CN102727230A - CT (computed tomography) scanning image rebuilding method and device - Google Patents

Abstract

The invention relates to the technical field of CT (computed tomography) and discloses a CT scanning image rebuilding method and a CT scanning image rebuilding device. The method comprises the following steps that scanning data is recombined into data formats required by projection; for each selected target pixel point, passage positions of the projection passing through the target pixel points are calculated; for each selected target pixel point, and layer positions of the projection passing through the target pixel points are determined according to sample points of rays of each type of focus in the z axis; and the obtained passage positions and layer positions of the projection passing through the target pixel points are subjected to weighted accumulation onto the target pixel points. When the CT scanning image rebuilding method and the CT scanning image rebuilding device are utilized, the influence on the z-axis direction resolution ratio caused by the two-step interpolation in the prior art can be avoided.

Description

CT scan image rebuilding method and device

Technical field

The present invention relates to CT (Computed Tomography, computer x ray layer scanning technology) technical field, more particularly, relate to a kind of CT scan image rebuilding method and device.

Background technology

It is a kind of raising CT (Computed Tomography that the z direction of principal axis flies focus (zDFS) technology; A kind of technology of the z direction of principal axis resolution computer tomography technology); Its x-ray focus position through periodically squinting; Make in single pass the data that obtain double sampling density, thereby obtain higher resolution at the z direction of principal axis.

3 D back projection method is a kind of of Image Reconstruction Technology among many row x ray CT; It projects to the coordinate on the two-dimensional ct detector through in three dimensions, calculating the target reconstruction point; Interpolation on two-dimensional detector then; The projection value that interpolation is gone out oppositely is added to target reconstruction point position, thereby obtains the CT value of image.It is generally acknowledged; In many row CT; 3 D back projection method can obtain than two-dimension image rebuild method better pictures quality, is mainly reflected in aspect the pseudo-shadow (the inaccurate pseudo-shadow that brings of image reconstruction that does not promptly cause at central plane owing to some layer of detector) of cone angle, the z direction of principal axis resolution.

Fly in the focus reconstruction technique at traditional z direction of principal axis; The general employing flies earlier the focus interpolation in the data for projection space, promptly the data recombination process utilizes the correction data that obtains to carry out backprojection reconstruction then; This makes that the interpolation in back projection's process is to be based upon on the basis that flies the focus interpolation result; In fact widening the interpolation width, making and to be differentiated, reducing z direction of principal axis resolution less than the wisp of this width.

Summary of the invention

The embodiment of the invention provides a kind of CT scan image rebuilding method and device to the problems referred to above that exist in the prior art, avoids in the prior art two step interpolation to the influence of z direction of principal axis resolution.

For this reason, the embodiment of the invention provides following technical scheme:

A kind of CT scan image rebuilding method comprises:

Scan-data is recombinated to the data format of back projection's requirement;

To each selected target pixel points, calculate projection and pass the channel position of said target pixel points;

To each selected target pixel points, confirm that at the sampled point of z shaft position said projection passes the layer position of said target pixel points according to the ray of all types of focuses;

The channel position that said target pixel points is passed in the said projection that obtains and layer position weighted accumulation are on said target pixel points.

Preferably, said the scan-data data format that requires to back projection of recombinating is comprised:

Scan-data is recombinated to single layer data; Perhaps

Scan-data is recombinated to multi-layer data.

Preferably, said ray according to all types of focuses confirms that at the sampled point of z shaft position said projection passes the layer position of said target pixel points and comprise:

The ray that calculates each focus respectively passes the theoretical layer position of said target pixel points;

The ray of each focus that obtains is passed the theoretical layer position of said target pixel points and carry out interpolation processing, obtain the layer position that said target pixel points is passed in said projection.

Preferably, the ray of said each focus that will obtain passes the theoretical layer position of said target pixel points and carries out interpolation processing, obtains said projection and passes the layer position of said target pixel points and comprise:

The theoretical layer position that the ray of each focus is passed said target pixel points rounds first interpolation point that obtains should focus downwards, and said first interpolation point is added 1 obtains second interpolation point that should focus;

All interpolation points to each focus carry out weighting, obtain the layer position that said target pixel points is passed in said projection.

