CN105787905B - A kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current - Google Patents
A kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current Download PDFInfo
- Publication number
- CN105787905B CN105787905B CN201610176008.2A CN201610176008A CN105787905B CN 105787905 B CN105787905 B CN 105787905B CN 201610176008 A CN201610176008 A CN 201610176008A CN 105787905 B CN105787905 B CN 105787905B
- Authority
- CN
- China
- Prior art keywords
- response
- tube current
- bad point
- current
- formula
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 claims abstract description 37
- 238000001514 detection method Methods 0.000 claims abstract description 24
- 239000011159 matrix material Substances 0.000 claims abstract description 16
- 238000012360 testing method Methods 0.000 claims abstract description 10
- 238000012417 linear regression Methods 0.000 claims abstract description 5
- 238000010219 correlation analysis Methods 0.000 claims abstract description 4
- 238000012937 correction Methods 0.000 claims description 19
- 230000001131 transforming effect Effects 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 3
- 230000005855 radiation Effects 0.000 claims description 3
- 239000013077 target material Substances 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 description 9
- 238000002591 computed tomography Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000007796 conventional method Methods 0.000 description 3
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 239000004411 aluminium Substances 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000002985 plastic film Substances 0.000 description 2
- 229920006255 plastic film Polymers 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The Cone-Beam CT ring artifact bearing calibration based on dynamic current that the invention discloses a kind of, overcomes in the prior art, there is ring artifact residue problem in reconstruction image.The invention contains following steps: whether step 1 meets linear relationship with tube current using the response of dynamic current test member, will visit member and be divided into bad point and first two classes of non_uniform response spy；Step 2 judges individually to visit whether the increment of member response is zero under different tube currents, detect to first kind bad point；Step 3, the related coefficient for calculating output response and tube current that each visits member, detect the second class bad point using the method for correlation analysis；Step 4 calculates the equation of linear regression visited between member response and tube current, on the basis of the mean values for visiting member response all under single tube current, calculates the Concordance parameter matrix for visiting member.It is difficult that the present invention solves the problems, such as that existing dead pixel detection method threshold value determines, the method for the present invention is preferable to ring artifact calibration result, and versatility is stronger.
Description
Technical field
The invention is related to a kind of Cone-Beam CT ring artifact bearing calibration, more particularly to a kind of cone-beam based on dynamic current
CT ring artifact bearing calibration.
Background technique
Cone-Beam CT (Computed Tomography) system is mainly by x-ray source, flat panel detector, precision machinery and meter
Calculation machine composition, the three-dimensional structure information of interior of articles can be obtained under the conditions of non-contact, non-destructive, is widely used to nothing
The fields such as damage detection and medical diagnosis.It in practical application, is influenced by flat panel detector manufacturing process, is shone in identical ray energy
Penetrate down, it is each to visit between member response there are different degrees of difference, be easy to cause in CT reconstruction image occur it is a series of to rebuild in
The heart is the concentric loop in the center of circle.These show as the subsequent analysis to image of the ring artifact severe jamming of annulus shape, place
The work such as reason.Therefore research ring artifact bearing calibration has important practical significance.
Since the spy member of flat panel detector has stability over a period to come, detected in existing ring artifact bearing calibration
The practicability of device correction method is most strong.It is embodied in: the correction parameter of detector being generated into template, loads template when being then imaged every time
It is corrected, after a period of use more new template.It is detection that detection detector bad point and then generation bad point, which are distributed template,
One committed step of device correction method.
