WO2009043232A1 - Method and device for detecting liquid article - Google Patents

Method and device for detecting liquid article Download PDF

Info

Publication number
WO2009043232A1
WO2009043232A1 PCT/CN2008/001654 CN2008001654W WO2009043232A1 WO 2009043232 A1 WO2009043232 A1 WO 2009043232A1 CN 2008001654 W CN2008001654 W CN 2008001654W WO 2009043232 A1 WO2009043232 A1 WO 2009043232A1
Authority
WO
WIPO (PCT)
Prior art keywords
liquid
image
inspected
energy
scan
Prior art date
Application number
PCT/CN2008/001654
Other languages
English (en)
French (fr)
Inventor
Li Zhang
Zhiqiang Chen
Xuewu Wang
Haifeng Hu
Hongxin Wu
Yuanjing Li
Yinong Liu
Ziran Zhao
Yuxiang Xing
Hu Tang
Yumin Yi
Jinyu Zhang
Original Assignee
Tsinghua University
Nuctech Company Limited
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 Tsinghua University, Nuctech Company Limited filed Critical Tsinghua University
Priority to EP08800646.5A priority Critical patent/EP2189785B1/en
Priority to ES08800646T priority patent/ES2531019T3/es
Publication of WO2009043232A1 publication Critical patent/WO2009043232A1/zh

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N9/00Investigating density or specific gravity of materials; Analysing materials by determining density or specific gravity
    • G01N9/24Investigating density or specific gravity of materials; Analysing materials by determining density or specific gravity by observing the transmission of wave or particle radiation through the material
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00Investigating 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
    • G01N23/02Investigating 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
    • G01N23/04Investigating 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
    • G01N23/046Investigating 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]
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00Investigating 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
    • G01N23/02Investigating 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
    • G01N23/06Investigating 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 measuring the absorption
    • G01N23/10Investigating 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 measuring the absorption the material being confined in a container, e.g. in a luggage X-ray scanners
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01VGEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V5/00Prospecting or detecting by the use of nuclear radiation, e.g. of natural or induced radioactivity
    • G01V5/20

