: 20181213