Dead pixel detection method usually first acquires the output image of detector when not placing any object under certain exposure levels
(hereinafter referred to as exposure image), then calculate exposure image certain parameters, then based on the certain threshold value of these parameter settings come
Identify bad point.Method is the direct mean value and variance for calculating and visiting member response in exposure image, then thinks that the variance of response is super
Crossing a certain range of spy member is bad point；It is to calculate the gain coefficient for visiting member response there are also method, is more than by gain coefficient later
A certain range of spy meta-tag is bad point；Above two method be easy by expose it is non-uniform influence, and more difficult determination is suitable
Threshold value.A kind of dead pixel detection method effect based on wavelet analysis that Tang et al. is proposed is preferable, not by the non-uniform shadow of exposure
Ring, but need to according to bad point distribution situation, will in dotted, tufted or the bad point separate detection of linear distribution, testing result obviously by
Bad point distribution form influences and the position of bad point cannot be accurately positioned.The present invention responds and tube current linear correlation according to member is visited
Feature proposes a kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current.It is first constant to being responded under different tube currents
Spy member detected；Then each related coefficient for visiting member response and tube current is calculated, to the random variation of response and rapidly
Reach maximum spy member to be detected.The spy member that will test out is used as bad point to charge to bad point template, row interpolation of going forward side by side correction, will
The mean value of each tube current test member response calculates as benchmark and respectively visits first response curve equation and benchmark under same group of tube current
The transforming relationship of curvilinear equation obtains the Concordance parameter matrix for all visiting member.Utilize bad point mould obtained by the above method
Plate and Concordance parameter are corrected the output response for visiting member in projection acquisition.
Summary of the invention
The present invention overcomes there is ring artifact residue problem in the prior art, in reconstruction image, a kind of significant effect is provided
The Cone-Beam CT ring artifact bearing calibration based on dynamic current.
The technical solution of the invention is as follows, and it is cyclic annular to provide a kind of Cone-Beam CT based on dynamic current having follow steps
Artifact correction method: contain following steps: whether step 1 meets linear pass with tube current using the response of dynamic current test member
System will visit member and be divided into bad point and first two classes of non_uniform response spy；Step 2 judges the increasing that member response is individually visited under different tube currents
Whether amount is zero, is detected to first kind bad point；Step 3, the output response for calculating each spy member are related to tube current
Coefficient detects the second class bad point using the method for correlation analysis；Step 4 is calculated and is visited between member response and tube current
Equation of linear regression calculates the Concordance parameter for visiting member on the basis of the mean values for visiting member response all under single tube current
Matrix.
The utilization method of dynamic current is as follows in the step 1: the tube voltage of fixed X-ray emitter, in detector
Tube current is adjusted in optimal dynamic response range, acquires multiple detectors output image under current tube current, and take it average
It is worth and exports image as the detector under the electric current；Finally multiple and different tube currents is continuously set, one can be obtained and is somebody's turn to do
The corresponding detector of current sequence exports image sequence.
It is as follows to visit first classification method for bad point and non_uniform response in the step 1: by the working principle of radiation detector
It is found that the signal that detector generates in a certain range is directly proportional to incident ray intensity；Wherein, the intensity E of X-ray0It can be with
It is indicated with following formula
E0=kZIVm, formula (1)
In formula, k is proportionality coefficient, and Z is the atomic number of target material, and I, V respectively indicate x-ray tube current and tube voltage,
Parameter m is approximately equal to 2；It therefore deduces that under fixed tube voltage, being visited between member response and tube current I in certain time for exposure has line
Property correlativity
G (x, y)=a (x, y) I+b (x, y), formula (2)
Wherein, g (x, y) represents the response visited at first (x, y), and a (x, y) is the slope of relation curve, and b (x, y) is relationship
The intercept of curve, and which includes the dark current outputs that member is visited when no radiation exposure；First response and tube current relationship is visited to deviate
The spy member of linear relationship is bad point；The spy member that response and tube current relationship meets linear relationship is non_uniform response
Visit member.
The feature of first kind bad point is as follows in the step 2: the output valve for visiting member is constant, including when ray tube current increases
When output valve it is constant be 0, the spy of maximum value or other constants member, this kind of response for visiting member do not change with the variation of tube current.
The feature of second class bad point is as follows in the step 3: visiting the output valve variation abnormality of member, including ray tube current connects
The fast-response of increasing speed of continuous output valve when increasing is rapidly reached the spy member that maximum value and output valve change at random, this kind of member of visiting
Response-tube current relationship is unsatisfactory for formula (2).