Preferably, said method also comprises:

Interpolation point according to corresponding each focus is set the weights that said interpolation point carries out weighting to the distance of the actual integer layer of the ray of this focus.

A kind of CT scan equipment for reconstructing image comprises:

The data recombination unit is used for the data format that scan-data is recombinated and required to back projection;

The channel position computing unit is used for each selected target pixel points is calculated projection and passed the channel position of said target pixel points;

Layer position calculation unit is used for each selected target pixel points, confirms that at the sampled point of z shaft position said projection passes the layer position of said target pixel points according to the ray of all types of focuses;

The unit that adds up, the channel position that is used for said target pixel points is passed in the said projection that obtains and layer position weighted accumulation are to said target pixel points.

Preferably, said data recombination unit specifically is used for scan-data is recombinated to single layer data; Perhaps scan-data is recombinated to multi-layer data.

Preferably, said layer position calculation unit comprises:

First computation subunit, the ray that is used for calculating respectively each focus passes the theoretical layer position of said target pixel points;

The interpolation processing subelement is used for the theoretical layer position that ray with each focus that obtains passes said target pixel points and carries out interpolation processing, obtains the layer position that said target pixel points is passed in said projection.

Preferably, said interpolation processing subelement comprises:

Interpolation point is confirmed subelement, and the theoretical layer position that the ray of each focus is passed said target pixel points rounds first interpolation point that obtains should focus downwards, and said first interpolation point is added 1 obtains second interpolation point that should focus;

The weighting subelement is used for all interpolation points of each focus are carried out weighting, obtains the layer position that said target pixel points is passed in said projection.

Preferably, said interpolation processing subelement also comprises:

The weight setting subelement is used for setting the weights that said interpolation point carries out weighting according to the interpolation point of corresponding each focus to the distance of the actual integer layer of the ray of this focus.

Embodiment of the invention CT scan image rebuilding method and device; The z direction of principal axis is flown the focus interpolation to be attached in back projection's process of image reconstruction; Make to fly the focus interpolation,, confirm that at the sampled point of z shaft position said projection passes layer position of said target pixel points according to the ray of each focus promptly to each selected target pixel points with back projection handles disposable completion; Thereby reduced actual interpolation width, avoided in the prior art two step interpolation the influence of z direction of principal axis resolution.

Description of drawings

Fig. 1 is the principle schematic that the z direction of principal axis flies focus in the prior art;

Fig. 2 is that two focus modes realize that the z direction of principal axis flies the sketch map of focus in the prior art;

Fig. 3 is based on the interpolation processing flow chart of two focus flying methods in the prior art;

Fig. 4 is based on the flow chart that flies focus interpolation realization image reconstruction in the prior art;

Fig. 5 is the flow chart of embodiment of the invention CT scan image rebuilding method;

Fig. 6 is the sketch map of target pixel points in the embodiment of the invention;

Fig. 7 is a kind of structural representation of embodiment of the invention CT scan equipment for reconstructing image.

The specific embodiment

In order to make those skilled in the art person understand the scheme of the embodiment of the invention better, the embodiment of the invention is done further to specify below in conjunction with accompanying drawing and embodiment.

At first the principle that flies focus in the prior art is done simple declaration below.

In flying focal spot techniques, mainly use focus A, C, its geometric position is designated as respectively:

A(z-)、C(z+)。

It is as shown in Figure 1 that the z direction of principal axis flies the principle of focus, and wherein, A representes two different focuses among the zDFS respectively with C.R representes the radius of turn of frame, R _FDExpression standard focal position is to the distance of detector, R _ODRepresent the distance of center of rotation, and use R to detector _C=R+d and R _A=R-d representes the radius of turn of focus C and focus A respectively.

Beam squints on the z direction, deflects to focus C from focus A, makes each layer of focus C at the center of rotation place; Pass the centermost of focus A adjacent two layers; Promptly on the residing straight line of center of rotation, each passage of focus C passes the half storey point of focus A, thereby obtains to be twice in the sampling of original layers.

Utilize the coordinate of each focus and each layer of detector, can obtain under this coordinate system, the intercept of each layer ray on the z axle is the sampling location of each layer at center of rotation, is respectively:

$C:Z_C\left(m\right)=\frac{(m-\frac{M-1}{2})\mathrm{\Δ}(R+d)+R_{\mathrm{OD}}d\tan \mathrm{\φ}}{R_{\mathrm{FD}}+d}$

$A:Z_A\left(m\right)=\frac{(m-\frac{M-1}{2})\mathrm{\Δ}(R-d)-R_{\mathrm{OD}}d\tan \mathrm{\φ}}{R_{\mathrm{FD}}-d}---\left(1\right)$

So in last Fig. 1, each layer arrangement position is from right to left:

Z _A(0)，Z _C(0)，Z _A(1)，Z _C(1)，…，Z _A(M-2)，Z _C(M-2)，Z _A(M-1)，Z _C(M-1)

The actual slice thickness of focus A (on the z axle) is:

$s_A=Z_A(m+1)-Z_A\left(m\right)=\frac{\mathrm{\Δ}(R-d)}{R_{\mathrm{FD}}-d}=\frac{R_{\mathrm{FD}}}{R}\·\frac{(R-d)}{R_{\mathrm{FD}}-d}s---\left(2\right)$

The actual slice thickness (center) of focus C is:

$s_C=Z_C(m+1)-Z_C\left(m\right)=\frac{\mathrm{\Δ}(R+d)}{R_{\mathrm{FD}+}d}=\frac{R_{\mathrm{FD}}}{R}\·\frac{(R+d)}{R_{\mathrm{FD}+}d}s---\left(2\right),$

Each of cross arrangement layer ray increased the sampling density on the z axle in fact, thereby can improve the z directional resolution.

In the prior art, the z direction flies the following mode of the general sampling of focal spot techniques:

Fig. 2 realizes that the z direction of principal axis flies the sketch map of focus.

The interpolation processing process is as shown in Figure 3, may further comprise the steps:

Step 301 obtains scan-data;

Step 302, the scan-data that the flight one that obtains is enclosed is that basic operation unit carries out circular treatment according to the number with each cycle period, the viewdata of the circle of recombinating out, each processing procedure may further comprise the steps:

Step 302-1, each destination layer position of calculating detector.

Particularly, establishing the actual physical detector number of plies is N, and layer thickness is s, and the number that makes destination layer so is 2N, and layer thickness is s/2, and then the position of each layer of target is: (k-(2N-1)/2) * (s/2), and k=0,1 ..., 2N-1.

Step 302-2 utilizes A, the pairing detector physical location of C focus, and interpolation goes out the data of the destination layer position that calculates among the above-mentioned steps 302-1.

Particularly; The data cross of A, C focus is arranged (according to the layer sequence of positions of reality),, the destination layer position that 302-1 calculates is exported as interpolation as the interpolation input; Utilize any interpolation method interpolation calculation to go out the data of destination layer position, promptly z flies focus interpolation data afterwards.

Fig. 4 realizes may further comprise the steps the flow chart of image reconstruction based on flying the focus interpolation in the prior art:

Step 401 carries out flying the focus interpolation processing based on scan-data;

Step 402 is carried out data recombination.

Be about to the data format of data recombination to the back projection requirement of scanning acquisition, like the parallel beam data.If adopt two-dimentional back projection method, then recombinate to single layer data; If the employing 3 D back projection method is then recombinated to multi-layer data.

Step 403 adds cone angle power and filtering to data.

(this disperses and causes image reconstruction inaccurate at the disperse function of z direction for each layer ray of offsetting multilayer detector; Bring the pseudo-shadow of cone angle); Need multiply by a cone angle power (being the cos value of cone angle) to each layer data, thereby each layer data is projected on the central plane.Adopt suitable convolution kernel and data to do convolution then, to strengthen the effective frequency composition in the data, perhaps to weaken the idling frequency composition in the data.

Step 404, definition pixel sequence number variable k=0 promptly selects first pixel.

Step 405 judges that whether k is less than total number of pixels N; If then execution in step 406; Otherwise execution in step 410.

Step 406 is calculated the channel position that the ray of above-mentioned pixel is passed in certain projection, and promptly this is projected in the projection value on the x axle.

Step 407 is calculated the theoretical layer position of the ray that certain projection passes above-mentioned pixel, and utilizes the integer layer position that flies after the focus interpolation, and interpolation goes out the data of theoretical layer position, and promptly this is projected in the projection value on the z axle.

Step 408, with the layer that calculates and the projection value weighted accumulation of passage to this pixel, i.e. back projection's process.

Step 409, k=k+1 promptly selects next pixel to carry out back projection's process.

Step 410 finishes.

It is thus clear that in above-mentioned image reconstruction process, the process of the projection value of the calculating pixel of back projection is based upon on the basis that flies the focus interpolation result, has widened the interpolation width, make and can't be differentiated less than the wisp of this width, reduce the z direction of principal axis and distinguished rate.

For this reason, embodiment of the invention CT scan image rebuilding method and device to the problems referred to above, fly focus correction with the z direction of principal axis and are attached in back projection's process, make to fly focus interpolation and the disposable completion of back projection's process, the adverse effect of avoiding two step interpolation to bring.

As shown in Figure 5, be the flow chart of embodiment of the invention CT scan image rebuilding method, may further comprise the steps:

Step 501 recombinates scan-data to the data format of back projection's requirement.

Be about to the data format of data recombination to the back projection requirement of scanning acquisition, like the parallel beam data.If adopt two-dimentional back projection method, then recombinate to single layer data; If the employing 3 D back projection method is then heavy and to multi-layer data.

The concrete grammar of data recombination can adopt processing methods more of the prior art, does not do detailed description at this.

Step 502 to each selected target pixel points, is calculated projection and is passed the channel position of said target pixel points.

Step 503 to each selected target pixel points, confirms that at the sampled point of z shaft position said projection passes the layer position of said target pixel points according to the ray of all types of focuses.

Particularly; The ray that can calculate each focus respectively passes the theoretical layer position of said target pixel points; The ray of each focus that will obtain then passes the theoretical layer position of said target pixel points and carries out interpolation processing, obtains the layer position that said target pixel points is passed in said projection.

When carrying out interpolation processing; The theoretical layer position that the ray of each focus is passed said target pixel points rounds first interpolation point that obtains should focus downwards; And said first interpolation point is added 1 obtain second interpolation point that should focus; All interpolation points to each focus carry out weighting then, obtain the data that the layer position of said target pixel points passed in said projection.The weights of each interpolation point can be set to the distance of the actual integer layer of the ray of the corresponding focus of this interpolation point according to this interpolation point.

Step 504, the channel position that said target pixel points is passed in the said projection that obtains and layer position weighted accumulation on said target pixel points, i.e. back projection's process.

Said back projection is a kind of method that the perspective geometry principle is carried out image reconstruction of using, and the principle of 3 D back projection process is following:

If p is for through data recombination (step 402) with add the cone angle data after the filtering (step 403) for the time being, then for target reconstruction point X (x1, x2, x3), its pixel value adding up that be all directions through the ray projection value of this impact point:

$f(x,y,z)=\underset{{\mathrm{\θ}}_{\mathrm{in}}}{\overset{{\mathrm{\θ}}_{\mathrm{out}}}{\∫}}w\left(\mathrm{\θ}\right)p(\mathrm{\θ},t(\mathrm{\θ},x,y,z),m(\mathrm{\θ},x,y,z))\mathrm{d\θ}$

Wherein, θ is a projected angle, θ _InAnd θ _OutBe respectively the initial projected angle and end projected angle that pass impact point, w (θ) is the weights of each projection, and (y z) is the ray position of x direction of principal axis (being channel direction) to t for θ, x, and (y z) is the ray position of z direction of principal axis (i.e. layer direction) to m for θ, x.T and m not necessarily drop on certain bar ray just, but drop between some sampling rays, at this moment, need obtain real m and t as interpolation with these two rays, and hereinafter hypothesis t adopts simple linear interpolation, mainly pays close attention to t.

Description by the front can be known, data for projection p (θ, t; M) need obtain through contiguous ray interpolation; In embodiments of the present invention, utilize through the projection ray behind the focal shift and come these data of interpolation, increased sampling density owing to fly focus; Utilize it to come interpolation can effectively reduce the interpolation width, thereby improve z direction of principal axis resolution.

Be example with the 3 D back projection mode below, be attached to the process that back projection handles and do further explain in the embodiment of the invention z direction of principal axis being flown the focus interpolation processing.

As shown in Figure 6, send ray respectively from focus A and focus C and pass target pixel points 700, because detector is the discrete sampling to the x ray; So sampling is not necessarily arranged just through target pixel points; At this moment need interpolation, shown in figure, on the yz plane, see; Four real x ray sample distribution are arranged in the both sides of impact point, they are to come from the sampled point 72,74 of focus A and the sampled point 71,73 of focus C respectively.

Below to utilize these four sampled point interpolation to go out the value of target pixel points, establish the interpolation basic function and describe Interpolation Process in detail below for

.

Can obtain through geometrical relationship, pass said target pixel points ray the layer position m theoretical value be:

$m(\mathrm{\θ},x,y,R,s)=\frac{R(Z_{\mathrm{img}}-(\mathrm{\θ}+\mathrm{\γ})\frac{\mathrm{pMs}}{2\mathrm{\π}})}{R\cos \mathrm{\γ}-(x\sin (\mathrm{\θ}+\mathrm{startAgl})+y\cos (\mathrm{\θ}+\mathrm{startAgl}))}---\left(3\right)$

Wherein, Z _ImgThe expression target image is with respect to the z shaft position of the initial projection of entire image; θ representes the projection angle with respect to initial projection, and p is a pitch, and M is total number of plies of detector; S is a central slice thickness; StartAgl representes the projection angle (rotary code that is equivalent to initial projection) of initial projection, and γ representes the angle (be commonly referred to as and fan the angle) of ray and ray through center of rotation through pixel

Wherein, t is the vertical dimension of ray to center of rotation.

Utilize above-mentioned formula (2), (2) ', (3), can calculate the theoretical layer position that each focus ray passes target pixel points:

m _c＝m(θ，x，y，R _C，S _C)，m _A＝m(θ，x，y，R _A，S _A)；

Make

Then, reuse m ₀+ 1, m ₀, m ₁+ 1, m ₁These four point interpolations go out the value of target pixel points.Weights are respectively:

Make W _All=W _BackC+ W _ForeC+ W _BackA+ W _ForeA

Theoretical channel position is t if channel direction is calculated gained, makes

weights

then calculate the data for projection of target pixel points at last:

$p(\mathrm{\θ},t,m)=p_{\mathrm{fore}}^{\mathrm{chn}}*W_{\mathrm{fore}}^{\mathrm{chn}}+p_{\mathrm{back}}^{\mathrm{chn}}*W_{\mathrm{back}}^{\mathrm{chn}}---\left(4\right)$

Wherein:

$p_{\mathrm{fore}}^{\mathrm{chn}}=\frac{(w_{\mathrm{backC}}*p_C(\mathrm{\&theta;},t_{\mathrm{fore}},{\mathrm{<figure-callout\; id="0"\; label="m"\; filenames="HDA0000053796780000031.png"\; state="\{\{state\}\}">m</figure-callout>}}_0+1)+w_{\mathrm{backA}}*p_A(\mathrm{\&theta;},t_{\mathrm{fore}},{\mathrm{<figure-callout\; id="1"\; label="m"\; filenames="HDA0000053796780000031.png"\; state="\{\{state\}\}">m</figure-callout>}}_1+1)+w_{\mathrm{foreC}}*p_C(\mathrm{\&theta;},t_{\mathrm{fore}},m_0)+w_{\mathrm{foreA}}*p_A(\mathrm{\&theta;},t_{\mathrm{fore}},m_1)}{W_{\mathrm{All}}}$

$p_{\mathrm{fore}}^{\mathrm{chn}}=\frac{(w_{\mathrm{backC}}*p_C(\mathrm{\&theta;},t_{\mathrm{back}},{\mathrm{<figure-callout\; id="0"\; label="m"\; filenames="HDA0000053796780000031.png"\; state="\{\{state\}\}">m</figure-callout>}}_0+1)+w_{\mathrm{backA}}*p_A(\mathrm{\&theta;},t_{\mathrm{back}},{\mathrm{<figure-callout\; id="1"\; label="m"\; filenames="HDA0000053796780000031.png"\; state="\{\{state\}\}">m</figure-callout>}}_1+1)+w_{\mathrm{foreC}}*p_C(\mathrm{\&theta;},t_{\mathrm{back}},m_0)+w_{\mathrm{foreA}}*p_A(\mathrm{\&theta;},t_{\mathrm{back}},m_1)}{W_{\mathrm{All}}}$

In order to reduce actual interpolation width; Interpolating function utilizes high-order moment;

(generally speaking; Selecting m is 2, if the narrower interpolation width of expectation, optional m＞2).

Above with have only the z direction to fly focus, 3 D back projection method is that example is described in detail embodiment of the invention CT scan image rebuilding method; Need to prove; The embodiment of the invention is not only applicable to have only the z direction to fly the flying method of focus, and other fly the focus mode and are suitable for too.

Visible by foregoing description; Embodiment of the invention CT scan image rebuilding method flies the focus interpolation with the z direction of principal axis and is attached in back projection's process of image reconstruction, makes to fly the focus interpolation and back projection handles disposable completion; Promptly to each selected target pixel points; Confirm that at the sampled point of z shaft position said projection passes the layer position of said target pixel points according to the ray of each focus, thereby reduced actual interpolation width, avoided in the prior art two step interpolation the influence of z direction of principal axis resolution

Correspondingly, the embodiment of the invention also provides a kind of CT scan equipment for reconstructing image, and is as shown in Figure 7, is a kind of structural representation of this device.

In this embodiment, said device comprises: data recombination unit 701, channel position computing unit 702, the layer position calculation unit 703 and the unit 704 that adds up.Wherein:

The data format of recombinating scan-data to require to back projection in data recombination unit 701; Channel position computing unit 702 calculates projection and passes the channel position of said target pixel points each selected target pixel points; Layer position calculation unit 703 be to each selected target pixel points, confirms that at the sampled point of z shaft position said projection passes the layer position of said target pixel points according to the ray of all types of focuses; The channel position that said target pixel points is passed in the said projection that unit 704 will obtain of adding up and layer position weighted accumulation are on said target pixel points.

In this embodiment, can recombinate scan-data to single layer data in said data recombination unit 701, perhaps scan-data is recombinated to multi-layer data.

Said layer position calculation unit 703 comprises: first computation subunit 731 and interpolation processing subelement 732.Wherein:

The ray that said first computation subunit 731 is calculated each focus respectively passes the theoretical layer position of said target pixel points; The ray of each focus that said interpolation processing subelement 732 will obtain passes the theoretical layer position of said target pixel points and carries out interpolation processing, obtains the layer position that said target pixel points is passed in said projection.

Said interpolation processing subelement 732 comprises: interpolation point is confirmed subelement and weighting subelement, also can further comprise: the weight setting subelement.Wherein:

Said interpolation point confirms that theoretical layer position that subelement passes said target pixel points to the ray of each focus rounds first interpolation point that obtains should focus downwards, and said first interpolation point is added 1 obtains second interpolation point that should focus; The weighting subelement carries out weighting to all interpolation points of each focus, obtains the layer position that said target pixel points is passed in said projection; The weight setting subelement is set the weights that said interpolation point carries out weighting according to the interpolation point of corresponding each focus to the distance of the actual integer layer of the ray of this focus.

The detailed process that interpolation processing is carried out in the theoretical layer position that utilizes the ray of each focus that interpolation processing subelement 732 obtains first computation subunit 731 to pass said target pixel points can be not described in detail at this with reference to the description in the embodiment of the invention CT scan image rebuilding method of front.

Identical similar part is mutually referring to getting final product between each embodiment in this description, and each embodiment stresses all is the difference with other embodiment.Especially, for device embodiment, because it is basically similar in appearance to method embodiment, so describe fairly simplely, relevant part gets final product referring to the part explanation of method embodiment.Device embodiment described above only is schematic; Wherein said unit as the separating component explanation can or can not be physically to separate also; The parts that show as the unit can be or can not be physical locations also; Promptly can be positioned at a place, perhaps also can be distributed on a plurality of NEs.Can realize the purpose of present embodiment scheme according to the needs selection some or all of module wherein of reality.Those of ordinary skills promptly can understand and implement under the situation of not paying creative work.