Definitions

  • the invention relates to the technical field of radiation inspection, in particular to a method and a device for performing rapid and safe inspection of liquid articles by dual-energy CT imaging of liquid articles.
  • the electromagnetic method adopts an active measurement method, which distinguishes liquid substances according to the dielectric constants of electromagnetic waves of different liquid substances.
  • the electromagnetic method itself is susceptible to the adverse effects of metal packaging and thicker material packaging. Therefore, in the actual situation where the packaging materials are complicated, the electromagnetic method has certain limitations.
  • the X-ray two-dimensional imaging technique is to integrate the three-dimensional information of the object to be inspected in the direction of the X-ray beam to obtain a two-dimensional information projection image of the object. These images represent objects in grayscale or pseudo-color representations.
  • the internal structure provides an intuitive display to the equipment operator.
  • X-ray two-dimensional imaging technology lacks information on one dimension of the object to be inspected, so the inspection of liquid objects by this technique is particularly affected by the shape and size of the liquid object being inspected.
  • Three-dimensional CT scanning imaging technology is the promotion and application of CT technology.
  • CT technology was first used in medicine and is a tool to assist doctors in their diagnosis. It is achieved by multi-angle projection of each fault of the object under inspection.
  • the reconstruction of the multi-angle projection data of each of the above-mentioned faults is performed by a computer, and the reconstructed image of each fault is calculated, and different attenuation coefficient information in the reconstructed image is displayed in different gradations, thereby displaying the internal structure of the object to be inspected.
  • industrial CT for non-destructive testing and baggage CT for safety inspection have emerged. These CT expansion techniques are still aimed at obtaining tomographic images showing the internal structure of objects, relying on artificial pairs of structures. Perform an analysis to determine whether the object is a dangerous item, not for automatic, quick inspection.
  • the CT type safety inspection equipment mentioned above has a wide range of inspection objects, which results in high cost, heavy equipment, and difficulty in promotion and application.
  • a method for inspecting a liquid article by dual energy CT comprising the steps of: performing a dual energy CT scan on the liquid object to be inspected to obtain dual energy CT projection data; CT reconstruction, obtaining a CT image indicating a physical property value of the liquid object to be inspected; extracting a physical property value of the liquid object to be inspected according to the CT image; and determining whether the liquid object to be inspected is a dangerous product based on the physical property value .
  • the physical property value comprises a density and an atomic number of the liquid item to be inspected.
  • the dual energy CT scan is performed by a planar tomography CT scan.
  • the dual energy CT scan is performed in a conventional spiral CT scan.
  • the dual energy CT scan uses a large pitch spiral CT scan.
  • a set of scanning positions is preset before performing a planar tomographic CT scan.
  • a DR scan is performed to obtain a transmission image of the object to be inspected, and then the CT scan position is determined based on the transmission image.
  • the operator specifies at least one of the transmission images as the CT scanning position by the input means.
  • At least one of the transmission images is automatically determined by the image processing technique as the CT scanning position.
  • the step of forming a transmission image comprises: emitting a chirped ray and a low energy ray from the ray source, penetrating the object to be inspected, forming a high-energy transmission image and a low-energy transmission image; merging the high-energy transmission image and the low-energy transmission image to form The transmission image.
  • the step of forming a transmission image comprises: emitting high energy rays and low energy rays from the radiation source, penetrating the object to be inspected, forming a high energy transmission image and a low energy transmission image; selecting one of a high energy transmission image and a low energy transmission image, As the transmission image.
  • the step of reconstructing a CT image representing physical property values of the liquid object to be inspected from the dual energy projection data comprises: generating projection data on the two base material coefficients according to the high and low energy projection data; Projection data of the base material coefficients are reconstructed to obtain CT images representing the two base material coefficients corresponding to the liquid to be tested; and generating the physical property values representing the liquid objects to be inspected according to the CT image representing the base material coefficients CT image.
  • the step of obtaining a physical property value of the liquid property to be inspected according to the CT image representing the physical property value of the liquid property to be inspected comprises: extracting a pixel corresponding to the liquid portion from the CT image; Part of the pixel calculates the density mean and the atomic number mean as the density and atomic number of the liquid object to be inspected.
  • the step of determining whether the liquid object to be inspected is a dangerous product based on the physical property value comprises: determining whether the point determined by the density and the atomic number falls within a density-atomic number two-dimensional In a predetermined area in the coordinate space; in the case where the point falls in the predetermined area, the liquid object to be inspected is considered to be a dangerous product.
  • the CT image of the liquid object being inspected is rotated to align with the image formed by the first dual energy CT scan.
  • the liquid object to be inspected is rotated to be the same as the position before the scan.
  • the liquid item is placed in a bucket that is divided into a plurality of spaces.
  • the method further comprises the steps of: automatically detecting the presence of the bucket using a predetermined template; detecting a specific marker in the CT image in the presence of the bucket; rotating the bucket based on the specific marker Go to the scheduled location.
  • the method further comprises the step of: displaying the judgment result of the liquid item to be inspected on the display screen.
  • the method further comprises the step of: printing the judgment result of each of the liquid objects to be inspected.
  • the method further comprises the step of: colorizing the CT image of each of the liquid objects to be inspected.
  • the dual energy CT scan is based on a predetermined location.
  • an apparatus for inspecting a liquid item using a dual energy CT comprising: a source of radiation for emitting radiation; a detection and acquisition device for detecting and collecting at least one piece of penetration a ray signal of the liquid object to be inspected; a controller that controls the ray source and the detecting and collecting device to perform dual-energy CT scanning on the liquid object to be inspected to obtain projection data; and reconstruct at least one physics representing the liquid object to be inspected from the projection data Means for the CT image of the attribute value; and means for determining whether the liquid object to be inspected is a dangerous product based on the position of the physical attribute value in the space of the corresponding dimension.
  • the dual energy CT scan is based on a predetermined location.
  • the detecting and collecting device detects and collects a ray signal that penetrates at least one of the liquid objects to be inspected to form a transmission image; wherein the device further comprises means for specifying at least one of the transmission images;
  • the dual energy CT scan is based on the specified line.
  • the physical attribute value includes at least a density and a prono number of the liquid item to be inspected.
  • a x-ray ray and a low-energy ray are emitted from a ray source, and the object to be inspected is formed to form a high-energy transmission image and a low-energy transmission image
  • the apparatus further comprising: merging the high-energy transmission image and the low-energy transmission image to form the A device for transmitting images.
  • high energy rays and low energy rays are emitted from the radiation source, penetrating the object to be inspected to form a high energy transmission image and a low energy transmission image, the apparatus further comprising: selecting a high energy transmission image and low energy One of the transmission images as a means of transmitting the image.
  • the means for specifying at least one of the transmission images comprises: means for the operator to select at least one of the transmission images using the input means.
  • the means for specifying at least one of the transmission images comprises: means for analyzing pixel values of the transmission image to layer the transmission image; designating an intermediate line in each layer as A device that performs a line of dual energy CT scans.
  • the apparatus for reconstructing a CT image representing a physical attribute value of a liquid object to be inspected from projection data includes: fusing a density image identified by a density of the liquid object to be inspected and an atomic number of the liquid substance to be inspected An atomic number image of the identification as a device of the CT image; extracting a pixel corresponding to the liquid portion from the CT image; calculating a density mean and an atomic number mean for the pixel of the liquid portion as a density of the liquid object to be inspected A device of atomic number.
  • the means for determining whether the liquid object to be inspected is a dangerous product based on the physical property value comprises: determining whether the point determined by the density and the atomic number falls in a density-atomic number two-dimensional A device in a predetermined area in the coordinate space, wherein, in the case where the point falls in the predetermined area, the liquid object to be inspected is considered to be a dangerous article.
  • the apparatus further comprises: after performing a dual energy CT scan for each row, rotating the CT image of the liquid object being inspected to align with the image formed by the first dual energy CT scan Device.
  • the apparatus further comprises means for rotating the liquid item being inspected to the same position as before scanning after performing a dual energy CT scan for each row.
  • the apparatus further comprises a tub divided into a plurality of spaces for respectively placing the liquid items.
  • the apparatus further includes: means for automatically detecting the presence of the bucket using a predetermined template; means for detecting a specific marker in the CT image in the presence of the bucket; based on the specific marker, A device that rotates the bucket to a predetermined position.
  • the apparatus further includes: a display device that displays a determination result of the liquid item to be inspected.
  • the apparatus further includes: means for printing out the judgment result of each of the liquid objects to be inspected.
  • the apparatus further comprises: a CT image of each liquid object to be inspected Colorized device.
  • the apparatus further includes a carrying mechanism carrying the liquid object to be inspected, the carrying structure carrying a plurality of areas identifiable by an operator on a surface of the liquid object to be inspected.
  • an apparatus for inspecting a liquid article by dual energy CT comprising: a radiation source for emitting radiation; and a detecting and collecting device for detecting and collecting at least one penetration a ray signal of the object to be inspected; a controller that controls the ray source and the detecting and collecting device to perform a helical CT scan on the liquid object to be inspected to form one of each of the physical property values representing the liquid object to be inspected a set of spiral CT images; means for analyzing the set of spiral CT images to determine a portion of the spiral CT image of the liquid; and determining the detected position based on the position of the physical attribute value contained in the liquid CT-based portion of the liquid in the space of the corresponding dimension Whether the liquid item is a dangerous goods device.
  • the detection speed is improved without deteriorating the detection accuracy, and it is also possible to judge whether or not the liquid article has a sandwich by the transmission image.
  • the liquid object to be inspected is an explosive.
  • the operator can perform a dual-energy CT scan at any given location, facilitating the review operation.
  • Figure 2 is a block diagram showing the structure of a computer data processor 60 as shown in Figure 1;
  • FIG. 3 is a block diagram showing the structure of a controller according to a first embodiment of the present invention.
  • FIGS. 4A and 4B are schematic views showing a method of inspecting a liquid article according to the present invention.
  • Figure 5 is a schematic view for explaining the relationship between DR imaging and CT imaging
  • Figure 6 shows an example of the DR imaging result
  • Figure 7 shows another example of the DR imaging result
  • Figure 8 is a general flow chart showing a liquid article inspection method according to a first embodiment of the present invention
  • Figure 9 is a flow chart showing a DR imaging process
  • Figure 10 illustrates the arrangement of DR image data acquired by the detection and acquisition device 30 during DR imaging
  • Figure 11 is a flow chart showing processing of a DR image to determine a CT scan position
  • Figure 12 shows the CT imaging process
  • Figure 13 shows the arrangement of CT projection data during CT imaging
  • Figure 14 shows the process of measuring the properties of a liquid
  • 15A and 15B are views showing a CT image reconstructed in the case of detecting a plurality of liquid articles according to a second embodiment of the present invention.
  • Figures 16A through 16K illustrate the process of how the CT reconstructed image and/or the carrier mechanism is rotated to coincide with the relative position of the object prior to the CT scan after CT imaging is completed;
  • Figure 17 is a flow chart showing an inspection performed in the case of a plurality of objects to be inspected.
  • Figure 18 shows a plan view of a carrier mechanism in accordance with a second embodiment of the present invention.
  • Figure 19 shows a side view of a compartmental bucket in accordance with the present embodiment
  • Figure 20 shows a top view of the compartmental bucket
  • Figure 21 shows a bottom view of the compartmental bucket
  • Figure 22 shows how the automatic detection of the compartments and markers during the inspection process
  • 23A to 23D show schematic views of the rotating tub during the detecting process
  • Figure 24 shows a flow chart of an inspection process according to a third embodiment
  • Figure 25 is a schematic view showing a spiral CT scan of a liquid article
  • 26A to 26M are images showing a spiral CT scan of a liquid article. detailed description
  • First Embodiment 1 is a schematic structural view of an inspection apparatus according to an embodiment of the present invention.
  • the inspection apparatus includes: a radiation source 10 that emits dual-energy X-rays for inspection, such as an X-ray machine; and a carrier mechanism 40 that carries the liquid substance to be inspected for rotation about the axis Z, and Lifting, the liquid object to be inspected enters the detection area, so that the radiation emitted by the radiation source 10 can pass through the liquid object to be inspected; the detecting and collecting device 30, which is a detector and data collector having an integral module structure for detecting Transmitting the dual-energy rays of the liquid object to be inspected, obtaining an analog signal, and converting the analog signal into a digital signal, thereby outputting scanned data of the liquid object for high-energy X-rays and low-energy X-rays; and controller 50 for controlling the entire system
  • the various parts operate synchronously; and a computer data processor 60 that processes the data collected by the data collector and outputs the inspection results.
  • the source 10 is placed on the side of the carrier 40 on which the liquid item to be inspected can be placed, and the detection and collection device 30 is placed on the other side of the carrier 40, including the detector and data collector, for Obtain the liquid item DR data and multi-angle projection data.
  • the data collector includes a data amplification forming circuit that operates in either (current) integration mode or pulse (count) mode.
  • the data output cable of the detection and acquisition device 30 is coupled to computer data processor 60, which stores the acquired data in computer data processor 60 in accordance with a trigger command.
  • the inspection apparatus further includes a cylindrical body passage 20 made of metal which is disposed on the carrier mechanism 40 to shield the X-ray radiation from the outside.
  • the liquid object to be inspected is placed in the channel of the object to be inspected.
  • Fig. 2 is a block diagram showing the structure of a computer data processor 60 as shown in Fig. 1.
  • the data collected by the data collector is stored in the memory 61 via the interface unit 68 and the bus 64.
  • the read-only memory (ROM) 62 stores configuration information and programs for the computer data processor.
  • a random access memory (RAM) 63 is used to temporarily store various data during the operation of the processor 66.
  • a computer program for performing data processing is also stored in the memory 61.
  • the internal bus 64 is connected to the above-described memory 61, read only memory 62, random access memory 63, input device 65, processor 66, display device 67, and interface unit 68.
  • the instruction code of the computer program commands the processor 66 to execute a predetermined data processing algorithm, and after obtaining the data processing result, displays it on, for example, an LCD display.
  • the processing result is outputted on the display device 67 of the class, or directly in the form of a hard copy such as printing.
  • FIG. 3 shows a block diagram of a controller in accordance with an embodiment of the present invention.
  • the controller 50 includes: a control unit 51 that controls the radiation source 10 and the carrier mechanism according to an instruction from the computer 60. 40 and detecting and collecting device 30; a trigger signal generating unit 52 for generating a trigger command for triggering the action of the radiation source 10, the detecting and collecting device 30, and the carrying mechanism 40 under the control of the control unit; the first driving motor 55, which drives the carrying mechanism 40 up or down according to a trigger command generated by the trigger signal generating unit 52 under the control of the control unit 51; the height information acquiring unit 53, which feeds back the bearing unit 51 with the movement of the carrying mechanism 40 The height information of the mechanism; the second drive motor 56, which drives the carrier mechanism 40 to rotate according to the trigger command generated by the trigger signal generating unit 52 under the control of the control unit 51; the angle information acquiring unit 54, which is acquired during the rotation of the carrier mechanism 40 The angle of rotation of the carrier mechanism 40 is fed back to the control unit 51
  • the above-described intensity information acquisition unit 53 and angle information acquisition unit 54 are both optical code disks, which have the advantage of being anti-interference.
  • the inspection method of the present invention can be used to detect dangerous goods by measuring the density and atomic number of the liquid in the liquid article, and judging whether the measured density and the atomic number fall within a predetermined danger zone.
  • safe liquids such as beverages, skin care products, cosmetics, etc.
  • hazardous liquids such as flammable liquids, liquid explosives, highly corrosive liquids, etc.
  • water as a safe liquid has a density of about 1 (characteristic density 1.11) and an atomic number of about 7.51.
  • Alcohol which is a dangerous liquid, has a density of about 0.79 (characteristic density of 0.89) and an atomic number of about 6.47.
  • pattern recognition safe and hazardous liquids can be separated based on density and atomic number.
  • FIG. 4A and 4B are schematic views showing a method of inspecting a liquid article in accordance with the present invention.
  • the two-dimensional space composed of density and atomic number one or more regions are drawn. When the density and atomic number of the object fall within these regions, it is judged as a safe liquid, and vice versa. .
  • the safe liquid area is determined by measuring a large number of safe liquids and dangerous liquids, and taking a safer liquid area where the safe liquid is concentrated and the dangerous liquid is sparse.
  • Fig. 4A by actually measuring the density and atomic number of various liquids, the area where the safe liquid falls and the area where the dangerous liquid is located are determined in the two-dimensional space of density and atomic number. Then, as shown in Fig.
  • the area where the safe liquid is located is shrunk to increase the sensitivity of the detection.
  • Fig. 5 is a schematic diagram for explaining the relationship between DR imaging and CT imaging.
  • a liquid article is first subjected to DR imaging to determine a portion of the liquid in the liquid article, and then only CT imaging is performed for a location in the liquid portion to increase the speed of the inspection.
  • Figures 6 and 7 show examples of DR imaging, respectively.
  • the position of the liquid in the liquid article is determined by analyzing the pixel values as described below.
  • the liquid article contains only one liquid.
  • the pixels of the DR image obtained by DR imaging are subjected to pixels. Analysis to determine the location of the interface between the liquids. Then, CT imaging is performed for different liquids.
  • Fig. 8 shows a general flow chart of a liquid article inspection method according to a first embodiment of the present invention. As shown in Figure 8, for example, during customs clearance, it is necessary to carry out a safety inspection of the liquid items carried by the passengers.
  • step S111 the operator places the liquid object to be inspected on the carrier mechanism 40.
  • step S112 the operator presses the start button to start the DR scan to generate a DR image, as shown in Figs.
  • the purpose of DR scanning is to: firstly obtain a fluoroscopic image of the liquid object to be inspected, so that the operator can see the internal structure of the liquid to be inspected; second, the software can automatically recognize the liquid position according to the DR image, and guide the next step.
  • CT imaging Third, the user can specify the location where CT imaging and examination is required in the DR image to guide the next CT imaging. The detailed process of DR imaging will be described in detail later. • It must be stated that DR scanning is not required. In order to speed up the inspection, the CT scan can be guided without DR scanning, but the CT scan can be directly scheduled at several locations.
  • a height of 5 cm from the bottom of the bottle can be used as the predetermined scanning height.
  • the operator can visually check the size of the object and set the appropriate height by experience, such as setting the scanning height of the can-like cola to 3 cm, and setting the scanning height of the thicker wine to 10 cm.
  • the DR image is analyzed in step S113A to automatically determine the scanning position of the CT image, or in step S113B, the operator operates an input device 65 such as a mouse to specify the position at which the CT scan is to be performed.
  • an input device 65 such as a mouse
  • a CT scan process is performed in step S114, that is, a CT scan of the liquid item is performed for the above-described determined scan position, CT scan data is obtained, and the CT image is reconstructed according to the reconstruction calculation.
  • Each pixel of the CT image represents the density and atomic number of the corresponding portion of the liquid article.
  • step S115 the computer analyzes the CT image by performing an analysis program to obtain the measured Density and atomic number. Further, in step S116, it is judged whether or not the point determined by the measured density and the atomic number falls in a predetermined region in the two-dimensional coordinate space of the density-atomic number. In step S117, if the point determined by the density and the atomic number falls within the predetermined danger zone, it indicates that the liquid item is suspicious or is likely to be dangerous, an alarm is issued to the operator, or the inspection result is printed.
  • FIG. 10 shows the arrangement of the DR image data collected by the detecting and collecting device 30 during the DR imaging process.
  • step S210 a command is sent from the computer 60 to the controller 50 to drive the carrier mechanism 40 to move vertically along the object path 20.
  • the controller 50 monitors the height of the carrier mechanism in real time by the height information acquisition unit 53 during the vertical movement of the carrier mechanism.
  • the controller 50 sends a trigger signal to the detecting and collecting device 30 at a certain height (e.g., 1 mm).
  • the detecting and collecting device 30 After receiving the trigger signal, the detecting and collecting device 30 performs a rounding of the output signals of each detector to obtain high-energy detecting data and low-energy detecting data, and saves them in the built-in buffer.
  • step S212 it is judged whether or not the carrier mechanism 40 reaches a specified height, for example, 500 mm. If it has not been reached, the flow proceeds to step S210 to continue the above operation.
  • the controller 50 no longer sends a trigger signal to the detecting and collecting device 30.
  • the computer 60 reads the acquired high and low energy detector signals from the detecting and collecting device 30 and arranges them in the form of a data matrix to become a DR image. Each pixel of the DR image records the intensity remaining after the line penetrates the object, including the low energy ray intensity and the high energy ray intensity.
  • the scan positions of subsequent CT imaging are determined on the DR image.