The detection method of second class bad point is specific as follows in the step 3: setting and visits first quantity on detector as M × N, X is penetrated
The sequence that is increased continuously of spool electric current is I=(I1,I2,…,In), it generates the pseudo random number r that one group of length is m and I is rearranged
Obtain random sequence Ir=(Ir1,Ir2,…,Irm), then visit first (x, y) corresponding response sequenceThen I is measured with Pearson correlation coefficient ρ (x, y)rWithIt
Between degree of correlation, it is specific as follows:
In formula,It is sequence IrAverage value,It is sequenceAverage value, it is available using formula (3)
The first corresponding related coefficient square ρ of all spies, recycles formula (4) to carry out bad point detection,
In formula, q is the threshold value of bad point detection, and the point that value is 0 in matrix D is bad point, according to the position of bad point in D,
Then bad point is corrected using interpolation method in projected image, the first kind bad point detected in step 2 also charges to template
D。
The Concordance parameter matrix solution procedure that member is visited in the step 4 is as follows: carrying out interpolation school to whole bad points
After just, all visits first response-tube current relationship and be all satisfied the equation form in formula (2), derived by formula (2) it is found that any two
Visit the corresponding response-tube current relation curve g (x of member1,y1)=a (x1,y1)I+b(x1,y1) and g (x2,y2)=a (x2,y2)I
+ b (x2,y2) also there is following transforming relationship
g(x2,y2)=g (x1,y1)P1+ Q1, formula (5)
Wherein, P1=a (x2,y2)/a(x1,y1), Q1=b (x2,y2)-b(x1,y1)·a(x2,y2)/a(x1,y1), due to complete
The mean value that member response is visited in portion represents the average level for visiting member response, therefore with all spies under different tube currents in Concordance
The average value of member response enables all average values for visiting member response under certain electric current as benchmark
The then corresponding sequence of average for visiting member response of sequence I
It can be calculatedEquation of linear regression between IWithFor benchmark sequence, by other spies
The response of member exports gI(x, y) using formula (5) toDo Concordance, the value after correction
G ' I (x, y)=gI(x, y) P (x, y)+Q (x, y), formula (8)
Wherein, P (x, y)=a0/ a (x, y), Q (x, y)=b0-b(x,y)·a0It is equal all to visit member response in this way by/a (x, y)
It is corrected to mean value level, it is hereby achieved that all Concordance coefficient matrix P and matrix Q for visiting member.
Compared with prior art, the present invention is based on the Cone-Beam CT ring artifact bearing calibrations of dynamic current with following excellent
Point: the present invention will visit member and carry out classification and Detection and school using this linearly related feature between member response and x-ray tube current of visiting
Just, a kind of ring artifact minimizing technology based on dynamic current is proposed, existing dead pixel detection method threshold value is solved and determines and be stranded
Difficult problem.Carry out actual experiment on different detector platforms and show that the method for the present invention is preferable to ring artifact calibration result,
Versatility is stronger.
Detailed description of the invention
Fig. 1 is detector correction flow chart in the Cone-Beam CT ring artifact bearing calibration the present invention is based on dynamic current；
Fig. 2 is different detector tube current tests in the Cone-Beam CT ring artifact bearing calibration the present invention is based on dynamic current
Member response grey scale change curve graph, such as the detector that PerkinElmer company model is XRD1621xN ES；
Fig. 3 is different detector tube current tests in the Cone-Beam CT ring artifact bearing calibration the present invention is based on dynamic current
Member response grey scale change curve graph, such as the detector that Varian company model is PaxScan 4030E；
Fig. 4 is PerkinElmer company model in the Cone-Beam CT ring artifact bearing calibration the present invention is based on dynamic current
For the bad point distribution on the detector of XRD1621xN ES；
Fig. 5 is that Varian company model is in the Cone-Beam CT ring artifact bearing calibration the present invention is based on dynamic current
Bad point distribution on the detector of PaxScan 4030E；
Fig. 6 is PerkinElmer company model in the Cone-Beam CT ring artifact bearing calibration the present invention is based on dynamic current
The reconstructed results of aluminium column are tested for XRD1621xN ES detector, (a) is conventional method correction；(b) it is corrected for the method for the present invention；
Fig. 7 is that Varian company model is in the Cone-Beam CT ring artifact bearing calibration the present invention is based on dynamic current
The reconstructed results of the detector experiment plastic film of PaxScan 4030E, (a) are after conventional method corrects；It (b) is present invention side
After method correction.
