Calibration method and device for back scattering imaging system

Abstract

The invention discloses a back scattering imaging system calibration method and device, and relates to the technical field of nuclear radiation imaging, and the method comprises the steps: enabling a pencil beam of a front collimator and a detector plane to be equivalent to straight lines in a two-dimensional plane, taking a slit of a rear collimator as a fixed point, and constructing a projective geometry optical path mapping model; a reference point which can be realized physically is introduced to establish an imaging system coordinate reference system, and projective mapping of an object space beam straight line and a detector plane is converted into a reciprocal linear relation under a local coordinate system; taking a tungsten plate as a scattering target and controlling the tungsten plate to move along the depth, collecting back scattering response data of each depth, and calculating a response center coordinate of the detector through a gravity center method; a mapping function is established through linear fitting, and geometric mapping calibration in the depth direction and depth geometric distortion correction of the back scattering image are achieved. Measurement and data collection are simplified, efficient calibration and deviation correction are achieved, and the problems that a traditional method depends on high-precision equipment and is poor in adaptability are solved.

Description

Technical Field

[0001] This invention relates to the field of nuclear radiation imaging technology, and in particular to a calibration method and apparatus for a backscatter imaging system. Background Technology

[0002] X-ray backscattering imaging technology uses the backscattered signals generated by the interaction of X-rays with matter to create images. By detecting Compton scattered photons, it can accurately obtain information about the near-surface structure of a target. After establishing the correlation between detector pixels and imaging spatial depth, it can also reconstruct the shallow three-dimensional structure of the object's surface.

[0003] Such imaging schemes typically use a front collimator to confine the X-ray beam into a pencil shape, a rear collimator to constrain the incident angle of the backscattered signal, and then a pixel detector to collect the scattered signals at different depths. Three-dimensional backscattered imaging requires establishing a correspondence between "detector pixel position and spatial depth position." Depth calibration is a key step in achieving three-dimensional depth-resolved imaging and also a core technical challenge.

[0004] Existing depth calibration methods are mainly divided into three categories: First, analytical / inverse methods based on physical models, which rely on Compton scattering physics and radiative transfer equations to construct a forward model and fit key parameters through experimental data. This method is theoretically sound but sensitive to model simplification and the inversion solution is unstable. Second, empirical calibration methods based on experimental data, which establish an empirical relationship between signal characteristics and depth by measuring standard calibration blocks. However, this method heavily relies on calibration samples and specially designed targets, resulting in large sampling volumes and poor versatility. Third, machine learning-driven methods, which use simulated or measured data to train models to learn depth features. This method can handle complex nonlinear relationships but requires massive amounts of high-quality data, has weak model interpretability, and requires the re-collection of a large amount of test data after device adjustments.

[0005] Furthermore, during actual installation, the relative poses of the front collimator, rear collimator, and detector inevitably deviate, causing the theoretical geometric mapping relationship to differ from the actual situation and resulting in imaging distortion. Existing technologies mostly rely on precision assembly or complex 3D pose solving methods, making it difficult to achieve high-precision, low-complexity, and high-efficiency depth calibration in engineering scenarios.

[0006] Therefore, there is an urgent need to propose a backscatter imaging depth calibration method with low measurement requirements, high calibration efficiency, and strong engineering adaptability to address the aforementioned shortcomings of existing technologies. Summary of the Invention

[0007] The main objective of this invention is to provide a calibration method for a backscatter imaging system.

[0008] Another object of the present invention is to provide a calibration device for a backscatter imaging system.

[0009] The third objective of this invention is to provide an electronic device.

[0010] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0011] To achieve the above objectives, a first aspect of the present invention provides a calibration method for a backscatter imaging system, comprising: S1, regard the pencil beam formed by the front collimator and the detector plane as two straight lines in a two-dimensional plane, and the slit of the rear collimator as a fixed point in the two-dimensional plane to construct a light path mapping model based on projective geometry. S2, introduce a physically achievable reference point, establish an imaging system coordinate reference system through this reference point, and transform the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system; S3, based on the imaging system coordinate reference system, uses a front collimator to collimate the X-ray source into an approximate pencil beam, uses a tungsten plate as a scattering target, and controls the scattering target to move along the depth to a preset position, collects backscattering response data at each depth, and calculates the corresponding detector response center coordinates using the centroid method; S4, based on the acquired depth position and detector response center coordinate data, establishes a mapping function between depth and detector position through linear fitting, realizes the geometric mapping relationship calibration in the depth direction, and completes the depth geometric distortion correction of the backscattered image.

