CN2919379Y - Image rebuilding device adopting straight trace scanning method - Google Patents

Image rebuilding device adopting straight trace scanning method Download PDF

Info

Publication number
CN2919379Y
CN2919379Y CN 200520114671 CN200520114671U CN2919379Y CN 2919379 Y CN2919379 Y CN 2919379Y CN 200520114671 CN200520114671 CN 200520114671 CN 200520114671 U CN200520114671 U CN 200520114671U CN 2919379 Y CN2919379 Y CN 2919379Y
Authority
CN
China
Prior art keywords
projection
data
image
scanning
utility
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
CN 200520114671
Other languages
Chinese (zh)
Inventor
张丽
高河伟
陈志强
康克军
程建平
李元景
刘以农
邢宇翔
赵自然
肖永顺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tsinghua University
Nuctech Co Ltd
Original Assignee
Tsinghua University
Qinghua Tongfang Weishi Tech Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tsinghua University, Qinghua Tongfang Weishi Tech Co Ltd filed Critical Tsinghua University
Priority to CN 200520114671 priority Critical patent/CN2919379Y/en
Application granted granted Critical
Publication of CN2919379Y publication Critical patent/CN2919379Y/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Abstract

The utility model discloses an image rebuilding device of adopting linear track scan, to avoid the reduction of image resolution due to the angle direction insert value and the detector insert value. This device comprises a projection data conversion section which is used to convert the projection data under the linear track scanning into the projection data under the parallel bunch scanning, a filter section for gaining the projection date through the preconcerted roll integral function check and the parallel electron scanning, and a reverse projection section which rebuilds the image through making weighted reverse projection. By adopting the device of the utility model, the image rebuilding resolution is enhanced, and the influence rising from the data breaking is thus decreased. The utility model adopts a filter reverse projective mode, thereby having the advantages of common filter reverse projection, such as speedy rebuilding, and being beneficial for parallel process, and convenient in increasing speed and etc.