Specific embodiment
With reference to the accompanying drawings and detailed description to the present invention is based on the Cone-Beam CT ring artifact correction sides of dynamic current
Method is described further:
Visit the relationship between member response and tube current: by the working principle of radiation detector it is found that detecting in a certain range
The signal that device generates is directly proportional to incident ray intensity.Wherein, the intensity E of X-ray0It can be indicated with following formula
E0=kZIVm, formula (1)
In formula, k is proportionality coefficient, and Z is the atomic number of target material, and I, V respectively indicate x-ray tube current and tube voltage,
Parameter m is approximately equal to 2.It therefore deduces that under fixed tube voltage, being visited between member response and tube current I in certain time for exposure has line
Property correlativity
G (x, y)=a (x, y) I+b (x, y), formula (2)
Wherein, g (x, y) represents the response visited at first (x, y), and a (x, y) is the slope of relation curve, and b (x, y) is relationship
The intercept of curve, and which includes the dark current outputs that member is visited when no radiation exposure.In practical applications, dark current output is made
For noise, need to be deducted from projected image.
Visit the classification of member: according to the relationship analysis between visiting member response tube current show the response of normal condition test member with
Tube current is linearly related.However, being influenced by flat panel detector manufacturing process, member is visited individually, X-ray is responded in the presence of exception, this
Invention thinks that visiting first response-tube current relationship deviation linear relationship is all bad point, and bad point is divided by difference according to response
Following two class:
(1) the constant spy member of output valve, including when ray tube current increases output valve is constant is 0, maximum value or other
The spy member of constant, this kind of response for visiting member do not change with the variation of tube current；
(2) the spy member of output valve variation abnormality, including the excessive response of curve of output slope is rapid when ray tube current increase
Reach the spy member that maximum value and output valve change at random, this kind of response-tube current relation curve for visiting member is unsatisfactory for formula (2).
In addition, there are also the linear relationships that the spy member response of several situations and tube current meet formula (2), but relation equation is not
Together, this kind of spy member is known as non_uniform response spy member.The inconsistent appearance for visiting member is that there are individual differences due to visiting member, to intensity phase
Same X-ray non_uniform response, so indicating that the equation of each spy member response and tube current linear relationship curve also has randomness.
Bad point and non_uniform response, which visit member, can all lead to occur ring artifact in reconstruction image, but the output response of the two
It is different from tube current relationship, it needs to be corrected respectively.
Detector correction: front has carried out features of response analysis to member is visited, and will visit member and be divided into bad point and inconsistent spy member
Next two classes introduce the detector correction method of classification and Detection and Concordance based on bad point, side proposed by the present invention
Method process is shown in attached drawing 1.Method needs to obtain bad point template D, and according to the position in D to two class bad point separate detections of definition
First bad point is corrected, then calculates spy first Concordance parameter matrix P, Q and non_uniform response spy member is corrected.
Bad point detection and correction:
It, can be by judging individually to visit whether the increment of member response is zero progress under different tube currents for first kind bad point
Detection, and bad point position is denoted as 0 in bad point logging template D.For the second class bad point, response is discontented with tube current
The linear relationship of sufficient formula (2) proposes a kind of detection method based on correlation analysis for this kind of bad point present invention.
