HK1140009B - Device and method for real-time mark in substance identification system - Google Patents

Description

Real-time calibration apparatus and method in substance identification system

Technical Field

The present invention relates generally to radiation imaging, and more particularly to a real-time calibration apparatus and method for a substance identification system, which can simplify the calibration process of the substance identification system and improve the stability of the material resolution of the system.

Background

The transportation is carried out by taking the container as a unit, and is a modern and advanced transportation mode. Containerization has become a major trend in international cargo transportation. Meanwhile, the use of container smuggling, the steal of firearms, Weapons, drugs, explosives, and even Weapons of Mass Destruction (WMDs) and Radioactive Distribution Devices (RDDs) has become an international nuisance that plagues governments and interferes with the normal order of international cargo transportation.

After the 911 event in the united states in 2001, the united states government began to appreciate the potential risks of shipping, with the greatest concern that WMDs and RDDs were shipped in containers into the united states. To combat this risk, the united states customs issued the "container security Initiative" (SCI) on day 17/1/2001, requiring that all foreign ports with direct access to united states ports must be equipped with non-intrusive X (gamma) ray scanning imaging equipment to perform ray scanning inspections of containers shipped to the united states. 1 year after the publication of the CSI, 18 major ports around the world joined the initiative and began operating. Under the large environment of increasing international transportation security requirements, the world customs organization has been in a concerted pass and has required 161 member countries to develop relevant container security inspection plans along the CSI model — container security inspection has become a subject of common attention all over the world.

The existing X (gamma) ray safety inspection equipment for containers mainly adopts transmission imaging, and adopts X rays to directly transmit goods to obtain transmission images of all articles covered by an X-ray path. The standard transmission imaging technology solves the problem of container visualization and is widely applied. The dual-energy transmission technology realized on the basis utilizes two X-rays with different energy spectrums to penetrate through the detected object, and the difference of output signals of the X-rays is processed to obtain the material atomic number information of the detected object. Therefore, the safety inspection level is effectively improved to a certain extent, and the container inspection requirement which is provided by CSI and developed from inspection of smuggled goods (abbreviated as 'inspection of privacy') to inspection of dangerous goods (abbreviated as 'inspection of danger') is met. The material identification system realized by the technology is successfully applied to an actual high-energy X-ray dual-energy imaging container inspection system at present, and has the advantages of high operation speed, high material resolution accuracy and the like.

However, the material identification subsystem of the current high-energy X-ray dual-energy imaging container inspection system (hereinafter referred to as the high-energy dual-energy system) has the following disadvantages.

Because the dual-energy states of each set of system are different, the system needs to be calibrated independently before being put into use so as to train a set of classification parameters suitable for the system. The calibration work is one of the most complicated works for debugging the dual-energy system at present.

The accelerator subsystem in a high energy dual energy system may experience fluctuations in the dual energy state for a variety of reasons, such as long haul transport, replacement of accessories, manual dose adjustments, and the like. If the fluctuation degree is large, recalibration is necessary to put the device into use again.

Due to the limitation of the current accelerator technology, the dual-energy states of the accelerator at different moments have fluctuation for the same accelerator, such as pulse current jitter and state fluctuation caused by AFC. Such fluctuations can cause differences in the material discrimination effect of the same substance at different times, thereby affecting image quality.

Disclosure of Invention

The invention aims to provide a real-time calibration device and a corresponding method in a substance identification system such as a high-energy dual-energy system, which simplify the calibration process of the substance identification subsystem of the high-energy dual-energy system and improve the stability of the material resolution effect of the system.

The system is mainly improved aiming at the defects of a substance identification subsystem such as a high-energy dual-energy system. The system can enable different sets of systems with energy and dosage which are not greatly different to use the same set of classification parameters, can avoid recalibration after the state of the accelerator subsystems of the same set of systems is changed, and can improve the influence on the material distinguishing effect caused by the fluctuation of the dual-energy state of the accelerator at different moments to a certain extent.

In one aspect of the present invention, a real-time calibration method in a substance identification system for identifying a substance of an object to be detected based on a set of classification parameters is provided, the method comprising the steps of: emitting a first main beam and a first auxiliary beam of rays having a first energy, and a second main beam and a second auxiliary beam of rays having a second energy; transmitting the first main ray beam and the second main ray beam through the object to be detected; enabling the first auxiliary ray beam and the second auxiliary ray beam to transmit at least one real-time calibration material block; collecting values of a first main ray beam and a second main ray beam which penetrate through a detected object to serve as dual-energy data; collecting values of a first auxiliary beam and a second auxiliary beam which transmit the real-time calibration material block as adjustment parameters; adjusting the set of classification parameters based on the adjustment parameter; and carrying out material identification processing on the dual-energy data according to the adjusted classification parameters.

According to an embodiment of the invention, the classification parameter is adjusted with an adjustment parameter every predetermined number of scans.

According to an embodiment of the invention, the at least one real-time calibration material block comprises a first block representing an organic substance, a second block representing a mixture, a third block representing an inorganic substance and a fourth block representing a heavy metal.

According to an embodiment of the invention, the first block is made of graphite, the second block is made of aluminum, the third block is made of iron, and the fourth block is made of lead.

According to the embodiment of the invention, the real-time calibration method further comprises the following steps: the energy spectrum modulation device is used for carrying out energy spectrum modulation on each ray bundle.

According to an embodiment of the invention, the first auxiliary beam is part of a first main beam and the second auxiliary beam is part of a second main beam.

According to an embodiment of the invention, the first auxiliary beam is independent of the first main beam and the second auxiliary beam is independent of the second main beam.

According to an embodiment of the invention, at least one real-time block of calibration material is arranged above, at the bottom or at the side of the first or second main beam.

According to an embodiment of the invention, each of the at least one real-time block of calibration material has a single thickness.

According to an embodiment of the invention, the at least one real-time block of calibration material has at least two respective thicknesses.

According to an embodiment of the invention, the classification parameters constitute a discretized classification curve for discriminating between the at least two substances.

In another aspect of the present invention, a real-time calibration apparatus in a substance identification system that identifies a substance of an object to be detected based on a set of classification parameters is provided, the real-time calibration apparatus including: the ray generating device emits a first main ray beam and a first auxiliary ray beam with first energy, and a second main ray beam and a second auxiliary ray beam with second energy, wherein the first main ray beam and the second main ray beam transmit the object to be detected, and the first auxiliary ray beam and the second auxiliary ray beam transmit at least one real-time calibration material block; the acquisition device acquires values of the first main ray beam and the second main ray beam which are transmitted through the detected object as dual-energy data, and acquires values of the first auxiliary ray beam and the second auxiliary ray beam which are transmitted through the real-time calibration material block as adjustment parameters; and the data processing device is used for adjusting the group of classification parameters based on the adjustment parameters and carrying out material identification processing on the dual-energy data according to the adjusted classification parameters.

According to an embodiment of the invention, the acquisition device comprises: the main detector module is approximately vertical to the central line of the first main ray beam or the second main ray beam and is used for detecting the first main ray beam or the second main ray beam after the object to be detected is transmitted; and the auxiliary detector module is approximately vertical to the central line of the first auxiliary beam or the second auxiliary beam and is used for detecting the first auxiliary beam or the second auxiliary beam after the real-time calibration material block is transmitted.

The device can be embedded in a substance identification subsystem of a high-energy dual-energy system. For a high-energy dual-energy system with higher material identification capacity, the same set of classification parameters which are calibrated in advance and are suitable for the system is used, and then the classification parameters are adjusted in real time through the system. Therefore, each set of dual-energy system is not required to be calibrated independently before being put into use, the hardware cost of the automatic calibration device is saved, and the system debugging time is saved. For a high-energy dual-energy system with slightly poor material identification capability, although calibration work before the system is put into use is not recommended to be omitted in order to achieve the best material distinguishing effect, when the dual-energy state of an accelerator of the system changes, initial classification parameters of the system can be used, and then the real-time calibration system is assisted to carry out real-time adjustment on the classification parameters. Therefore, the calibration work is avoided to be carried out again, and the maintenance time is saved.

Drawings

The above features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a schematic diagram of a high energy dual energy system;

FIG. 2 is a schematic diagram of a high-energy dual-energy system with a real-time calibration device embedded therein, wherein the dimensions of the various components are only schematic and have no practical meaning;

FIG. 3 shows a mass attenuation coefficient curve;

FIG. 4 is a schematic diagram of an energy spectrum shaping device according to an embodiment of the present invention, in which black arrows represent high energy level rays, gray arrows represent low energy level rays, and black regions represent a block of shaping material;

FIG. 5 is a schematic diagram of an automatic calibration apparatus according to an embodiment of the present invention, in which black arrows indicate high-energy range rays and gray arrows indicate low-energy range rays;

FIG. 6 is a schematic diagram of a real-time calibration apparatus;

FIG. 7 shows a flow diagram of an automatic calibration process according to an embodiment of the invention;

FIG. 8 is a schematic diagram of a real-time adjustment process for classification parameters;

FIG. 9 illustrates an alpha graph coordinate definition;

FIG. 10A shows a schematic of calibration material training data used in an automatic calibration process;

FIG. 10B shows a schematic of an alpha curve generated from calibration material training data;

FIG. 10C shows the statistical results of the calibration material training data;

FIG. 11 shows a comparison of alpha curves before and after adjustment; and

FIG. 12 is a graph showing the effect of real-time adjustment of classification parameters before and after the accelerator status is changed.

Detailed Description

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same reference numerals are used to designate the same or similar components, although they are shown in different drawings.

Methods and apparatus according to embodiments of the present invention are based on high energy dual energy data. For convenience of explanation, the high-energy X-ray dual-energy system will be referred to as high-energy X-ray and the low-energy X-ray as low-energy X-ray. The method and the equipment according to the embodiment of the invention are suitable for high-energy dual-energy systems with the energy section ranging from 3MeV to 10 MeV.

Fig. 1 is a schematic diagram of a high energy dual energy system. As shown in fig. 1, the high-energy dual-energy system includes a radiation generating device 10, an energy spectrum shaping device 40, an automatic calibration device 50, a mechanical transmission device (not shown), a data acquisition subsystem 30, a scan control computer and a data processing computer (not shown), wherein the radiation generating device 10, the mechanical transmission device, the data acquisition subsystem 30, the scan control computer and the data processing computer are basic components of the high-energy X-ray dual-energy imaging container inspection system; the energy spectrum shaping device 40 and the automatic calibration device 50 belong to a substance identification subsystem. The invention provides a real-time calibration device, which also belongs to a material identification subsystem.

According to one embodiment of the present invention, the radiation generating apparatus 10 comprises a dual energy X-ray accelerator and corresponding auxiliary equipment. The radiation generating device 10 is capable of alternately generating X-ray beams of two energies, for example, a first beam of 3MeV and a second beam of 8MeV, at a high frequency. The center line of the ray bundle is approximately perpendicular to the detection surface of the detector module of the data acquisition subsystem.

The spectral shaping means 40 comprises a spectral shaping material and corresponding auxiliary equipment. The energy spectrum shaping device should be placed between the radiation generating device 10 and the object 20 to shape the energy spectrum of the radiation output from the radiation generating device 10, such as an accelerator, before the radiation penetrates the object 20, in order to make the energy spectrum distribution more favorable for material discrimination.

As shown in fig. 4, the spectral shaping material is characterized by a large attenuation for low-energy rays and a small attenuation for high-energy rays. The better this characteristic is, the better the effect of the spectral shaping is. As long as this property can be satisfied, the material can be used as a spectrum shaping material. Based on the characteristics of the energy spectrum shaping material, the equivalent energy of the ray is improved after the energy spectrum shaping. If only the shaping material acts on the high-energy-level ray, the equivalent energy of the high-energy-level ray is improved, while the equivalent energy of the low-energy-level ray is unchanged, so that the energy difference between the double energies is enlarged, and the material resolution capability of the system is improved.

Based on this characteristic, a graphite material is selected as the shaping material. From a pure theoretical point of view, the thicker the shaping material is, the better the resolving power of the material is. However, considering statistical fluctuations, the thicker the shaping material, the greater the attenuation of the radiation, the lower the dose received by the detector, and the lower the signal-to-noise ratio of the data. Therefore, there is an optimum value for the thickness of the shaping material. This optimum value needs to be determined according to the actual conditions of the system. The spectral shaping is determined only for a certain energy level according to the distribution of the energy of the high-energy level and the low-energy level, and the left side of fig. 4 shows a schematic diagram of a spectral shaping device in the form of a turntable. Alternatively, both dual energy gears are spectrally shaped, and the right side of fig. 4 shows a spectral shaping device that can spectrally shape both dual energy gears.

The design of the spectrum shaping device 40 should be determined according to the requirements of spectrum shaping. The method can only shape the high-energy level rays, and can improve the material resolution of the system by improving the equivalent energy of the high-energy level rays to enlarge the energy difference between double energies. The energy spectrum shaping can also be carried out on high and low energy levels at the same time, which is special, and generally, the low energy level ray is near 3 MeV. It can be seen from the mass attenuation coefficient curve shown in fig. 3 that near the 3MeV energy band, the attenuation coefficients of low Z materials are close and the transformation trend is very slow. Therefore, near this energy band, the energy variation has little effect on the resolving power of the low Z material, while the attenuation coefficient of the high Z material is an inflection point near 3 MeV. This phenomenon will result in that the lead material is indistinguishable from the other materials under this energy selection. Therefore, the 3MeV low-energy level energy is also subjected to energy spectrum shaping, the low-energy part in the low-energy level energy is absorbed by the energy spectrum shaping material, the distinguishability of the high-Z material can be improved, and the low-Z material is not negatively influenced.

As shown in fig. 5, the automatic calibration device 50 includes a calibration material having a step shape and corresponding auxiliary equipment. The automatic calibration device is suitable for collecting calibration data, and classification parameters matched with the system state are obtained in real time through the processing of an automatic calibration module in a computer and are used as the input of the automatic identification module.

The calibration material includes typical materials of various classes, and at least one typical material is prepared for each class in order to ensure the calibration precision, and a plurality of materials with different equivalent atomic numbers can be prepared for each class. If the material is not ready or if the space in which the automatic calibration device 50 is placed is limited, the material of the middle category may be omitted and the automatic calibration algorithm replaced with interpolation of the data of the adjacent categories. The calibration material selection is related to the material resolution requirements of the system. Since the high-energy X-ray dual energy is required to be able to distinguish four types of organic matter, light metal, inorganic matter, and heavy metal, four typical materials are selected from the four types, namely graphite (Z ═ 6), aluminum (Z ═ 13), iron (Z ═ 26), and lead (Z ═ 82). The four materials are selected for two reasons, namely that the materials are common and both belong to simple substances and have stable properties.

Each material is designed with several steps from thin to thick. The thinnest and thickest thicknesses are determined by the material resolution range of the system. The number of the step stages is determined by the calibration precision and the space for placing the automatic calibration device.

The auxiliary device mainly provides mechanical transmission to realize positioning scanning so as to obtain dual-energy transmission data of each step of each material. At each positioning point, a plurality of columns of dual-energy transmission data are required to be scanned continuously, and more than 256 columns are recommended to be scanned, so that the influence of signal statistical fluctuation can be eliminated to a large extent.

In the height direction, the angular distribution of the X-rays received by different detectors on the detector arm support is different. The spectral distribution differs from angular distribution to angular distribution, resulting in different material resolution parameters. Therefore, considering the influence of the angular distribution of the X-rays, all detection heights can be divided into a plurality of regions, and each region is independently counted to generate classification parameters. This requires that the calibration material in the automatic calibration device 50 should cover all the detection zones of interest.

If the height of the calibration material is limited by objective factors (processing capacity, equipment space, etc.) and cannot cover all the detector modules on the arm support, a simplified way is as follows: typically, the detection heights of most interest are the locations within the container where the cargo is located, and typically the system will adjust the main beam of X-rays to be near that location. Therefore, the main beam direction of the rays is the key calibration object. The calibration material can be designed to cover only the region, the obtained dual-energy transmission data is used as parameters to be input into an automatic calibration algorithm, and classification parameters corresponding to the energy spectrum distribution in the X-ray main beam direction are generated and used as classification parameters of all detection regions. This simplified approach is within the tolerance range in the case of a small X-ray angular distribution.

The calibration material in the automatic calibration device 50 may be designed into any shape as long as the above requirements can be satisfied. In fig. 3, the number of steps and the thickness of the steps are for illustrative purposes only and do not represent an actual meaning.

The mechanical transmission device can cause the ray generation device 10 and the data acquisition subsystem 30 to move together relative to the object to be inspected in the horizontal direction. The radiation generating device 10 and the data collecting subsystem 30 may be stationary, and the object to be detected may be moving. Alternatively, the object to be examined may be stationary and the radiation generating apparatus 10 and the data collecting subsystem 30 may move together.

The data acquisition subsystem 30 mainly includes a linear array detector for detecting the rays of the dual-energy X-ray beam generated by the ray generation device 10 after passing through the object 20 to be detected, generating dual-energy transmission data, and transmitting the data to the scan control and data processing computer. The data acquisition subsystem 30 further includes a projection data readout circuit and a logic control unit on the detector. The detector can be a solid detector, a gas detector or a semiconductor detector.

The scanning control and data processing computer is responsible for the main control of the operation process of the inspection system, including mechanical control, electrical control, safety control and the like, and is responsible for processing and displaying the dual-energy transmission data obtained by the data acquisition subsystem.

In order to perform real-time calibration on the classification parameters, thereby simplifying the calibration process of the material identification system and improving the material resolution effect, a real-time calibration device 60 is introduced, and fig. 2 shows a system schematic diagram of the material identification system with the real-time calibration device 60.

As shown in fig. 6, the real-time calibration device includes real-time calibration materials 61, 62, 63, 64, a real-time calibration detector module 65 and corresponding auxiliary equipment (not shown).

The principle of the installation position of the real-time calibration device 60 is that the device can be covered by rays and does not block the ray beam of the detected object.

According to one embodiment of the invention, the installation position of the real-time calibration device should be close to the accelerator. Depending on the type of high-energy dual-energy system, the beam can be placed on the top, bottom, or side of the fan beam as required for normal scanning.

Thus, the radiation generating device 10, such as an accelerator, is required to add an auxiliary beam current at the top, bottom or side of the fan beam required for normal scanning.

In addition, according to the embodiment of the present invention, the real-time calibration material block generally includes four categories of organic matter, mixture, inorganic matter and heavy metal, and the thickness thereof is determined according to the actual system. Fig. 6 shows an organic mass 61, a mixture mass 62, an inorganic mass 63, and a heavy metal mass 64. In the high-energy dual-energy system, graphite, aluminum, iron and lead are generally selected as typical materials in the four categories. If the classification requirement on a certain material is not high or the space is too small, only 1-3 of the four materials of graphite, aluminum, iron and lead can be designed, and the missing information is obtained in a weighting mode. The real-time calibration material block needs to be matched with an automatic control device, so that the displacement of the real-time calibration material block in the horizontal direction can be controlled. The purpose is to move air and background away when the system is collecting it, preventing it from blocking the secondary beam.

The real-time calibration detector module 65 is used for collecting the attenuated ray information of all the materials of the real-time calibration material block. The detector module 65 is mounted in an orientation that requires the detection plane to be perpendicular to the secondary beam centerline. The data (hereinafter referred to as auxiliary image) collected by the detector module 65 is merged with the data of the data collection subsystem 30, and then transmitted to the data processing computer by the data collection subsystem 30 for real-time adjustment of the classification parameters for classifying the image data.