[0012] Optionally, the optical path mapping model based on projective geometry also includes a complete optical path structure for backscatter imaging. This model uses a preset point-to-point projective mapping relationship to make a point on the beamline in the object space correspond to a point on the line in the image space. The correspondence logic is determined by the projection mapping between lines defined by a fixed point in projective geometry.

[0013] Optionally, the step of introducing a physically achievable reference point, establishing an imaging system coordinate reference system through this reference point, and transforming the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system, further includes: Introducing a physically achievable benchmark ,in This is the reference depth position of the scattering target. Number the detector pixel column corresponding to the reference depth position, based on the reference point. Establish a local coordinate reference system for the imaging system with the origin as the reference point. In the local coordinate reference system, unit vectors are defined respectively. , Distance parameters , auxiliary vector , ,in, The unit vector representing the direction of the detector. Represents the unit vector in the depth direction. This indicates the distance of the detector's position relative to the reference point. This represents the distance between the depth of the scattering target and the reference point. This represents the fixed point of the projection of the slit collimator onto a two-dimensional plane. Calculation using two-dimensional cross product , , , To obtain the fractional relation Taking the reciprocal of the expression and rearranging, we get... and combined The reciprocal linear relationship in the local coordinate system is obtained. .

[0014] Optionally, the pencil beam is defined from a geometrical optical perspective as having a lateral dimension. Angular divergence A highly directional photon stream in which the energy and direction within the beam remain almost unchanged laterally; The engineering criterion is that, at any depth within the imaging range, the diameter r of the spot on the target surface illuminated by the pencil beam satisfies... ,in To determine the detector's position resolution, this calibration method is used as an aid in the determination. If the mapping curve between depth d and detector position t during the calibration process conforms to a linear model, it is considered to be an approximate pencil beam.

[0015] Optionally, the step of using a tungsten plate as a scattering target and controlling the scattering target to move along the depth to a preset position further includes: A scattering target is prepared using a thin tungsten plate. The thickness of the tungsten plate in the depth direction is 2 mm, and the area in the vertical depth direction covers the irradiation range of the pencil beam. The area in front of the tungsten plate is set as an air region. By utilizing the high attenuation coefficient and high scattering cross section of the tungsten plate, backscattered photons are mainly generated at the corresponding positions on the front surface of the tungsten plate. The tungsten plate scattering target is fixedly mounted on a guide rail structure arranged along the depth direction of backscattering imaging. A drive mechanism drives the tungsten plate scattering target to move along the depth direction and precisely position it to a preset depth. Location.

[0016] Optionally, the step of collecting backscattered response data at various depths and calculating the corresponding detector response center coordinates using the centroid method further includes: When the tungsten plate scattering target is at depth When positioned, the backscattered photons generated by the pencil beam illumination are filtered and confined by the back collimator slit structure and then transmitted to the long strip linear array detector and received. The linear array detector's response to a single particle event is divided into four readout channels, arranged in a counter-clockwise direction starting from the upper left corner of the detector. , , , ; The backscattered response signal amplitudes from the four channels were collected, and the depth was calculated based on the centroid method. Below, the response center coordinates of the linear array detector region corresponding to the tungsten plate scattering target. The calculation formula is:

[0017]

[0018] in, The coordinates of the detector response center, i.e., the corresponding depth Detector response center coordinates .

[0019] Optionally, the step of establishing a mapping function between depth and detector position based on the acquired depth position and detector response center coordinate data through linear fitting, thereby calibrating the geometric mapping relationship in the depth direction and completing the depth geometric distortion correction of the backscattered image, further includes: Based on preset depth Location obtained Calculate separately 、 Through linear functions , fitting , To obtain the relationship between the depth of the tungsten plate scattering target and the detector position , Parameters required for mapping relationship Construct mapping function ,in, , , Represented as ; Using the mapping function Instead of traditional theoretical geometric relationships, this serves as the basis for converting detector coordinates to depth coordinates. Changes in depth cause a non-linear change in the distance of a particle event relative to a reference point in the detector coordinate system. First, the detector response center coordinates t of the particle event are obtained using the centroid method. Then, t is substituted into the mapping function. By obtaining the true depth d, the precise calibration of the geometric mapping relationship in the depth direction is finally completed, realizing the depth geometric distortion correction of the backscattered image.