If visiting first quantity on detector is M × N, the sequence that is increased continuously of x-ray tube current is I=(I1,I2,…,In),
The pseudo random number r that one group of length is m is generated I is rearranged to obtain random sequence Ir=(Ir1,Ir2,…,Irm), then visit member (x,
Y) corresponding response sequence
The present invention selects to measure I with Pearson correlation coefficient ρ (x, y)rWithBetween degree of correlation, specifically such as
Under:
In formula,It is sequence IrAverage value,It is sequenceAverage value.It can be obtained using formula (4)
To the corresponding related coefficient square ρ of all spy members, formula (5) is recycled to carry out bad point detection.
In formula, q is the threshold value of bad point detection, and experiment shows that q takes 0.9 to can satisfy the bad of a variety of different type detectors
Point detection needs.The point that value is 0 in matrix D is bad point, and according to the position of bad point in D, interpolation is utilized in projected image
Method is corrected bad point.After carrying out interpolation correction to whole bad points, all visits first response-tube current relationship and be all satisfied formula
(2) equation form in is derived by formula (2) it is found that any two visit the corresponding response-tube current relation curve g (x of member1,y1)
=a (x1,y1)I+b(x1,y1) and g (x2,y2)=a (x2,y2)I+b(x2,y2) also there is following transforming relationship
g(x2,y2)=g (x1,y1)P1+Q1, formula (5)
Wherein, P1=a (x2,y2)/a(x1,y1), Q1=b (x2,y2)-b(x1,y1)·a(x2,y2)/a(x1,y1)。
Concordance: first response-tube current relationship is all visited after bad point correction and is all satisfied the equation shape in formula (2)
Formula, the mean value due to all visiting member response represent the average level for visiting member response, therefore with different pipe electricity in Concordance
All average values for visiting member response are flowed down as benchmark.Enable all average values for visiting member response under certain electric current
The then corresponding sequence of average for visiting member response of sequence I
It substitutes into known to formula (2)WithFor benchmark sequence, other responses for visiting member are exported into gI(x, y) benefit
With formula (5) toDo Concordance, the value after correction
g′I(x, y)=gI(x, y) P (x, y)+Q (x, y), formula (8)
Wherein, P (x, y)=a0/ a (x, y), Q (x, y)=b0-b(x,y)·a0It is equal all to visit member response in this way by/a (x, y)
It is corrected to mean value level, it is hereby achieved that all Concordance coefficient matrix P and matrix Q for visiting member.
For the efficiency and applicability for verifying the method for the present invention, carry out experiment on a few money different type detectors, and
It is compared respectively with the conventional correction methods based on biasing, gain and detector bad point template.Specific experiment parameter is shown in Table
1。
1 experiment parameter of table
It acquires 50 exposure images under each tube current to be averaged, the spy member response for obtaining a few money detectors becomes with tube current
It is as shown in Fig. 2 to change relation curve.As can be seen that visiting the mean value and X-ray tube of member response under each group experiment parameter of setting
All meet good linear relationship between electric current.
In terms of bad point detection effect, attached drawing 3 and attached drawing 4 are respectively shown and are examined in not homologous ray using the method for the present invention
The bad point of a few money detectors measured is distributed, and the white ' * ' in figure indicates that the bad point detected using the method for the present invention is being visited
Survey the position on device.It can be seen that all there is a large amount of bad points of the present invention on a few money detectors, and its distribution form is more
Kind multiplicity, however the method for the present invention is not influenced the detectability of bad point by bad point distribution form.
In order to more intuitively observe the calibration result of the method for the present invention, under the experiment parameter in table 1,1,2 difference of experiment
CT scan has been carried out to diameter 4cm aluminium column under 2.0mA electric current, CT scan has been carried out to plastic film under 200 μ A electric currents, it is real
Test that the results are shown in attached figure 5 and attached drawing 6.
It compares the reconstructed results (a) corrected in attached drawing 5 and attached drawing 6 using conventional method and utilizes the method for the present invention school
The reconstructed results (b) just obtained, it can be seen that the method for the present invention to as non_uniform response visit slight ring artifact caused by member (see
Attached drawing 5) and the obvious ring artifact (see attached drawing 6 (a)) as caused by bad point can reach good removal effect, compared to tradition
Bearing calibration corrects more thorough to ring artifact.
Claims (2)
1. a kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current, it is characterised in that: contain following steps:
Whether step 1 meets linear relationship with tube current using the response of dynamic current test member, will visit member and is divided into bad point and response
First two classes of inconsistent spy；
Wherein the utilization method of dynamic current is as follows: the tube voltage of fixed X-ray emitter, responds in the optimal dynamic of detector
Tube current is adjusted in range, acquires multiple detectors output image under current tube current, and take its average value as the electric current
Under detector export image；Finally multiple and different tube currents is continuously set, can obtain one it is corresponding with the current sequence
Detector export image sequence；
Wherein it is as follows to visit first classification method for bad point and non_uniform response: by the working principle of radiation detector it is found that certain
The signal that detector generates in range is directly proportional to incident ray intensity；Wherein, the intensity E of X-ray0Following formula table can be used
Show
E0=kZIVm, formula (1)
In formula, k is proportionality coefficient, and Z is the atomic number of target material, and I, V respectively indicate x-ray tube current and tube voltage, parameter m
It is approximately equal to 2；It therefore deduces that under fixed tube voltage, member response is visited in certain time for exposure and there is linear phase between tube current I
Pass relationship
G (x, y)=a (x, y) I+b (x, y), formula (2)
Wherein, g (x, y) represents the response visited at first (x, y), and a (x, y) is the slope of relation curve, and b (x, y) is relation curve
Intercept, and which includes when no radiation exposure visit member dark current output；First response and tube current relationship is visited to deviate linearly
The spy member of correlativity is bad point；Response and tube current relationship meets the spy member of linear relationship for non_uniform response spy
Member；
Step 2 judges individually to visit whether the increment of member response is zero under different tube currents, detect to first kind bad point；
Wherein the feature of first kind bad point is as follows: the output valve for visiting member is constant, including output valve is permanent when ray tube current increases
It is set to 0, maximum value or the spy member of other constants, this kind of response for visiting member does not change with the variation of tube current；
Step 3, the related coefficient for calculating output response and tube current that each visits member, using the method for correlation analysis to the
Two class bad points are detected；
Wherein the feature of the second class bad point is as follows: visiting the output valve variation abnormality of member, including when ray tube current increases continuously is defeated
Value fast-response of increasing speed out is rapidly reached the spy member that maximum value and output valve change at random, this kind of response for visiting member-pipe electricity
Flow relation is unsatisfactory for formula (2)；
Detection method is specific as follows:
If visiting first quantity on detector is M × N, the sequence that is increased continuously of x-ray tube current is I=(I1,I2,…,In), it generates
The pseudo random number r that one group of length is m rearranges I to obtain random sequence Ir=(Ir1,Ir2,…,Irm), then it is right to visit first (x, y)
The response sequence answeredThen it is measured with Pearson correlation coefficient ρ (x, y)
IrWithBetween degree of correlation, it is specific as follows:
In formula,It is sequence IrAverage value,It is sequenceAverage value, utilize the available institute of formula (3)
There is the corresponding related coefficient square ρ of spy member, formula (4) recycled to carry out bad point detection,
In formula, q is the threshold value of bad point detection, and the point that value is 0 in matrix D is bad point, according to the position of bad point in D, with
Bad point is corrected using interpolation method in projected image afterwards, the first kind bad point detected in step 2 also charges to template D；
Step 4 calculates the equation of linear regression visited between member response and tube current, visits the equal of member response with all under single tube current
On the basis of value, the Concordance parameter matrix for visiting member is calculated.
2. the Cone-Beam CT ring artifact bearing calibration according to claim 1 based on dynamic current, it is characterised in that: described
The Concordance parameter matrix solution procedure that member is visited in step 4 is as follows: after carrying out interpolation correction to whole bad points, all visiting member
Response-tube current relationship is all satisfied the equation form in formula (2), is derived by formula (2) it is found that any two visit the corresponding sound of member
Answer-tube current relation curve g (x1,y1)=a (x1,y1)I+b(x1,y1) and g (x2,y2)=a (x2,y2)I+b(x2,y2) also have
Following transforming relationship
g(x2,y2)=g (x1,y1)P1+Q1, formula (5)
Wherein, P1=a (x2,y2)/a(x1,y1), Q1=b (x2,y2)-b(x1,y1)·a(x2,y2)/a(x1,y1), due to whole spies
The mean value of member response represents the average level for visiting member response, therefore is rung in Concordance with spy members all under different tube currents
The average value answered enables all average values for visiting member response under certain electric current as benchmark
The then corresponding sequence of average for visiting member response of sequence I
It can be calculatedEquation of linear regression between IWithFor benchmark sequence, first sound are visited by other
G should be exportedI(x, y) using formula (5) toDo Concordance, the value after correction
g'I(x, y)=gI(x, y) P (x, y)+Q (x, y), formula (8)
Wherein, P (x, y)=a0/ a (x, y), Q (x, y)=b0-b(x,y)·a0/ a (x, y) all visits member response by school in this way
Mean value level is just arrived, it is hereby achieved that all Concordance coefficient matrix P and matrix Q for visiting member.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610176008.2A CN105787905B (en) | 2016-03-24 | 2016-03-24 | A kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610176008.2A CN105787905B (en) | 2016-03-24 | 2016-03-24 | A kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105787905A CN105787905A (en) | 2016-07-20 |
| CN105787905B true CN105787905B (en) | 2019-03-26 |
Family
ID=56390831
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610176008.2A Active CN105787905B (en) | 2016-03-24 | 2016-03-24 | A kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105787905B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108078577B (en) * | 2017-11-22 | 2020-10-16 | 上海奕瑞光电子科技股份有限公司 | Digital X-ray radiation system, attitude detection method and attitude detection system |
| CN108918559B (en) * | 2018-07-28 | 2021-08-17 | 北京纳米维景科技有限公司 | X-ray image detector for realizing image self-correction and method thereof |
| CN109712084B (en) * | 2018-12-10 | 2021-01-19 | 上海奕瑞光电子科技股份有限公司 | Image restoration method, image restoration system and flat panel detector |
| CN109697476B (en) * | 2019-02-01 | 2023-06-23 | 重庆大学 | X-ray photon counting detector consistency calibration method based on deep learning |
| CN109965896A (en) * | 2019-04-25 | 2019-07-05 | 江苏摩科特医疗科技有限公司 | A kind of body mould and bearing calibration for the correction of CT ring artifact |
| CN110400358B (en) * | 2019-07-17 | 2022-10-04 | 郑州信大先进技术研究院 | Annular artifact removing method for flat panel detector |
| CN111053568B (en) * | 2019-12-30 | 2021-10-08 | 苏州瑞派宁科技有限公司 | Method and device for correcting ring artifact in CT image and computer storage medium |
| CN113223109B (en) * | 2021-05-27 | 2022-08-12 | 明峰医疗系统股份有限公司 | Dead pixel correction method based on CT axis scanning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390285A (en) * | 2013-07-09 | 2013-11-13 | 中国人民解放军信息工程大学 | Cone beam computed tomography (CT) incomplete angle rebuilding method based on edge guide |
| CN103714578A (en) * | 2014-01-24 | 2014-04-09 | 中国人民解放军信息工程大学 | Single-layer rearrangement filtered backprojection reconstruction method aiming to half mulching helical cone beam CT |
| CN103813355A (en) * | 2014-02-21 | 2014-05-21 | 厦门大学 | Identification method for anomalous points of cooperative synchronization in distributed network |
| CN104835125A (en) * | 2015-05-08 | 2015-08-12 | 上海联影医疗科技有限公司 | Defect pixel correction method for flat-panel detector |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105556507B (en) * | 2013-09-18 | 2020-05-19 | 美国西门子医疗解决公司 | Method and system for generating a reconstructed image of a target object from an input signal |
-
2016
- 2016-03-24 CN CN201610176008.2A patent/CN105787905B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390285A (en) * | 2013-07-09 | 2013-11-13 | 中国人民解放军信息工程大学 | Cone beam computed tomography (CT) incomplete angle rebuilding method based on edge guide |
| CN103714578A (en) * | 2014-01-24 | 2014-04-09 | 中国人民解放军信息工程大学 | Single-layer rearrangement filtered backprojection reconstruction method aiming to half mulching helical cone beam CT |
| CN103813355A (en) * | 2014-02-21 | 2014-05-21 | 厦门大学 | Identification method for anomalous points of cooperative synchronization in distributed network |
| CN104835125A (en) * | 2015-05-08 | 2015-08-12 | 上海联影医疗科技有限公司 | Defect pixel correction method for flat-panel detector |
Non-Patent Citations (3)
| Title |
|---|
| Assessment of three dead detector correction methods for cone-beam computed tomography;David W. Nelms et al.;《Medical Physics》;20091031;第36卷(第10期);第4569-4576页 |
| CT图像环状伪影校正方法;李俊江 等;《北京航空航天大学学报》;20071130;第33卷(第11期);第1378-1382页 |
| 体积CT系统中的平板探测器校正方法;张定华 等;《仪器仪表学报》;20050228;第26卷(第2期);第157-163页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105787905A (en) | 2016-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105787905B (en) | A kind of Cone-Beam CT ring artifact bearing calibration based on dynamic current | |
| CN101262819B (en) | Direct measuring and correction of scatter for CT | |
| CN105025796B (en) | X-ray computed tomograohy apparatus and bearing calibration | |
| CN100587391C (en) | Projection rotating center measurement method suitable for 2D-CT scanning system | |
| US8000513B2 (en) | System and method for 3D time of flight PET forward projection based on an exact axial inverse rebinning relation in fourier space | |
| Braun et al. | A versatile noise performance metric for electrical impedance tomography algorithms | |
| US9801595B2 (en) | Count-weighted least squares parameter estimation for a photon-counting detector | |
| CN107110988B (en) | Pet detector timing alignment | |
| US11099141B2 (en) | Method and apparatus for processing photon counting-type X-ray detection data and X-ray apparatus | |
| JP5853123B2 (en) | Dynamic modeling of imperfections for photon counting detectors | |
| CN103814291A (en) | X-ray detector of grid-based phase-contrast x-ray device and method for operating grid-based phase-contrast x-ray device | |
| US7894568B2 (en) | Energy distribution reconstruction in CT | |
| CN104155677B (en) | Line of response localization method and device is met for pet detector | |
| CN108027447A (en) | The correction of photon counting in photon counting X-ray radiation detection system | |
| CN102917642A (en) | X-ray ct device | |
| WO2012120791A1 (en) | Positron ct device and reconstruction method | |
| Zhang et al. | Iterative geometric calibration in circular cone-beam computed tomography | |
| CN106471392B (en) | Image reconstruction process method | |
| US8411814B2 (en) | Method of and apparatus for measuring center detector index on CT scanner with limited field of view access | |
| US9355437B2 (en) | Image position compensation method and apparatus | |
| CN108065950A (en) | A kind of radiological imaging method and its system | |
| CN103405241A (en) | Detector afterglow correction method for ray imaging | |
| Levine et al. | Consistency conditions for cone-beam CT data acquired with a straight-line source trajectory | |
| CN110400358A (en) | A kind of flat panel detector ring artifact minimizing technology | |
| CN102759538B (en) | Method for measuring and modeling point spread function of cone beam CT system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |