Inspection system and method for recognizing the material composition of inspected objects

MYPI2025000725A0Undetermined Publication Date: 2024-02-01OBSHHESTVO S OGRANICHENNOJ OTVETSTVENNOSTYU INDIKOM (OOO INDIKOM)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
MYPI2025000725
Authority / Receiving Office
MY · MY
Patent Type
Applications
Current Assignee / Owner
Priority Date
2022-08-01
Filing Date
2023-07-26
Publication Date
2024-02-01

AI Technical Summary

Technical Problem

Existing X-ray inspection systems are inefficient in identifying the material composition of objects due to limitations in energy spectrum analysis, low energy resolution, and slow processing speeds, leading to inaccurate material identification and long processing times.

Method used

The system employs an array of Nal(Tl) scintillation spectrometers with high energy resolution and efficient detection, coupled with a processor unit for synchronous data processing, allowing for the measurement of backscattered radiation energy distributions and comparison against reference databases using the Pearson 'chi-square' criterion to determine material composition.

Benefits of technology

This approach significantly enhances the information content and accuracy of radiographic images, enabling precise identification and display of material composition, with improved energy resolution and detection efficiency, allowing for the differentiation of materials within the energy range of 20-160 keV.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure PI2025000725_DRAWING_001
    Figure PI2025000725_DRAWING_001
  • Figure PI2025000725_DRAWING_002
    Figure PI2025000725_DRAWING_002
  • Figure PI2025000725_DRAWING_003
    Figure PI2025000725_DRAWING_003
Patent Text Reader

Abstract

An invention is proposed, that consists of a device for obtaining a radiographic image in backscattered radiation and a method for determining the material composition of an inspected object. The device contains an x-ray source with a mechanical scanning system, a detection system, a processor unit and a radiographic image display device. The detection system of the system is formed by an array of NaI(Tl)-based scintillation spectrometers and has a large total sensitive surface area. Minimization of the array elements number is achieved by coupling of scintillation crystals and photodetectors through a light guide. The material composition is recognized as a result of comparing the measured energy distributions with references and is displayed on the radiographic image by means of pixel color coding. The technical result achieved with the claimed invention consists in the increase of the radiographic image information content and in significant expansion of its functionality, which allows one, along with obtaining a radiographic image of the inspected object, to identify and display its material composition in the form of a specific image. Fig 1.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] INSPECTION SYSTEM AND METHOD FOR RECOGNIZING THE

[0002] MATERIAL COMPOSITION OF INSPECTED OBJECTS

[0003] Field of the Invention

[0004] The claimed group of inventions relates to X-ray inspection systems designed to detect objects hidden behind an optically opaque barrier.

[0005] Background of the Invention

[0006] Among x-ray inspection systems, there are devices in which the x-ray emitter and the detection system are located on the same side in relation to the inspected object. In such inspection systems, the surface of the inspected object is scanned by a pencil X-ray beam, and the intensity of the radiation scattered (reflected) by the object is registered by the detection system synchronously with the displacement of the probing beam and is used for creating a radiographic image. Inspection systems of such type can be designed both in mobile and stationary versions and have the common name “backscatter” [1], [2]-

[0007] The radiographic image of an inspected object in such inspection systems is created on the basis of the measured backscattered radiation intensity. The image contrast is determined by both the material composition of the object and the specific geometry of each region of this object, for example, by the extent of this region along the direction of the probing beam. For this reason, the material composition of an object cannot be definitely identified on the basis of intensity measurements alone. It is known [3], [4] that information about the material composition of each area of the inspected object is carried out by the energy spectrum of the scattered radiation, in addition to its intensity.

[0008] An inspection system capable of measuring not only the intensity, but also the spectrum of scattered radiation, provides, in fact, relevant information for identifying the material composition of each area of the inspected object.

[0009] Hereinafter, the term “inspected object area” denotes a fragment of an object illuminated by a probing radiation beam during the exposure time of one image element (pixel). Distinguishing the material of an object involves identification of the material in each pixel.

[0010] There are a system and a method based on the analysis of the energy spectrum of backscattered radiation for determining the relative position of materials with high and low atomic number relative to the probing beam known in the art [3]. The inspection system scans the inspected object with a pencil X- ray beam, and the radiation scattered by this object is registered by two sensors which differ by sensitivity to high-energy and low-energy parts of the spectrum. The method for determining the relative position of materials is based on the fact that the combination of the signals of these two sensors allows one to roughly estimate the energy characteristics of the scattered radiation, and as a final result, to conclude about the relative position of materials with a high and low effective atomic number along a particular line of sight. This conclusion is based on the fact that the spectrum of backscattered radiation varies depending on whether an object with a low atomic number (low Z) is located behind a layer of matter with a high atomic number (high Z) or, conversely, in front of it.

[0011] These system and method known in the art are not very informative, because they do not identify the material composition of the object, but only allow one to conclude about the relative position of objects that significantly differ in atomic number. Estimation of the energy characteristics of scattered radiation, which is based on a comparison of only two of its parts - the high- energy one and the low-energy one, is too rough and does not allow for identification of the material composition. The lack of sensitivity of the known method to variations in the energy spectrum due to the material composition of the object is additionally evidenced by the necessity for taking into account absorption in air that is shown in this invention.

[0012] A system and a method are known in the art [4], that are intended, besides a number of functions declared therein, for determining the atomic number of the material of the inspected object, and that are the closest ones to the claimed invention in terms of technical solution and principle of operation and that are taken as a prototype.

[0013] Disclosure of the Invention

[0014] The inspection system is an x-ray mobile inspection complex containing at least one sensor for determining the distance from the detectors to the surface of an object and a set of inorganic scintillation detectors, such as Nal(Tl) or organic scintillation detectors, such as polyvinyl toluene, where each detector is coupled with one or more photosensitive readers such as a photomultiplier tube or photodiode. In the very same device the use of a number of detectors based on high-purity germanium HPGe or on wide-gap dielectrics such as CdTe, CdZnTe, Hgl and the like is suggested as an option.

[0015] Signals from a set of detectors are fed to an analog-to-digital converter (ADC) and then to a processor for processing the obtained radiographic image with use of a certain distance from the surface of the object for obtaining the atomic number of the material contained in the object.

[0016] The principal disadvantage of the known device is the unacceptably large time required to identify the material composition of the inspected object by the inspection system for obtaining the expected result. The reason for this is that signals from a set of detectors are transmitted to a single analog-to-digital converter (ADC). Such a technical solution fundamentally limits the speed of a set of detectors to the speed of only one of them. Among the detectors listed in the device, only the polyvinyltoluene one has a high speed, but the energy resolution of such a detector is extremely low, which makes it unsuitable for performing spectral measurements. The spectrometric properties of other types of detectors described in the prototype are good enough for performing spectral measurements, but the speed of any of them is not sufficient for use in an inspection system.

[0017] Description of the Claimed Invention

[0018] The claimed method for identifying the material composition of the inspected object consists in the following: for the energy spectrum measured by a set of detectors, the values of the average energy and the standard deviation are calculated, which are compared with a predetermined scale, where, as stated in the prototype, a low result value corresponds to a material with low Z, and a high value of the result corresponds to a material with high Z. A disadvantage of the known method is the nonmonotonic dependence of the energy characteristics of the backscattered radiation spectrum for heavy elements on their atomic number due to the significant contribution of their characteristic radiation. For this reason, the stated claim that "a low result value corresponds to a low Z material and a high result value corresponds to a high Z material" is not generally true.

[0019] The detection system of the known device comprises a set of inorganic scintillation detectors such as Nal(Tl) or organic scintillation detectors such as polyvinyltoluene, where each detector is coupled to one or more photodetectors such as a photomultiplier tube or photodiode. Signals from a set of detectors are fed to an analog-to-digital converter (ADC) and then to a processor for processing the obtained radiographic image using a certain distance from the surface of the object for a the atomic number of the material the object consists of. In the prototype, as an option, the usage a set of detectors based on high- purity germanium HPGe or based on wide-gap dielectrics such as CdTe, CdZnTe, Hgl and the like is proposed, and the recognition of the atomic number of the material is proposed to be performed for each pixel of the image based on the calculation of the average value and the standard deviation from the measured energy spectrum of the backscattered radiation, with the results of the calculations to be compared to a predetermined scale, where their low value corresponds to a low Z material and their high value corresponds to a high Z material.

[0020] The main disadvantages of the prototype, which do not allow identifying the material composition of the inspected objects with sufficiently high accuracy, are due to the following: 1. Polyvinyltoluene used in it is characterized by too small photoelectric effect cross section like any other organic scintillation materials, therefore it has an extremely low energy resolution and, for this reason it is fundamentally unsuitable as a detector for scintillation spectrometer in the energy range (20- 160) keV ; 2. The detectors mentioned in the patent that are based on wide-gap dielectrics CdTe, CdZnTe, Hgl and the like have been known for decades, but are used in laboratory practice mainly. This is because the parameters of such detectors are not sufficiently stable, and the area of their sensitive surface does not exceed 1 cm2. These limitations are related to fundamental reasons and are well known to specialists. Moreover, tens of thousands of such detectors would be needed to build a detection system with a large sensitive surface area.

[0021] 3. The detector used that is based on high-purity germanium HPGe can operate only at temperature of liquid nitrogen. The maximum area of the sensitive surface of such detectors, that are designed to operate in the energy range below 200 keV, does not exceed 10 cm2with a detection efficiency below 10% and a cost of tens of thousands of US dollars. The main advantage of this detector over other types of detectors is its very high energy resolution, which is fractions of a percent. It is not practical to use thousands of such detectors for continuous X-ray spectrum spectrometry.

[0022] 4. The employed principle of signal transmission from a set of Nal(Tl) detectors to one analog-to-digital converter (ADC) is unsuitable for spectrometers operating with a total count rate above 106pulses / s. At a high count rate, individual signals would “overlap” each other, thus not allowing one to determine the amplitude of each signal and making usage of multiple detectors for scattered radiation spectrometry pointless.

[0023] 5. The indicated possibility of determining the atomic number of a material and the accuracy of its identification have not been disclosed.

[0024] 6. According to the Applicant, the stated claim that a low result value corresponds to a material with low Z, and a high result value to a material with high Z, is not true as well, since it is known that the average energy of the backscattered radiation spectrum for heavy elements depends nonmonotonically on their atomic numbers due to the significant contribution of their characteristic radiation.

[0025] The claimed inspection system and the method for recognizing the material composition are based on the fundamental knowledge stating that the energy dependence of the absorption and scattering of X-ray radiation is unique for each material and manifests itself both in the intensity and in the energy spectrum of the radiation scattered by the inspected object. The difference between the claimed inspection system and the already known ones is that the detection system of the former, together with the signal processing technology of this system, allows one to recognize the material composition of the inspected object, in addition to constructing a radiographic image.

[0026] For solving of most of the problems that arise during the inspection process, it is sufficient that the boundary energy of the probing beam of X-ray radiation of the source is (200 - 250) keV. In this case, the maximum energy of backscattered radiation quanta does not exceed (150 - 160) keV. Quanta with energies below (20-30) keV are strongly absorbed in the air and structural materials of the detection system and provide limited information about the inspected object. Thus, the detection system of the inspection system must ensure registration of back-scattered X-ray radiation in the energy range of (20- 160) keV.

[0027] The information content of a radiographic image increases with the increase in the number of scattered radiation quanta registered by the detection system. However, the number of such quanta, that is proportional to the intensity of the probing X-ray beam, is limited by the maximum allowable radiation dose of the object. In practice, this means that given a characteristic distance between the inspection system and an object of (1-1.5) m and a typical speed of displacement of the inspection unit relative to the object of (5-10) km / h, the maximum flux density of scattered radiation on the surface of the detection system will be approximately 107l / m2s.

[0028] The main flux of backscattered radiation, generated by the probing beam and directed towards the emitter and the detection system, lies within the solid angle (1-1.5) steradians. The optimal area of the sensitive surface of the detection system, which ensures the detection of radiation scattered within this solid angle, can be estimated at (1-2) m2.

[0029] For obtaining the best quality radiographic image, the detection system that is part of the inspection installation, must ensure high efficiency of X-ray quanta registration in the range (20 - 160) keV, in addition to the requirement of possessing sufficient area. Moreover, the detection system have to be an energy spectrometer ensuring an energy resolution of at least 20% for line 122 keV of Co57radioisotope source in order to allow one to recognize the material composition of an object.

[0030] The claimed group of inventions (inspection system and method for recognizing the material composition of inspected objects) is free from the indicated disadvantages.

[0031] The technical result achieved by the claimed invention consists in increasing the information content of the radiographic image and expanding its functionality, which allows identifying and displaying the material composition of the inspected object in the form of a specific image, along with obtaining its radiographic image.

[0032] The specified technical result is achieved due to the fact that in the inspection system containing an X-ray source with a mechanical system for scanning an object with a pencil beam of this X-ray radiation, a processor unit, a device for displaying radiographic image and a backscattered radiation registration system located next to the source, consisting of an array of detectors, in accordance with the claimed invention, each of the elements of the detector array is made as a scintillation spectrometer with an energy resolution of at least 20% for line 122 keV of Co57radioisotope source with an X-ray detection efficiency of at least 80% in the energy range of 30-150 keV and contains a Nal(Tl) scintillation detector, a photodetector, an electronic path, an amplitude analyzer that converts a sequence of scintillation signals into a sequence of digital codes, and an interface that provides transmission of digital codes from each element of the array into the processor unit synchronously with the formation of pixels of the radiographic image during the process of scanning the surface of the inspected object with a probing x-ray beam.

[0033] Moreover, the specified technical result is achieved due to the value of a sensitive surface area of the registration system that covers the X-ray flux scattered by an object towards the direction of the emitter within a solid angle of at least 1-1.5 steradians.

[0034] In addition, the specified technical result is achieved by the fact that each element of the array of spectrometers has such a sensitive surface area for which the flux of backscattered radiation quanta does not exceed (2-4) • 105 / s anyway.

[0035] At the same time, the specified technical result is achieved by coupling of Nal(Tl) scintillation detector of each spectrometer of the registration system with its photodetector through a light guide, which provides uniform collection of scintillation flash light from an arbitrary point of the Nal(Tl) detector.

[0036] The specified technical result is also achieved by the claimed method for recognizing the material composition of the inspected object, which consists in measuring the energy distributions of backscattered radiation synchronously with the formation of pixels of the radiographic image and then comparing the measurement results for each image pixel with each of the reference database results available in the processor unit, in which, in accordance with the claimed invention, for each comparison, Pearson's “chi-square” goodness-of-fit criterion is calculated, the minimum value of which determines whether the material of a given pixel belongs to one or another group of materials.

[0037] In order to ensure the fundamental possibility of identifying the material composition of an inspected object, the inspection system should include a detection system with a sensitive surface area of (1-2) m2that has high efficiency of detection of X-ray radiation in the energy range (20- 160) keV and that is capable of measuring of such radiation energy spectrum at a count rate of at least 1071 / s. The combination of such parameters of the detection system is necessary also to ensure the statistical significance of the characteristics of the registered backscattered radiation.

[0038] Within the energy range (20-160) keV mentioned above, it is detectors based on inorganic scintillation materials that can provide detection efficiency close to 100%. Based on the combination of such properties as a high atomic number, speed, high conversion efficiency, transparency to intrinsic radiation, cost, availability of large-area samples, etc., Nal(Tl) is the optimal material for a detection system. This material is widely used for constructing X-ray spectrometers and is well known to those skilled in the art. For ensuring the detection efficiency close to 100% in the energy range below 150 keV, it is sufficient to use a Nal(Tl) crystal with a thickness of at least 10 mm.

[0039] Since the maximum radiation flux density on the surface of the detection system can reach 107- l / (m2s), and the maximum count rate of the spectrometer with the Nal(Tl) detector does not exceed (2-4) IO51 / s, a detection system of an inspection system with sensitive surface of 1 m2or more, that is capable of measuring the spectral composition of X-rays with high detection efficiency, should be an array of scintillation spectrometers based on Nal(Tl) detectors. The number of elements of this array is chosen in such a way that the maximum flux of detected radiation quanta on the sensitive surface of each element would not exceed (2-4)- 1051 / s. This condition limits the maximum area of the sensitive surface of the spectrometer to (300-400) cm2.

[0040] The detection system of the claimed system is an array of scintillation spectrometers based on Nal(Tl), combined into a single system (see the example of an array of three spectrometers in the scheme shown in Fig. 1.). Each of the spectrometers (1) included in the detection system is connected to an interface unit (2). Also the rotary encoder (3) of the mechanical device forming pencil X- ray beam is connected to the interface unit (2). A processor unit (4) is connected to the interface unit (2).

[0041] The design of the scintillation spectrometer ( 1 ), which is a single element of the detection system, is schematically shown in Fig.2.

[0042] The scintillation detector (5) is in optical contact with the photomultiplier tube (6). The output of the photomultiplier is connected to the input of the electronic path (7), which is designed to amplify the amplitude of the pulse signals from the output of the photomultiplier. The output of the electronic path is connected to the input of the amplitude analyzer (8). The amplitude analyzer is designed to converts the amplitude of each pulse signal into a digital code that is proportional to this amplitude. Other functional units of a typical scintillation spectrometer, such as a control unit, a communication line transceiver, power supplies, etc., are not shown in Fig. 2 explicitly, since a detailed description of their operation and a specific implementation are not essential for the claimed device.

[0043] The operation of the claimed device carried out as follows.

[0044] An X-ray quantum, due to interaction with the working material of the detector (5), creates a flash of light in it, that is proportional to the energy released in the detector. The photodetector (6) converts the light flash into an electrical impulse. The duration of this pulse is determined by the duration of the light flash, and its amplitude is proportional to the flash amplitude. The electrical impulse is amplified by the electronic path (7) and fed to the input of the amplitude analyzer (8). The amplitude analyzer converts the amplitude of each pulse signal into a digital code proportional to the amplitude of this signal. This digital code is transmitted to the output of the spectrometer (1).

[0045] The rotary encoder (3) of the device for scanning an object with a probing X-ray beam generates logical electrical signals that are determined by the direction of the probing X-ray beam at each moment of time. The signals from the rotation angle sensor are transmitted to the processor unit (4) via the interface unit (2). The processor unit (4), while receiving these signals, continuously generates successive time intervals corresponding to the periods of scanning of a single element of the inspected object - a pixel of the radiographic image. Each digital code from any of the spectrometers (1), independently of each other, enters through the interface unit (2) into the processor unit (4). The set of digital codes accumulated in the processor unit (4) during the scanning of each pixel of the radiographic image is an energy spectrum that is used to identify the material composition of each single element of the inspected object. The maximum count rate of the detection system shown in Fig. 1 equals the sum of the limiting count rates of the spectrometers included in this system. At the same time the limiting count rate of each of the spectrometers (1) is defined by the speed of operation of the amplitude analyzer (8) located in the spectrometer. In turn, the speed of an amplitude analyzer, regardless of its circuit or software implementation, is limited by the duration of the scintillation flash in the detector, which imposes a limit on the maximum count rate of the entire spectrometer. For a spectrometer with a Nal(Tl) detector, regardless of the area of the sensitive surface of the detector, the limiting count rate does not exceed (2-4) • 1051 / s.

[0046] Moreover, the mentioned result is achieved by covering of the X-ray flux scattered by the object in the direction of the emitter within a solid angle of at least 1-1.5 steradians by the area of the sensitive surface of the detection system of the inspection unit.

[0047] In addition, the claimed technical result is achieved by the fact that the energy resolution of each of the scintillation spectrometers that form the detection system is not worse than 20% for line 122 keV of Co57radioisotope source with an X-ray detection efficiency of at least 80% in the energy range of 30-150 keV.

[0048] Moreover, the specified technical result is achieved by optimization of the energy resolution and the sensitive surface area of the spectrometers included in the detection system. The best energy resolution in scintillation spectrometers is achieved if the area of the output surface of the scintillator does not exceed the area of the sensitive surface of the photodetector. In this case, for Nal(Tl) crystals coupled with a photomultiplier, the typical value of the energy resolution measured for line 122 keV can reach (8.5 - 9)%.

[0049] The maximum area of the sensitive surface of industrially produced photodetectors (photomultipliers) does not exceed 200 cm2, which is significantly less than the maximum allowed area of the sensitive surface of a single element of the detection system (300-400) cm2. Direct coupling (Fig. 3a) of the scintillation detector (5) with an area significantly larger than the area of the sensitive surface of the photomultiplier tube (6) leads to a significant nonuniformity of the scintillation signal over the spectrometer surface and, as a result, to an unacceptable degradation of the energy resolution. Reducing the sensitive surface of the scintillation crystal to the size of the sensitive surface of the photodetector would lead to an increase in the total number of spectrometers in the detection system and, accordingly, to a significant increase in the cost of this system.

[0050] The number of spectrometers in the detection system can be optimized by coupling the large-area scintillation detector (5) to the photodetector (6) via the light guide (9) (Fig. 3b). Despite a decrease in the amplitude of the scintillation signal during the use of a light guide, the light guide provides better uniformity of the scintillation signal over the entire area of the detector and reduces the degradation of energy resolution due to employing a large detector area.

[0051] The area of the input surface of the light guide should not be less than the area of the exit surface of the scintillator, and the area of the exit surface of the light guide should not exceed the area of the sensitive surface of the photodetector.

[0052] Curve (10) in Fig. 4 illustrates the non-uniformity that occurs when a Nal(Tl) detector with a diameter of 200 mm is coupled to a photodetector with a diameter of 160 mm.

[0053] Curve (11) in Fig. 4 illustrates the improvement in uniformity due to the use of a light guide.

[0054] By making a rational choice of the ratio of the detector and the photodetector areas, as well as of the size of the light guide, one can reduce the number of spectrometers in the detection system to an acceptable number of (20- 30) pieces, while maintaining the energy resolution at an acceptable level - no worse than 20% for line 122 keV. The technical result achieved in the claimed device is also achieved by the method for recognizing the material composition of the inspected object.

[0055] The energy distributions accumulated in the processing unit (4) shown in Fig. 1, for each pixel of the radiographic image of the object, carry information about the material composition of the pixel, but the statistical significance of these distributions is insufficient for determining the material with an accuracy of one atomic number.

[0056] The low statistical significance of the measurement results does not allow one to distinguish between materials with close atomic numbers, but leaves the possibility of distinguishing groups of materials, such as organic materials, inorganic materials, and heavy metals.

[0057] For each of these groups of materials, calibration measurements with known materials are performed. The averaged results of calibration measurements are reduced to a library of references, are stored in the processor unit of the inspection system and used to evaluate the measurement results.

[0058] Identification of the material composition for each pixel of the radiographic image of the object is carried out by a non-parametric method of statistical hypothesis testing, which consists in comparing the measurement result with each of the library references and calculating the Pearson “chi- square” goodness-of-fit criterion for each reference. Based on the numerical value of the goodness-of-fit criterion, for a given level of significance, a conclusion is made about the probability that the material of a given pixel belongs to one or another group of materials.

[0059] The digital code of the identified group of materials is assigned to the pixel, together with a digital code proportional to the intensity of the detected radiation. Upon visualizing a radiographic image of an object, the brightness of a pixel is proportional to the intensity of the radiation, and its hue is determined by the digital code of the identified group of materials. An experimental verification of the claimed method for identifying the material composition of the inspected object has been implemented on a prototype of the claimed inspection system with a detection system of 24 spectrometers with a total area of 0.7 m2. The supply voltage of the X-ray tube during all measurements was 225 kV.

[0060] For creating a library of references, measurements have been performed with test samples made of polystyrene, aluminum alloy, steel, tungsten and lead. The test samples were placed at distance of 1.5 m from the inspection system and moved relative to the system at different velocity within the range (1.5 - 3) m / s. The reference energy distributions measured for each material and averaged over 106pixels are shown in Fig. 5. In this figure, curve (12) corresponds to a polystyrene test sample, curve (13) corresponds to an aluminum alloy test sample, curve (14) corresponds to a steel test sample, curve (15) corresponds to a tungsten test sample, and curve (16) to the lead test sample.

[0061] Table 1 summarizes the average values calculated from the energy distributions shown in Fig. 5. The table indicates that the average energy over the spectrum varies nonmonotonically in dependence on the atomic number of the material.

[0062] Table 1

[0063] Average energies of backscattering spectra in dependence on material of the test sample

[0064] Identification of the material composition using the claimed device and the claimed method had been experimentally verified in several series of measurements. In each of these series, the simulator of inspected object was located at distance of (1-2) m from the inspection system and moved relative to it at velocity within the range of (1.5-3) m / s.

[0065] Embodiment No. 1

[0066] The practical implementation of the claimed method for identifying the material composition of an inspected object consists in placing plates of various thicknesses from a particular material - polystyrene or aluminum or steel or lead - on the surface of this object. The measurement results were processed automatically in accordance with the claimed method. Each material, regardless of the thickness of the plates, was displayed on the radiographic image in its own color, corresponding to each of the 3 identified groups - organic materials, inorganic materials, heavy metals.

[0067] Embodiment No. 2

[0068] Demonstrates the implementation of the claimed method for identifying the material composition of an inspected object, which is expressed in simultaneous placing samples made of various materials: polystyrene, aluminum, steel, tungsten, lead - on the surface of the inspected object. On the radiographic image, as a result of the claimed method’s performance, each material was displayed in its own color.

[0069] Embodiment No. 3

[0070] An illustration of the claimed method for identifying the material composition of an inspected object implementation is presented in Table 2, which provides the typical numerical results of the “chi-square” goodness-of-fit criterion, averaged for each material over 9 pixels of the radiographic image. These values were obtained automatically in accordance with the claimed method. The energy spectra shown in Fig. 5 were used as reference distributions, and the measurement results are taken from Embodiment No. 2.

[0071] Table 2

[0072] The values of the “chi-square” criterion

[0073] Table 2 shows that for each material, the lowest goodness-of-fit criterion values correspond to cases where the reference measurements are compared with a test sample from the same material. This means that the test material and the standard material are the same, thus confirming that the claimed technical result allows one to identify the inspected material composition by the claimed invention.

[0074] Embodiment No. 4

[0075] Shows the practical implementation of the claimed method for identifying the material composition of the inspected object, which consists in series of measurements with samples of polystyrene, wood, aluminum, steel, tungsten and lead, that were covered from the probing X-ray beam with a plastic sheet 2 mm thick. On the radiographic image, each material was displayed in color corresponding to its attribution to each of the 3 selected groups - organic materials, inorganic materials, heavy metals. List of information sources used

[0076] 1. Daniel J. Strom, W.Wade Sapp, Josepf Callerame Physics, Health Physics, and Applications of Backscatter X-ray Imaging Technology. PNNL- SA-36973.

[0077] 2. Swift R. D. Mobile X-ray backscatter imaging system for inspection of vehicles / / Physics-Based Technologies for the Detection of Contraband. - SPIE, 1997. - V. 2936. - P. 124-132.

[0078] 3. US8442186, B2

[0079] 4. WO2012109273A3 (prototype for the device and for the method)

[0080] Note:

[0081] In the domestic and foreign scientific and technical literature, a list of specific terms related to the processes of measuring parameters of ionizing radiation fields has been formed to date.

[0082] Terms used in the claimed invention:

[0083] Detector (of ionizing radiation) - a sensitive element of a measuring instrument designed for converting the energy of ionizing radiation into another form of energy suitable for registration or further conversion and (or) measurement of one or more quantities that characterize the radiation affecting the detector.

[0084] Scintillation detector (of ionizing radiation) - a radioluminescent detector that uses a material that emits light quanta under the action of ionizing radiation.

[0085] Ionizing radiation (energy) spectrometer - a device designed to measure the energy distributions of ionizing radiation fields.

[0086] Scintillation spectrometer (of energy) - a spectrometer, the sensitive element of which is a scintillation detector.

Claims

Claims1. An inspection system containing an X-ray source with a mechanical device for scanning an object with a pencil beam of this X-ray radiation, a processor unit, a device for displaying radiographic image and a backscattered radiation registration system located next to the source, consisting of an array of detectors, wherein each of the elements of the detector array is made as a scintillation energy spectrometer with an energy resolution of at least 20% for line 122 keV of Co57radioisotope source with an X-ray detection efficiency of at least 80% in the energy range of 30-150 keV and contains a Nal(Tl) scintillation detector, a photodetector, an electronic path, an amplitude analyzer that converts a sequence of scintillation signals into a sequence of digital codes, and an interface that provides transmission of digital codes from each element of the array into the processor unit synchronously with the formation of pixels of the radiographic image during the process of scanning the surface of the inspected object with a probing x-ray beam.

2. The inspection system of claim 1, wherein the registration system has a sensitive surface area that covers the X-ray flux scattered by the object towards the emitter within a solid angle of at least 1-1.5 steradians.

3. The inspection system of claim 2, wherein each element of the array of spectrometers has such a sensitive surface area for which the flux of backscattered radiation quanta does not exceed (2-4) ■ 105 / s anyway.

4. The inspection system of claim 1, wherein the Nal(Tl) scintillation detector of each spectrometer of the recording system is coupled with its photodetector through a light guide, which provides uniform collection of scintillation flash light from an arbitrary point of the Nal(Tl) detector.

5. A method for recognizing the material composition of the inspected object, which consists in measuring the energy distributions of backscattered radiation synchronously with the formation of pixels of the radiographic imageand then comparing the measurement results for each image pixel with each of the reference database results available in the processor unit, wherein for each comparison, Pearson's “chi-square” goodness-of-fit criterion is calculated, the minimum value of which determines whether the material of a given pixel belongs to one or another group of materials.