[0020] To achieve the above objectives, a second aspect of the present invention provides a calibration device for a backscatter imaging system, comprising: The projective geometry modeling module is used to treat the pencil beam formed by the front collimator and the detector plane as two straight lines in a two-dimensional plane, and to treat the slit of the rear collimator as a fixed point in the two-dimensional plane, thus constructing a light path mapping model based on projective geometry. The linear relationship transformation module is used to introduce a physically achievable reference point, establish an imaging system coordinate reference system through this reference point, and transform the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system. The data acquisition and calculation module is used to collimate the X-ray source into an approximate pencil beam using a front collimator based on the imaging system coordinate reference system, use a tungsten plate as a scattering target, control the scattering target to move along the depth to a preset position, acquire backscattering response data at each depth, and calculate the corresponding detector response center coordinates using the centroid method. The distortion correction module is used to establish a mapping function between depth and detector position based on the acquired depth position and detector response center coordinate data through linear fitting, thereby calibrating the geometric mapping relationship in the depth direction and completing the depth geometric distortion correction of the backscattered image.

[0021] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0022] To achieve the above objectives, a third aspect of this application provides an electronic device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing a backscattering imaging system calibration method as described in the first aspect embodiment.

[0023] To achieve the above objectives, a fourth aspect of this application provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a backscattering imaging system calibration method as described in the first aspect embodiment.

[0024] The embodiments of the present invention have the following beneficial effects: 1. It eliminates the need for complex instruments to measure the precise spatial geometric pose of detectors, collimators, or X-ray sources, thus reducing the requirements for system measurement; 2. Only a small number of target movements and detector data acquisitions are required to restore the depth-direction geometric mapping relationship of the calibrated backscatter imaging system, resulting in high calibration efficiency; 3. It can be directly used in engineered systems without relying on dedicated testing equipment or high-precision assembly structures; 4. It can effectively correct for installation deviations, structural deformations, or geometric drift caused by long-term use in backscatter imaging systems; 5. Applicable to a variety of backscatter imaging systems, including fixed, scanning, and array structures. Attached Figure Description

[0025] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart of a backscatter imaging system calibration method provided in an embodiment of the present invention; Figure 2 This is an equivalent schematic diagram of the projective geometry optical path mapping of the backscatter imaging system provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the X-ray backscatter imaging system provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the depth calibration steps provided in an embodiment of the present invention; Figure 5 This is a schematic diagram showing the relationship between the depth sampling points of the thin-film tungsten plate scattering target and the detector position, provided in an embodiment of the present invention. Figure 6 This is a structural diagram of a backscatter imaging system calibration device provided in an embodiment of the present invention. Detailed Implementation

[0026] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] A backscatter imaging system calibration method and apparatus according to an embodiment of the present invention are described below with reference to the accompanying drawings.

[0029] Example 1 This invention provides a calibration method for a backscatter imaging system. Figure 1 This is a schematic flowchart illustrating a backscattering imaging system calibration method provided in an embodiment of the present invention. Figure 1 As shown, the method includes the following steps: Step S1: Treat the pencil beam formed by the front collimator and the detector plane as two straight lines in a two-dimensional plane, and the slit of the rear collimator as a fixed point in the same two-dimensional plane to construct a light path mapping model based on projective geometry.

[0030] In one embodiment of the present invention, a light path mapping model based on projective geometry is constructed based on the above-mentioned geometric equivalence method, so that when the corresponding point of the scattering target at the depth position in the pencil beam direction is projected onto the detector plane through the slit, a one-to-one corresponding detector pixel point is obtained, thereby realizing the accurate conversion of object spatial depth information into image spatial detection signal.