  • the line number is first obtained on the DR image, and then the computer converts the line number into the carrier mechanism, and commands the controller 50 to drive the carrier mechanism 40 to move to After the position is specified, CT imaging is performed.
  • each row of the DR image corresponds to a particular carrier 40 height.
  • the height of the carrier mechanism is 0 when the DR imaging process starts, and the carrier mechanism is continuously decreasing during the imaging process, triggering a set every hour, then the m-th row in the DR image, the corresponding carrier height is -m*h .
  • Figure 11 shows a flow chart for processing a DR image to determine the CT scan position.
  • the liquid items to be inspected are generally divided into the bottom of the bottle, the liquid part, the bottle neck, the bottle cap, etc.
  • the liquid portion (possibly with multiple layers) can be extracted and the CT scan position of each layer can be determined.
  • step S310 the high and low energy data of the DR image is fused and smoothed to obtain a single-valued DR image with less noise.
  • the specific method of high and low energy fusion may be directly selecting one of high and low energy data as a fusion result, or a weighted combination of high and low energy data.
  • the smoothing method can be to filter the image using a Gaussian filter.
  • step S311 the liquid object to be inspected (foreground) in the smoothed DR image is extracted, and the air (background) is removed.
  • the specific method may be setting a threshold, classifying pixels whose values are below the threshold as foreground pixels, and other pixels as background pixels. The reason why the threshold is used to remove the background is that the liquid object to be inspected blocks the ray, so the corresponding DR map pixel value is lower (the DR image records the remaining intensity of the ray).
  • horizontal edge pixels in the smoothed DR image are extracted.
  • the specific method may be: calculating, for each pixel of the DR image, a difference value between the pixels adjacent to the vertical direction; if the difference value is greater than a certain threshold, the pixel is considered to be a horizontal edge pixel.
  • a horizontal edge line in the smoothed DR image is extracted, the horizontal edge line corresponding to the interface between the bottom of the bottle and the liquid, the interface of the liquid and air, the interface of the cap and the air, or the interface between the layers of the liquid in the container.
  • the specific method may be: counting the ratio of the number of horizontal edge pixels to the number of foreground pixels for each row in the horizontal direction in the DR image, and if the ratio is larger than a certain threshold (for example, 50%), the line is classified as Horizontal edge line.
  • step S314 the DR image is vertically partitioned, and the non-liquid area is excluded.
  • Each horizontal edge line in the DR image divides the DR image into multiple regions, from bottom to top, including: bottom, liquid (multiple layers depending on density), air inside the bottle (if any), Bottle caps, etc.
  • Non-liquid areas can be excluded by developing screening criteria, such as:
  • the area where the number of rows is less than a certain threshold is excluded.
  • the area with a small number of rows may be the bottom of the bottle, the cap, or a small gap between the liquid in the container and the top of the container (such as the air at the top of the can).
  • the determination of the specific threshold can be determined by investigating the thickness of the bottom of the liquid packaging container, the cap and the thickness of the air layer inside the container.
  • the determination of the specific threshold value can be determined by achieving the width of the bottleneck of various liquid packaging containers.
  • the CT scan position of the liquid region (possibly multiple) is determined to complete the hierarchical positioning of the liquid. After the non-liquid area is excluded, the remaining area (possibly multiple) is classified as a liquid area. Take the center line of these areas in the height direction as the CT scan position.
  • Fig. 12 shows the CT imaging process
  • Fig. 13 shows the arrangement of the CT projection data during CT imaging.
  • performing a CT imaging process that is, performing CT imaging at the CT scan position performed, generating a density and atomic number image of a slice of the object to be performed. Measurement of liquid density and atomic number.
  • the customs clearance time can be greatly saved.
  • the computer 60 sends a command to the controller 50 to drive the carrier mechanism 40 to rotate by a predetermined angle, for example, 1 degree.
  • the controller 50 monitors the angle of the carrier mechanism in real time by the angle information acquisition unit 54 during the rotational movement of the carrier mechanism.
  • the controller 50 sends a trigger signal to the detecting and collecting device 40.
  • the detection and collection device 40 collects the output signal of each detector once and saves it into its built-in buffer.
  • step S412 it is judged whether or not the cumulative rotation angle has reached one week. If it has not reached one week, the flow proceeds to step S410 to continue the above operation.
  • the rotation motion is stopped in step S413, and the controller 50 no longer transmits a trigger signal to the detection and collection device 30.
  • the computer 60 reads the acquired high and low energy detector signals from the detecting and collecting device 30, and arranges them in the form of a data matrix to become CT projection data, as shown in FIG.
  • Each pixel of the CT projection data records the remaining intensity after the ray penetrates the object, including the low energy ray intensity and the high energy ray intensity.
  • step S414 the computer 60 reconstructs the density and atomic number image on the fault, that is, the CT image, based on the high and low energy CT projection data using the dual energy reconstruction algorithm.
  • Each pixel in the CT image records the density and atomic number of the object under test at the corresponding position of the pixel.
  • the two-dimensional distribution u(x, y) is plotted along a certain direction to obtain a one-dimensional function p e (t), which is called the projection of u(x, y) at the ⁇ angle. If the projection p e (t) in each direction can be obtained, the two-dimensional distribution u(x, y) can be accurately calculated from the Radon transform.
  • the process of obtaining a two-dimensional distribution from a projection is called a practical application.
  • the projection of the attenuation coefficient of a certain slice of the object is measured in all directions, so that the CT principle can be
  • the reconstruction obtains a two-dimensional distribution of attenuation coefficients of the object slice.
  • the material line attenuation coefficient can be approximated by the following analytical expression (1).
  • the scattering cross section has a known analytical expression for f p (E) and f KN (E) as a function of energy.
  • the constants &1 and a 2 are related to the atomic number, mass and density of the substance, and the expressions are as shown in equations (2) and (3), where Z represents the atomic number, M represents the mass number, and p represents the density (g/cm). 3 ), n is a constant.
  • the linear attenuation coefficient of each substance can be determined by the two coefficients ai and uniquely in the formula (1), two base materials, such as carbon and aluminum, can be selected, and the linear combination of the line attenuation coefficients of the base material is used to indicate other All materials, as shown in the following formula (4) -
  • ⁇ (E) / , (E) + 2 (E) (4)
  • /(E) is the line attenuation coefficient of any material
  • / 2 (E) is the line of the two selected base materials Attenuation coefficient, ! ⁇ and ⁇ are called base material coefficients.
  • the product of the ratio of the atomic number of 2 times to the mass number and the density is defined as the feature density.
  • the energy spectrum produced by an X-ray tube is usually a continuum, and the energy response function of the detector to X-rays is not constant.
  • the product of the energy spectrum N(E) and the energy response function P d (E) is S(E), and S(E) is normalized.
  • Io and I respectively represent the detector readings before and after the attenuation of the radiation by the object, E m represents the maximum energy of the radiation, and 1 represents the path through which the radiation passes.
  • Equation (9) is the relationship between the measured projection value p of the actual system and the two-dimensional distribution ⁇ . It can be seen that due to the X-ray pleochroism, the formula (9) does not represent the line integral along a certain line, and thus does not satisfy the requirements of the CT mathematical principle. The conventional reconstruction algorithm ignores this inconsistency, and the reconstructed ⁇ c image contains a cup-shaped artifact called a hardening artifact.
  • the existing dual-energy CT method first obtains two sets of data by using a conventional reconstruction algorithm / (x, and then calculates information such as atomic number and density. However, this does not eliminate the influence of ray polychromaticity.
  • the present invention solves the problem of decomposition of the base material. This problem.
  • the dual-energy CT collects the projection data at two energies, and the following dual-energy projection data is obtained:
  • Figure 14 shows the process of measuring the properties of a liquid.
  • step S510 the density image and the atomic number image are fused and smoothed to obtain a single-valued CT image with less noise.
  • the specific method of fusion may be directly selecting one of the density image and the atomic number image as a fusion result, or a weighted combination of the two.
  • the smoothing method can be to filter the image using a Gaussian filter.
  • step S511 the object to be examined (foreground, including the liquid and its packaging) in the smoothed CT image is extracted, and the air (background) is removed.
  • the specific method may be setting a threshold, classifying pixels whose values are above the threshold as foreground pixels, and other pixels as background pixels. This is because the density and atomic number of air are almost zero, while the density and atomic number of liquid objects being inspected are relatively large.
  • step S512 liquid pixels are extracted in the foreground pixels.
  • the specific method may be: firstly establishing a binary image corresponding to the CT image, the foreground pixel has a value of 1, and the background pixel has a value of 0; and then the binary image is subjected to multiple etching operations, one pixel at a time.
  • the last pixel with a value of 1 is a liquid pixel. This is because the liquid is always inside the package and the package can be removed step by step through multiple corrosions. The number of times of corrosion can be set by implementing the thickness of the package.
  • step S513 the average density and the average atomic number of all liquid pixels in the CT map are counted as the output of this measurement.
  • the steps as described above are performed for each layer of liquid, thereby judging whether or not the liquid of each layer is suspicious. Finally, report the final inspection results to the operator.
  • the first embodiment above describes the case of detecting a single piece of liquid article each time.
  • 15 ⁇ 18 describes in detail the operation process of simultaneously detecting multiple pieces of liquid articles.
  • the second embodiment differs from the first embodiment in that, since there are a plurality of liquid articles, it is necessary to position the imaging result displayed on the display after the CT imaging is completed, corresponding to the position of the object on the carrier mechanism, thereby facilitating the operation.
  • the staff figured out that the liquid item was dangerous.
  • 15A and 15B are views showing a CT image reconstructed in the case of detecting a plurality of liquid articles in accordance with a second embodiment of the present invention.
  • the operator observes the object on the carrier in a plan view, which requires that the position of each object on the CT image of each layer after the end of the inspection must be consistent with the top view of the carrier mechanism.
  • 16A-16K illustrate how the CT reconstructed image and/or the carrying mechanism are rotated to coincide with the relative position of the object prior to the CT scan after CT imaging is completed.
  • Figure 16A shows a top view of the carrier mechanism 40 at the beginning of the first CT imaging, with the arrows indicating the angle of the carrier.
  • Fig. 16B shows a CT image before the rotation of the first CT image, which coincides with the top view of the carrier mechanism at the beginning of the first imaging, wherein the broken line identifies the range of rotation of the carrier mechanism.
  • Fig. 16C shows the image after the rotation of the first layer CT image, since the rotation angle is zero, there is virtually no need to rotate the layer CT image.
  • Figure 16D shows a top view of the carrier mechanism at the beginning of the nth CT imaging.
  • the carrier mechanism has an angle compared to the case before the first CT imaging.
  • Figure 16E shows the image before the nth CT image is rotated, which is the same as the nth time.
  • the top view of the load bearing mechanism at the beginning is the same.
  • Figure 16F shows a schematic diagram of the rotation of the image of Figure 16D aligned with the first layer CT image.
  • Fig. 16G shows a top view of the carrier mechanism at the beginning of the last CT imaging, and it can be seen that the carrier mechanism at this time has an angle compared to the beginning of the first CT imaging.
  • Fig. 16H shows an image before the rotation of the Nth CT image, which is identical to the top view of the carrier mechanism at the start of the Nth imaging.
  • Figure 161 shows a schematic diagram of the rotation of the image of Figure 16H aligned with the first layer CT image.
  • Fig. 16J shows a plan view of the carrier mechanism after the end of the last CT imaging, in which there is a rotation angle compared to the case before the first CT imaging.
  • Fig. 16K shows that the carrier mechanism is rotated after the end of imaging to coincide with the image obtained by the first CT imaging.
  • an angle adjustment is made to each layer of the CT image and the carrier mechanism.
  • the CT images of the respective layers are rotated to maintain the position of the same object in the CT images of the layers. Consistent, such as alignment to the first layer CT.
  • adjust the angle of the carrier mechanism so that the top view of the carrier mechanism is consistent with the CT image.
  • the angle of the carrier mechanism at the beginning of the nth CT imaging is , and the angle of the carrier mechanism at the end is that the carrier mechanism rotates in a counterclockwise direction in a plan view.
  • After rotating the image in order to keep the object in the top view of the bearing mechanism and the CT image, simply rotate the bearing mechanism by 360 in the counterclockwise direction.
  • Fig. 17 is a flow chart showing an inspection performed in the case of a plurality of objects to be inspected.
  • the operator powers up the system and logs in to the inspection interface. After the system self-test, it enters the ready state. Then, the operator places a plurality of objects, such as object A and object B, directly on the carrier mechanism 40 and presses the check button.
  • object A is placed at the upper right corner of the carrying mechanism, and the object B is placed at the lower left corner of the carrying mechanism.
  • step S712 the operator presses the start button to start performing DR scanning in the same manner as described above to generate a DR image.
  • step S713A the DR image is analyzed to automatically determine the scan position of the CT image, or in step S713B, the operator operates an input device 65 such as a mouse to specify the position where the CT scan is to be performed. . In this way, because only the representative position in the liquid item is CT scans are performed, which speeds up the inspection without degrading the quality of the inspection.
  • the CT scan process is performed in step S714, that is, the liquid article is subjected to CT scan for the determined position, the CT scan data is obtained, and the CT image is reconstructed according to the reconstruction algorithm.
  • Each pixel of the CT image represents the density and atomic number of the corresponding portion of the liquid item.
  • the angle of the bearing mechanism and the angles of the CT images of the layers are adjusted according to the method as described above, so that the positions of the objects in the CT images of the layers are consistent (for example, the first layer is aligned).
  • the CT image) is consistent with the actual position of the object on the support mechanism (eg, consistent with the top view) to facilitate distinguishing between object B and object A.
  • step S717 image segmentation processing is performed on each layer of CT images (for example, using a watershed algorithm) to obtain a liquid region of each of the objects, and in step S718, the density and atomic number mean of each pixel in each liquid region are counted.
  • Step S719 determines whether the point determined by the density and the atomic number measured for each liquid region falls within a predetermined region in the density-atomic coefficient coordinate space, and accordingly, in step S720, it is determined whether the liquid of the layer of the object is It is dangerous.
  • step S721 the inspection results of each layer are summarized and presented to the operator.
  • a summary method is: The conclusion of this inspection is "safe” only if all liquid areas in all CT images are classified as safe and dangerous; otherwise the conclusion of this inspection is "suspicious".
  • each layer of the CT image is color-coded and displayed to the user, and the object to be judged as a suspicious object in the CT map uses a color (such as red), which is judged to be a safe liquid using another color (such as green). ).
  • the above description is to check the condition of two liquid articles at the same time.
  • the liquid carrying material is carried by the supporting mechanism.
  • the surface is divided into a plurality of areas, such as the A area, the B area, the C area, and the D area in FIG. 18, so that the operator can place each liquid item in the corresponding area, and input each of the respective areas. Liquid identification information.
  • the third embodiment proposes to place a plurality of objects in the compartment.
  • the third embodiment differs from the second embodiment in that in the process of inspecting a plurality of objects A compartment bucket is used.
  • the specific operation of the detecting apparatus according to the present embodiment will be described in detail below with reference to FIGS. 19 to 24.
  • Fig. 19 shows a side view of the compartmental bucket according to the present embodiment
  • Fig. 20 shows a plan view of the compartmental bucket
  • Fig. 21 shows a bottom view of the compartmental bucket.
  • the compartmental bucket 70 includes a bottom portion and a side wall coupled to the bottom portion.
  • a plurality of conical or other shaped projections are provided on the lower surface of the bottom. The three conical projections can be inserted into corresponding positioning holes in the carrier mechanism 40 such that during rotation of the carrier mechanism, the compartmental tub 70 rotates with the carrier mechanism 40 to prevent relative movement therebetween.
  • a flange is provided at the top of the side wall to facilitate the operator to grasp and carry it.
  • the side walls are cylindrical or conical and may be made of an elastic side material such as polyethylene (PE) or aluminum.
  • Figure 20 is a top plan view of three compartments.
  • one or more compartment members are provided in the space formed by the bottom and the side walls, and the space is divided into a plurality of spaces as needed for respectively placing the liquid articles.
  • the plurality of liquid articles are respectively placed in a space divided by the compartment members, and the liquid objects are simultaneously detected.
  • a mark may be provided on the outer surface of the side wall to facilitate positioning of the article in the test tub at the time of detection. For example, when it is detected that one of the four items is suspicious, it can be reported to the user which item is suspicious based on the relative position of the item relative to the mark placed on the side wall.
  • a circular mark having a predetermined cross section is provided on the upper side of the division member, or the length of a certain division member is smaller than the others, thereby serving as a positioning liquid article.
  • Figure 21 is a bottom view of the compartment bucket. Three projections are evenly distributed on the lower surface of the bottom. However, the three convex distributions may also be uneven.
  • Figure 22 shows how the automatic detection of the compartments and markers during the inspection process.
  • the cell has a specific structure and can be detected by a template matching method commonly used in image processing.
  • a template image with a cross may be first established, and the center of the template is placed at the center of the CT image to be recognized to obtain a matching value.
  • step S811 it is judged whether the matching value is greater than a predetermined threshold. If the result is negative, the template image is rotated after the template image and the CT image are matched to the maximum. If the degree of matching is greater than a predetermined threshold, it is considered that there is a bin in the CT chart, otherwise it is considered not in the CT chart. There is a compartment tube.
  • the bit mark can be further determined based on the characteristics of the positioning mark in step S812. For example, in the first type of cylinder with a cross, the positioning mark is located at the top of a grid line, making this grid line shorter than the other three grid lines. After using the template matching method to confirm that there is a bin in the CT image, the cross line in the template image coincides with the ruled line when the matching degree is maximum; by comparing the length of the four ruled lines, the longest one is extracted. Get the positioning tag.
  • each layer of CT images is subjected to binar detection and localization mark detection. If no cell is detected in each layer of the CT image, the operator is considered not to use the cell; if a cell is detected in at least one CT image, the operator is considered to have used a cell; The position marks detected by the layer are different, and the one with the highest signal strength can be used as the final positioning mark.
  • One way to describe the signal strength of a positioning mark in a layer of CT is to subtract the average value from the maximum value of the length of the four dividing lines of the layer. The larger the difference, the stronger the positioning mark signal.
  • Figures 23A-23D show schematic views of a rotating bucket during the inspection process.
  • the polar coordinate system is established with the center of the CT image as the coordinate system origin, and the position of each division is uniquely determined by the angular coordinate of the positioning mark.
  • the system determines the polar coordinate angle corresponding to the positioning mark in various CT images, and the preset adjustment target of the system is ⁇ , then only the CT image of each layer and the rotation angle of the supporting mechanism can be adjusted, as shown below. 24 is shown.
  • the compartment bucket is rotated by an angle to achieve the desired position.
  • the image obtained by imaging each layer of CT is rotated to match the angle of the rotated compartment.
  • Fig. 24 shows a flow chart of the inspection process according to the present embodiment. Since step S911 to step S921 are the same as step S711 to step S721 of the above-described third embodiment, only step S922 to step S926 are described herein.
  • step S922 the CT image is analyzed, and it is judged at step S923 whether or not the division detecting cylinder is provided. If there is no test cartridge, this check is over.
  • step S924 the positioning mark of the detection cylinder is identified in step S924, and the position of each division is determined. Then, in step S925, the bearing mechanism angle and the CT image angles of the layers are adjusted a second time, and the position of each of the cells in the detection cylinder on the bearing mechanism and the position of each of the divisions on the CT images of the liquid layers are adjusted to At the scheduled location.
  • step S926 the system lists the inspection results of each liquid region falling within each division portion of each layer of the CT image, and summarizes the inspection results for each of the divisions and presents them to the user.
  • a summary method is: For a certain division, only when all liquid regions falling within the division in all CT images are judged as safe liquids, the inspection result of the original grid is "safe"; Otherwise, the conclusion of the check of the original is "suspicious.”
  • a spiral CT scan of a liquid article results in a set of spiral CT images.
  • the position of the liquid in the liquid article can be determined, and whether the liquid is stratified, and then utilized as described above.
  • a similar method obtains the physical property values of the liquid at each layer location, such as density and atomic number. For example, when performing spiral CT imaging on a liquid article as shown in Fig. 25, if scanning is performed with a spiral pitch of 2 cm, a set of CT images as shown in Figs. 26A-26M will be obtained.
  • Spiral CT imaging here can be performed with large pitch CT imaging or conventional pitch CT imaging.
  • the present invention can also be carried out using only one of density or atomic number, or using other physical properties for identification of dangerous goods.
  • it is determined whether the liquid is dangerous by determining whether a physical property value falls within a predetermined range.
  • Whether the points determined by the plurality of physical attribute values fall within a predetermined area in the multi-dimensional space defined by the attributes determines whether the liquid is a dangerous product.
  • DR imaging is performed first, and then dual-energy CT imaging is performed to obtain the density and atomic number of the liquid, DR imaging here is not necessary because the predetermined position can be specified for different liquid objects for dual-energy CT. Imaging to obtain the density and atomic number of the liquid.