More than disclosedly be merely preferred implementation of the present invention; But the present invention is not limited thereto; Any those skilled in the art can think do not have a creationary variation, and, all should drop in protection scope of the present invention not breaking away from some improvement and the retouching of being done under the principle of the invention prerequisite.

Claims (10)

1. a CT scan image rebuilding method is characterized in that, comprising:

Scan-data is recombinated to the data format of back projection's requirement;

To each selected target pixel points, calculate projection and pass the channel position of said target pixel points;

To each selected target pixel points, confirm that at the sampled point of z shaft position said projection passes the layer position of said target pixel points according to the ray of all types of focuses;

The channel position that said target pixel points is passed in the said projection that obtains and layer position weighted accumulation are on said target pixel points.

2. the method for claim 1 is characterized in that, said the scan-data data format that requires to back projection of recombinating is comprised:

Scan-data is recombinated to single layer data; Perhaps

Scan-data is recombinated to multi-layer data.

3. the method for claim 1 is characterized in that, said ray according to all types of focuses confirms that at the sampled point of z shaft position said projection passes the layer position of said target pixel points and comprise:

The ray that calculates each focus respectively passes the theoretical layer position of said target pixel points;

The ray of each focus that obtains is passed the theoretical layer position of said target pixel points and carry out interpolation processing, obtain the layer position that said target pixel points is passed in said projection.

4. method as claimed in claim 3 is characterized in that, the ray of said each focus that will obtain passes the theoretical layer position of said target pixel points and carries out interpolation processing, obtains said projection and passes the layer position of said target pixel points and comprise:

The theoretical layer position that the ray of each focus is passed said target pixel points rounds first interpolation point that obtains should focus downwards, and said first interpolation point is added 1 obtains second interpolation point that should focus;

All interpolation points to each focus carry out weighting, obtain the layer position that said target pixel points is passed in said projection.

5. method as claimed in claim 4 is characterized in that, said method also comprises:

Interpolation point according to corresponding each focus is set the weights that said interpolation point carries out weighting to the distance of the actual integer layer of the ray of this focus.

6. a CT scan equipment for reconstructing image is characterized in that, comprising:

The data recombination unit is used for the data format that scan-data is recombinated and required to back projection;

The channel position computing unit is used for each selected target pixel points is calculated projection and passed the channel position of said target pixel points;

Layer position calculation unit is used for each selected target pixel points, confirms that at the sampled point of z shaft position said projection passes the layer position of said target pixel points according to the ray of all types of focuses;

The unit that adds up, the channel position that is used for said target pixel points is passed in the said projection that obtains and layer position weighted accumulation are to said target pixel points.

7. device as claimed in claim 6 is characterized in that,

Said data recombination unit specifically is used for scan-data is recombinated to single layer data; Perhaps scan-data is recombinated to multi-layer data.

8. device as claimed in claim 6 is characterized in that, said layer position calculation unit comprises:

First computation subunit, the ray that is used for calculating respectively each focus passes the theoretical layer position of said target pixel points;

The interpolation processing subelement is used for the theoretical layer position that ray with each focus that obtains passes said target pixel points and carries out interpolation processing, obtains the layer position that said target pixel points is passed in said projection.

9. device as claimed in claim 8 is characterized in that, said interpolation processing subelement comprises:

Interpolation point is confirmed subelement, and the theoretical layer position that the ray of each focus is passed said target pixel points rounds first interpolation point that obtains should focus downwards, and said first interpolation point is added 1 obtains second interpolation point that should focus;

The weighting subelement is used for all interpolation points of each focus are carried out weighting, obtains the layer position that said target pixel points is passed in said projection.

10. device as claimed in claim 9 is characterized in that, said interpolation processing subelement also comprises:

The weight setting subelement is used for setting the weights that said interpolation point carries out weighting according to the interpolation point of corresponding each focus to the distance of the actual integer layer of the ray of this focus.

Images