CN110517330A - A Reconstruction Method of Industrial Cone Beam CT in Offset Scanning Mode - Google Patents

Abstract

The present invention provides a kind of industrial cone-beam CT reconstruction method biased under scan pattern, in order to realize that the CT under biasing sample stage scan pattern is rebuild, by establishing dummy detector, practical biasing sample stage CT scan mode is converted into biased detector CT scan mode, existing Wang-FDK biased detector scan pattern method for reconstructing is finally utilized, the accurate CT image of test sample is obtained.The CT Problems of Reconstruction that the present invention solves under biasing sample stage scan pattern substantially increases the visual field of CT system under the premise of not changing standard cone-beam CT scan geometric layout, has preferable engineering application value.

Description

A Reconstruction Method of Industrial Cone Beam CT in Offset Scanning Mode

technical field

Embodiments of the present invention relate to the technical field of computed tomography, and in particular to an industrial cone-beam CT reconstruction method in offset scan mode.

Background technique

CT (Computed Tomography, computer tomography) is a non-destructive testing method that uses the projection information of objects under different viewing angles, combined with image reconstruction algorithms, to obtain internal quality status and structural information of objects. Has a wide range of applications.

The schematic diagram of standard cone beam CT scanning is shown in Figure 1. When performing standard cone beam CT scanning, the X-ray source 1 emits cone beam rays 2, and the sample 3 to be inspected is placed on the sample rotating platform 4, and the cross section of the sample is completely covered by the cone beam. Ray 2 envelope. The tested sample 3 rotates 360° stepwise around the rotation center 5. During this process, the real detector 6 collects the projection image 7 at each step angle, and finally calculates the final image of the tested sample 3 according to the classic FDK reconstruction algorithm. Tomographic image. However, the detected sample often has great uncertainty in terms of material, shape, size, etc. When the width of the detected sample 3 is greater than the width of the real detector 6, it will appear that the CT imaging field of view cannot completely cover the detected sample. The transverse projection data truncation problem caused by detecting the cross section of sample 3. The central plane of the standard cone-beam CT system is shown in Figure 2(a). It can be seen that due to the large size of the detected sample 3, which exceeds the imaging field of view 8 of the standard cone-beam CT system, the projection data collected by the real detector 6 Bilateral truncation occurs. This kind of bilateral truncated projection data will bring serious artifacts to the reconstruction results, which seriously affects the accurate interpretation of the fault structure information of the tested sample. At this time, the offset detector CT scanning method becomes a practical solution.

As shown in Figure 2(b), when performing offset CT scanning, the real detector 6 moves to one side of the sample 3 to be tested, and at this time the center of rotation 5 is still within the range of the X-ray beam FEG, and only one side of the sample exceeds The range of the ray beam FEG, and the line connecting the X-ray source 1 and the rotation center 5 is perpendicular to the real detector 6, and the projection data collected by the real detector 6 is actually unilaterally truncated projection data. The unilateral truncated projection data obtained by the scanning method in Fig. 2(b) can be reconstructed through the Wang-FDK reconstruction algorithm to obtain the complete tomographic reconstruction result of the tested sample 3. In the offset detector CT scanning mode, the imaging field of view of CT scanning can be up to twice the imaging field of view of ordinary cone-beam CT scanning.

It should be noted that the Wang-FDK reconstruction algorithm is mainly aimed at the offset detector CT scanning geometry shown in Fig. 2(b), where the line connecting the X-ray source focus and the rotation center is perpendicular to the real detector 6 . In reality, the scanning geometry layout of the offset sample stage shown in Fig. 2(c) is more widely used. The layout shown in Fig. 2(c) not only simplifies the geometric layout of offset CT scanning, but also makes the difference between the standard cone-beam CT scanning shown in Fig. 2(a) and the offset sample stage CT scanning mode shown in Fig. 2(c) The switching is more flexible, which greatly facilitates engineering applications. However, it can be clearly seen from Fig. 2(c) that the line connecting the X-ray source 1 and the rotation center 5 in the geometric layout of the CT scan is not perpendicular to the real detector 6, so the Wang-FDK reconstruction method will no longer be suitable for Fig. CT reconstruction in offset scan mode shown in 2(c).

Contents of the invention

An embodiment of the present invention provides an industrial cone-beam CT reconstruction method in an offset scanning mode, which is used to solve the problem of accurate complete tomographic reconstruction of the inspected sample in the existing offset sample stage scanning mode.

The invention provides an industrial cone beam CT reconstruction method in offset scan mode, comprising:

S1, start the CT scanning device, and the real detector collects the original sequence projection image of the detected sample in the offset sample stage scanning mode;

S2, establishing a virtual detector and determining the spatial position of the virtual detector;

S3. According to the spatial coordinate relationship between the real detector and the virtual detector, the original sequence projection image acquired by the real detector is converted to the virtual detector to form a sequence projection image on the virtual detector;

S4, according to the spatial coordinate relationship between the real detector CT system and the virtual detector CT system, transforming the reconstructed geometric parameters in the real detector CT system into the virtual detector CT system;

S5, adjusting the angle of the sequence projection image on the virtual detector to correct the torsion angle of the reconstructed image;

S6, based on the angle-adjusted sequential projection images on the virtual detector and the transformed CT reconstruction geometric parameters, the CT image of the sample under inspection is calculated by using the Wang-FDK reconstruction method.

On the basis of the above technical solutions, the present invention can also be improved as follows.

Further, the CT scanning device includes an X-ray source, an inspected sample, a sample rotating platform and a real detector, and the X-ray source, the inspected sample, a sample rotating platform and a real detector are all placed in an X-ray shielding lead room wherein, the inspected sample is placed on the sample rotating platform, the X-ray source emits X-rays to the inspected sample, and the real detector receives the attenuated X-ray signal to form a projection image;

The step S1 specifically includes:

The sample rotation platform is rotated step by step at fixed angular intervals, so that the detected sample rotates 360°, and the real detector collects the projection image of the detected sample at each step angle to obtain the original sequence of projection images.

Further, the step S2 specifically includes:

The central plane of the cone-beam CT system is FEG. On the central plane, the projection of the rotation center of the sample rotating platform on the real detector is point S ₀ , which passes through point S ₀ and is perpendicular to the straight line FS ₀ to establish a virtual detection device, wherein, F is the focal point of the X-ray source, and the central plane of the cone-beam CT system is a horizontal plane passing through the focal point of the X-ray source;

The spatial coordinates of the four boundary points of the virtual detector are calculated to determine the spatial position of the virtual detector.

Further, the step S3 specifically includes:

S31, according to the horizontal coordinates and vertical coordinates of any detector unit P' on the virtual detector on the virtual detector and the spatial position of the virtual detector, calculate the space of the detector unit P' coordinate;

S32. Calculate the spatial coordinates of the detector unit P on the real detector corresponding to the detector unit P′ on the virtual detector, and then determine the horizontal and vertical coordinates of the detector unit P on the real detector. direct coordinates;

S33. Obtain the projection value of the virtual detector unit P′ according to the projection value of the real detector unit P;

When the tested sample is at one of the rotation positions, the projection value of each detector unit on the virtual detector constitutes a projection image on the virtual detector, and each projection image under all rotation angles constitutes the Sequential projection images on a virtual detector.

Further, the step S4 specifically includes:

S41, calculating the rotation angle of the virtual detector CT system relative to the real detector CT system;

S42. Calculate reconstruction geometric parameters of the virtual detector CT system according to the spatial geometric relationship between the real detector CT system and the virtual detector CT system.

Further, the reconstruction geometric parameters include the distance from the focal point of the X-ray source to the detector, the distance from the focal point of the X-ray source to the rotation center of the rotating platform, and the projection coordinates of the rotation center on the detector.

Further, the step S5 specifically includes:

According to the calculated rotation angle of the virtual detector CT system relative to the real detector CT system, the angle of the sequence projection images on the virtual detector is adjusted.

The beneficial effects of the present invention are: in order to realize the CT reconstruction under the offset sample stage scanning mode, by establishing a virtual detector, the actual offset sample stage CT scanning mode is converted into offset detector CT scanning, and finally the existing Wang - The reconstruction method under the FDK offset detector scanning mode can obtain accurate CT images of the inspected sample. The invention solves the CT reconstruction problem under the offset sample stage scanning mode without changing the geometric layout of the standard cone beam CT scan Under the circumstances, the imaging field of view of the CT system is greatly increased, and it has good engineering application value.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is the schematic diagram of standard cone beam CT scan of prior art;

Figure 2(a) is a schematic diagram of standard cone-beam CT scanning;

Figure 2(b) is a schematic diagram of biased detector scanning;

Figure 2(c) is a schematic diagram of offset sample stage scanning;

Fig. 3 is a schematic diagram of offset sample stage scanning;

Fig. 4(a) is a schematic diagram of transforming the geometric parameters of CT reconstruction into the virtual detector CT system;

Figure 4(b) is a schematic diagram of the reconstruction image torsion angle correction;

Figure 5(a) is the virtual detector projection data sequence;

Figure 5(b) is the angle-adjusted virtual detector projection data sequence.

In the accompanying drawings, the names of the components represented by each label are as follows:

1. X-ray source, 2. Cone beam ray, 3. Sample to be inspected, 4. Sample rotation platform, 5. Rotation center, 6. Real detector, 7. Projection image, 8. Standard cone beam CT scanning imaging field of view , 9. Virtual detector.

Detailed ways

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

Aiming at the geometric layout of the offset sample stage scanning shown in Fig. 2(c), the present invention adopts the projection data conversion method to convert the offset sample stage scanning mode shown in Fig. 2(c) to the offset sample stage shown in Fig. 2(b). The scanning method of the detector is set, and then the Wang-FDK reconstruction algorithm is used to realize the accurate CT reconstruction of the inspected sample 3, and finally the complete tomographic information of the inspected sample 3 is obtained.

The industrial cone beam CT reconstruction method in the offset scanning mode provided by the embodiment of the present invention includes: S1, start the CT scanning device, and obtain the original sequence projection image of the detected sample in the offset sample stage scanning mode through the real detector 6; S2, establish the virtual detector 9 and determine the spatial position of the virtual detector 9; S3, according to the spatial coordinate relationship between the real detector 6 and the virtual detector 9, convert the original sequence projection image acquired by the real detector 6 into the virtual On the detector 9, a sequence projection image on the virtual detector 9 is formed; S4, according to the spatial coordinate relationship between the real detector CT system and the virtual detector CT system, the reconstruction geometric parameters in the real detector CT system are transformed into In the virtual detector CT system; S5, adjust the angle of the sequence projection image on the virtual detector 9 to correct the torsion angle of the reconstructed image; S6, based on the angle-adjusted sequence projection image on the virtual detector 9 and the converted CT reconstruction geometry Parameters, using the Wang-FDK reconstruction method to calculate the CT image of the inspected sample 3.

In order to realize CT reconstruction under offset sample stage scanning mode, the present invention converts the actual offset sample stage CT scanning mode into offset detector CT scanning by establishing a virtual detector 9, and finally utilizes the existing Wang-FDK offset The detector scanning mode reconstruction method obtains accurate CT images of the inspected sample 3. The present invention solves the problem of CT reconstruction in the offset sample stage scanning mode, and greatly increases the scanning geometry layout of the standard cone beam CT without changing the It improves the imaging field of view of the CT system and has good engineering application value.

In one embodiment of the present invention, the CT scanning device includes an X-ray source 1, an inspected sample 3, a sample rotating platform 4 and a real detector 6, wherein the X-ray source 1, the inspected sample 3, the sample rotating platform 4 and The real detectors 6 are all placed in the X-ray shielding lead room. The step S1 specifically includes: stepping and rotating the sample rotation platform 4 at fixed angular intervals, so that the tested sample is rotated 360°, and the real detector 6 collects the projection image of the tested sample at each step angle to obtain the original Sequence projected images.

In one embodiment of the present invention, the method for establishing the virtual detector 9 and determining the spatial position of the virtual detector 9 is as follows: the projection of the rotation center 5 of the sample rotating platform 4 on the central plane of the cone beam CT system is point S ₀ , Through the point S ₀ and perpendicular to the straight line FS ₀ , establish a virtual detector 9, wherein, F is the focus of the X-ray source 1; calculate and determine the space coordinates of the four boundary points of the virtual detector 9 to determine the virtual The spatial position of the detector 9.

Specifically, the schematic diagram of the cone beam CT system in the offset sample stage scanning mode is shown in Figure 3, the focus of the X-ray source 1 and the real detector 6 are point F and plane ABCD respectively, and the width and height of the real detector 6 are respectively W and H pixels. The projections of the center of the real detector 6 and the focal point F of the X-ray source 1 on the plane ABCD where the real detector 6 is located are both point O, and the straight line FO is perpendicular to the plane where the real detector 6 is located. In order to realize the CT detection of large-sized samples, the sample rotating platform 4 moves to the side of the sample 3 to meet the needs of the offset scanning mode. The projection of the rotation center 5 of the sample rotation platform 4 on the plane where the real detector 6 is located is O'O", and the central plane of the cone beam CT system is FEG. On the central plane, the projection of the rotation center 5 on the plane where the real detector 6 is located For point S ₀ , through point S ₀ and perpendicular to straight line FS ₀ , establish a virtual detector 9, as shown in A'B'C'D' in Fig. 3. Since the focal point of X-ray source 1 and the center of rotation The line FS ₀ of 5 is perpendicular to the plane A'B'C'D', so in the virtual detector CT system, the Wang-FDK reconstruction method can be used for accurate CT reconstruction in offset scan mode.

The established CT system coordinate system is shown as XYZ in FIG. 3 , and the distance between the focal point F of the X-ray source 1 and the real detector 6 is FDD. Therefore the focal point space coordinate value of X-ray source 1 is F:(FDD,0,0), and the normal vector of the plane where real detector 6 is located is Therefore the equation of the plane where the real detector 6 is located is:

FDD x = 0 (1)

Through the point S ₀ and perpendicular to the straight line FS ₀ , a virtual detector 9 is established, and the normal vector of the plane where the virtual detector 9 is located is defined as The projected coordinate of the center of rotation 5 on the Y axis is defined as COR, so the spatial coordinate of point S ₀ is defined as (0,COR,0), and The equation of the plane where the virtual detector 9 is located is:

FDD x-COR (y-COR) = 0 (2)

As shown in Figure 3, the boundary points of the virtual detector 9 are A', B', C', D', and the spatial coordinates of A', B', C', D' are defined by the plane A'B'C'D ' and X-rays FA, FB, FC and FD intersections are determined. Assuming that the space coordinates of any detector unit M on the real detector 6 are p _M : (x _M , y _M , z _M ), the X-rays passing through the detector unit M are defined as L _M , L _M etc. formula can be expressed as:

The intersection of X-ray L _M and the plane where the virtual detector 9 is located can be expressed as:

The coordinates of A, B, C, and D are respectively p _A :(0,-W/2,H/2), p _B :(0,W/2,H/2), p _C :(0,W/ 2,-H/2) and p _D : (0,-W/2,-H/2), by substituting p _A , p _B , p _C , and p _D into formula (4), the virtual detection The spatial coordinates of the boundary points p _A' , p _B' , p _C' and p _D' of the device 9 are:

According to formula (5), the geometric shape of virtual detector 9 is an isosceles trapezoid, and A'B'=C'D', B'C'＞A'D', However, the shape of the detector in reality is generally a rectangle, therefore, set z _B' = z _A' and z _C' = z _D' , adjust the geometric shape of the virtual detector 9 to a rectangle, like this, the virtual detector 9 The boundary point coordinates of are adjusted to:

The width and height of virtual detector 9 can be calculated by formula (7):

In formula (7), [·] is rounding operation.

In an embodiment of the present invention, the step S3 specifically includes: S31, according to the horizontal and vertical coordinates of any detector unit P' on the virtual detector on the virtual detector and the virtual detector Calculate the spatial position of the detector to obtain the spatial coordinates of the detector unit P'; S32, calculate the spatial coordinates of the detector unit P on the real detector corresponding to the detector unit P' on the virtual detector, and then determine The horizontal and vertical coordinates of the detector unit P on the real detector; S33, obtaining the projection value of the virtual detector unit P' according to the projection value of the real detector unit P; When a position is rotated, the projection value of each detector unit on the virtual detector constitutes a projection image on the virtual detector. Each projection image under all rotation angles constitutes a sequence of projection images on the virtual detector.

Specifically, the coordinates of the detector units on the virtual detector 9 may be determined by the spatial coordinates of the virtual detector 9 . As shown in Figure 3, the horizontal and vertical coordinates of the detector unit P' on the virtual detector 9 are VD _i and VD _j respectively, and the real space coordinates of the virtual detector unit P' (that is, in The spatial coordinates in the CT coordinate system XYZ) are defined as p _P′ : (x _P′ , y _P′ , z _P′ ). According to the spatial positions of the virtual detector 9 and the real detector 6, it can be obtained:

The real space coordinates of the virtual detector unit P' can be obtained by the above formula (8), and the X-ray passing through the virtual detector unit P' is defined as:

As shown in FIG. 3 , the intersection of the plane where the real detector 6 is located and the straight line L _P′ is point P, which is the real detector unit corresponding to the detector unit P′ on the virtual detector 9 . The x-axis coordinates of all points on the plane ABCD are 0, so according to formula (9), the spatial coordinates of point P are Then the horizontal and vertical coordinates of the detector unit P on the real detector 6 are respectively: and Finally, the projection value of each detector unit P' on the virtual detector 9 is calculated by formula (10).

In formula (10), is the projection value of the detector unit P whose horizontal and vertical coordinates are RD _i and RD _j respectively for the tested sample 3 on the real detector 6, is the projection value of the detector unit P′ whose horizontal and vertical coordinates are VD _i and VD _j respectively for the tested sample 3 on the virtual detector 9 . Since RD _i and RD _j are decimals, so It is obtained by interpolation, so that the projection value of each detector unit on the virtual detector 9 can be accurately calculated. When the tested sample is at one of the rotation angles, the projection value of each detector unit on the virtual detector constitutes a projection image on the virtual detector, and each projection image under all rotation angles constitutes the virtual detection sequence projected images on the monitor.

In one embodiment of the present invention, the step S4 specifically includes: S41, calculating the rotation angle of the virtual detector CT system relative to the real detector CT system; S42, according to the rotation angle of the real detector CT system and the virtual detector CT system Spatial geometric relationship, calculating the reconstruction geometric parameters of the virtual detector CT system.

Specifically, the reconstruction geometric parameters of the virtual detector CT system include the distance FDD' from the focal point of the X-ray source 1 to the detector, the distance FOD' from the X-ray source 1 to the center of rotation 5, and the projection COR of the center of rotation 5 on the detector '. The schematic diagram of the center plane in the offset sample stage scanning mode is shown in Fig. 4(a). On the central plane, the geometric layouts of the real detector CT system and the virtual detector CT system are FEG and FE′G′ respectively, where EG and E′G′ represent the real detector 6 and the virtual detector 9 respectively, points O and M' is the midpoint of EG and E'G' respectively. The lengths of FDD and FOD on the real detector CT system are FO and FK respectively, FDD′ and FOD′ on the virtual detector CT system are respectively FS ₀ and FR, the COR on the real detector 6 is the length of OS ₀ , and the virtual detector COR' on 9 is the length of M'S ₀ , and COR is negative when the center of rotation 5 is shifted to the left.

According to Figure 4(a), we can get:

On the virtual detector CT system, since COR′ is the distance between the rotation center projection point S ₀ and the point M′, in the triangle S ₀ EE′, it can be obtained:

According to the above formula (11) and formula (12), the reconstruction geometric parameters FDD, FOD and COR of the real detector CT system can be converted into the virtual detector CT system, and the transformed reconstruction geometric parameters are FDD′, FOD′ and COR ',

As shown in Figure 4(b), compared with the geometric layout FEG of the real detector CT system, the geometric layout of the virtual detector CT system is rotated by an angle α counterclockwise, that is, the coordinate system X′S ₀ Y′ rotates counterclockwise relative to XOY angle α. At this time, the pose of the sample in the reconstructed image is rotated by an angle α clockwise relative to the pose of the inspected sample 3 placed in the coordinate system XOY. Therefore, in order to identify the details of the reconstructed image more conveniently, the torsion angle of the reconstructed image needs to be corrected accurately.

In one embodiment of the present invention, the step S5 specifically includes: adjusting the angle of the sequential projection images on the virtual detector 9 according to the calculated rotation angle of the geometric layout of the virtual detector CT system, and correcting the twist angle of the reconstructed image.

Specifically, during a complete 360° CT scan, the sample rotating platform 4 is rotated step by step at a fixed angle interval, and N projection images of the sample 3 to be inspected are obtained. Therefore, the angles of the projected images are 0, step,...., (N-1)·step respectively. As shown in Figure 5(a), the N projection images on the virtual detector are respectively Assuming that α is one of the real step angles, then when the step angle is α, the projected image of the tested sample 3 on the virtual detector 9 is Then the original sequence projection image PS on the virtual detector 9 contains two sub-sequence projection images, namely In order to correct the reconstructed image torsion angle, the sequential projection images on the virtual detector 9 are adjusted as As shown in Figure 5(b). Since each projection image in PS' can find the corresponding projection image in the original sequence projection image PS, each projection image in PS' is accurate, so that the angle-adjusted projection image will not lead to the reconstructed image error. When the angle α is not one of the real step angles, that is, the angle α is not equal to any step angle, then each projection image in the sequence projection image PS′ needs to be projected at two adjacent real projection angles Data interpolation obtained. In this case, although the projection images in PS' are obtained by interpolation, as long as the projection angle interval is small enough, these interpolated projection data will not have a significant impact on the reconstructed image result.

After the reconstruction geometric parameters FDD', FOD' and COR' are accurately calibrated and the torsion angle of the reconstructed image is corrected, the Wang-FDK reconstruction method can be used for tomographic image reconstruction based on the corrected sequence projection images and the transformed reconstruction geometric parameters. Accurate reconstruction.

In the following, a specific example is used to describe the industrial cone-beam CT reconstruction method in the offset scan mode provided by the present invention.

The parameters of the scanning device used in the experiment are as follows:

(1) X-ray source: tube voltage 150kV, tube current 0.145mA;

(2) Real detector: the number of pixels of the real detector is 900 (width)×766 (height), the pixel size is 0.2mm, and the effective imaging area is 200×153.2mm ² .

(3) The distance from the focal point of the X-ray source to the real detector is FDD=852mm, the distance from the focal point of the X-ray source to the rotation center of the sample stage is FOD=291.8mm, and the projection COR of the rotation center=-391.0 pixels.

Experimental steps:

(1) Place a cylindrical sample of plexiglass to be tested on the sample rotating platform, turn on the X-ray source and the real detector, start the sample rotating platform to drive the tested sample to rotate 360° step by step, the step angle is 0.5°, the real The detector collected a total of 720 projection images at different step angles to form a sequence of projection images.

(2) According to formulas (1) to (6), establish a virtual detector and locate the spatial position of the virtual detector, and calculate the width and height of the virtual detector according to formula (7), and obtain the width and height of the virtual detector as W'=912 pixels, H'=765 pixels.

(3) Based on the spatial position relationship between the virtual detector and the real detector, the projection value of each detector unit on the virtual detector is determined by formulas (8)-(10).

(4) According to the spatial coordinate relationship between the real detector CT system and the virtual detector CT system, the reconstruction geometric parameters FDD, FOD and COR of the real detector CT system are converted to the virtual detector CT system through formulas (11)-(12) Middle; the reconstruction geometric parameters of the final virtual detector CT system are shown in Table 1.

(5) Adjust the angle of the sequence projection image of the virtual detector CT system to correct the twist angle of the reconstructed image, and obtain an angle-adjusted sequence projection image.

Table 1 Geometric parameters of virtual detector CT system

(6) Based on the angle-adjusted sequential projection images on the virtual detector and the geometric parameters of CT reconstruction (Table 1), the Wang-FDK reconstruction method is used to calculate the accurate CT image of the sample under inspection.

The invention provides an industrial cone beam CT reconstruction method under the offset scanning mode. In order to realize the CT reconstruction under the offset sample stage scanning mode, the actual offset sample stage CT scanning mode is converted into an offset sample stage CT scanning mode by establishing a virtual detector. CT scanning with a biased detector, and finally using the existing Wang-FDK biased detector scanning mode reconstruction method to obtain accurate CT images of the sample under inspection. The invention solves the problem of CT reconstruction in the scanning mode of the offset sample stage, greatly increases the imaging field of view of the CT system without changing the geometric layout of standard cone-beam CT scanning, and has good engineering application value.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (7)

1. An industrial cone beam CT reconstruction method under offset scan mode, characterized in that, comprising: S1, start the CT scanning device, and collect the original sequence projection image of the detected sample in the offset sample stage scanning mode through the real detector; S2, establishing a virtual detector and determining the spatial position of the virtual detector; S3. According to the spatial coordinate relationship between the real detector and the virtual detector, the original sequence projection image acquired by the real detector is converted to the virtual detector to form a sequence projection image on the virtual detector; S4, according to the spatial coordinate relationship between the real detector CT system and the virtual detector CT system, transforming the reconstructed geometric parameters in the real detector CT system into the virtual detector CT system; S5, adjusting the angle of the sequence projection image on the virtual detector to correct the torsion angle of the reconstructed image; S6, based on the angle-adjusted sequential projection images on the virtual detector and the transformed reconstruction geometric parameters, the CT image of the inspected sample is calculated by using the Wang-FDK reconstruction method. 2. the industrial cone-beam CT reconstruction method under the offset scanning mode according to claim 1, is characterized in that, described CT scanning device comprises X-ray source, tested sample, sample rotating platform and real detector, described The X-ray source, the sample to be inspected, the sample rotating table and the real detector are all placed in the X-ray shielding lead room, the sample to be inspected is placed on the sample rotating platform, and the X-ray source emits X-rays to the sample to be inspected. ray, the real detector receives the attenuated X-ray signal to form a projection image; The step S1 specifically includes: The sample rotation platform is rotated step by step at fixed angular intervals, so that the detected sample is rotated 360°, and the projection image of the detected sample at each step angle is collected by the real detector to obtain the original sequence of projection images. 3. The industrial cone beam CT reconstruction method under the offset scan mode according to claim 2, wherein said step S2 specifically comprises: The central plane of the cone-beam CT system is FEG. On the central plane, the projection of the rotation center of the sample rotating platform on the real detector is point S ₀ , which passes through point S ₀ and is perpendicular to the straight line FS ₀ to establish a virtual detection device, wherein, F is the focal point of the X-ray source, and the central plane of the cone-beam CT system is a horizontal plane passing through the focal point of the X-ray source; The spatial coordinates of the four boundary points of the virtual detector are calculated to determine the spatial position of the virtual detector. 4. The industrial cone beam CT reconstruction method under the offset scan mode according to claim 1, wherein said step S3 specifically comprises: S31, according to the horizontal coordinates and vertical coordinates of any detector unit P' on the virtual detector on the virtual detector and the spatial position of the virtual detector, calculate the space of the detector unit P' coordinate; S32. Calculate the spatial coordinates of the detector unit P on the real detector corresponding to the detector unit P′ on the virtual detector, and then determine the horizontal and vertical coordinates of the detector unit P on the real detector. direct coordinates; S33. Obtain the projection value of the virtual detector unit P′ according to the projection value of the real detector unit P; When the tested sample is at one of the rotation angles, the projection value of each detector unit on the virtual detector constitutes a projection image on the virtual detector, and each projection image under all rotation angles constitutes the Sequential projection images on a virtual detector. 5. The industrial cone beam CT reconstruction method under the offset scan mode according to claim 2, wherein said step S4 specifically comprises: S41, calculating the rotation angle of the virtual detector CT system relative to the real detector CT system; S42. Calculate reconstruction geometric parameters of the virtual detector CT system according to the spatial geometric relationship between the real detector CT system and the virtual detector CT system. 6. The industrial cone-beam CT reconstruction method under the offset scanning mode according to claim 1 or 5, wherein the reconstruction geometric parameters include the distance from the focus of the X-ray source to the detector, the distance from the focus of the X-ray source to the rotation The distance from the center of rotation of the platform and the projected coordinates of the center of rotation on the detector. 7. The industrial cone-beam CT reconstruction method under the offset scanning mode according to claim 5, characterized in that, the step S5 specifically comprises: According to the calculated rotation angle of the virtual detector CT system relative to the real detector CT system, the angle of the sequence projection images on the virtual detector is adjusted.