Description

A kind of equipment for reconstructing image that adopts straight path scanning
Technical field
The utility model relates to the radiant image field, and particularly, the utility model relates to a kind of equipment for reconstructing image that adopts straight path scanning, with the speed that scans in improving three-dimensional imaging.
Background technology
Along with the development of CT technology, under limited angle and data truncation situation, also can reconstruct the faultage image of certain mass, this makes incomplete scan rebuilding image use in practice becomes possibility.Knowing in theory, is the imaging system of straight line for scanning pattern, if the straight line endless, just could the accurate reconstruction faultage image.If scanning pattern is a finite length, then be equivalent to the CT scan pattern of limited angle (Limited-Angle).Therefore, use incomplete reconstruction algorithm, the data of the imaging system collection of rectilinear scanning are done reconstruction, just can obtain faultage image, realize three-dimensional imaging.
In the safety inspection of reality, speed passage through customs three-dimensional imaging and big object rotation are difficult problems.Because need in traditional CT imaging system inspected object is carried out circular orbit scanning, or the inspected object rotation, perhaps testedly have a medical check-up fixing and to justify track scanning inspection.That is to say, or scanning system rotation or inspected object rotation.Because need rotation, this is for large-scale inspected object, for example the vehicle of train and truck and so on is difficult.In addition, also there is big cone angle problem in rotary scanning.
At the problems referred to above, a kind of imaging system of straight path has been proposed---computing machine layering tomography (Computed Laminography) system, its ray subtended angle is very little, reconstruction algorithm is the mode that adopts chromatography, but the three-dimensional imaging ability is not enough, also is not applied in the safety inspection at present.Another kind of technology is to adopt shuffle algorithm, and the data for projection that is about to rectilinear scanning is reset to the data under the collimated beam situation, carries out the reconstruction of image then.But there is the low problem of spatial resolution in this algorithm, and this is because when the straight path scanning data rearrangement is circular orbit collimated beam scanning data, can cause systemic resolution to descend in the interpolation of angle direction and detector direction.
Summary of the invention
In view of the above problems, the utility model has been proposed.A purpose of the present utility model provides a kind of equipment for reconstructing image that adopts straight path scanning, it does not need to reset data for projection is collimated beam, thereby avoided in the data rearrangement because the image resolution ratio that angle direction interpolation and detector direction interpolation cause descends, and improved the quality of rebuilding tomography.
Aspect one of the present utility model, a kind of equipment for reconstructing image that adopts straight path scanning has been proposed, comprising: the data for projection conversion portion is used for converting the data for projection under the straight path scanning to intend under the collimated beam scanning data for projection; The filtering part by with predetermined convolution function nuclear and the data for projection convolution of intending under the collimated beam scanning, obtains filtered data for projection; And back projection's part, by being weighted back projection, filtered data for projection comes reconstructed image.
In addition, according to an embodiment of the present utility model, equipment for reconstructing image also comprises the detector array of a plurality of detector cells, is used to receive the transmission signal that penetrates the examine object from the ray of radiographic source emission, and converts transmission signal to data for projection.
In addition, according to an embodiment of the present utility model, described a plurality of detector cells are equidistant arrangements.
In addition, according to an embodiment of the present utility model, the anti-pleat translation of described data for projection conversion portion data for projection p (l, t, z) to obtain intending the data for projection q (l under the collimated beam scanning, t, z), wherein, data for projection p (l, t, z) expression is the projection value at t place when examine object of which movement when coordinate is the l position to straight line at detector array z layer coordinate position.Described filtering partly with predetermined convolution function nuclear along the l direction to intend under the collimated beam scanning data for projection q (l, t z) do the one dimension convolution, obtain filtered data for projection Q (l ', t, z).Described back projection part along directions of rays to filtered data for projection O (l ', t z) is weighted back projection's operation, to obtain reconstructed image, wherein weight factor is
Figure Y20052011467100041
Wherein D represents the distance of radiographic source to the rectilinear motion center line.
In addition, according to an embodiment of the present utility model, described a plurality of detector cells are equiangularly arranged about radiographic source.
In addition, according to an embodiment of the present utility model, the anti-pleat translation of described data for projection conversion portion data for projection p (l, γ, z) to obtain intending the data for projection q (l under the collimated beam scanning, γ, z), wherein, data for projection p (l, γ, z) table expression is the projection value of γ when object of which movement when coordinate is the l position to straight line in detector array z layer angle position.Described filtering partly with predetermined convolution function nuclear along the l direction to intend under the collimated beam scanning data for projection q (l, γ z) do the one dimension convolution, obtain filtered data for projection O (l ', γ, z).Described back projection part along directions of rays to filtered data for projection O (l ', γ z) is weighted back projection's operation, to obtain reconstructed image, wherein weight factor is 1/cos γ.
According to device of the present utility model, a kind of image rebuilding method that adopts straight path scanning has been proposed, comprising: the data for projection switch process converts the data for projection under the straight path scanning to the data for projection of intending under the collimated beam scanning; Filter step by with predetermined convolution function nuclear and the data for projection convolution of intending under the collimated beam scanning, obtains filtered data for projection; And back projection's step, by being weighted back projection, filtered data for projection comes reconstructed image.
In addition, according to an embodiment of the present utility model, method of the present utility model also comprises step: utilize the detector array that comprises a plurality of detector cells to receive the transmission signal that penetrates the examine object from the ray of radiographic source emission, and convert transmission signal to data for projection.
In addition, according to an embodiment of the present utility model, described a plurality of detector cells are equidistant arrangements.
In addition, according to an embodiment of the present utility model, the anti-pleat translation of described data for projection switch process data for projection p (l, t, z) to obtain intending the data for projection q (l under the collimated beam scanning, t, z), wherein, data for projection p (l, t, z) expression is the projection value at t place when examine object of which movement when coordinate is the l position to straight line at detector array z layer coordinate position.Described filter step with predetermined convolution function nuclear along the l direction to intend under the collimated beam scanning data for projection q (l, t z) do the one dimension convolution, obtain filtered data for projection O (l ', t, z).Described back projection step along directions of rays to filtered data for projection O (l ', t z) is weighted back projection operation, to obtain reconstructed image, wherein weight factor is Wherein D represents the distance of radiographic source to the rectilinear motion center line.
In addition, according to an embodiment of the present utility model, described a plurality of detector cells are equiangularly arranged about radiographic source.
In addition, according to an embodiment of the present utility model, the anti-pleat translation of described data for projection switch process data for projection p (l, γ, z) to obtain intending the data for projection q (l under the collimated beam scanning, γ, z), wherein, data for projection p (l, γ, z) table expression is the projection value of γ when object of which movement when coordinate is the l position to straight line in detector array z layer angle position.Described filter step with predetermined convolution function nuclear along the l direction to intend under the collimated beam scanning data for projection q (l, γ z) do the one dimension convolution, obtain filtered data for projection Q (l ', γ, z).Described back projection step along directions of rays to filtered data for projection Q (l ', γ z) is weighted back projection operation, to obtain reconstructed image, wherein weight factor is 1/cos γ.
The utility model is compared with shuffle algorithm, has improved image reconstruction resolution, reduces the influence of data truncation to reconstructed image.The utility model adopts the form of filtered back projection, thereby has the advantage of general filtered back projection, such as rebuilding fast, is beneficial to parallelly, is convenient to quicken etc.
Like this, compare with traditional perspective imaging, device of the present utility model can obtain fluoroscopy images, can obtain faultage image again, can solve the overlapped object problem in the perspective imaging, realizes the quick stereo imaging in the safety check system.Imaging is compared with traditional CT, and device of the present utility model has simplicity of design, and inspection speed is fast, does not need rotation, does not have the advantages such as " big cone angle " problem in the circular orbit Cone-Beam CT.
Description of drawings
Fig. 1 is the floor map according to imaging device straight path scanning of the present utility model;
Fig. 2 is the formation synoptic diagram according to the imaging device of the utility model embodiment;
Fig. 3 is the control in as shown in Figure 2 the imaging device and the functional block diagram of image processing apparatus;
Fig. 4 shows the synoptic diagram of the geometric relationship of equivalent detector between Z direction and reconstruction object point;
Fig. 5 shows and is used for explaining straight line filtered back-projection method geometric relationship synoptic diagram of the present utility model;
Fig. 6 shows the process flow diagram according to straight line filtered back-projection method of the present utility model;
Fig. 7 shows method of the present utility model and rebuilds the data of being gathered with rearrangement parallel beam reconstruction stereo imaging system, and the analog image of acquisition (X-Y plane) effect relatively.
Embodiment
The contrast accompanying drawing is described embodiment of the present utility model in detail below.
Fig. 1 is illustrated in the synoptic diagram that imaging device of the present utility model carries out rectilinear scanning in the three-dimensional imaging safety inspection.Fig. 2 shows the formation synoptic diagram of imaging device of the present utility model.
As shown in Figure 1, according to rectilinear motion, in the process of motion, radiographic source A sends ray according to the order of control system to the examine object, penetrates the examine object between radiographic source A and detector.Detector receives transmission signal, and under the control of control system recording projection data, and data for projection is stored in the storer.
Device as shown in Figure 2 comprises: radiographic source 110, and for example X-ray tube, accelerator radiographic source or isotope source or the like are selected according to the size and the application background of examine object 120; Carrying and the steady drive apparatus 130 that transmits examine object 120 are used for making the examine object 120 that is carried along rectilinear motion in checking process; The detector array 140 that comprises a plurality of detector cells, be positioned at radiographic source 110 opposites, perpendicular to drive apparatus 130, subtended angle with radiographic source 110 is the angle of spending greater than 90 in the horizontal direction, for example between 90~180 the degree between angle, and cover object at vertical direction, this detector array can be a planar array detector, also can be single row detector; Be used for transmitting control and data signal control and data signal line 150; And by controling and data signal control and the image processing apparatus 160 that line 150 is connected with radiographic source 110, drive apparatus 130 and detector array 140, it controls drive apparatus 130 along rectilinear motion and order radiographic source 110 to launch ray in scanning process, and control detector array 140 begins to receive transmission signal, produce data for projection, and the data for projection that obtains is carried out aftertreatment.
Like this, examine object 120 is at the uniform velocity mobile according to straight path shown in Figure 1, and sampling in detector array 140 synchronous constant duration ground, obtains data for projection.
Fig. 3 is the control in as shown in Figure 2 the imaging device and the functional block diagram of image processing apparatus 160.As shown in Figure 3, control and image processing apparatus 160 comprise: be used for storing memory of data 161, for example the storage medium of hard disk and so on; Input block 162, for example keyboard and so on makes things convenient for the input media of user's input parameter or order; Controller 163, it is after the user gives an order by input block 162, and instruction drive apparatus 130 begins examine object 120 along the straight line uniform motion, and radiographic source 110 and detector array 140 start working, to obtain data for projection; The internal bus 164 that connects various piece and transmission of control signals and data; And image reconstruction unit 165, be used for the data for projection that detector array 140 obtains is rebuild.
Be described in detail in below in conjunction with Fig. 4 and carry out image reconstruction process in the image reconstruction unit 165.Fig. 4 shows the synoptic diagram of the geometric relationship of equivalent detector between Z direction and reconstruction object point.
If (a kind of approximate evaluation z) is expressed as examine object f for r, φ
Figure Y20052011467100081
Following formula is then arranged:
f ^ ( r , φ , z ) = ∫ - t m t m 1 D 2 + t 2 Q ( l ′ , t , z D D + r sin φ ) dt - - - ( 1 )
Wherein,
Q(l′,t,z)=q(l,t,z)*h(l) (2)
q(l,t,z)=p(-l+t,t,z) (3)
l ′ = - r cos φ + tr sin φ D - - - ( 4 )
Here, detector is equidistant the arrangement, and (z) expression is the projection value at t place at detector array 140 z layer coordinate position when object 120 moves to that coordinate is the l position on the straight line to data p for l, t.It should be noted that t, z is each detector cells equivalence of detector array 140 numerical value after to the straight-line center line of object.
In addition, in formula (1)~(4), D is the distance of radiographic source 110 to the rectilinear motion center line; ± t mCharacterize minimum and the maximum position of detector array 140 on the examine movement direction of object; H is a convolution function nuclear, and theoretical value is h ( l ) = ∫ - ∞ ∞ | ω | e j 2 πωl dω , The general S-L filter function that adopts, the discrete form of this function is:
h ( n ) = - 2 π 2 ( 4 n 2 - 1 ) , n=0,±1,±2,… (5)
Therefore, in image reconstruction unit 165, (z) (z), (z) data for projection under the collimated beam scanning is intended in expression to the q here to data for projection conversion portion 1651 anti-pleat translation data for projection p for l, t for l, t to obtain q for l, t.The implication of here ' plan collimated beam scanning ' is, the detector equivalent sampling spacing under all angles is inequality, and the scanning angle sampling also may be inhomogeneous.
Then, filtering part 1652 usefulness convolution functions nuclear h along the l direction to intend under the collimated beam scanning data for projection q (l, t z) do the one dimension convolution, obtain filtered data for projection Q (l, t, z).
Next, (l, t z) are weighted back projection's operation to back projection's part 1653, and to obtain reconstructed image, wherein weight factor is to filtered data for projection Q along directions of rays
Figure Y20052011467100086
Here, need to prove, the purpose of carrying out anti-pleat translation is for the data for projection with rectilinear scanning, transform to the data for projection of intending under the collimated beam scanning, the plan collimated beam scanning here is not the collimated beam during the CT of standard rebuilds, because the detector equivalent sampling spacing under each scanning angle is all different, the angle sampling also may be uneven.
In addition, identical with the purpose of convolution function nuclear h filtering with the filtering in the standard FBP reconstruction algorithm, filtered data for projection Q (l ', t z), just can obtain reconstructed image through weighted back projection, and the weight factor here Reflected the size that the data for projection under the different detectors is contributed reconstructed image.
Therefore, in the utility model, along data acquisition direction l filtering, along directions of rays back projection, to compare for the algorithm of collimated beam with rearrangement, the utility model can make full use of each valid data, improved image resolution ratio, and the sensitivity that data are blocked has been lower than shuffle algorithm.
Contrast Fig. 1,4 and 5 above-mentioned formula (1) of deriving below.Before deriving, the process that the rectilinear scanning data rearrangement is the circular orbit collimated beam scanning is described at first.
Scan mode as shown in Figure 1, each surveys all corresponding angle scanning angle, and (x y) in moving process, is equivalent to the collimated beam scanning under this angle to object f.Referring to the perspective view of Fig. 5, for the detector array of equidistant arrangement, the rectilinear scanning data rearrangement is that the rearrangement formulae of circular orbit collimated beam scanning is:
g ( θ , s ) = p ( l , t ) | s = D ( - l + t ) D 2 + t 2 θ = π - tan - 1 ( t D ) - - - ( 6 )
Here g (θ, s)=(x, y) (dxdy of xcos θ+ysin θ-s) is illustrated in the circular orbit collimated beam scanning δ ∫ f, and scanning angle is θ, is the data for projection of s apart from rotation center.(l, t) when coordinate was the l position to straight line, the detector array coordinate position was the projection value of t to the expression detector array to p at object of which movement.
Utilize formula (6) just can realize that the rearrangement of straight path scanning data for projection is the data for projection under the circular orbit collimated beam scanning.But in the real system, straight line can not be an endless, so the data after resetting also can not be the collimated beam scanning data of 180 degree under the circular orbit, also is to say here to rebuild for CT that data are incomplete.
With regard to rectilinear scanning, though the sampling of l and t can be uniformly, angle θ under the scanning of corresponding circle parallel track bundle and detector position s sampling all are uneven.Therefore, rearrangement need be done interpolation at angle direction and detector direction, causes the reduction of reconstructed image resolution.
Next, describe the direct filtered back projection of rectilinear scanning data of the present utility model process of reconstruction in detail.
Filtered back projection's reconstruction formula under the circular orbit collimated beam scanning is
f ( r , φ ) = ∫ 0 π ∫ - s m s m g ( θ , s ) h ( r cos ( θ - φ ) - s ) dsdθ - - - ( 7 )
For endless straight path, equidistant detector, utilize formula (7), (l, t) (θ s), obtains in replacement with parameter
f ( r , φ ) = ∫ - ∞ ∞ ∫ - ∞ ∞ 1 D 2 + t 2 p ( - l + t , t ) h ( l ′ - l ) dldt - - - ( 8 )
Here, l ′ = - r cos φ + tr sin φ D .
Prove as follows:
f ( r , φ ) = ∫ ∞ - ∞ ∫ ∞ - ∞ g ( π - tan - 1 ( t D ) , D ( - l + t ) D 2 + t 2 )
· h ( r cos ( π - tan - 1 ( t D ) - φ ) - D ( - l + t ) D 2 + t 2 ) D 2 ( D 2 + t 2 ) 3 / 2 dldt - - - ( 9 )
Here, θ = π - tan - 1 ( t D ) , s = D ( - l + t ) D 2 + t 2 , dsdθ = D 2 ( D 2 + t 2 ) 3 / 2 dldt .
In straight path scanning, (l t) replaces with p Simultaneously, according to the geometry of Fig. 5, can draw
r cos ( π - tan - 1 ( t D ) - φ ) - D ( - l + t ) D 2 + t 2
= - r cos φ D D 2 + t 2 + r sin φ t D 2 + t 2 - D ( - l + t ) D 2 + t 2
= ( l ′ - tr sin φ D - t ) D D 2 + t 2 + r sin φ t D 2 + t 2 - D ( - l + t ) D 2 + t 2
= D D 2 + t 2 ( l ′ + l - 2 t ) - - - ( 10 )
Here, l ′ = - r cos φ + tr sin φ D + t , Its expression through point (r, φ) and the data for projection of t detector, the spatial sampling position under rectilinear scanning.
With (10) substitution (9), utilize h [ D D 2 + t 2 ( l ′ + l - 2 t ) ] = D 2 + t 2 D 2 h ( l ′ + l - 2 t ) ,
And with l=l-t, l '=l '-t substitution obtains reconstruction formula (8) immediately.
For formula (8), if with q (l, t)=p (l+t, t) substitution formula (8), then
f ( r , φ ) = ∫ - ∞ ∞ 1 D 2 + t 2 Q ( l ′ , t ) dt - - - ( 11 )
Here, and Q (l ', t)=q (l, t) * h (l).
In the reality, if straight path is [L, L], the detector total length is [t m, t m], the image of rebuilding according to formula (8) just is not that (x, y), it is a kind of approximate to accurate f.Simultaneously, if consider three-dimensional situation, inspected object f (x, y, a kind of being similar to z)
Figure Y20052011467100112
Can be expressed as shown in (1) formula.
Fig. 6 shows the process flow diagram according to straight line filtered back-projection method of the present utility model.As shown in Figure 6, after detector array 140 obtains data for projection and is stored in the storer 161, in the time will carrying out image reconstruction, at step S110, from storer, read the data for projection p that the straight line surface sweeping obtains (l, t, z).
Then, at step S120, (z) (z), (z) data for projection under the collimated beam scanning is intended in expression to the q here for l, t for l, t to obtain q for l, t by data for projection conversion portion 1651 anti-pleat translation data for projection p.
Then, at step S130, filtering part 1652 usefulness convolution functions nuclear h along the l direction to intend under the collimated beam scanning data for projection q (l, t z) do the one dimension convolution, obtain filtered data for projection Q (l ', t, z).
Next, at step S140, back projection's part 1653 along directions of rays to filtered data for projection Q (l ', t z) is weighted back projection operation, to obtain reconstructed image, wherein weight factor is
At step S150, the image of rebuilding is stored in the storer 161, perhaps be presented on the screen of display 170.
The filtered back projection's reconstruction formula (1) of straight path scanning and the detailed implementation of method of the present utility model have been derived under the equidistant detector array above.In fact, can also be between the detector cells in the detector array 140 according to arranging about radiographic source 110 equal angular modes.Arrange if detector is an isogonism, be similar to top derivation, its filtered back projection's reconstruction formula is:
f ^ ( r , φ , z ) = ∫ - γ m γ m 1 cos γ Q ( l ′ , γ , z D D + r sin φ ) dγ - - - ( 12 )
Wherein,
Q(l′,γ,z)=q(l,γ,z)*h(l) (13)
q(l,γ,z)=p(-l+Dtanγ,γ,z) (14)
l′=-rcosφ+rsinφtanγ (15)
Here, detector cells is that isogonism is arranged, and (z) when coordinate was the l position to straight line, detector array z layer angle position was the projection value of γ to the expression detector array to data p at object of which movement for l, γ.It should be noted that γ, z is the numerical value of detector array equivalence after to the straight-line center line of object.± γ mThe expression detector array is listed in minimum and the maximum angle on the examine movement direction of object.
Therefore, for equiangularly arranged detector, the process of reconstruction of straight line filtered back projection is same as described above, and wherein anti-pleat translation is carried out according to formula (14), under the implication of convolution operation and the equidistant situation identical with, and the weight factor that adopts in the weighted back projection operation is 1/cos γ.
In other words, in data for projection conversion portion 1651, anti-pleat translation data for projection p (l, γ, z) to obtain intending the data for projection q (l under the collimated beam scanning, γ, z), wherein, data for projection p (l, γ, z) table expression is the projection value of γ when object of which movement when coordinate is the l position to straight line in detector array z layer angle position.
In filtering part 1652, with predetermined convolution function nuclear h along the l direction to intend under the collimated beam scanning data for projection q (l, γ z) do the one dimension convolution, obtain filtered data for projection Q (l ', γ, z);
In back projection's part 1653, along directions of rays to filtered data for projection Q (l ', γ z) is weighted back projection operation, to obtain reconstructed image, wherein weight factor is 1/cos γ.
For arriving image reconstruction accurately, following systematic parameter should be able to accurately be measured or demarcate to x-ray imaging system: the X ray source point to detector apart from T, the X ray source point is to rectilinear motion centre distance D, gearing point-to-point speed v, detector array sampling interval Δ t, the detector physical size comprises the physical size of single detector physical size and detector array etc.
Fig. 7 shows with method of the present utility model and resets the collimated beam algorithm and compare rebuild faultage image (X-Y plane) simulation drawing that is obtained by the data of imaging system collection, wherein simulate ray subtended angle 150 degree, detector width 6mm, the equidistant arrangement, detector array equivalence spatial sampling is 3mm at interval.Reconstructed image 300*300 pixel, single Pixel Dimensions 3mm*3mm.This figure is the simulated experiment result who utilizes a luggage model to carry out, what rebuild is the image of central core, (A) expression original analog image wherein, (B) image of expression circular orbit parallel sweep accurate reconstruction, (C) expression is reset the sub-picture that the collimated beam method is rebuild with the data for projection that straight path scanning obtains, (D) image of expression method reconstruction of the present utility model, (E) data that the collimated beam method rebuilds are reset in expression another width of cloth image under the situation of blocking in detector direction, and (F) the expression method of the present utility model data of rebuilding have another width of cloth image under the situation of blocking in detector direction.By Fig. 7 more as can be seen, method of the present utility model obviously improves than resetting collimated beam method resolution, the influence of data truncation obviously reduces.
The above; it only is the embodiment in the utility model; but protection domain of the present utility model is not limited thereto; anyly be familiar with the people of this technology in the disclosed technical scope of the utility model; the conversion that can expect easily or replacement all should be encompassed in of the present utility model comprising within the scope.Therefore, protection domain of the present utility model should be as the criterion with the protection domain of claims.

Claims (4)

1, a kind of equipment for reconstructing image that adopts straight path scanning is characterized in that it comprises:
The data for projection conversion portion is used for converting the data for projection under the straight path scanning to intend under the collimated beam scanning data for projection;
The filtering part by with predetermined convolution function nuclear and the data for projection convolution of intending under the collimated beam scanning, obtains filtered data for projection; And
Back projection's part is come reconstructed image by filtered data for projection is weighted back projection.
2, equipment for reconstructing image as claimed in claim 1, it is characterized in that, it also comprises: comprise the detector array of a plurality of detector cells, be used to receive the transmission signal that penetrates the examine object from the ray of radiographic source emission, and convert transmission signal to data for projection.
3, equipment for reconstructing image as claimed in claim 2 is characterized in that, described a plurality of detector cells are equidistant arrangements.
4, equipment for reconstructing image as claimed in claim 2 is characterized in that, described a plurality of detector cells are equiangularly arranged about radiographic source.
CN 200520114671 2005-11-22 2005-11-22 Image rebuilding device adopting straight trace scanning method Expired - Lifetime CN2919379Y (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200520114671 CN2919379Y (en) 2005-11-22 2005-11-22 Image rebuilding device adopting straight trace scanning method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200520114671 CN2919379Y (en) 2005-11-22 2005-11-22 Image rebuilding device adopting straight trace scanning method

Publications (1)

Publication Number Publication Date
CN2919379Y true CN2919379Y (en) 2007-07-04

Family

ID=38217196

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200520114671 Expired - Lifetime CN2919379Y (en) 2005-11-22 2005-11-22 Image rebuilding device adopting straight trace scanning method

Country Status (1)

Country Link
CN (1) CN2919379Y (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009127118A1 (en) * 2008-04-17 2009-10-22 清华大学 Imaging system using a straight-line trajectory scan and method thereof
CN109313723A (en) * 2018-01-15 2019-02-05 深圳鲲云信息科技有限公司 Artificial intelligence convolution processing method, device, readable storage medium storing program for executing and terminal
CN109475336A (en) * 2016-06-09 2019-03-15 爱克发有限公司 Correction method for geometric distortion and device for chest tomography converged reconstruction
US11874898B2 (en) 2018-01-15 2024-01-16 Shenzhen Corerain Technologies Co., Ltd. Streaming-based artificial intelligence convolution processing method and apparatus, readable storage medium and terminal

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009127118A1 (en) * 2008-04-17 2009-10-22 清华大学 Imaging system using a straight-line trajectory scan and method thereof
CN109475336A (en) * 2016-06-09 2019-03-15 爱克发有限公司 Correction method for geometric distortion and device for chest tomography converged reconstruction
CN109313723A (en) * 2018-01-15 2019-02-05 深圳鲲云信息科技有限公司 Artificial intelligence convolution processing method, device, readable storage medium storing program for executing and terminal
CN109313723B (en) * 2018-01-15 2022-03-15 深圳鲲云信息科技有限公司 Artificial intelligence convolution processing method and device, readable storage medium and terminal
US11874898B2 (en) 2018-01-15 2024-01-16 Shenzhen Corerain Technologies Co., Ltd. Streaming-based artificial intelligence convolution processing method and apparatus, readable storage medium and terminal

Similar Documents

Publication Publication Date Title
CN100495439C (en) Image rebuilding system and method using straight track scanning
US7672427B2 (en) Imaging system
US7512213B2 (en) Multiple-view-angle cargo security inspection method and system
CN1711561A (en) Computed tomography and method with three-dimensional backprojection
CN100464707C (en) Processing system and method for reconstructing 3D pyramidal CT image
CN1865954A (en) Wide view-field three-D CT imaging method
CN102456227B (en) Reconstruction method and device for CT (computerized tomography) image
EP2629083A2 (en) 3d backscatter imaging system
CN105717145B (en) Multi-joint dress 3 D cone-beam computer tomography method and device
CN101034479A (en) Image rebuilding method and X-ray CT equipment
CN1251975A (en) Nutating slice CT image reconstruction
JP2013253969A (en) Handheld x-ray system for 3d scatter imaging
CN104809750A (en) Linear scanning CT system and image reconstructing method
CN104502382B (en) A kind of new core three-dimensional CT imaging devices and purposes
CN101672806B (en) Wide visual field cone-beam X ray oblique scanning three-dimension digital imaging method based on algebraic reconstruction algorithm
CN104757988B (en) A kind of electronic line of sight scans micro-nano focus CT scan system and method
CN2919379Y (en) Image rebuilding device adopting straight trace scanning method
CN101056584A (en) Ct method for the examination of a cyclically moving object
CN1489976A (en) Method and device for arranging images of movable objects for CT
CN104254786B (en) Computerized axial tomography imaging method and system
CN103808740B (en) Detection method based on computer hierarchal scanning imaging CL (Computer Laminography) system
CN1489977A (en) Method and device for arranging images of movable objects for CT
CN202950664U (en) X-ray imaging unit
CN107192726A (en) The quick high-resolution 3 D cone-beam computer tomography method of plate shell object and device
CN2860349Y (en) Processing system three-dimensional conical bundle CT image rebuilding

Legal Events

Date Code Title Description
C14 Grant of patent or utility model
GR01 Patent grant
C56 Change in the name or address of the patentee
CP03 Change of name, title or address

Address after: A, block 2907, Tsinghua Tongfang Technology Plaza, Beijing, 100083

Co-patentee after: Nuctech Company Limited

Patentee after: Tsinghua University

Address before: A, block 2907, Tsinghua Tongfang Technology Plaza, Beijing, 100083

Co-patentee before: Qinghua Tongfang Weishi Tech Co.,Ltd.

Patentee before: Tsinghua University

CX01 Expiry of patent term

Granted publication date: 20070704

EXPY Termination of patent right or utility model