According to the embodiment of the invention, the design of the auxiliary ray beam angle, the height of the real-time calibration material block and the number of units of the real-time calibration detector module should be considered in a matched manner, the design principle is that on the acquired auxiliary image, the data of each material has a plurality of pixels, and the effective pixel of each material is recommended to be not less than 6 pixels so as to reduce the statistical fluctuation in the statistical process.

FIG. 7 shows a flow chart of an automatic calibration process according to an embodiment of the invention. Typically, after the substance identification system is installed, the automatic calibration device 50 is operated by manual triggering to obtain the original calibration data and classification parameters of the system.

As shown in fig. 7, in step S10, the radiation generating apparatus 10 generates an X-ray beam. In step S11, the X-ray beam is shaped by the spectral shaping device 40 to obtain an X-ray beam that facilitates material resolution. In step S12, when the automatic calibration process is to be performed, the automatic calibration process is triggered and executed manually, and the original calibration data is acquired in real time.

Then, in step S13, data correction processing is performed on the original calibration data to eliminate the influence of detector background, detector inconsistency, radiation dose fluctuation, and the like. At step S14, an automatic calibration algorithm is run, and classification parameters are generated and saved in a file.

As described above, each time the system status changes, the automatic calibration process is triggered manually, the automatic calibration device 50 is started, and the original calibration data shaped by the energy spectrum is collected and sent to the data processing computer by the data collection subsystem. And designing a material resolution algorithm by adopting an alpha curve method. Therefore, the purpose of the automatic calibration algorithm is to calculate the alpha curve chart classification parameters matched with the system state. And (4) obtaining the system state matched alpha curve chart classification parameters by calling an automatic calibration algorithm, and storing the system state matched alpha curve chart classification parameters in a file as parameter input of a material distinguishing module. The alpha graph coordinate definition is shown in fig. 9.

As shown in fig. 9, alphaL and alphaH are defined as follows:

alphaL＝(1-log(T_L) 1000); wherein T is_LLow energy transparency;

alphaH＝(1-log(T_H) 1000); wherein T is_HIs of high energy transparency.

Taking alphaH as the abscissa alphax of the alpha curve, and taking the difference between alphaL and alphaH as the ordinate alphay of the alpha curve:

alphax＝alphaH＝(1-log(T_H))*1000；

alphay＝alphaL-alphaH＝(-log(T_L)+log(T_H))*1000。

as described above, in step S13, the data correction module is called to perform data correction on the original calibration data, so as to eliminate the influence of the detector background, the detector inconsistency, the radiation dose fluctuation, and the like, and obtain the calibration material training data. FIG. 10A is a graph of training data for a certain detection interval on an alpha graph.

The process of generating class boundaries between types of materials from calibration material training data is described in detail below.

(i) And within a certain detection interval range, carrying out mean value statistics on the corrected dual-energy data of a plurality of rows of steps of various materials in sequence, thereby obtaining a series of mean value points of the calibration material training data. FIG. 10C is a graph of the mean value point of the training data in a certain detection interval on an alpha curve.

(ii) On the alpha curve graph 10B, a plurality of training data mean points of a certain material are connected, and then an alpha dispersion curve of the material can be obtained. However, the number of steps of the calibration material is limited, and therefore, the accuracy of the directly connected alpha dispersion curve is low. For this purpose, a least square curve fitting method (a fitting polynomial of a given data point is solved by a least square method) is adopted for curve fitting, a plurality of training data mean value points are used as input parameters, curve fitting is carried out, and fitting parameters of the curve, namely coefficients of each order of the polynomial, are obtained, wherein the times of fitting the polynomial are selected according to actual conditions. Curve fitting may also use other fitting methods such as best fit polynomials in the chebyshev sense.

(iii) Discretizing the x axis of the alpha curve, wherein the discretization precision is determined according to the requirement. Then, using the curve fitting parameters, the y-axis data corresponding to each discrete point is calculated. By the operation, a discretization alpha curve of the material is obtained.

(iv) (iv) repeating steps (ii) (iii) until discretized alpha curves for all materials are obtained.

(v) As can be seen from FIG. 10B, the alpha curve is monotonous in the atomic number direction, which is the basis of the dual energy material resolution algorithm. Therefore, obtaining discretization alpha curves of various materials, the discretization boundary of two adjacent curves can be calculated in sequence, as shown in fig. 10C.

● the four categories are classified according to equivalent atomic number: dividing Z1-10 into organic categories; z is 10-18 divided into light metal categories; z is 18 to 57 divided into inorganic substances; z > 57 into the heavy metal class. Graphite (Z ═ 6), aluminum (Z ═ 13), iron (Z ═ 26), and lead (Z ═ 82) were used as the four typical materials.

And obtaining a discretization alpha curve with the atomic number Z of 10 by weighted averaging according to the discretization alpha curve of the graphite (Z ═ 6) material and the discretization alpha curve of the aluminum (Z ═ 13) material, and obtaining the category boundary of the organic matter and the light metal. Wherein the weighted average weight can be simply calculated according to the atomic number, i.e. it is assumed that the differentiability within different atomic number ranges is the same. Although, strictly speaking, the differentiability between different atomic number ranges is different, the material resolution is weak because the high energy dual energy is different from the low energy dual energy, and only materials belonging to different classes can be distinguished, but materials with different atomic numbers cannot be distinguished accurately, so that the difference is acceptable.

● similarly, a discretization alpha curve with an atomic number Z of 18, namely a classification boundary between a light metal and an inorganic substance, is obtained by weighted average of a discretization alpha curve of an aluminum (Z ═ 13) material and a discretization alpha curve of an iron (Z ═ 26) material; the discretization alpha curve with the atomic number Z of 57 is obtained by weighted averaging according to the discretization alpha curve of the iron (Z-26) material and the discretization alpha curve of the lead (Z-82) material, and the classification boundary of the inorganic substance and the heavy metal is obtained.

(vi) Repeating the steps (i), (ii), (iii), (iv), (v) until discretized class boundaries for all detection intervals are obtained.

And storing the classification boundary data of each detection interval and each typical material in a file according to an agreed format as a classification parameter of the material distinguishing module.

As mentioned above, material resolution is a characteristic of dual-energy X-ray systems as distinguished from single-energy X-ray systems. Because the material resolving power obtained by high-energy X-ray imaging is much worse than that of a low-energy dual-energy X-ray technology, the material resolving module not only needs to consider how to correctly classify the material, but also needs to consider how to improve the material resolving effect.

The real-time calibration algorithm module is embedded in the substance identification system, is installed in the data processing computer and is mainly responsible for adjusting the classification parameters in real time before the material resolution module of the substance identification system uses the classification parameters to perform material resolution on the dual-energy transparency data obtained by the high-energy dual-energy system, so that the classification parameters are suitable for the dual-energy state of the accelerator at the time.

As shown in fig. 6, according to the embodiment of the present invention, the thickness of the real-time calibration material block has a single thickness in the radial direction, so the auxiliary detector module 65 will obtain the classification parameter for the thickness.

When the calibration material with known material properties and thickness is used for classification parameter training, the alpha data information of various material blocks of the real-time calibration device is recorded as the reference information of real-time calibration. The dual-energy state during the training of the system classification parameters is referred to as a standard state.

Fig. 8 is a schematic diagram of a real-time adjustment process of classification parameters. The process of adjusting the classification parameters in real time is described in detail below.

In step S20, the radiation generating device 10 generates alternating dual energy radiation beams, such as a first primary X-ray beam of a first energy and a second primary X-ray beam of a second energy. As described above, the radiation generating apparatus 10 also generates the first auxiliary X-ray beam of the first energy and the second auxiliary X-ray beam of the second energy in synchronization with the first main X-ray beam and the second main X-ray beam. As a further embodiment, the first and second auxiliary X-ray beams are split from the first and second main X-ray beams, respectively. In step S21, the energy spectrum shaping device 40 shapes the first main beam and the first auxiliary beam and/or the second main beam and the second auxiliary beam to expand the distance between the energy spectrums of the two beams, so as to obtain a better material identification effect.

In step S22, the first and second radiation beams illuminate the real-time calibration material blocks 61, 62, 63, 64, the auxiliary detector module 65 collects the radiation beams penetrating through the respective real-time calibration material blocks, and in step S25, the auxiliary image data for real-time adjustment of the classification curve is acquired and transmitted to the data processing computer.

In step S23, the object 20 is irradiated with the first and second primary beams, the data collection subsystem 30 collects the first and second primary beams transmitted through the object 20, and the dual-energy data of the object is obtained in step S24 and transmitted to the data processing computer.

In step S26, a real-time adjustment process of the classification parameters is performed in the data processing computer.

As described above, the n-th column of normal images is adjusted before being automatically recognized using the classification parameters. Firstly, counting the dual-energy attenuation coefficients of various materials in the auxiliary images corresponding to the normal images of the (n-m) th column to the (n + m) th column (the m value is determined by the system requirement). Then, its corresponding alpha data is calculated.

Next, the classification parameters to be used for the n normal images are adjusted in real time.

As used herein, classification parameters generally refer to alpha curve parameters for various materials in accordance with a real-time embodiment of the present invention, and a typical alpha curve is shown in FIG. 10C. There are two main ways of adjustment, and the following description will take the organic matter category as an example to illustrate the adjustment process.

First, the real-time calibration device calculates the alpha data of the graphite material block in real time to be (alphax1, alphay1), and in the standard state, the alpha data of the graphite having a low energy alpha value of alphax1 obtained from the classification curve should be (alphax1, alphay2), and the adjusted coefficient is coff _ alpha ═ alphay1/alphay2, and the alpha curve of the entire graphite is adjusted using the coefficient.

Secondly, when the automatic calibration device is used for calibration, the real-time calibration device is used for collecting data, when a classification curve is generated, the average dual-energy attenuation coefficient of the graphite material block in the auxiliary image is counted at the same time, and then corresponding alpha data (alphax2, alphay2) are calculated and stored in advance; when the image is normally acquired, the real-time calibration device is used to calculate the alpha data of the graphite material block in real time to be (alphax1, alphay1), and the adjustment coefficient is coff _ alpha (alphay1/alphax1)/(alphay2/alphax2), and the alpha curve of the whole graphite is adjusted by using the coefficient. In addition, if there are two mass thicknesses of blocks of graphite material, the calculated adjustment parameter for the smaller mass thickness 1 is coff _ alpha1, and the adjustment parameter for the thicker thickness 2 is coff _ alpha2, the adjustment principle is: adjusting the value corresponding to the mass thickness less than 1 on the alpha curve by coff _ alpha 1; adjusting the value corresponding to the mass thickness 2 on the alpha curve by coff _ alpha 2; the adjustment coefficients required for values between mass thicknesses 1 and 2 are then derived from linear interpolation of coff _ alpha1 and coff _ alpha 2. Thus, the alpha curves of aluminum, iron, and lead may be adjusted, respectively, so that adjusted classification parameters are obtained at step S27. Fig. 11 shows a schematic diagram of the adjusted classification curve.

In step S28, material resolution is performed on the nth column normal image using the adjusted new alpha curve. In order to increase the recognition speed, the classification parameters of each column are not adjusted, but several columns are adjusted once, and the adjustment method is similar according to the system requirements.

In addition, the material discrimination is performed and the gray scale fusion is performed on the dual-energy image at step S29, and the final material discrimination result is presented to the user in a colorized form by colorizing the material discrimination result and the gray scale fusion result as input to the colorizing block at step S30.

In addition, according to another embodiment of the present invention, the thickness of the real-time calibration material block may have two thicknesses, so that the classification parameters for the first thickness and the classification parameters for the second thickness may be adjusted separately, instead of adjusting all the classification parameters at different thicknesses by a single coefficient as described above, thereby further improving the accuracy of material identification.

In addition, for a system with a large space in an accelerator cabin, the real-time calibration device provided by the embodiment of the invention can be arranged below a fan-shaped beam required by normal scanning, and the lower part of a beam outlet of the accelerator is expanded, so that an auxiliary beam is positioned below the normal beam.

For a system with a small space in the accelerator cabin, the real-time calibration device provided by the embodiment of the invention can be arranged above a fan-shaped beam required by normal scanning, and the upper part of the beam outlet of the accelerator is expanded, so that the auxiliary beam is positioned above the normal beam. Or an auxiliary beam outlet is additionally arranged on the side of the beam outlet of the accelerator, and meanwhile, the material block in the real-time calibration device provided by the embodiment of the invention is arranged in the ray coverage range of the auxiliary beam outlet.

As another embodiment, the auxiliary beam can use a part of the rarely used normal beam, generally at the top of the beam, and the real-time calibration detector directly uses the detector module at the top of the arm frame of the scanning detector. In this case, the auxiliary detector module 65 is not required. The data acquisition subsystem 30 directly transmits the acquired auxiliary image data and dual-energy data to the data processing computer.

While the normal beam collects air data, the auxiliary beam also needs to collect air data. When air data is collected, the real-time calibration material block needs to be moved away so as not to shield the auxiliary ray beam.

Generally, the real-time calibration detector module needs to be installed between the accelerator and the object to be detected, and therefore, the distance between the real-time calibration detector module and the target point of the accelerator is much smaller than that between the real-time calibration detector module and the scanning detector module in the data acquisition subsystem. Saturation is easily reached if the same detector modules are used. Therefore, the dynamic range of the auxiliary detector module in the embodiment of the invention is larger than that of the detector module in the data acquisition subsystem.

As described above, each column of normal data is adjusted before being identified by the classification parameters. This results in the best recognition, but the corresponding speed of operation is slow.

According to another embodiment of the present invention, the operation speed can be increased by performing the classification parameter adjustment once every several columns, for example, 4 columns.

FIG. 12 illustrates the use of the raw classification parameters after a change in the dual energy state of the accelerator, and the material discrimination results adjusted using the present system. As can be seen from the upper part of fig. 12, after the dual-energy state of the accelerator is changed, the original classification parameters are used to perform material discrimination, and the discrimination result is erroneous. After the calibration system is implemented, the fluctuation of the dual-energy state of the accelerator can be adapted to obtain a correct material resolution result, as shown in the lower part of fig. 12. Therefore, the method and the device according to the embodiment of the invention can improve the stability of the material distinguishing effect.

The above description is only for implementing the embodiments of the present invention, and those skilled in the art will understand that any modification or partial replacement without departing from the scope of the present invention shall fall within the scope defined by the claims of the present invention, and therefore, the scope of the present invention shall be subject to the protection scope of the claims.

Claims (22)

1. A real-time calibration method in a substance identification system that identifies a substance of a detected object based on a set of classification parameters, the method comprising the steps of:

emitting a first main beam and a first auxiliary beam of rays having a first energy, and a second main beam and a second auxiliary beam of rays having a second energy;

transmitting the first main ray beam and the second main ray beam through the object to be detected;

enabling the first auxiliary ray beam and the second auxiliary ray beam to transmit at least one real-time calibration material block;

collecting values of a first main ray beam and a second main ray beam which penetrate through a detected object to serve as dual-energy data;

collecting values of a first auxiliary beam and a second auxiliary beam which transmit the real-time calibration material block as adjustment parameters;

adjusting the set of classification parameters based on the adjustment parameter; and

and carrying out material identification processing on the dual-energy data according to the adjusted classification parameters.

2. A real-time calibration method as claimed in claim 1, wherein the classification parameters are adjusted with adjustment parameters every predetermined number of scans.

3. The real-time calibration method of claim 1, wherein the at least one block of real-time calibration material comprises a first block representing an organic substance, a second block representing a mixture, a third block representing an inorganic substance, and a fourth block representing a heavy metal.

4. A real-time calibration method as defined in claim 3, wherein the first block is composed of graphite, the second block is composed of aluminum, the third block is composed of iron, and the fourth block is composed of lead.

5. The real-time calibration method according to claim 1, further comprising the steps of:

the energy spectrum modulation device is used for carrying out energy spectrum modulation on each ray bundle.

6. The real-time calibration method of claim 1, wherein the first auxiliary beam is a portion of a first main beam and the second auxiliary beam is a portion of a second main beam.

7. The real-time calibration method of claim 1, wherein the first auxiliary beam is independent of the first main beam, and the second auxiliary beam is independent of the second main beam.

8. The real-time calibration method of claim 6, wherein at least one real-time calibration material block is disposed at an upper portion, a bottom portion, or a side surface of the first or second main beam.

9. The real-time calibration method of claim 1, wherein the at least one real-time calibration material block has a single thickness.

10. The real-time calibration method of claim 1, wherein the at least one real-time calibration material block has at least two thicknesses, respectively.

11. A real-time calibration method as claimed in claim 1, wherein the classification parameters constitute a discretized classification curve for distinguishing at least two substances.

12. A real-time calibration apparatus in a substance identification system that identifies a substance of an object to be inspected based on a set of classification parameters, the real-time calibration apparatus comprising:

the ray generating device emits a first main ray beam and a first auxiliary ray beam with first energy, and a second main ray beam and a second auxiliary ray beam with second energy, wherein the first main ray beam and the second main ray beam transmit the object to be detected, and the first auxiliary ray beam and the second auxiliary ray beam transmit at least one real-time calibration material block;

the acquisition device acquires values of the first main ray beam and the second main ray beam which are transmitted through the detected object as dual-energy data, and acquires values of the first auxiliary ray beam and the second auxiliary ray beam which are transmitted through the real-time calibration material block as adjustment parameters;

and the data processing device is used for adjusting the group of classification parameters based on the adjustment parameters and carrying out material identification processing on the dual-energy data according to the adjusted classification parameters.

13. The real-time calibration apparatus of claim 12, wherein the acquisition device comprises:

the main detector module is approximately vertical to the central line of the first main ray beam or the second main ray beam and is used for detecting the first main ray beam or the second main ray beam after the object to be detected is transmitted;

and the auxiliary detector module is approximately vertical to the central line of the first auxiliary beam or the second auxiliary beam and is used for detecting the first auxiliary beam or the second auxiliary beam after the real-time calibration material block is transmitted.

14. The real-time calibration device of claim 12, wherein at least one real-time calibration material block is disposed at an upper portion, a bottom portion, or a side of the first or second main beam.

15. Real-time calibration arrangement according to claim 12, wherein the data processing means are adapted to adjust the classification parameters with an adjustment parameter every predetermined number of scans.

16. The real-time calibration apparatus of claim 12, wherein the at least one block of real-time calibration material includes a first block representing an organic substance, a second block representing a mixture, a third block representing an inorganic substance, and a fourth block representing a heavy metal.

17. The real-time calibration apparatus of claim 16, wherein the first block is composed of graphite, the second block is composed of aluminum, the third block is composed of iron, and the fourth block is composed of lead.

18. The real-time calibration device of claim 12, wherein the first auxiliary beam is part of a first main beam and the second auxiliary beam is part of a second main beam.

19. The real-time calibration device of claim 12, wherein the first auxiliary beam is independent of the first main beam, and the second auxiliary beam is independent of the second main beam.

20. The real-time calibration apparatus of claim 12, wherein the energy spectrum modulation device is used to perform energy spectrum modulation on each ray bundle.

21. Real-time calibration apparatus as claimed in claim 12, said at least one block of real-time calibration material each having a single thickness.

22. The real-time calibration apparatus of claim 12, wherein the at least one block of real-time calibration material has at least two thicknesses, respectively.