Description

液态物品检査方法和设备 技术领域
本发明涉及辐射检查技术领域, 特别涉及一种通过对液态物品进行双能 CT 成像来对液态物品进行快速安全检査的方法及设备。
背景技术
美国的 9· 11事件发生以后,航空领域的安全检査工作越来越受到重视。在以 往进行的行李包裹安全检査的基础上,增加了对旅客随身携带的液态物品的安全 检査要求。 因此, 急需有效的方式和手段进行行李物品中液态物品的快速安全检 查。
当前, 有如下四类的方法可以用于液态物品的安全检査: 化学方法、 电磁方 法、 中子方法和射线方法, 具体描述如下:
1 ) 化学方法可以细分为气味识别、 离子扫描探测和物质分析。 气味识别在 实际应用中常常因为液态物品被密封包装而无法实现检査。离子扫描探测以高敏 感性著称, 但是其缺点在于误报率高, 常常受到背景环境的影响。 物质分析具有 精度高和准确性高的特点, 但是这种方法需要一定的时间对样品进行分析, 不能 满足现场快速检査的需求。
2 ) 电磁方法采取主动的测量方式, 其根据不同液态物品对电磁波的介电常 数不同从而将液态物品区分开来。 电磁方法本身容易受到金属包装和较厚材料包 装的不利影响。 因此, 在包装材料复杂的实际情况下, 电磁方法具有一定的局限 性。
3 ) 中子检査方法的使用会出现 "中子活化"的现象, 即经过中子检查的被检 查液态物品会有辐射残留现象。 并且, 由于中子的穿透能力更强, 故其辐射屏蔽 更为复杂、 设备占地面积大, 因而不适合在民航的安全检査系统中使用。
4 ) 当前, 航空领域的安全检査装置多为射线装置, 这些装置中, 目前釆用 最多的技术是 X射线二维成像技术和三维 CT扫描成像技术。这些技术主要用于 对行李物品进行安全检查, 不能专门针对行李物品中的液态物品进行安全检査。
X射线二维成像技术是将被检查物体的三维信息在 X射线束方向上积分而得 到物体的二维信息投影图像。这些图像采用灰度或伪彩色的表现形式表现出物体 内部的结构, 给设备操作人员以直观的显示。 但是, X射线二维成像技术毕竟缺 少了被检查物体一个维度的信息, 因此用该技术对液态物品进行检査受到被检査 液态物品的外形和尺寸的影响特别严重。
三维 CT扫描成像技术是 CT技术的推广和应用。 CT技术最早应用于医学, 是辅助医生进行诊断的工具。它是通过对被检査物的各个断层进行多角度的投影 来实现的。 通过计算机对上述各个断层的多角度投影数据的重建, 计算得到每个 断层的重建图像, 以不同的灰度显示重建图像中不同的衰减系数信息, 从而显示 被检査物的内部结构。 随着 CT技术的发展, 出现了用于无损探伤的工业 CT和 用于安全检查的行李 CT等, 这些 CT拓展技术仍然是以获得显示物体内部结构 的断层图像为目的的, 依赖于人工对结构进行分析, 以判断被检物是否属于危险 物品, 不适用于自动、 快速检査。 此外, 上述 CT型安全检查设备, 因其所覆盖 的被检查物范围广, 而导致造价高昂, 设备笨重, 推广应用困难。
综上所述, 对于液态物品进行快速检查, 化学方法、 电磁方法和中子方法存 在着本身不适合快速安全检查的特点, 采用 X射线二维成像技术和三维 CT扫描 成像技术, 只能获得表现物品内部结构的图像, 不能为液态物品的自动、 安全检 查提供充分依据。 发明内容
为了克服上述现有技术中存在的不足, 本发明的目的是提供了一种用射线对 液态物品进行安全检查的方法和设备, 它可以在不破坏液态物品包装的情况下对 其进行快速检查, 得到被检查液态物品的定量信息。
在本发明的一个方面, 提出了一种用双能 CT对液态物品进行检查的方法, 包括步骤: 对被检液态物品进行双能 CT扫描, 得到双能 CT投影数据; 根据所 述投影数据进行 CT重建, 得到表示被检液态物品的物理属性值的 CT图像; 根 据所述 CT图像提取被检液态物品的物理属性值; 以及基于所述物理属性值判断 所述被检液态物品是否是危险品。
根据本发明的实施例, 所述物理属性值包括被检液态物品的密度和原子序 数。
根据本发明的实施例, 所述双能 CT扫描釆用平面断层 CT扫描的方式。 根据本发明的实施例, 所述双能 CT扫描釆用常规螺旋 CT扫描的方式。 根据本发明的实施例, 所述双能 CT扫描采用大螺距螺旋 CT扫描的方式。 根据本发明的实施例, 在进行平面断层 CT扫描之前, 预先设定一组扫描位 置。
根据本发明的实施例, 在进行平面断层 CT扫描之前, 先进行 DR扫描得到 被检物品的透射图像, 然后根据透射图像确定 CT扫描位置。
根据本发明的实施例, 得到透射图像后, 操作员通过输入装置指定透射图像 中的至少一行, 作为 CT扫描位置。
根据本发明的实施例, 得到透射图像后, 通过图像处理技术自动确定透射图 像中的至少一行, 作为 CT扫描位置。
根据本发明的实施例, 形成透射图像的步骤包括: 从射线源发出髙能射线和 低能射线, 穿透被检物体, 形成高能透射图像和低能透射图像; 融合高能透射图 像和低能透射图像, 形成所述透射图像。
根据本发明的实施例, 形成透射图像的步骤包括: 从射线源发出高能射线和 低能射线, 穿透被检物体, 形成高能透射图像和低能透射图像; 选择高能透射图 像和低能透射图像之一, 作为所述透射图像。
根据本发明的实施例, 从双能投影数据重建表示被检液态物品的物理属性值 的 CT图像的步骤包括: 根据高低能投影数据, 生成关于两种基材料系数的投影 数据; 根据所述两种基材料系数的投影数据进行重建, 得到表示被检液体所对应 的两种基材料系数的 CT图像; 以及根据所述表示基材料系数 CT图像, 生成所 述表示被检液态物品物理属性值的 CT图像。
根据本发明的实施例, 根据表示被检液态物品的物理属性值的 CT图像, 得 到被检液态物品物理属性值的步骤包括: 从所述 CT图像中提取与液体部分相对 应的像素; 对液体部分的像素计算密度均值和原子序数均值, 作为所述被检液态 物品的密度和原子序数。
根据本发明的实施例, 基于所述物理属性值来判断所述被检液态物品是否是 危险品的步骤包括:判断所述密度和原子序数所确定的点是否落在密度-原子序数 二维坐标空间中的预定区域中; 在所述点落在所述预定区域中的情况下, 认为所 述被检液态物品是危险品。
根据本发明的实施例, 在针对每个位置进行双能 CT扫描之后, 旋转被检査 液态物品的 CT图像, 使其与第一次双能 CT扫描所形成的图像对齐。 根据本发明的实施例, 在针对各行进行双能 CT扫面之后, 旋转被检液态物 品, 使其与扫描之前的位置相同。
根据本发明的实施例, 所述液态物品放置在被分成了多个空间的桶内。
根据本发明的实施例, 所述的方法还包括步骤: 利用预定的模板来自动检测 桶的存在; 在存在桶的情况下, 检测 CT图像中的特定标记; 基于所述特定标记, 将桶旋转到预定的位置。
根据本发明的实施例, 所述的方法还包括步骤: 将被检液态物品的判断结果 显示在显示屏上。
根据本发明的实施例, 所述的方法还包括步骤: 将各个被检液态物品的判断 结果打印出来。
根据本发明的实施例, 所述的方法还包括步骤: 将各个被检液态物品的 CT 图像彩色化。
根据本发明的实施例, 所述双能 CT扫描是基于预定的位置进行的。
在本发明的另一方面, 提出了一种用双能 CT对液态物品进行检査的设备, 包括: 射线源, 用于发出射线; 探测和采集装置, 用于探测并采集穿透至少一件 被检液态物品的射线信号; 控制器, 控制所述射线源和探测和采集装置对被检液 态物品进行双能 CT扫描, 得到投影数据; 从投影数据重建表示被检液态物品的 至少一种物理属性值的 CT图像的装置; 以及基于所述物理属性值在相应维度的 空间中的位置来判断所述被检液态物品是否是危险品的装置。
根据本发明的实施例, 所述双能 CT扫描是基于预定的位置进行的。
根据本发明的实施例, 所述探测和采集装置探测并采集穿透至少一件被检液 态物品的射线信号以形成透射图像; 其中所述设备还包括指定透射图像中的至少 一行的装置; 所述双能 CT扫描是基于所指定的行而进行的。
根据本发明的实施例, 中所述物理属性值至少包括被检液态物品的密度和原 子序数。
根据本发明的实施例, 从射线源发出髙能射线和低能射线, 穿透被检物体, 形成高能透射图像和低能透射图像, 所述设备还包括: 融合高能透射图像和低能 透射图像形成所述透射图像的装置。
根据本发明的实施例, 从射线源发出高能射线和低能射线, 穿透被检物体, 形成高能透射图像和低能透射图像, 所述设备还包括: 选择高能透射图像和低能 透射图像之一, 作为所述透射图像的装置。
根据本发明的实施例, 所述指定透射图像中的至少一行的装置包括: 操作员 利用输入装置从透射图像中选择至少一行的装置。
根据本发明的实施例, 所述指定透射图像中的至少一行的装置包括: 分析所 述透射图像的像素值, 以将所述透射图像分层的装置; 将各层中的中间行指定为 要进行双能 CT扫描的行的装置。
根据本发明的实施例, 从投影数据重建表示被检液态物品的物理属性值的 CT 图像的装置包括: 融合由被检液态物品的密度所标识的密度图像和由被检液 态物品的原子序数所标识的原子序数图像作为所述 CT图像的装置; 从所述 CT 图像中提取与液体部分相对应的像素; 对液体部分的像素计算密度均值和原子序 数均值作为所述被检液态物品的密度和原子序数的装置。
根据本发明的实施例, 基于所述物理属性值来判断所述被检液态物品是否是 危险品的装置包括:判断所述密度和原子序数所确定的点是否落在密度-原子序数 二维坐标空间中的预定区域中的装置,其中,在所述点落在预定区域中的情况下, 认为所述被检液态物品是危险品。
根据本发明的实施例,所述的设备还包括:在针对每行进行双能 CT扫描之后, 旋转被检查液态物品的 CT图像, 使其与第一次双能 CT扫描所形成的图像对齐 的装置。
根据本发明的实施例,所述的设备还包括在针对各行进行双能 CT扫面之后, 旋转被检液态物品, 使其与扫描之前的位置相同的装置。
根据本发明的实施例, 所述的设备还包括桶, 被分成了多个空间, 分别用于 放置所述液态物品。
根据本发明的实施例, 所述的设备还包括: 利用预定的模板来自动检测桶的 存在的装置; 在存在桶的情况下, 检测 CT图像中的特定标记的装置; 基于所述 特定标记, 将桶旋转到预定的位置的装置。
根据本发明的实施例, 所述的设备还包括: 显示装置, 显示被检液态物品的 判断结果。
根据本发明的实施例, 所述的设备还包括: 将各个被检液态物品的判断结果 打印出来的装置。
根据本发明的实施例, 所述的设备还包括: 将各个被检液态物品的 CT图像 彩色化的装置。
根据本发明的实施例, 所述的设备还包括承载所述被检液态物品的承载机 构, 所述承载结构承载所述被检液态物品的表面上被划分成操作人员可识别的多 个区域。
在本发明的又一方面, 提出了一种用双能 CT对液态物品进行检査的设备, 包括: 射线源, 用于发出射线; 探测和釆集装置, 用于探测并采集穿透至少一件 被检物体的射线信号; 控制器, 控制所述射线源和探测和采集装置对被检液态物 品进行螺旋 CT扫描, 来形成每个均表示被检液态物品的至少一种物理属性值的 一组螺旋 CT图像;分析该组螺旋 CT图像以确定液体的螺旋 CT图像部分的装置; 以及基于液体的螺旋 CT图像部分中包含的物理属性值在相应维度的空间中的位 置来判断所述被检液态物品是否是危险品的装置。
利用本发明的方法和设备,由于釆用透射图像作为引导来进行双能 CT扫描, 在提高了检测速度的同时, 并不降低检测准确率, 并且通过透射图像还可以判断 液态物品是否存在夹层。
另外, 通过判断测量的密度与原子序数是否落在二维空间的预定区域中, 可 以判断被检液态物品是否是爆炸物。
另外, 操作员可以在任何指定的位置进行双能 CT扫描, 方便了复査操作的 进行。
另外, 在多件同时检测的情况下, 采用分格桶, 可以方便地确定是那件被检 液态物品是否是爆炸物。 附图说明
从下面结合附图的详细描述中, 本发明的上述特征和优点将更明显, 其中: 图 1是根据本发明实施方式的检查设备的结构示意图;
图 2示出了如图 1所示的计算机数据处理器 60的结构框图;
图 3示出了根据本发明第一实施方式的控制器的结构框图;
图 4 A和 4B示出了根据本发明的检查液态物品的方法的示意图;
图 5是用于说明 DR成像和 CT成像的关系的示意图;
图 6示出了 DR成像结果的一个例子;
图 7示出了 DR成像结果的另一例子; 图 8示出了根据本发明第一实施方式的液态物品检査方法的总体流程图; 图 9示出了 DR成像过程的流程图;
图 10示出了在 DR成像过程中探测和采集装置 30所采集的 DR图像数据的 排列方式;
图 11示出了对 DR图像进行处理以确定 CT扫描位置的流程图;
图 12示出了 CT成像过程;
图 13示出了在 CT成像过程中 CT投影数据的排列方式;
图 14示出了测量液体属性的过程;
图 15 A和图 15B示出了根据本发明第二实施方式的检测多件液态物品的情 况下重建的 CT图像的示意图;
图 16 A到 16K示出了在 CT成像完成之后, 如何对 CT重建图像和 /或承载 机构进行旋转使其与 CT扫描之前物体的相对位置相一致的过程;
图 17示出了在多件被检物体的情况下进行检査的流程图;
图 18示出了根据本发明第二实施方式的承载机构的俯视图;
图 19示出了根据本实施方式的分格桶的侧视图;
图 20示出了分格桶的俯视图;
图 21示出了分格桶的底视图;
图 22示出了在检査过程中如何自动检测分格桶和标记的过程;
图 23 A到 23D示出了在检测过程中旋转桶的示意图;
图 24示出了根据第三实施方式的检查过程的流程图;
图 25是说明对液态物品进行螺旋 CT扫描的示意图;
图 26A到 26M是说明对液态物品进行螺旋 CT扫描所得到的图像。 具体实施方式
下面, 参考附图详细说明本发明的优选实施方式。 在附图中, 虽然示于不同 的附图中, 但相同的附图标记用于表示相同的或相似的组件。 为了清楚和简明, 包含在这里的已知的功能和结构的详细描述将被省略, 否则它们将使本发明的主 题不清楚。 【第一实施方式】 图 1是根据本发明实施方式的检査设备的结构示意图。
如图 1所示, 根据本实施方式的检查设备包括: 发出检査用双能 X射线的射 线源 10, 诸如 X光机; 承载机构 40, 其承载被检液态物品围绕轴 Z转动, 并且 可以升降, 使得被检液态物品进入检测区域, 从而由射线源 10发出的射线能够 透过被检液态物品; 探测和采集装置 30, 它是具有整体模块结构的探测器及数据 采集器, 用于探测透射被检液态物品的双能射线, 获得模拟信号, 并且将模拟信 号转换成数字信号, 从而输出液态物品针对高能 X射线和低能 X射线的扫描数 据; 控制器 50, 它用于控制整个系统的各个部分同步工作; 以及计算机数据处理 器 60, 它用来处理由数据釆集器采集的数据, 并输出检査结果。
如图 1所示, 射线源 10置于可放置被检液态物品的承载机构 40—侧, 探测 和釆集装置 30置于承载机构 40的另一侧, 包括探测器和数据采集器, 用于获取 被检液态物品 DR数据和多角度投影数据。数据采集器中包括数据放大成形电路, 它可工作于 (电流) 积分方式或脉冲 (计数) 方式。 探测和采集装置 30 的数据 输出电缆与计算机数据处理器 60连接, 根据触发命令将采集的数据存储在计算 机数据处理器 60中。
另外, 检查设备还包括由金属制成的筒状物体通道 20, 它设置在承载机构 40上, 能屏蔽 X射线向外的辐射。 被检液态物品放置在被检物体通道中。
图 2示出了如图 1所示的计算机数据处理器 60的结构框图。 如图 2所示, 数据采集器所采集的数据通过接口单元 68和总线 64存储在存储器 61中。 只读 存储器 (ROM) 62 中存储有计算机数据处理器的配置信息以及程序。 随机存取 存储器(RAM) 63用于在处理器 66工作过程中暂存各种数据。 另外, 存储器 61 中还存储有用于进行数据处理的计算机程序。内部总线 64连接上述的存储器 61、 只读存储器 62、 随机存取存储器 63、 输入装置 65、 处理器 66、 显示装置 67和 接口单元 68。
在用户通过诸如键盘和鼠标之类的输入装置 65 输入的操作命令后, 计算机 程序的指令代码命令处理器 66执行预定的数据处理算法, 在得到数据处理结果 之后, 将其显示在诸如 LCD显示器之类的显示装置 67上, 或者直接以诸如打印 之类硬拷贝的形式输出处理结果。
图 3示出了根据本发明实施方式的控制器的结构框图。 如图 3所示, 控制器 50包括: 控制单元 51, 根据来自计算机 60的指令, 来控制射线源 10、 承载机构 40和探测和采集装置 30; 触发信号产生单元 52, 用于在控制单元的控制下产生 用来触发射线源 10、探测和釆集装置 30以及承载机构 40的动作的触发命令; 第 一驱动电机 55,它在根据触发信号产生单元 52在控制单元 51的控制下产生的触 发命令驱动承载机构 40上升或者下降; 高度信息获取单元 53, 它随着承载机构 40的运动, 向控制单元 51反馈承载机构的高度信息; 第二驱动电机 56, 它根据 触发信号产生单元 52在控制单元 51 的控制下产生的触发命令驱动承载机构 40 旋转; 角度信息获取单元 54, 它在承载机构 40旋转过程中获取承载机构 40的旋 转角度, 反馈给控制单元 51。
根据本发明的实施方式, 上述的髙度信息获取单元 53和角度信息获取单元 54都是光电码盘, 它具备抗干扰的优点。
根据本发明的实施例, 通过测量液态物品中液体的密度和原子序数, 并且判 断测量的密度和原子序数是否落在预先设定的危险区域中, 本发明的检查方法可 以用来检测危险品。 对于密度和原子序数, 安全液体 (比如饮料、 护肤品、 化妆 品等) 和危险液体 (比如易燃液体、 液体炸药、 强腐蚀性液体等) 是具有很大差 异的。 比如, 作为安全液体的水, 密度大约是 1 (特征密度 1.11 ), 原子序数大约 是 7.51。 而作为危险液体的酒精, 其密度大约是 0.79 (特征密度 0.89), 原子序 数大约是 6.47。 利用模式识别的方法, 可以根据密度和原子序数将安全液体和危 险液体分开。
图 4A和 4B示出了根据本发明的检查液态物品的方法的示意图。 在密度和原 子序数所构成的二维空间内, 划出一个或多个区域, 当被检物的密度和原子序数 落在这些区域内时, 则被判为安全液体, 反之被判为危险液体。 而安全液体区域 的确定方法是, 实测大量的安全液体和危险液体, 取安全液体较集中而危险液体 较稀疏的区域, 作为安全液体区域。 如图 4A所示, 通过实际测量各种液体的密 度和原子序数, 在密度和原子序数二维空间中确定安全液体所落的区域, 和危险 液体所在的区域。 然后, 如图 4B所示, 将安全液体所在的区域进行收缩以便提 高检测的灵敏度。 这样, 在通过上述的方法获得了液态物品的密度和原子序数之 后, 就可以根据其在二维空间上的位置来判断其是否为危险品。
图 5是用于说明 DR成像和 CT成像的关系的示意图。 根据本发明的实施方 式, 首先对液态物品进行 DR成像, 以便判断液态物品中的液体部分, 然后仅仅 针对液体部分中的某个位置进行 CT成像, 以便提高检查的速度。 图 6和图 7分别示出了 DR成像的例子。 如图 6所示, 在对某液态物品进行 DR成像之后,通过如下所述的对像素值进行分析来判断液态物品中液体的位置。 如图 6所示, 该液态物品中仅仅包含一种液体。 但是, 如图 7所示, 由于不同液 体对液体的吸收系数不同, 当液态物品包含两种或者两种以上的液体, 并且液体 中出现分层时, 通过对 DR成像后获得的 DR图像进行像素分析来判断液体之间 的分界面的位置。 然后, 针对不同的液体进行 CT成像。
图 8示出了根据本发明第一实施方式的液态物品检查方法的总体流程图。如 图 8所示, 例如在通关期间, 需要对乘客携带的液态物品进行安全检查。 首先在 步骤 S 111 , 操作员将被检液态物品放在承载机构 40上。
接下来, 在步骤 S112, 操作员按下启动按钮开始执行 DR扫描, 以生成 DR 图像, 如图 6和 7所示。
如上所述, 进行 DR扫描的目的在于: 一是获取被检液态物品的透视图像, 使操作员能够看清被检液体的内部结构; 二是软件能够根据 DR图像自动识别液 体位置, 引导下一步的 CT成像; 三是用户可以在 DR图像指定需要进行 CT成 像和检查的位置, 引导下一步的 CT成像。后面将详细描述 DR成像的详细过程。 • 必须说明的是, DR扫描并不是必须的。 为了加快检査速度, 可以不通过 DR 扫描来引导 CT扫描, 而是直接预定若干位置进行 CT扫描。 比如, 经过调研, 可知在距离瓶底 5厘米的进行 CT扫描时,可检査到大部分液体商品的液体部位, 则可使用离瓶底 5厘米的高度作为预定扫描高度。 另外, 操作员可以通过目测被 检物尺寸, 凭经验设置合适的高度, 比如设置罐状可乐的扫描高度为 3cm, 而设 置瓶底较厚的葡萄酒的扫描高度为 10cm。
在获得 DR图像后, 在步骤 S113A, 对 DR图像进行分析, 自动确定 CT成 像的扫描位置, 或者在步骤 S113B, 由操作员操作诸如鼠标之类的输入装置 65 来指定要进行 CT扫描的位置。 这样, 由于仅仅对液态物品中有代表性的位置进 行 CT扫描, 从而加快了检查速度, 而不会降低检査质量。
在确定了 CT扫描的位置之后, 在步骤 S114执行 CT扫描过程, 也就是针对 上述确定的扫描位置对液态物品进行 CT扫描, 获得 CT扫描数据, 并且根据重 建算重建 CT图像。 该 CT图像的每个像素代表了液态物品中相应部分的密度和 原子序数。
然后, 在步骤 S115, 计算机通过执行分析程序来分析 CT图像, 获取测量的 密度和原子序数。进而, 在步骤 S116, 判断实测的密度和原子序数所确定的点是 否落在了由密度-原子序数二位坐标空间中的预定区域中。 在步骤 S117, 如果密 度和原子序数所确定的点落在了预定的危险区中, 则表明该液态物品是可疑的或 者很可能是危险品, 向操作员发出报警, 或者打印出检査结果。
下面结合附图 9〜14详细说明上述各个步骤的详细操作过程。 图 9示出了
DR成像过程的流程图, 而图 10示出了在 DR成像过程中探测和采集装置 30所 釆集的 DR图像数据的排列方式。
如图 9所示, 在 DR成像过程中, 在步骤 S210, 从计算机 60向控制器 50发 送命令, 以驱动承载机构 40沿物体通道 20垂直运动。 控制器 50在承载机构垂 直运动的过程中, 通过高度信息获取单元 53实时监控承载机构的高度。
在步骤 S2i l, 控制器 50每隔一定的高度 (比如 1毫米) 给探测和釆集装置 30发送一个触发信号。 探测和采集装置 30接收到该触发信号后, 对每个探测器 的输出信号进行一次釆集, 获得高能探测数据和低能探测数据, 并保存到其内置 的缓冲器中。
在步骤 S212, 判断承载机构 40是否达到指定高度, 比如 500毫米。 如果仍 未达到, 则流程转到步骤 S210, 继续上述操作。
如果承载机构 40到达上述预定的高度, 则控制器 50不再给探测和采集装置 30发送触发信号。计算机 60从探测和采集装置 30中读取采集到的高低能探测器 信号, 排列为数据矩阵的形式, 成为 DR图像。 DR图像的每个像素, 记录了射 线穿透物体后剩余的强度, 包括低能射线强度和高能射线强度。
如上所述,后续的 CT成像的扫描位置都是在 DR图像上确定的。无论是使用 自动识别的方式, 还是人工指定的方式, 都是首先在 DR图像上取得一个行号, 然后计算机将这个行号转换为承载机构髙度, 并命令控制器 50驱动承载机构 40 运动到指定位置后, 再进行 CT成像。
由 DR成像流程可知, DR图像的每一行, 对应于一个特定的承载机构 40高 度。 假设 DR成像过程开始时承载机构高度值为 0, 成像过程中承载机构不断下 降, 每隔 h毫米触发一次釆集, 那么 DR图像中的第 m行, 其对应的承载机构高 度为 -m*h。
图 11示出了对 DR图像进行处理以确定 CT扫描位置的流程图。
在 DR图像中, 被检液态物品一般分为瓶底、 液体部分、 瓶颈、 瓶盖等几部 分, 通过图像分析技术可以将其中的液体部分 (可能有多层) 提取出来, 并确定 各层 CT扫描位置。
在步骤 S310, 对 DR图像的高低能数据进行融合和平滑, 得到噪声较小的单 值 DR图像。 例如, 高低能融合的具体方法可以是直接选择高低能数据中的一种 作为融合结果, 也可以是高低能数据的加权组合。 平滑的方法可以是使用高斯滤 波器对图像进行滤波处理。
在步骤 S311, 提取平滑后 DR图像中的被检液态物品(前景), 去除空气(背 景)。 具体方法可以是设定一个阈值, 将取值在该阈值以下的像素归为前景像素, 其他像素归为背景像素。 釆用阈值来去除背景的原因在于: 被检液态物品遮挡了 射线, 因此其对应的 DR图像素值较低 (DR图像记录的是射线的剩余强度)。
在步骤 S312, 提取平滑后 DR图像中的水平边缘像素。 具体方法可以是: 对 DR 图像的每个像素, 计算其与垂直方向上相邻像素的差异值; 如果这个差异值 大到某个阈值以上, 就认为这个像素属于水平边缘像素。
在步骤 S313 , 提取平滑后 DR图像中的水平边缘行, 水平边缘行对应于瓶底 和液体的界面、 液体和空气的界面、 瓶盖和空气的界面或者容器中多层液体之间 的界面。 具体方法可以是: 对 DR图像中沿水平方向的每一行, 统计其水平边缘 象素数和前景像素数的比例, 如果比例大到某个阈值以上 (比如 50%), 则将这 一行归为水平边缘行。
在步骤 S314, 对 DR图像进行垂直分区, 并排除非液体区域。 DR图像中的 各条水平边缘行将 DR图像分成了多个区域,从下往上依次包括:瓶底、液体(根 据密度的不同, 可能有多层)、 瓶内空气 (如果有的话)、 瓶盖等。 通过制定筛选 准则, 可以排除非液体区域, 具体准则如:
a) 在垂直方向上, 行数小于某一阈值的区域, 排除。 行数较小的区域, 也就是厚度很小的区域, 有可能是瓶底、 瓶盖、 或容器内液体和容器 顶部之间狭小的空隙(比如易拉罐内顶部的空气)。 具体阈值的确定, 可以通过调研各种液体包装容器的瓶底、瓶盖和容器内空气层厚度来 确定。
b) 在水平方向上, 各行的平均前景像素数小于某个阈值的区域, 排除。
这些区域往往对应于细长的瓶颈。 具体阈值值的确定, 可以通过实现 获取各种液体包装容器瓶颈的宽度来确定。 在步骤 S315, 确定液体区域 (可能有多个) 的 CT扫描位置, 完成液体的分 层定位。 排除非液体区域后, 剩下的区域 (可能有多个) 归为液体区域。 取这些 区域在高度方向上的中心行, 作为 CT扫描位置。
以上描述的是自动确定 CT扫描位置的过程。 但是, 在手动指定扫描位置的 情况下,操作员通过输入装置 65在显示的 DR图像上直接指定要进行 CT扫描的 行, 作为 CT扫描位置。
图 12示出了 CT成像过程,而图 13示出了在 CT成像过程中 CT投影数据的 排列方式。
如图 12所示, 在确定了 CT扫描的位置之后, 执行 CT成像过程, 也就是在 所执行的 CT扫描位置, 进行 CT成像, 生成关于被检物一个断层的密度和原子 序数图像, 以进行液体密度和原子序数的测量。 如上所述, 由于仅仅对典型的位 置进行 CT扫描, 所以可以大大节省通关时间。
在步骤 S410, 计算机 60给控制器 50发送命令, 驱动承载机构 40旋转预定 角度, 例如 1度。 控制器 50在承载机构旋转运动的过程中, 通过角度信息获取 单元 54实时监控承载机构角度。
在步骤 S411 , 当转过 1度, 控制器 50给探测和采集装置 40发送一个触发信 号。 探测和釆集装置 40接收到触发信号后, 对每个探测器的输出信号迸行一次 采集, 并保存到其内置的缓冲器中。
然后, 在步骤 S412, 判断累计转动角度是否达到一周。 如果未达到一周, 则 流程转到步骤 S410, 继续上述的操作过程。
如果累计转动角度达到指定角度 (比如 360度) 后, 在步骤 S413旋转运动 停止, 控制器 50不再给探测和釆集装置 30发送触发信号。 计算机 60从探测和 采集装置 30 中读取采集到的高低能探测器信号, 排列为数据矩阵的形式, 成为 CT投影数据, 如图 13所示。 CT投影数据的每个像素, 记录了射线穿透物体后 剩余的强度, 包括低能射线强度和高能射线强度。
在步骤 S414, 计算机 60利用双能重建算法, 根据高低能 CT投影数据, 重 建得到断层上的密度和原子序数图像, 即 CT图像。 CT图像中的每一个像素, 记 录了被检物体在该像素对应位置的密度和原子序数。
下面说明从高低能 CT投影数据重建断层图像的过程。
· CT数学原理 将二维分布 u(x,y)沿着某个方向 Θ求线积分, 便得到一维的函数 pe(t) , 该函 数称为 u(x,y)在 Θ角度的投影。 如果能够得到各个方向的投影 pe(t), 那么可以根 据 Radon变换精确计算得到二维分布 u(x,y)。 从投影得到二维分布的过程称为重 实际应用中, X光机和探测器围绕物体旋转一圈, 便测量得到物体的某个切 片的衰减系数分布在各个方向的投影, 从而可以根据 CT原理重建得到物体切片 的衰减系数二维分布。
• 基材料分解模型
在小型 X射线安全检査系统所涉及的能量范围内 (<200keV), 物质线衰减系 数可以用下面的解析表达式 (1)来近似表示。
μ(Ε) = α, Ε) + α2/κη (Ε) (1)
Figure imgf000016_0001
散射截面随着能量的变化, fp(E) 与 fKN(E)均有已知的解析表达式。 常数 &1和 a2 与物质的原子序数、 质量数和密度有关, 其表达式如 (2)和 (3)式所示, 其中 Z表 示原子序数, M表示质量数, p表示密度 (g/cm3 ), n为常数。
由于每种物质的线衰减系数都可以被公式 (1 )中的两个系数 ai和 唯一确定, 因此可以选取两种基材料, 比如碳和铝, 用基材料的线衰减系数的线性组合表示 其他所有材料, 如下式 (4)所示-
〃(E) = / , (E) + 2 (E) (4) 其中, /(E)为任意一种材料的线衰减系数, 和 / 2(E)为所选的两种基材料 的线衰减系数, !^和^称为基材料系数。
按照公式 (5),定义原子序数的 2倍与质量数的比值与密度的乘积为特征密度。
Figure imgf000016_0002
假设两种基材料的原子序数和特征密度分别为 和 (Z2, ),那么根据上 面的公式 (1)~(4)可以推导出任意一种材料的原子序数和特征密度的表达式如下 - p* = blpl* + b2p2* (6)
Figure imgf000017_0001
• 基材料投影模型
X光管产生的能谱通常为连续谱, 探测器对 X射线的能量响应函数也不是常 数。 假设能谱 N(E)与能量响应函数 Pd(E)的乘积为 S(E), 并且将 S(E)归一化,
^ S(E)dE = l (8) 那么一条投影线上的投影值的表达式为如下的积分式: χ, y)dl)dE (9)
Figure imgf000017_0002
公式 (9)中, Io和 I分别表示射线被物体衰减前和衰减后的探测器读数值, Em表示 射线的最大能量, 1表示射线穿过的路径。
公式 (9)即为实际系统的测量投影值 p与二维分布 ^χ, 的关系。 可以看出, 由于 X射线多色性, 公式 (9)并不表示 沿着某条直线的线积分, 因而并不 满足 CT数学原理的要求。常规重建算法忽略这个不一致性,重建得到的 ^ c, 图 像含有杯状的伪影, 称为硬化伪影。
现有的双能 CT方法先釆用常规重建算法得到两组 / (x, , 然后计算原子序 数和密度等信息。 但是这样不能消除射线多色性的影响。 本发明采用基材料分解 的思路解决了这个问题。
将基材料分解模型代入到公式 (9)中, 可以得到基于基材料系数的投影值表达 式: /7 = - In f m S(E) exp (- |[ (E)bx (x, y) + μ2 (E)b2 (x, y)]dl)dE (10) 将上式中的沿着路径 1的积分用下面的式子表达:
[b {x, y)dl = B, (11) [b2(x, y)dl = B2 (12) 按照上述 (1 1)、 (12)两式定义, 称81和 为基材料系数61 (^, 和 (;0)的投 影。 假设获取到了每个角度下的完整的基材料系数投影, 那么便可以根据 CT重 建理论重建得到基材料系数 1^和 b2的分布, 从而根据基材料分解模型计算出物 体的原子序数和特征密度分布, 以及任意能量下的线衰减系数值。
• 基材料系数投影的求解
双能 CT在两个能量下釆集投影数据, 便得到如下的双能投影数据:
Pl (B B2) = ~]n f Sx (E) (E) - Β2μ2 (E)]dE (13) p2 (Bv B2) = -ln f S2 (E) cxp[-B^ (E) - Βτμτ (E)]dE (14) 在测量得到 (p1 ; p2)后, 可以根据方程 (13)和(14)求解得到 (Bl 5 B2)。但是由于上 述两个方程均为对数积分方程, 无法解析求解。 非线性迭代求解方法通常计算量 很大, 并且不容易得到稳定解。
注意到, 当射线穿过厚度为 和^的基材料 1和 2后, 测量到的双能投影如 下面的表达式:
^ = -ln f 5, (E) exp[-ci1/a1 (E) - ά2μ2 (E)]dE (15)
^2 = - In f 25"2 (E) expt-c/^ (E) - ά2μ2 (E)]dE ( 16) 对比 (13)、 (14)和 (15)、 (16)可以看出, 假设测量到的投影数据对 (ph p2)相同, 那 么基材料投影数据对 (Β, , B2)与基材料的厚度组合 (φ , d2)是完全相同的。
测量不同厚度组合下的双能投影, 就可以得到双能投影数据对 (Pi, p2)与基材 料系数投影数据对 (B l 5 B2)的对应关系, 形成査找表。根据 (pl 5 p2)计算 (B1 B2)就可 以依据査找表、 利用线性插值实现, 代替复杂的求解过程。
图 14示出了测量液体属性的过程。
如图 14所示, 在步骤 S510, 对密度图像和原子序数图像进行融合和平滑, 得到噪声较小的单值 CT图像。 融合的具体方法可以是直接选择密度图像和原子 序数图像中的一种作为融合结果, 也可以是两者的加权组合。 平滑的方法可以是 使用高斯滤波器对图像进行滤波处理。
在步骤 S511, 提取平滑后 CT图像中的被检物 (前景, 包括液体及其包装), 去除空气 (背景)。 具体方法可以是设定一个阈值, 将取值在该阈值以上的像素 归为前景像素, 其他像素归为背景像素。 这是因为空气的密度和原子序数几乎为 0, 而被检液态物品的密度和原子序数则相对较大。 在步骤 S512, 在前景像素中提取液体像素。具体方法可以是: 首先建立一个 和 CT图对应的二值图像, 前景像素的取值为 1, 背景像素的取值为 0; 然后对此 二值图像进行多次腐蚀运算, 每次腐蚀一个像素, 最后取值为 1的像素为液体像 素。 这是因为液体总是在包装物的内部, 通过多次腐蚀可以逐步地去除包装物。 腐蚀的次数可以通过实现确定包装物的厚度来设定。
在步骤 S513 , 统计 CT图中所有液体像素的平均密度和平均原子序数, 作为 本次测量的输出结果。
另外, 如果在 DR图像分析过程中确定液态是分层的, 则针对每层液体执行 如上所述的步骤, 进而判断各层液体是否是可疑的。 最终, 向操作员报告最终的 检査结果。
【第二实施方式】
上面的第一实施方式描述的是每次检测单件液态物品的情况, 下面结合图
15〜18详细描述同时检测多件液态物品的操作过程。第二实施方式与第一实施方 式的区别在于, 因为存在多件液态物品, 需要在 CT成像完成之后, 在显示器上 显示的成像结果的位置与承载机构上的物体的位置相对应, 从而便于操作员弄清 那件液态物品是危险品。 图 15A和图 15B示出了根据本发明第二实施方式的检 测多件液态物品的情况下重建的 CT图像的示意图。
例如, 操作员以俯视的方式来观察承载机构上的被检物, 这就要求检查结束 后各层 CT图像上各物体的位置必须与承载机构俯视图一致。
图 16A— 16K示出了在 CT成像完成之后, 如何对 CT重建图像和 /或承载机 构进行旋转使其与 CT扫描之前物体的相对位置相一致的过程。
图 16A示出了在第一次 CT成像开始时, 承载机构 40的俯视图, 其中用箭 头指示承载机构的角度。 图 16B示出了在第一次 CT图像旋转之前的 CT图像, 它与第一次成像开始时的承载机构的俯视图是一致的, 其中的虛线标识承载机构 的旋转范围。图 16C示出了在第一层 CT图像旋转后的图像, 由于旋转角度是零, 因此实际上无需对该层 CT图像进行旋转。
图 16D示出了在第 n次 CT成像幵始时, 承载机构的俯视图。从图 16D中可 以看出, 在第 n次 CT成像开始时, 承载机构相比于第一次 CT成像前的情形, 具有一个角度。 图 16E示出了在第 n次 CT图像旋转之前的图像, 它与第 n次成 像幵始时承载机构的俯视图是一致的。图 16F示出了对图 16D中的图像进行旋转, 使其与第一层 CT图像对齐的示意图。
图 16G示出了在最后一次 CT成像开始时, 承载机构的俯视图, 可以看出此 时的承载机构相比于第一次 CT成像开始时具有一个角度。 图 16H示出了在第 N 次 CT图像旋转之前的图像,它与第 N次成像开始时承载机构的俯视图是一致的。 图 161示出了对图 16H中的图像进行旋转,使其与第一层 CT图像对齐的示意图。
图 16J示出了最后一次 CT成像结束后, 承载机构的俯视图, 其中相比于第 一次 CT成像幵始之前的情况, 具有一个旋转角度。 图 16K示出了在成像结束后 对承载机构进行旋转使其与第一次 CT成像所获得的图像一致。
在所有 CT成像过程结束后, 对各层 CT图像和承载机构进行一次角度调整。 首先, 根据各次 CT成像开始时的承载机构角度 (这个角度是可以通过角度信息 获取单元 54来得到的), 对各层 CT图像进行旋转处理, 使同一物体在各层 CT 图像中的位置保持一致, 比如对齐到第一层 CT。 然后, 调整承载机构的角度, 使承载机构的俯视图与 CT图像保持一致。
例如, 假设一共进行了 N次 CT成像, 第 n次 CT成像开始时承载机构角度 为 , 结束时承载机构角度为 , 在俯视图上承载机构以逆时针方向转动。 为 了使物体在第 n层 CT图像中的位置都与第 1层上的位置保持一致, 只需将第 n 层 CT图像按逆时针方向旋转角度 即可。 旋转图像后, 为了使物体在承载 机构俯视图与 CT 图保持一致, 只需将承载机构按逆时针方向旋转角度 360 即可。
图 17示出了在多件被检物体的情况下进行检査的流程图。 如图 17所示, 在 步骤 S711 , 操作员给系统上电, 并登录检查界面, 系统自检后进入就绪状态。 然 后, 操作员将多件被检物, 例如物体 A和物体 B直接放在承载机构 40上并按下 检查按扭。 这里, 假设物体 A放置在承载机构的右上角位置, 而物体 B放置在 承载机构的左下角位置。
接下来, 在步骤 S712,操作员按下启动按钮开始按照与如上所述相同的方式 执行 DR扫描, 以生成 DR图像。
在获得 DR图像后, 在步骤 S713A, 对 DR图像迸行分析, 自动确定 CT成 像的扫描位置, 或者在步骤 S713B , 由操作员操作诸如鼠标之类的输入装置 65 来指定要进行 CT扫描的位置。 这样, 由于仅仅对液态物品中有代表性的位置进 行 CT扫描, 从而加快了检查速度, 而不会降低检査质量。
在确定了 CT扫描的位置之后, 在步骤 S714执行 CT扫描过程, 也就是针对 上述确定的位置对液态物品进行 CT扫描, 获得 CT扫描数据, 并且根据重建算 法重建 CT图像。 该 CT图像的每个像素代表了液态物品中相应部分的密度和原 子序数。 当液体存在多层的情况下, 针对各层进行 CT扫描。
在步骤 S715和 S716, 最后一次 CT成像结束后, 按照如上所述的方法对承 载机构角度和各层 CT图像角度进行调整, 使各层 CT图像中各物体的位置保持 一致(比如对齐第一层 CT图像), 同时与物体在承载机构上的实际位置保持一致 (比如和俯视图一致), 以便于区分物体 B和物体 A。
在步骤 S717, 对每层 CT图像进行图像分割处理 (比如使用分水岭算法), 得到其中每件被检物的液体区域, 在步骤 S718统计每个液体区域内各像素的密 度和原子序数均值, 在步骤 S719判断针对各个液体区域所测量的密度和原子序 数所确的点是否落在了密度-原子系数坐标空间中的预定区域中, 据此在步骤 S720判别该件被检物的该层液体是否是危险品。
在步骤 S721 , 汇总各层检查结论并提示给操作员。一种汇总方法是: 仅当所 有 CT图像中所有液体区域都被判为安全危险品时, 本次检査的结论为"安全"; 否则本次检査的结论即为"可疑"。 另外, 对每层 CT图像进行彩色化处理并显示 给用户, CT图中被判为可疑物品的被检物使用一种颜色(比如红色〉, 判为安全 液体的使用另一种颜色 (比如绿色)。
以上说明的是同时检查两件液态物品的情况, 当要同时检査更多件液态物品 时, 为了便于操作人员看弄清楚各个液态物品的位置, 如图 18所示, 对承载机 构承载液态物品的表面上划分出多个区域, 例如图 18中的 A区、 B区、 C区和 D区, 这样操作人员可以将各件液态物品放置在相应的区域中, 并且针对各个区 域来输入各自的液体标识信息。
【第三实施方式】
为了进一步提高检测效率以及防止细长的被检液态物品直接置于承载机构 上导致检测时容易产生不稳定而影响检测效果的问题,第三实施方式提出将多件 物体放置在分格桶中。
因此, 第三实施方式与第二实施方式的区别在于, 在检査多件物体的过程中 使用了分格桶。 下面结合附图 19〜24来详细说明根据本实施方式的检测设备的 具体操作过程。
图 19示出了根据本实施方式的分格桶的侧视图, 图 20示出了分格桶的俯视 图, 而图 21示出了分格桶的底视图。
如图 19所示, 分格桶 70包括底部和与底部耦合的侧壁。 在底部的下表面上 设置有多个锥形或者其他形状的凸起。 这三个锥形凸起可以插入到承载机构 40 上对应的定位孔中, 从而在承载机构旋转过程中, 分格桶 70与承载机构 40—起 旋转, 防止在二者之间产生相对运动。
另外, 如图 19所示, 在侧壁的顶部一周设置有凸缘, 方便操作员抓握和搬 运。侧壁是圆柱状的或者是锥形的,并且可以由具有弹性侧材料,例如聚乙烯 (PE) 或者铝制成。
图 20是三种分格桶的俯视图。 如图所示, 在由底部和侧壁形成的空间中设 置有一个或者多个分格部件, 将该空间按照需要分成多个空间, 分别用于放置液 态物品。 这样, 当要同时检测多件液态物品时, 将这多件液态物品分别放置在由 分格部件所划分的空间中, 对这些液态物品同时进行检测。 在这种情况下, 可以 在侧壁的外表面上设置标记, 便于在检测时对检测桶中物品的定位。 例如, 当检 测到四件物品中有一瓶可疑时, 可以根据该件物品相对于侧壁上设置的标记的相 对位置, 来向用户报告是哪一件物品是可疑的。
例如, 如图 20的左侧图所示, 在分格部件的上侧设置有截面为预定尺寸的 圆形的标记, 或者某个分格部件的长度比其他的小, 由此作为定位液态物品所用 的标记。
图 21 是分格桶的底视图。 在底部的下表面上均匀分布有三个凸起。 但是这 三个凸起分布也可以是不均匀的。
图 22示出了在检查过程中如何自动检测分格桶和标记的过程。 如上所述, 分格筒具有特定的结构, 可以用图像处理中常用的模版匹配的方法检测。 以带十 字分格的第一种桶为例, 在步骤 S810可以首先建立一幅带有十字的模板图像, 并将此模板的中心置于待识别的 CT图像中心, 得到匹配值。
然后在步骤 S811, 判断该匹配值是否大于预定的阈值, 如果结果是否定的, 则后旋转模板图像, 直到模板图像与 CT图像的匹配程度最大为止。 如果匹配程 度大于某个预先设定的阈值, 则认为 CT图中存在分格筒, 否则认为 CT图中不 存在分格筒。
如果 CT图中存在分格筒, 可以在步骤 S812进一步根据定位标记的特点来检 测定位标记。 仍以带十字分格的第一种筒为例, 其定位标记位于一条分格线的顶 端, 使这条分格线较其他三条分格线短。 使用模板匹配法确认 CT图中存在分格 筒后, 匹配程度最大时模板图像中的十字线正好与分格线重合; 通过比较四条分 格线的长度, 提取其中的最长的一条, 即可得到定位标记。
在多层 CT成像后, 首先对每一层 CT图像进行分格筒检测和定位标记检测。 如果各层 CT图像中都没有检测到分格筒, 则认为操作员没有使用分格筒; 如果 在至少一层 CT图像中检测到分格筒, 则认为操作员使用了分格筒; 如果各层检 测到的定位标记位置不一, 可以其中信号强度最大的一个作为最终的定位标记。 描述某一层 CT中定位标记信号强度的一个方法是: 用该层四条分割线长度的最 大值减去平均值, 其差值越大, 定位标记信号越强。
图 23A-23D示出了在检测过程中旋转桶的示意图。只需将分格检测筒的定位 标记调整到预定位置,即可将各分格调整到指定位置。以第一种十字分格筒为例, 以 CT图像中心为坐标系原点建立极坐标系, 各分格的位置由定位标记的角度坐 标唯一确定。 假设系统在各种 CT图像中最终确定的定位标记所对应的极坐标角 , 而系统预先设定的调整目标是^ 那么只需将各层 CT 图像以及承载机 构旋转角度 即可完成调整, 如下图 24所示。
如图 23A和 23B所示, 将分格桶旋转一个角度, 以达到预先要求的位置。 同样, 对每层 CT成像所得到的图像进行旋转, 使其与旋转后的分格桶的角度一 致。
图 24示出了根据本实施方式的检査过程的流程图。 由于从步骤 S911到步骤 S921与上述第三实施方式的步骤 S711到步骤 S721相同,这里仅仅描述步骤 S922 到步骤 S926.
在步骤 S922, 分析 CT图像, 在步骤 S923判断中是否带有分格检测筒。 如 果没有检测筒, 则本次检査结束。
如果使用了检测桶, 则在步骤 S924识别检测筒的定位标记, 确定每个分格 的位置。 然后在步骤 S925对承载机构角度和各层 CT图像角度进行第二次调整, 将承载机构上检测筒中每个分格的位置, 以及液体各层 CT图像上每个分格的位 置, 都调整到预定的位置上。 在步骤 S926, 系统列出每层 CT图像中落在每个分格部分内的各液体区域的 检查结果, 汇总成每个分格的检查结论并提示给用户。 根据本实施方式的一种汇 总方法是: 对于某个分格, 仅当所有 CT图像中落在该分格内所有液体区域都被 判为安全液体时, 本格的检査结论为"安全"; 否则本格的捡查结论即为"可疑"。
【变型 1】
虽然以上以先进行 DR成像然后进行 CT成像为例对本发明的实施例进行了 说明, 但是也可以直接进行螺旋 CT成像, 来对液态物品进行检查。
对液态物品进行螺旋 CT检查所得到的是一组螺旋 CT图像, 通过比较和分 析该组 CT图像中的像素, 就可以确定液态物品中液体的位置, 以及液体是否分 层, 然后利用与如上描述的类似的方法获得各层位置的液体的物理属性值, 例如 密度和原子序数。例如在对如图 25所示的液态物品进行螺旋 CT成像时, 如果以 2cm为螺旋节距来进行扫描,将会得到如图 26A-26M所示的一组 CT图像。这样, 通过分析各个螺旋 CT图像中的像素, 就可以得知液态物品中液体的位置。 这里 的螺旋 CT成像可以采用大螺距 CT成像或者常规螺距 CT成像。
同样, 如本领域的普通技术人员所清楚的那样, 上述的变型实施例也可以如 第二实施方式和第三实施方式那样应用于多瓶检测。
另外, 虽然上面以密度和原子序数为例进行了说明, 但是本发明同样可以仅 仅使用密度或者原子序数之一来进行, 或者使用其他更多的物理属性来进行危险 品的识别。 在前一情况下, 在检査过程中, 通过判断一种物理属性值是否落在预 定的范围中来判定液体是否是危险品, 在后一种情况下, 在检査过程中, 通过判 断由多种物理属性值所确定的点是否落在由这些属性所限定的多维空间中的预 定区域中拉判断液体是否是危险品。 【变型 2】
虽然以上以先进行 DR成像, 然后进行双能 CT成像来获取液体的密度和原 子序数, 但是这里的 DR成像并不是必须的, 因为可以针对不同的液态物品来事 先指定预定的位置进行双能 CT成像, 来获得液体的密度和原子序数。
上面的描述仅用于实现本发明的实施方式, 本领域的技术人员应该理解, 在 不脱离本发明的范围的任何修改或局部替换, 均应该属于本发明的权利要求来限 定的范围, 因此, 本发明的保护范围应该以权利要求书的保护范围为准 <

Claims

权利要求
1、 一种用双能 CT对液态物品进行检査的方法, 包括步骤:
对被检液态物品进行双能 CT扫描, 得到双能 CT投影数据;
根据所述投影数据进行 CT重建, 得到表示被检液态物品的物理属性值的 CT 图像;
根据所述 CT图像提取被检液态物品的物理属性值; 以及
基于所述物理属性值判断所述被检液态物品是否是危险品。
2、 如权利要求 1所述的方法, 其中所述物理属性值包括被检液态物品的密 度和原子序数。
3、 如权利要求 1所述的方法, 其中所述双能 CT扫描采用平面断层 CT扫描的 方式。
4、 如权利要求 1所述的方法, 其中所述双能 CT扫描采用常规螺旋 CT扫描的 方式。
5、 如权利要求 1所述的方法, 其中所述双能 CT扫描采用大螺距螺旋 CT扫描 的方式。
6、 如权利要求 3所述的方法, 在进行平面断层 CT扫描之前, 预先设定一组 扫描位置。
7、 如权利要求 3所述的方法, 在进行平面断层 CT扫描之前, 先进行 DR扫描 得到被检物品的透射图像, 然后根据透射图像确定 CT扫描位置。
8、 如权利要求 7所述的方法, 得到透射图像后, 操作员通过输入装置指定 透射图像中的至少一行, 作为 CT扫描位置。
9、 如权利要求 7所述的方法, 得到透射图像后, 通过图像处理技术自动确 定透射图像中的至少一行, 作为 CT扫描位置。
10、 如权利要求 7所述的方法, 其中形成透射图像的步骤包括:
从射线源发出高能射线和低能射线, 穿透被检物体, 形成高能透射图像和 低能透射图像;
融合高能透射图像和低能透射图像, 形成所述透射图像。
11、 如权利要求 7所述的方法, 其中形成透射图像的步骤包括:
从射线源发出高能射线和低能射线, 穿透被检物体, 形成高能透射图像和 低能透射图像;
选择高能透射图像和低能透射图像之一, 作为所述透射图像。
12、 如权利要求 1所述的方法, 其中从双能投影数据重建表示被检液态物品 的物理属性值的 CT图像的步骤包括:
根据髙低能投影数据, 生成关于两种基材料系数的投影数据;
根据所述两种基材料系数的投影数据进行重建, 得到表示被检液体所对应 的两种基材料系数的 CT图像; 以及
根据所述表示基材料系数 CT图像, 生成所述表示被检液态物品物理属性值 的 CT图像。
13、 如权利要求 1所述的方法, 其中根据表示被检液态物品的物理属性值的
CT图像, 得到被检液态物品物理属性值的步骤包括:
从所述 CT图像中提取与液体部分相对应的像素;
对液体部分的像素计算密度均值和原子序数均值, 作为所述被检液态物品 的密度和原子序数。
14、 如权利要求 1所述的方法, 其中基于所述物理属性值来判断所述被检液 态物品是否是危险品的步骤包括:
判断所述密度和原子序数所确定的点是否落在密度-原子序数二维坐标空间 中的预定区域中;
在所述点落在所述预定区域中的情况下, 认为所述被检液态物品是危险品。
15、 如权利要求 3所述的方法, 其中在针对每个位置进行双能 CT扫描之后, 旋转被检查液态物品的 CT图像, 使其与第一次双能 CT扫描所形成的图像对齐。
16、 如权利要求 15所述的方法, 其中在针对各行进行双能 CT扫面之后, 旋 转被检液态物品, 使其与扫描之前的位置相同。
17、 如权利要求 3所述的方法, 其中所述液态物品放置在被分成了多个空间 的桶内。
18、 如权利要求 17所述的方法, 包括步骤:
利用预定的模板来自动检测桶的存在;
在存在桶的情况下, 检测 CT图像中的特定标记;
基于所述特定标记, 将桶旋转到预定的位置。
19、 如权利要求 16或 18所述的方法, 还包括步骤: 将被检液态物品的判断结果显示在显示屏上。
20、 如权利要求 16或 18所述的方法, 还包括步骤:
将各个被检液态物品的判断结果打印出来。
21、 如权利要求 16或 18所述的方法, 还包括步骤:
将各个被检液态物品的 CT图像彩色化。
22、 一种用双能 CT对液态物品进行检査的设备, 包括:
射线源, 用于发出射线;
探测和采集装置, 用于探测并采集穿透至少一件被检液态物品的射线信号; 控制器, 控制所述射线源和探测和采集装置对被检液态物品进行双能 CT扫 描, 得到投影数据;
从投影数据重建表示被检液态物品的至少一种物理属性值的 CT图像的装 置; 以及
基于所述物理属性值在相应维度的空间中的位置来判断所述被检液态物品 是否是危险品的装置。
23、如权利要求 22所述的设备, 所述双能 CT扫描是基于预定的位置进行的。
24、 如权利要求 22所述的设备, 其中所述探测和采集装置探测并釆集穿透 至少一件被检液态物品的射线信号以形成透射图像; 其中所述设备还包括指定透射图像中的至少一行的装置;
所述双能 CT扫描是基于所指定的行而进行的。
25、 如权利要求 22所述的设备, 其中所述物理属性值至少包括被检液态物 品的密度和原子序数。
26、 如权利要求 24所述的设备, 其中从射线源发出高能射线和低能射线, 穿透被检物体, 形成高能透射图像和低能透射图像, 所述设备还包括:
融合高能透射图像和低能透射图像形成所述透射图像的装置。
27、 如权利要求 24所述的设备, 其中从射线源发出高能射线和低能射线, 穿透被检物体, 形成高能透射图像和低能透射图像, 所述设备还包括:
选择高能透射图像和低能透射图像之一, 作为所述透射图像的装置。
28、 如权利要求 24所述的设备, 所述指定透射图像中的至少一行的装置包 括- 操作员利用输入装置从透射图像中选择至少一行的装置。
29、 如权利要求 24所述的设备, 其中所述指定透射图像中的至少一行的装 置包括:
分析所述透射图像的像素值, 以将所述透射图像分层的装置;
将各层中的中间行指定为要进行双能 CT扫描的行的装置。
30、 如权利要求 24所述的设备, 其中从投影数据重建表示被检液态物品的 物理属性值的 CT图像的装置包括:
融合由被检液态物品的密度所标识的密度图像和由被检液态物品的原子序 数所标识的原子序数图像作为所述 CT图像的装置;
从所述 CT图像中提取与液体部分相对应的像素;
对液体部分的像素计算密度均值和原子序数均值作为所述被检液态物品的 密度和原子序数的装置。
31、 如权利要求 24所述的设备, 其中基于所述物理属性值来判断所述被检 液态物品是否是危险品的装置包括:
判断所述密度和原子序数所确定的点是否落在密度-原子序数二维坐标空间 中的预定区域中的装置, 其中, 在所述点落在预定区域中的情况下, 认为所述被 检液态物品是危险品。
32、 如权利要求 24所述的设备, 还包括: 在针对每行进行双能 CT扫描之后, 旋转被检查液态物品的 CT图像, 使其与第一次双能 CT扫描所形成的图像对齐的 装置。
33、 如权利要求 24所述的设备, 还包括在针对各行进行双能 CT扫面之后, 旋转被检液态物品, 使其与扫描之前的位置相同的装置。
34、 如权利要求 24所述的设备, 还包括桶, 被分成了多个空间, 分别用于 放置所述液态物品。
35、 如权利要求 34所述的设备, 包括:
利用预定的模板来自动检测桶的存在的装置;
在存在桶的情况下, 检测 CT图像中的特定标记的装置;
基于所述特定标记, 将桶旋转到预定的位置的装置。
36、 如权利要求 33或 35所述的设备, 还包括:
显示装置, 显示被检液态物品的判断结果。
37、 如权利要求 33或 35所述的设备, 还包括: 将各个被检液态物品的判断结果打印出来的装置。
38、 如权利要求 33或 35所述的设备, 还包括:
将各个被检液态物品的 CT图像彩色化的装置。
39、 如权利要求 24所述的设备, 还包括承载所述被检液态物品的承载机构, 所述承载结构承载所述被检液态物品的表面上被划分成操作人员可识别的多个 区域。
40、 一种用双能 CT对液态物品进行检査的设备, 包括:
射线源, 用于发出射线;
探测和采集装置, 用于探测并采集穿透至少一件被检物体的射线信号; 控制器, 控制所述射线源和探测和釆集装置对被检液态物品进行螺旋 CT扫 描, 来形成每个均表示被检液态物品的至少一种物理属性值的一组螺旋 CT图像; 分析该组螺旋 CT图像以确定液体的螺旋 CT图像部分的装置; 以及 基于液体的螺旋 CT图像部分中包含的物理属性值在相应维度的空间中的位 置来判断所述被检液态物品是否是危险品的装置。
41、 如权利要求 40所述的设备, 其中所述物理属性值至少包括被检液态物 品的密度和原子序数。
PCT/CN2008/001654 2007-10-05 2008-09-25 Method and device for detecting liquid article WO2009043232A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP08800646.5A EP2189785B1 (en) 2007-10-05 2008-09-25 Method and device for detecting a liquid article
ES08800646T ES2531019T3 (es) 2007-10-05 2008-09-25 Método y dispositivo para la detección de un artículo líquido