[0031] In this embodiment of the application, to simplify the geometric analysis process of the optical path structure, the core components of the backscatter imaging system are projected onto the same two-dimensional plane, such as... Figure 2 As shown, the slit collimator projects onto a fixed point P on this plane. This fixed point P serves as the core projection reference point for the entire projective mapping relationship, uniquely defining the effective transmission path of backscattered photons. Simultaneously, the projections of the pencil beam formed by the pre-collimator and the sensitive plane of the detector onto this two-dimensional plane are both straight lines. This geometric setting provides the necessary premise for subsequently constructing the projective mapping relationship. Based on the above equivalent geometric model, this application abstracts the entire optical path structure of backscattered imaging. Through the constructed point-based projective mapping relationship, discrete points representing different depth positions on the beam line in object space are all projected onto the detector projection line in image space. Then, through the projection mapping rules between lines defined by fixed points in projective geometry, this mapping relationship is appropriately parameterized using affine parameters. Finally, the relationship between the depth of object space and the pixel position of the detector is expressed as a Möbius fractional linear transformation. This transformation form can accurately characterize the nonlinear mapping characteristics between depth position and detector position, providing crucial theoretical support for subsequently transforming this nonlinear relationship into a computationally pleasing linear relationship.

[0032] In this embodiment of the application, to further clarify the mathematical expression of the projective mapping relationship, the following notation is defined: , , , At the same time, define auxiliary vectors , Based on the above notation and auxiliary vectors, the following relationship can be obtained: ,in , , , The operation here is the cross product, which is equivalent to the determinant operation in two dimensions. This expression intuitively reflects the projective mapping relationship between the object's spatial depth-related parameters and the image spatial detector's related parameters, laying the theoretical foundation for simplifying the fractional linear transformation into a reciprocal linear relationship that is easier to fit and calculate.

[0033] Step S2: Introduce a physically achievable reference point, establish an imaging system coordinate reference system through this reference point, and transform the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system.

[0034] In one embodiment of the invention, a physically realizable reference point is introduced. ,in This is the reference depth position of the scattering target. This reference point is used to number the corresponding detector pixel column. This application simplifies the original Möbius fractional linear transformation using this reference point, reducing the complex point-based projective mapping transformation to a linear form that is easier to calculate and fit. A local coordinate reference system dedicated to the imaging system is established with this reference point as the origin. This allows the projective mapping relationship between scattering points on the beam line in object space and detector pixels in imaging space to exist as a simple reciprocal linear relationship within this local coordinate system, thereby achieving accurate calibration of the depth-direction geometric mapping relationship of the backscattering imaging system based on a moving scattering target.

[0035] In this embodiment of the application, the aforementioned obtained Converting it into a linear expression, we can obtain... , combined The above reciprocal linear relationship can be expressed as: , where k and b are constants related to the geometry of the backscatter imaging device.

[0036] The reference point introduced in this application has significant error compensation advantages, effectively absorbing overall translational and angular deviations generated during system assembly and use. Even in a non-ideal assembly state, the mapping relationship between the object space and the imaging space can still maintain a stable reciprocal linear structure, thus ensuring the stability and repeatability of the depth calibration process. This characteristic is unavailable in traditional similar triangle calibration models, which have stringent requirements for device assembly accuracy. Even a small installation deviation can lead to a large error in the calibration results, failing to meet the requirements of high-precision imaging. Furthermore, the reference point used in this application is highly flexible, allowing it to be any depth position within the imaging range. A fixed reference point can be selected before the system leaves the factory via a preset program, or the optimal reference point can be selected in real-time based on the movement range of the scattering target and the response characteristics of the detector during actual calibration, further improving calibration accuracy. In addition, the mapping function representing the correspondence between depth and detector pixels has diverse implementation forms, including linear fitting of the reciprocal variable and a parameterized form equivalent to projective geometry principles, allowing for flexible selection based on the actual calibration scenario and data processing requirements.

[0037] Step S3: Based on the imaging system coordinate reference system, the X-ray source is collimated into an approximate pencil beam using a front collimator. A tungsten plate is used as a scattering target, and the scattering target is controlled to move along the depth to a preset position. Backscattering response data at each depth are collected, and the coordinates of the corresponding detector response center are calculated using the centroid method.

[0038] In this embodiment, from a strictly geometrical optical perspective, a pencil beam refers to a stream of photons with extremely small lateral dimensions, extremely low divergence, and highly concentrated directionality; its spatial shape can be approximated as a straight ray without lateral widening. Specifically, a pencil beam means that its lateral dimension... Angular divergence The energy / direction distribution within the beam remains almost unchanged laterally. In engineering, the determination of a "pencil beam" is relative to the imaging model; as long as it falls within the allowable error range of the imaging model, it is acceptable. In backscattered imaging calibration, the following condition must be met: at any depth within the imaging range, the diameter *r* of the spot on the target surface illuminated by the pencil beam satisfies: , This refers to the detector's position resolution. This calibration method can also be used to help determine whether it approximates a pencil beam. During the calibration process... If the mapping curve conforms to the linear model, it can be determined to be an approximate pencil-shaped bundle.

[0039] In this embodiment, a thin tungsten plate is used as the scattering target. Specifically, the thickness of the thin tungsten plate scattering target in the depth direction is set to 2 mm, and the effective area in the vertical depth direction completely covers the irradiation range of the pencil beam emitted from the front collimator, ensuring that the pencil beam can completely act on the tungsten plate target surface to generate a stable backscattered signal. The tungsten plate material selected in this application has the dual characteristics of high attenuation coefficient and high scattering cross section. This characteristic allows photons to interact with the tungsten plate material and generate backscattered photons when the X-ray pencil beam is incident on the tungsten plate surface. Moreover, the backscattering effect is mainly concentrated at the corresponding incident position on the front surface of the tungsten plate, effectively ensuring the spatial directivity and positional correlation of the scattered signal. At the same time, the area in front of the tungsten plate is set as an air region to avoid additional scattering or attenuation interference from the air medium on the incident pencil beam and backscattered photons, ensuring the stability of the optical path transmission.

[0040] This application fixes the aforementioned tungsten plate scattering target onto a high-precision guide rail structure that can move along the backscattering imaging depth direction. This guide rail structure, in conjunction with a servo drive mechanism, can precisely move the tungsten plate scattering target to any preset depth within the imaging depth range according to preset control commands. Position, enabling controllable adjustment and precise positioning of the scattering target's depth. When the tungsten plate scattering target is at the target depth... When positioned, the backscattered photons generated by the pencil beam incident on the front surface of the tungsten plate will be limited by the angle of the back collimator slit structure. Only backscattered photons that meet the angle conditions can pass through the slit and be incident on the sensitive surface of the elongated linear array detector, where they are effectively received by the detector and converted into electrical signals.

[0041] In this embodiment, the structure of the back collimator is not limited to a single slit; it can also be an array aperture, pinhole, or other structure capable of equivalently limiting the scattering angle, as long as the structure can form a fixed projection point on the two-dimensional equivalent plane, satisfying the requirement of constructing a point-projective mapping relationship. Correspondingly, the selection of the detector is also flexible. In addition to a long linear array detector, a planar array detector or other detector structures with equivalent column coordinates can also be used to adapt to different imaging system layouts and accuracy requirements. After the detector receives the backscattered photon signal, the system acquires the electrical signal data output by the linear array detector and uses the centroid method to calculate the center position of the tungsten plate scattering target corresponding to the linear array detector region. This provides data support for fitting the subsequent depth mapping relationship.

[0042] Specifically, assuming the linear array detector's response to a single particle event is divided into four readout channels, their spatial arrangement starting from the upper left corner of the detector and proceeding counterclockwise is denoted as follows: , , , When a particle generates a signal on the detector, the amplitude obtained by different channels is related to the particle's incident position. By treating the signal magnitudes of the four channels as weights, the position coordinates of the event on the detector can be calculated using the centroid method. The calculation formula is: The position coordinates calculated using this formula can accurately characterize the tungsten plate scattering target at the current depth. The corresponding detector center position .

[0043] Step S4: Based on the acquired depth position and detector response center coordinate data, a mapping function between depth and detector position is established through linear fitting to calibrate the geometric mapping relationship in the depth direction and complete the depth geometric distortion correction of the backscattered image.

[0044] In one embodiment of the present invention, after completing multiple sets of depth positions Coordinates of the corresponding detector response center After data acquisition, this application preprocesses the acquired discrete data, calculating the reciprocal of the depth for each group of data. Reciprocal of the detector position ,in, , Based on the reciprocal linear mapping relationship in the local coordinate system constructed above in this application, a linear function is adopted. Regarding the above and The data is linearly fitted, and the key parameters, i.e., constants, required to obtain the mapping relationship between the depth of the tungsten plate scattering target and the detector position are obtained by fitting algorithms such as the least squares method. ,constant Reference depth position and the corresponding detector pixel column number Based on the parameters obtained from the solution, this application further constructs a mapping function that can accurately characterize the correspondence between depth position and detector position. The specific expression for this mapping function is: ,in For the detector sampling position and reference point The difference.

[0045] This application is based on the mapping function obtained through fitting. This paper establishes a precise correspondence between the spatial depth of an object and the position of the image space detector in a backscatter imaging system, thereby calibrating the geometric mapping relationship in the depth direction of the backscatter imaging system. This mapping relationship has flexible acquisition and updating methods; parameters can be solved and functions constructed through pre-calibration before the system's formal imaging, or the function parameters can be dynamically updated based on real-time acquired data during online calibration to adapt to the effects of structural shifts or environmental changes caused by long-term system operation. In the backscatter imaging image reconstruction process, this application abandons traditional theoretical geometric relationships and uses the fitted mapping function... It replaces the theoretical geometric model in image reconstruction operations, thereby completing the depth geometric distortion correction of backscattered images and improving the depth resolution and spatial positioning accuracy of the images.

[0046] Specifically, after entering imaging mode, the array detector can independently separate and record the signal of each backscattered photon event, ensuring that the signal characteristics of a single photon event are not disturbed. For any photon event, the detector first outputs the response distribution of the photon event on the detector's sensitive surface, calculates the spatial position t of the photon event on the detector, since a mapping function between depth d and detector position t has been established during calibration. The system can directly convert t into the spatial depth d corresponding to the incident particle. After solving for the depth, the system backprojects the particle's depth information to the corresponding depth layer in three-dimensional space according to the imaging geometry, realizing the spatial localization of photon events. As a large number of incident particle events are processed one by one, their backprojection results accumulate and superimpose in three-dimensional space, forming complete voxelized structural information, thereby reconstructing the true depth distribution of the measured object or scene. Throughout the entire imaging process, all detector sampling positions are converted into the corresponding true depth through the fitted mapping function, thereby obtaining a distortion-free depth reconstruction structure and effectively eliminating spatial three-dimensional distortion caused by device assembly errors, component wear, or environmental vibrations, ensuring the accuracy and reliability of the imaging results.

[0047] In the application of one embodiment of the present invention, the implementation process is as follows: like Figure 3As shown, an X-ray backscattering imaging system using depth-direction geometric calibration includes: an X-ray source 1 positioned on one side; a pre-collimator 2 positioned in front of the X-ray source 1; a thin, movable tungsten plate scattering target 3 positioned in front of the pre-collimator 2 along the X-ray propagation direction; a post-collimator 4 positioned between the tungsten plate scattering target 3 and a detector 5; and a long strip array detector 5 positioned on the other side of the system. The cone-shaped X-ray beam emitted from the X-ray source 1 is collimated by the pre-collimator 2 to form an approximately pencil-shaped beam, which then irradiates the tungsten plate scattering target 3. The area in front of the tungsten plate scattering target 3 is air. The tungsten plate scattering target 3 has a high attenuation coefficient and a high scattering cross-section for X-rays, and backscattering mainly occurs at the front surface of the tungsten plate scattering target 3. Backscattered photons, after passing through the slit structure of the post-collimator 4, are incident on a specific pixel position of the long strip array detector 5. When the tungsten plate scattering target 3 is located at different... At different depth positions (e.g., tungsten plate scattering target 3a at depth 1 position d1 and tungsten plate scattering target 3b at depth 2 position d2 as shown in the figure), the incident positions of the corresponding backscattered photons on detector 5 are different, thus forming a geometric correspondence between the depth position and the detector position.

[0048] like Figure 4 Calibration flowchart and depth correspondence diagram Figure 5 As shown, during the calibration process, the tungsten plate scattering target is moved sequentially to the preset depth position via a linear guide mechanism. At each depth position Below, a pencil beam irradiates the front surface of a tungsten plate scattering target, generating a backscattered signal. After being confined by a collimator, the signal is received by an array detector. The center position t of the corresponding linear array detector region of the tungsten plate scattering target is calculated and recorded using the centroid method. i Multiple sets were obtained ( , ) data. Calculate separately. , ,in, , Through linear functions Fitting , To obtain the relationship between the depth of the tungsten plate scattering target and the detector position and Mapping relationship, obtain mapping function .

[0049] After the system enters imaging mode, the array detector can independently separate and record the signal for each backscattered photon event. For any photon event, the detector first outputs the response distribution of the photon event on the detector's sensitive surface, and calculates the spatial position t of the photon event on the detector. Since a mapping function between depth d and detector position t has been established during calibration... The system can directly convert t into the spatial depth d corresponding to the incident particle.

[0050] After completing the depth calculation, the system backprojects the particle's depth information to the corresponding depth layer in 3D space according to the imaging geometry, achieving spatial localization of the photon event. As a large number of incident particle events are processed one by one, their backprojection results accumulate and superimpose in 3D space, forming complete voxelized structural information, thereby reconstructing the true depth distribution of the measured object or scene. During the system imaging process, all detector sampling positions are converted from fitting mapping to depth, thus obtaining a true depth reconstruction structure and effectively eliminating spatial 3D distortion caused by installation errors.

[0051] Example 2 This invention provides a calibration device for a backscatter imaging system. Figure 6 This is a schematic flowchart of a backscatter imaging system calibration device provided in an embodiment of the present invention. Figure 6 As shown, the device includes: The projective geometry modeling module 100 is used to treat the pencil beam formed by the front collimator and the detector plane as two straight lines in a two-dimensional plane, and to treat the slit of the rear collimator as a fixed point in the two-dimensional plane, thus constructing a light path mapping model based on projective geometry. The linear relationship transformation module 200 is used to introduce a physically achievable reference point, establish an imaging system coordinate reference system through this reference point, and transform the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system. The data acquisition and calculation module 300 is used to collimate the X-ray source into an approximate pencil beam using a front collimator based on the imaging system coordinate reference system, use a tungsten plate as a scattering target, control the scattering target to move along the depth to a preset position, acquire backscattering response data at each depth, and calculate the corresponding detector response center coordinates using the centroid method. The distortion correction implementation module 400 is used to establish a mapping function between depth and detector position based on the acquired depth position and detector response center coordinate data through linear fitting, realize the geometric mapping relationship calibration in the depth direction, and complete the depth geometric distortion correction of the backscattered image.

[0052] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0053] Example 3 To implement the methods of the above embodiments, the present invention also provides an electronic device, which includes a memory and a processor; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the various steps of the methods described above.

[0054] Example 4 To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in the foregoing embodiments.

[0055] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0056] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0057] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A calibration method for a backscatter imaging system, characterized in that, include: S1, regard the pencil beam formed by the front collimator and the detector plane as two straight lines in a two-dimensional plane, and the slit of the rear collimator as a fixed point in the two-dimensional plane to construct a light path mapping model based on projective geometry. S2, introduce a physically achievable reference point, establish an imaging system coordinate reference system through this reference point, and transform the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system; S3, based on the imaging system coordinate reference system, uses a front collimator to collimate the X-ray source into an approximate pencil beam, uses a tungsten plate as a scattering target, and controls the scattering target to move along the depth to a preset position, collects backscattering response data at each depth, and calculates the corresponding detector response center coordinates using the centroid method; S4, based on the acquired depth position and detector response center coordinate data, establishes a mapping function between depth and detector position through linear fitting, realizes the geometric mapping relationship calibration in the depth direction, and completes the depth geometric distortion correction of the backscattered image.

2. The method according to claim 1, characterized in that, The optical path mapping model based on projective geometry also includes a complete optical path structure for backscatter imaging. This model uses a preset point-to-point projective mapping relationship to make a point on the beam line in the object space correspond to a point on the line in the image space. The correspondence logic is determined by the projection mapping between lines defined by fixed points in projective geometry.

3. The method according to claim 2, characterized in that, The introduction of a physically achievable reference point, through which an imaging system coordinate reference system is established, transforms the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system, and also includes: Introducing a physically achievable benchmark ,in This is the reference depth position of the scattering target. Number the detector pixel column corresponding to the reference depth position, based on the reference point. Establish a local coordinate reference system for the imaging system with the origin as the reference point. In the local coordinate reference system, unit vectors are defined respectively. , Distance parameters , auxiliary vector , ,in, The unit vector representing the direction of the detector. Represents the unit vector in the depth direction. This indicates the distance of the detector's position relative to the reference point. This represents the distance between the depth of the scattering target and the reference point. This represents the fixed point of the projection of the slit collimator onto a two-dimensional plane. Calculation using two-dimensional cross product , , , To obtain the fractional relation Taking the reciprocal of the expression and rearranging, we get... and combined The reciprocal linear relationship in the local coordinate system is obtained. .

4. The method according to claim 3, characterized in that, The pencil beam is defined from a geometrical optical perspective as its lateral dimension. Angular divergence A highly directional photon stream in which the energy and direction within the beam remain almost unchanged laterally; The engineering criterion is that, at any depth within the imaging range, the diameter r of the spot on the target surface illuminated by the pencil beam satisfies... ,in To determine the detector's position resolution, this calibration method is used as an aid in the determination. If the mapping curve between depth d and detector position t during the calibration process conforms to a linear model, it is considered to be an approximate pencil beam.

5. The method according to claim 4, characterized in that, The method of using a tungsten plate as a scattering target and controlling the scattering target to move along the depth to a preset position also includes: A scattering target is prepared using a thin tungsten plate. The thickness of the tungsten plate in the depth direction is 2 mm, and the area in the vertical depth direction covers the irradiation range of the pencil beam. The area in front of the tungsten plate is set as an air region. By utilizing the high attenuation coefficient and high scattering cross section of the tungsten plate, backscattered photons are mainly generated at the corresponding positions on the front surface of the tungsten plate. The tungsten plate scattering target is fixedly mounted on a guide rail structure arranged along the depth direction of backscattering imaging. A drive mechanism drives the tungsten plate scattering target to move along the depth direction and precisely position it to a preset depth. Location.

6. The method according to claim 5, characterized in that, The process of collecting backscattered response data at various depths and calculating the corresponding detector response center coordinates using the centroid method also includes: When the tungsten plate scattering target is at depth When positioned, the backscattered photons generated by the pencil beam illumination are filtered and confined by the back collimator slit structure and then transmitted to the long strip linear array detector and received. The linear array detector's response to a single particle event is divided into four readout channels, arranged in a counter-clockwise direction from the upper left corner of the detector as A1, A2, A3, and A4. The backscattered response signal amplitudes from the four channels were collected, and the depth was calculated based on the centroid method. Below, the response center coordinates of the linear array detector region corresponding to the tungsten plate scattering target. The calculation formula is: in, The coordinates of the detector response center, i.e., the corresponding depth Detector response center coordinates .

7. The method according to claim 6, characterized in that, The method, based on the acquired depth position and detector response center coordinate data, establishes a mapping function between depth and detector position through linear fitting, thereby calibrating the geometric mapping relationship in the depth direction and completing the depth geometric distortion correction of the backscattered image. It also includes: Based on preset depth Location obtained Calculate separately , Through linear functions , fitting , To obtain the relationship between the depth of the tungsten plate scattering target and the detector position , Parameters required for mapping relationship Construct mapping function ,in, , , Represented as ; The geometric relationship is described by the mapping function d=f(t), which serves as the basis for converting detector coordinates to depth coordinates. Changes in depth cause a non-linear change in the distance of a particle event relative to a reference point in the detector coordinate system. First, the coordinates t of the detector response center of the particle event are obtained using the centroid method, and then t is substituted into the mapping function. By obtaining the true depth d, the precise calibration of the geometric mapping relationship in the depth direction is finally completed, realizing the depth geometric distortion correction of the backscattered image.

8. A calibration device for a backscatter imaging system, characterized in that, include: The projective geometry modeling module is used to treat the pencil beam formed by the front collimator and the detector plane as two straight lines in a two-dimensional plane, and to treat the slit of the rear collimator as a fixed point in the same two-dimensional plane, thus constructing a light path mapping model based on projective geometry. The linear relationship transformation module is used to introduce a physically achievable reference point, establish an imaging system coordinate reference system through this reference point, and transform the projective mapping relationship between the object space beam line and the detector plane into a reciprocal linear relationship in the local coordinate system. The data acquisition and calculation module is used to collimate the X-ray source into an approximate pencil beam using a front collimator based on the imaging system coordinate reference system, use a tungsten plate as a scattering target, control the scattering target to move along the depth to a preset position, acquire backscattering response data at each depth, and calculate the corresponding detector response center coordinates using the centroid method. The distortion correction module is used to establish a mapping function between depth and detector position based on the acquired depth position and detector response center coordinate data through linear fitting, thereby calibrating the geometric mapping relationship in the depth direction and completing the depth geometric distortion correction of the backscattered image.

9. An electronic device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images