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2007101806528A CN101403710B (zh) 2007-10-05 2007-10-05 液态物品检查方法和设备
CN200710180652.8 2007-10-05

Publications (1)

Publication Number Publication Date
WO2009043232A1 true WO2009043232A1 (en) 2009-04-09

Family

ID=40019947

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2008/001654 WO2009043232A1 (en) 2007-10-05 2008-09-25 Method and device for detecting liquid article

Country Status (11)

Country Link
US (2) US8036337B2 (zh)
EP (1) EP2189785B1 (zh)
JP (1) JP5164777B2 (zh)
CN (1) CN101403710B (zh)
DE (1) DE102008050306B4 (zh)
ES (1) ES2531019T3 (zh)
FR (1) FR2922020B1 (zh)
GB (1) GB2453642B (zh)
HK (1) HK1128960A1 (zh)
RU (1) RU2445609C2 (zh)
WO (1) WO2009043232A1 (zh)

Families Citing this family (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2666838C (en) 2006-09-18 2010-12-07 Optosecurity Inc. Method and apparatus for assessing characteristics of liquids
WO2009043145A1 (en) 2007-10-01 2009-04-09 Optosecurity Inc. Method and devices for assessing the threat status of an article at a security check point
CN101403711B (zh) * 2007-10-05 2013-06-19 清华大学 液态物品检查方法和设备
CN101647706B (zh) 2008-08-13 2012-05-30 清华大学 高能双能ct系统的图象重建方法
WO2010025539A1 (en) 2008-09-05 2010-03-11 Optosecurity Inc. Method and system for performing x-ray inspection of a liquid product at a security checkpoint
WO2010091493A1 (en) 2009-02-10 2010-08-19 Optosecurity Inc. Method and system for performing x-ray inspection of a product at a security checkpoint using simulation
EP2443441B8 (en) 2009-06-15 2017-11-22 Optosecurity Inc. Method and apparatus for assessing the threat status of luggage
CN101936924B (zh) 2009-06-30 2012-07-04 同方威视技术股份有限公司 物品检查系统
CN102175697B (zh) * 2009-06-30 2014-05-14 同方威视技术股份有限公司 物品检查系统
EP2459990A4 (en) 2009-07-31 2017-08-09 Optosecurity Inc. Method and system for identifying a liquid product in luggage or other receptacle
CN102455305B (zh) * 2010-11-01 2014-06-18 北京中盾安民分析技术有限公司 双能量x射线人体藏物检查设备中所使用的图像处理方法
US9247919B2 (en) * 2011-02-01 2016-02-02 Koninklijke Philips N.V. Method and system for dual energy CT image reconstruction
WO2012114199A2 (en) * 2011-02-25 2012-08-30 Smiths Heimann Gmbh Image reconstruction based on parametric models
GB2526471B (en) 2011-11-02 2016-03-02 Johnson Matthey Plc Scanning method and apparatus
RU2494377C1 (ru) * 2012-05-17 2013-09-27 Федеральное государственное бюджетное учреждение науки Институт океанологии им. П.П. Ширшова РАН Способ получения трехмерного образа пробы планктона
CN103900503B (zh) * 2012-12-27 2016-12-28 清华大学 提取形状特征的方法、安全检查方法以及设备
CN104346820B (zh) * 2013-07-26 2017-05-17 清华大学 一种x光双能ct重建方法
CN105092610B (zh) * 2014-05-14 2017-09-26 同方威视技术股份有限公司 扫描成像系统
CN105094725B (zh) * 2014-05-14 2019-02-19 同方威视技术股份有限公司 图像显示方法
CN105759318B (zh) * 2015-03-06 2020-11-06 公安部第一研究所 一种安全检测装置及其检测方法
RU2612058C1 (ru) * 2015-12-22 2017-03-02 Федеральное государственное бюджетное учреждение науки Институт ядерной физики им. Г.И. Будкера Сибирского отделения РАН (ИЯФ СО РАН) Рентгенографическая установка
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
CN105527654B (zh) 2015-12-29 2019-05-03 中检科威(北京)科技有限公司 一种检验检疫用检查装置
CN105911604B (zh) * 2016-05-30 2019-12-10 公安部第一研究所 一种通道式危险液体检测装置与方法
CN105807329B (zh) * 2016-05-30 2019-05-17 公安部第一研究所 一种用于识别包裹中危险液体的x射线检测装置与方法
FR3052555B1 (fr) * 2016-06-09 2019-06-28 Tiama Procede pour determiner la capacite de recipients en verre
US10809414B2 (en) * 2018-01-09 2020-10-20 Voti Inc. Methods for reconstructing an unknown object in a scanned image
US11335083B2 (en) * 2018-01-31 2022-05-17 Cyberdyne Inc. Object identification device and object identification method
US11977037B2 (en) 2018-10-22 2024-05-07 Rapiscan Holdings, Inc. Insert for screening tray
CN110349151B (zh) * 2019-07-16 2021-12-03 科大讯飞华南人工智能研究院(广州)有限公司 一种目标识别方法及装置
RU2758189C1 (ru) * 2021-02-01 2021-10-26 Государственное казенное образовательное учреждение высшего образования "Российская таможенная академия" Инспекционно-досмотровый комплекс
CN113159110A (zh) * 2021-03-05 2021-07-23 安徽启新明智科技有限公司 一种基于x射线液体智能检测方法
US11885752B2 (en) 2021-06-30 2024-01-30 Rapiscan Holdings, Inc. Calibration method and device therefor

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5132998A (en) * 1989-03-03 1992-07-21 Matsushita Electric Industrial Co., Ltd. Radiographic image processing method and photographic imaging apparatus therefor
US5367552A (en) * 1991-10-03 1994-11-22 In Vision Technologies, Inc. Automatic concealed object detection system having a pre-scan stage
CN1305610A (zh) * 1998-02-11 2001-07-25 模拟技术公司 用于分选目的物的计算机层析x射线照相术的装置和方法
CN1319759A (zh) * 2000-01-24 2001-10-31 模拟技术公司 使用双能量扫描信息的爆炸材料检测装置和方法
CN1779444A (zh) * 2004-11-26 2006-05-31 清华大学 一种用射线源对液体进行ct安全检测的方法及其装置
US7116751B2 (en) * 2003-04-02 2006-10-03 Reveal Imaging Technologies, Inc. System and method for resolving threats in automated explosives detection in baggage and other parcels

Family Cites Families (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5182764A (en) 1991-10-03 1993-01-26 Invision Technologies, Inc. Automatic concealed object detection system having a pre-scan stage
WO1994002839A1 (en) 1992-07-20 1994-02-03 Scientific Measurement Systems, Inc. A process for analyzing the contents of containers
US6018562A (en) * 1995-11-13 2000-01-25 The United States Of America As Represented By The Secretary Of The Army Apparatus and method for automatic recognition of concealed objects using multiple energy computed tomography
US5974111A (en) 1996-09-24 1999-10-26 Vivid Technologies, Inc. Identifying explosives or other contraband by employing transmitted or scattered X-rays
US6026171A (en) 1998-02-11 2000-02-15 Analogic Corporation Apparatus and method for detection of liquids in computed tomography data
US6078642A (en) 1998-02-11 2000-06-20 Analogice Corporation Apparatus and method for density discrimination of objects in computed tomography data using multiple density ranges
EP1062555A4 (en) 1998-02-11 2001-05-23 Analogic Corp TOMODENSITOMETRY APPARATUS AND METHOD FOR CLASSIFYING OBJECTS
RU2166749C1 (ru) * 1999-09-22 2001-05-10 Кульбеда Владимир Емельянович Рентгенографическое устройство
DE10053301A1 (de) 2000-10-27 2002-05-08 Philips Corp Intellectual Pty Verfahren zur farblichen Wiedergabe eines Graustufenbildes
JP2004177138A (ja) * 2002-11-25 2004-06-24 Hitachi Ltd 危険物探知装置および危険物探知方法
US7062011B1 (en) 2002-12-10 2006-06-13 Analogic Corporation Cargo container tomography scanning system
GB0525593D0 (en) 2005-12-16 2006-01-25 Cxr Ltd X-ray tomography inspection systems
JP4206819B2 (ja) 2003-05-20 2009-01-14 株式会社日立製作所 X線撮影装置
US7492855B2 (en) 2003-08-07 2009-02-17 General Electric Company System and method for detecting an object
US7366282B2 (en) 2003-09-15 2008-04-29 Rapiscan Security Products, Inc. Methods and systems for rapid detection of concealed objects using fluorescence
DE10347971B3 (de) 2003-10-15 2005-06-09 Siemens Ag Verfahren und Vorrichtung zur Bestimmung des Flüssigkeitstyps einer Flüssigkeitsansammlung in einem Objekt
US7634061B1 (en) 2004-03-26 2009-12-15 Nova R & D, Inc. High resolution imaging system
US7190757B2 (en) 2004-05-21 2007-03-13 Analogic Corporation Method of and system for computing effective atomic number images in multi-energy computed tomography
US7324625B2 (en) 2004-05-27 2008-01-29 L-3 Communications Security And Detection Systems, Inc. Contraband detection systems using a large-angle cone beam CT system
US7327853B2 (en) 2004-06-09 2008-02-05 Analogic Corporation Method of and system for extracting 3D bag images from continuously reconstructed 2D image slices in computed tomography
US7254211B2 (en) 2004-09-14 2007-08-07 Hitachi, Ltd. Method and apparatus for performing computed tomography
GB2452187B (en) 2004-11-26 2009-05-20 Nuctech Co Ltd Computed Tomography apparatus for detecting unsafe liquids
CN1779443B (zh) 2004-11-26 2010-09-15 清华大学 一种用放射源对液体进行安全检测的方法及其装置
DE202005004500U1 (de) 2005-03-17 2006-07-27 Tutech Innovation Gmbh Quantitativer Mikro-Computertomograph zur quantitativen Mikro-Schichtbildaufnahme
JP2007064727A (ja) 2005-08-30 2007-03-15 Ishikawajima Inspection & Instrumentation Co X線検査装置とx線検査方法
CN101071110B (zh) 2006-05-08 2011-05-11 清华大学 一种基于螺旋扫描立体成像的货物安全检查方法
US8000440B2 (en) 2006-07-10 2011-08-16 Agresearch Limited Target composition determination method and apparatus
CN1916611A (zh) 2006-09-08 2007-02-21 清华大学 集装箱多重dr/ct检测装置
CN101334369B (zh) 2007-06-29 2010-04-14 清华大学 检查液态物品中隐藏的毒品的方法和设备
CN101403711B (zh) * 2007-10-05 2013-06-19 清华大学 液态物品检查方法和设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5132998A (en) * 1989-03-03 1992-07-21 Matsushita Electric Industrial Co., Ltd. Radiographic image processing method and photographic imaging apparatus therefor
US5367552A (en) * 1991-10-03 1994-11-22 In Vision Technologies, Inc. Automatic concealed object detection system having a pre-scan stage
CN1305610A (zh) * 1998-02-11 2001-07-25 模拟技术公司 用于分选目的物的计算机层析x射线照相术的装置和方法
CN1319759A (zh) * 2000-01-24 2001-10-31 模拟技术公司 使用双能量扫描信息的爆炸材料检测装置和方法
US7116751B2 (en) * 2003-04-02 2006-10-03 Reveal Imaging Technologies, Inc. System and method for resolving threats in automated explosives detection in baggage and other parcels
CN1779444A (zh) * 2004-11-26 2006-05-31 清华大学 一种用射线源对液体进行ct安全检测的方法及其装置

Also Published As

Publication number Publication date
HK1128960A1 (en) 2009-11-13
RU2445609C2 (ru) 2012-03-20
CN101403710B (zh) 2013-06-19
DE102008050306B4 (de) 2019-09-05
EP2189785A1 (en) 2010-05-26
JP5164777B2 (ja) 2013-03-21
US20110261922A1 (en) 2011-10-27
GB2453642B (en) 2011-08-24
FR2922020B1 (fr) 2016-02-26
DE102008050306A1 (de) 2009-04-16
EP2189785A4 (en) 2011-01-26
FR2922020A1 (fr) 2009-04-10
US20100284514A1 (en) 2010-11-11
RU2008139276A (ru) 2010-04-10
JP2009092659A (ja) 2009-04-30
ES2531019T3 (es) 2015-03-09
EP2189785B1 (en) 2014-12-31
US8320523B2 (en) 2012-11-27
CN101403710A (zh) 2009-04-08
US8036337B2 (en) 2011-10-11
GB2453642A (en) 2009-04-15
GB0818056D0 (en) 2008-11-05

Similar Documents

Publication Publication Date Title
WO2009043232A1 (en) Method and device for detecting liquid article
JP5172582B2 (ja) 液体物の検査方法及び装置
JP4944892B2 (ja) 放射線を用いて液体物品に対する保安検査を行う方法及び装置
CN102608135B (zh) 在危险品检查系统中确定ct扫描位置的方法和设备
US8254656B2 (en) Methods and system for selective resolution improvement in computed tomography
CN102095663B (zh) 液态物品检查方法和设备
CN102565107B (zh) 液态物品检查方法和设备
CN102162798B (zh) 液态物品检查方法和设备
CN102435620A (zh) 液态物品检查方法和设备
CN102095664B (zh) 液态物品检查方法和设备
CN102095666B (zh) 液态物品检查方法和设备
CN102095665B (zh) 液态物品检查方法和设备
CN102590239B (zh) 在液态物品检查系统中确定ct扫描位置的方法和设备
CN102565102B (zh) 液态物品检查方法和设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08800646

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2008800646

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE