CN102542534A - Image distortion correcting method and device based on image contour - Google Patents
Image distortion correcting method and device based on image contour Download PDFInfo
- Publication number
- CN102542534A CN102542534A CN201010618547XA CN201010618547A CN102542534A CN 102542534 A CN102542534 A CN 102542534A CN 201010618547X A CN201010618547X A CN 201010618547XA CN 201010618547 A CN201010618547 A CN 201010618547A CN 102542534 A CN102542534 A CN 102542534A
- Authority
- CN
- China
- Prior art keywords
- image
- coefficient
- source images
- pixel
- outline
- 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.)
- Granted
Links
Images
Abstract
The embodiment of the invention discloses an image distortion correcting method and device. The method comprises the following steps of: extracting an image contour of a source image; obtaining distortion information according to the image contour; and correcting the source image by using the distortion information so as to register the source image with a target image. By using the method and device, disclosed by the invention, the influences of the contrast, b value, direction variation and the like in a diffusion weighted magnetic resonance image on the correction are avoided; moreover, additional reference scanning is not needed so that convenient, fast and accurate image distortion correction is realized.
Description
Technical field
The present invention relates to diffusion-weighted imaging field, be specifically related to the image that diffusion-weighted imaging forms is carried out the method and apparatus of distortion correction.
Background technology
Diffusion imaging is a kind of functional mri technology that has very much clinical value; It can be surveyed hydrone or can in MR imaging apparatus, realize the biological characteristics that other molecule of diffusion imaging spreads in tissue, thereby detects the pathology (non-patent literature 1) that many conventional medical image means can not be found.Diffusion imaging uses echo-planar imaging (EPI) sequence to realize usually; EPI sequence image taking speed is very fast; But receive the especially influence of pattern distortion of various pseudo-shadows (non-patent literature 2 and 3) easily; In the EPI sequence; Because the extra magnetic field Δ B that the heterogeneity of main field etc. causes can be expressed as (non-patent literature 4 and 5) respectively in the pattern distortion that causes on frequency coding direction and the phase-encoding direction: the distortion on
frequency coding direction is very little usually; Can ignore the distortion Δ y that general only consideration causes on phase-encoding direction.
Diffusion-weighted echo-planar imaging (DW-EPI) sequence realizes diffusion-weighted through in the EPI sequence, adding diffusion gradient.The diffusion gradient that increases can cause extra vortex field in imaging space, this extra vortex field is the main cause that pattern distortion produces.Order
The expression vortex field is at main field B
0Component on the direction can expand into (non-patent literature 3):
First b wherein
0(t) irrelevant with the locus, be commonly referred to the zeroth order item, second is linear term, perhaps is called the single order item, component is gx, gy, gz is illustrated respectively in x, y, the gradient of vortex field on three directions of z.More the item of high-order is smaller usually, can ignore.In addition,, thereby make the integral image signal intensity descend, reduce the contrast of image, can not cause pattern distortion though the z component in the linear term can cause choosing layer phase place to return the imperfection that gathers.Therefore,, only consider the x in zeroth order item and the single order item usually for the correction of pattern distortion, the y component, thus following formula can be rewritten as:
B
e(x，y)＝b
0+x·g
x+y·g
y (2)
In the expression formula of substitution pattern distortion Δ y, the expression formula of the pattern distortion that can obtain causing by the vortex field:
Can find out that from following formula the zeroth order vortex field can cause the displacement of integral image, can make image produce shearing deformation, make image produce flexible deformation along the single order item of phase-encoding direction along the single order item of frequency coding direction.
Need in imaging sequence, add diffusion gradient in the diffusion-weighted imaging, as measuring the isoparametric means of coefficient of diffusion.The b value is and the adding diffusion gradient pulse shape lumped parameter relevant with intensity that the b value is big more usually, the width of gradient pulse and highly also big good more to the picture contrast of last formation; Yet big b value needs that amplitude is big, the diffusion gradient pulse of longer duration, so can cause bigger vortex field, causes even more serious pattern distortion.Diffusion-weighted imaging need be gathered the different a plurality of images different with direction of b value usually; Thus; When in pulse train, using the different diffusion gradient parameter value of being correlated with b value, can cause different vortex field distributions, the pattern distortion degree and the type that cause thus are incomplete same.Therefore, need one by one to the correction of these pattern distortions that image carries out respectively, just can obtain diffusion-weighted clearly imaging (DWI) image, it is more accurate to make that apparent diffusion coefficient figure calculates (ADC mapping).
The method that suppresses the vortex field has the gradient coil of use self-shileding and the preparatory reinforcement (non-patent literature 6-8) of gradient pulse.Self-shield coil comprises a main coil and a potted coil; The direction and the main coil in the magnetic field that potted coil produces when imaging are opposite; Through particular design; Make this a pair of coil in imaging region, produce the gradient magnetic of expectation, and the magnetic field outside imaging region is approaching as far as possible zero, this has just fundamentally reduced the distribution and the size of vortex field.And strengthen technology in advance is through changing the shape of gradient waveform, offsetting the influence of vortex field.Through making in this way, can reduce the influence of vortex field, but can not eliminate fully.And diffusion-weighted imaging is very responsive to the vortex field, in order to improve the quality of imaging, must carry out distortion correction to the image that obtains.
Yet as stated, diffusion-weighted imaging need be gathered the different a plurality of images of b value usually, causes picture contrast to differ greatly; In addition, different b values and method cause different vortex fields to distribute, and the pattern distortion degree and the type that cause are incomplete same, have to one by one that image carries out image distortion correction respectively, and this can bring extra reference scan, increase sweep time.
The image distortion correction method that needs are a kind of easily and fast, error is little.
Non-references 1: Bao Shanglian, modern medicine Physics of Medicine Imaging, Beijing: medical science publishing house of Peking University, 2004
Non-references 2: an ancient sacrificial utensil woods, Magnetic resonance imaging is learned, Beijing: Higher Education Publishing House, 2004
Non-references 3:Bernstein M, King K, Zhou X, Fong W.Handbook of MRI pulse sequences.San Diego:Elsevier Academic Press; 2004.
Non-references 4:Jezzard P, Balaban RS.Correction for geometricdistortion in echo-planar images from B-0field variations.MagneticResonance in Medicine 1995; 34 (1): 65-73.
Non-references 5:Zeng HR, Constable RT.Image distortioncorrection in EPI:Comparison of field mapping with point spreadfunction mapping.Magnetic Resonance in Medicine 2002; 48 (1): 137-146.
Non-references 6:Mansfield P, Chapman B.Active magneticscreening of coils for static and time-dependent magnetic-field generationin NMR imaging.Journal of Physics E-Scientific Instruments1986; 19 (7): 540-545.
Non-references 7:Vanvaals JJ, Bergman AH.Optimization ofeddy-current compensation.Journal of Magnetic Resonance1990; 90 (1): 52-70.
Non-references 8:Jehenson P; Westphal M, Schuff N.Analyticalmethod for the compensation of eddy-current effects induced by pulsedmagnetic-field gradients in NMR systems.Journal of Magnetic Resonance1990; 90 (2): 264-278.
Summary of the invention
For addressing the above problem; The invention provides a kind of image distortion correction method and apparatus; Can obtain the distortion information that is used to proofread and correct based on image outline; Avoided the influence to proofreading and correct such as picture contrast, b value and direction variation, and do not needed extra reference scan, realized easily and fast, image distortion correction accurately.
According to an aspect of the present invention, a kind of image distortion correction method is provided, has comprised the steps:
A. the image outline of extraction source image;
B. obtain distortion information according to image outline;
C. adopt distortion information that source images is proofreaied and correct, with the target image registration.
According to the embodiment of the invention, step b comprises: the image outline of source images and the image outline of target image are compared, and displacement calculating coefficient and coefficient of dilatation are as distortion information.
According to the embodiment of the invention, this method also comprised before step a:
D. source images being carried out histogram equalization handles;
Wherein, in step a, extract the image outline of the source images after handling.
According to the embodiment of the invention, in step a, the usage level diversity method comes the image outline of extraction source image.
According to the embodiment of the invention, in step b, to all pixel columns of image outline, displacement calculating coefficient and coefficient of dilatation line by line are to obtain the pixel-shift figure of whole source images, as distortion information.
According to the embodiment of the invention, in step b, to the one part of pixel row of image outline, displacement calculating coefficient and coefficient of dilatation line by line;
Wherein, said method also comprises between step b and step c: the displacement coefficient and the coefficient of dilatation of said one part of pixel row are carried out match and be extrapolated to other pixel columns, to obtain the pixel-shift figure of whole source images, as distortion information.
According to the embodiment of the invention, source images is image or the image sequence that diffusion-weighted imaging forms.
According to a further aspect of the invention, a kind of image distortion correction device is provided, has comprised:
Extraction unit, the image outline of extraction source image;
Acquiring unit obtains distortion information according to image outline;
Correcting unit adopts distortion information that source images is proofreaied and correct, with the target image registration.
According to the embodiment of the invention, acquiring unit compares the image outline of source images and the image outline of target image, and displacement calculating coefficient and coefficient of dilatation are as distortion information.
According to the embodiment of the invention, this device also comprises:
The equalization unit carries out histogram equalization to source images and handles;
Wherein, the image outline of the source images after the extraction unit extraction equalization cell processing.
According to the embodiment of the invention, to all pixel columns of image outline, acquiring unit is displacement calculating coefficient and coefficient of dilatation line by line, to obtain the pixel-shift figure of whole source images, as distortion information.
According to the embodiment of the invention, to the one part of pixel row of image outline, acquiring unit is displacement calculating coefficient and coefficient of dilatation line by line;
Wherein, said device also comprises: the extrapolation unit, the displacement coefficient and the coefficient of dilatation of the said one part of pixel row that acquiring unit is calculated carry out match and are extrapolated to other pixel columns, to obtain the pixel-shift figure of whole source images, as distortion information.
Image distortion correction method and apparatus of the present invention obtains distortion information based on image outline, is used for proofreading and correct, and does not need image that similar contrast is arranged, thereby can be used in the bigger situation of b value; In addition, do not need the change of sequence and extra reference scan, can not increase sweep time.Image after the correction can be used to calculate isotropic DWI image and ADC figure, has eliminated the pseudo-shadow in these images.
Description of drawings
Following accompanying drawing has shown embodiment of the present invention.These accompanying drawings and embodiment provide some embodiments of the present invention with the mode of non-limiting, non exhaustive property.
Fig. 1 shows the process flow diagram according to the image distortion correction method of the embodiment of the invention;
Fig. 2 shows displacement and the convergent-divergent relation between target image and the source images;
Fig. 3 shows the computer mould graphoid that extracts according to image outline in the image distortion correction method of the embodiment of the invention;
Fig. 4 is the schematic configuration diagram of the image distortion correction device of the embodiment of the invention;
Fig. 5 shows the synoptic diagram of the correcting result of the image distortion correction method that adopts the embodiment of the invention; And
Fig. 6 shows the figure through the ADC behind the image distortion correction of the embodiment of the invention.
Embodiment
With reference to the accompanying drawings, specify the image distortion correction method and apparatus of the embodiment of the invention.In ensuing explanation, some concrete details, for example the concrete element among the embodiment, step, concrete parameter etc. all are used for to embodiments of the invention better understanding being provided.The invention is not restricted to these concrete elements, step or parameter.In addition, specific descriptions have been omitted to algorithm commonly known in the art, means etc.Even the technician in present technique field is appreciated that embodiments of the invention also can be implemented under the situation that lacks combinations such as some details or other elements, step, parameter.
For clear, briefly the embodiment of the invention described, below be that example is described mainly with echo-planar imaging (DW-EPI) technology diffusion-weighted in the magnetic resonance imaging (MRI).Yet; The embodiment of the invention is not limited thereto; But be applicable to that distort having of various diffusion-weighted imagings formation or the correction of the image of pseudo-shadow, comprise that the diffusion-weighted imaging (DWI) of other types, apparent diffusion coefficient calculate (ADC mapping), diffusion tensor imaging (DTI), fibrous bundle tracking etc.
Fig. 1 shows the process flow diagram according to the image distortion correction method of the embodiment of the invention, and Fig. 2 shows target image I
0(x is y) with source images I
1(x, displacement between y) and convergent-divergent relation.Use 100 pairs of source images of image distortion correction method to proofread and correct from the DW-EPI sequence, to eliminate in the source images since distortion diffusion-weighted or that diffusion gradient causes make it and the target image registration.Image distortion correction method 100 comprises: step 102, the image outline of extraction source image; Step 104 is obtained distortion information according to image outline; And step 106, adopt distortion information that source images is proofreaied and correct, with the target image registration.
Below in conjunction with Fig. 2 the image distortion correction method 100 of the embodiment of the invention is specifically described.Among Fig. 2, a has indicated target image I
0(x, y), for the b value is zero, distortionless image, b has indicated source images I
1(x y), is b value image non-vanishing, that have distortion, and wherein, x representes that along the pixel coordinate of frequency coding direction, y representes the pixel coordinate along phase-encoding direction.In the step 102 of bearing calibration 100, adopt level set (1evel set) method extraction source image I
1(x, profile y).For the specific descriptions of Level Set Method, can be referring to non-patent literature 9 (Chan TF, Vese LA.Active contours without edges.IEEETransactions on Image Processing, 2001; 10 (2): 266-277.).It will be understood by those skilled in the art that the embodiment of the invention is not limited to the level set contour extraction method, but can adopt known other any suitable contour extraction methods of art technology to extract image outline.
In step 102, target image I
0(x is predefined y), and target image I
0(x, profile y) can extract in advance.Alternatively, in step 102, also can with source images I
1(x, profile y) extracts together, extracts target image I
0(x, profile y).
With image shown in Figure 2 is example, and those skilled in the art can know and see, target image I
0(x is y) with source images I
1(x, profile y) are respectively circular and oval-shaped, source images I
1(x, profile phase y) is for target image I
0(x, displacement has taken place in profile y), shearing and flexible, promptly of the background technology part, diffusion-weighted middle vortex field has caused whole displacement, shearing and flexible.Of the background technology part, the distortion on the frequency coding direction is very little usually, can ignore, the distortion that general only consideration causes on phase-encoding direction.In addition, the image outline that obtains is handled line by line, eliminated the influence of shearing distortion thus, only need to consider the displacement and flexible two kinds of deformation of each pixel column, I
1(x) carry out an amount of flexible and translation, just can with I
0(x) alignment:
I
0(x)＝S(x)·I
1(x)+T(x) (4)
Wherein, S (x), T (x) representes coefficient of dilatation and the displacement coefficient that x is capable respectively.
Based on above analysis, in the present embodiment,, the image outline of source images and the image outline of target image are compared in step 104, displacement calculating coefficient and coefficient of dilatation as distortion information, are used for proofreading and correct.
Particularly, referring to Fig. 2, along the pixel column of phase-encoding direction, use I respectively in consideration target and the source images
0(x) and I
1(x) expression, the solid line among Fig. 2 a has been indicated I
0(x), the dotted line among Fig. 2 b has been indicated I
1(x), Fig. 2 c has indicated the pixel column I of target image
0(x) and the pixel column I of source images
1The comparison of the distortion information (x), the line segment y from about the pixel column shown in Fig. 2 c between two end points
0y
0' and y
1y
1' comparison can know and find out y
1y
1' with respect to y
0y
0' taken place to right translation, and convergent-divergent deformation taken place in length, and this is illustrated by " convergent-divergent " and " displacement " in Fig. 2 c.With y
0, y
0' and y
1, y
1' bring following formula (4) into, obtain S (x), T (x) is:
Like this, calculate flexible and displacement coefficient line by line, distortion in images is proofreaied and correct as distortion information.
According to one embodiment of the invention, in step 104, can be based on the image outline that extracts, to source images I
1(displacement calculating coefficient and coefficient of dilatation line by line are to obtain whole source images I for x, y) all pixel columns of image outline
1(x, pixel-shift figure y) is as distortion information.In conjunction with above example, with S (x) and T (x) calculate as follows pixel-shift figure psm (x, y):
psm(x，y)＝(y-y
center)·(S(x)-1)+T(x) (6)
Wherein, y
CenterIt is the coordinate of the capable center pixel of x.Obtain pixel-shift figure psm (x, y) after, in step 106, the method that just can use interpolation is to the source images I of distortion
1(x y) proofreaies and correct, and for example uses interpolation method.
According to another embodiment of the present invention, further contemplating can other distortion or the noise of existence except that the distortion that the vortex field causes in the image.For example, suppose that image has 256 row, several rows wherein are ground unrests, and for example 1-20 is capable capable with 240-256, preferably only uses 21-239 capable when obtaining distortion information so, to reduce the error that other distortion or noise bring.In addition, image outline extracts and also can introduce new noise or error.In order further to reduce the error of calculating line by line in flexible and the displacement coefficient process, under the condition of only considering zeroth order and first-order linear error, can use S (x), T (x) are carried out match, be extrapolated to row all in the image then.In view of the above, alternatively,, only be directed against the one part of pixel row of source images image outline, line by line displacement calculating coefficient and coefficient of dilatation in step 104.Then; Between step 104 and step 106; Bearing calibration 100 also comprises the steps: the displacement coefficient of this one part of pixel row and coefficient of dilatation are carried out match and be extrapolated to other pixel columns, to obtain the pixel-shift figure of whole source images, as distortion information.Here, match can be adopted any proper method well known in the art with extrapolation, considers the linear relationship of S (x) and x, preferably linear fit method, for example least square method.
With S ' (x), T ' (x) representes flexible and displacement coefficient after match and the extrapolation respectively, according to computes go out pixel-shift figure psm (x, y):
psm(x，y)＝(y-y
center)·(S′(x)-1)+T′(x) (7)
Wherein, y
CenterIt is the coordinate of the capable center pixel of x.Equally, obtain pixel-shift figure psm (x, y) after, in step 106, just can to the distortion source images I
1(x y) proofreaies and correct, for example with source images I
1(x, y) addition obtain and target image I
0(x, y) image of registration.
Owing to be based on image outline according to the image distortion correction method 100 of the embodiment of the invention, thus accurately image outline to extract be the prerequisite of accurate Calculation pixel-shift figure.For the importance of this prerequisite, describe in conjunction with example shown in Figure 3.
When using phantom and healthy volunteer that bearing calibration 100 is tested, utilize Level Set Method can accurately extract image outline.But; When the patient that brain tumor is arranged is scanned,, may have influence on the extraction of image boundary because tumour shows as high signal in diffusion-weighted figure; At this moment need use some image processing techniquess; Such as histogram equalization etc., reduce the contrast of tumour and normal tissues, so that correctly find image boundary.Fig. 3 shows the computer mould graphoid that extracts according to image outline in the image distortion correction method 100 of the embodiment of the invention; Fig. 3 a is the Shepp-Logan model that computer simulation produces; One of them ellipse is set to high signal; Be used for the analog spread tumor region in when imaging, added the noise of Gaussian distribution in addition in the image; Fig. 3 b is the profile (like the white arrow indication) that directly extracts with Level Set Method, and what obtained this moment is not to be the result who wants, because profile only comprises tumor region; Fig. 3 c carries out the result behind the histogram equalization to Fig. 3 a, and Fig. 3 d is that the image to Fig. 3 c carries out profile and extracts, and has obtained the image outline of actual needs, shown in white arrow among the figure.
Therefore; In the image distortion correction method 100 of the embodiment of the invention; In order to extract image outline exactly; Before step 102, source images and/or target image are carried out the histogram equalization processing, in step 102, source images after handling and/or target image are carried out the profile extraction then.Thus, can further reduce profile and extract the error that causes, improve the accuracy of pixel-shift and correction.
Main or alternative step according to the image distortion correction method 100 of the embodiment of the invention has more than been described.For each source images in the image sequence of diffusion-weighted imaging formation; Can be to all b value and all directions in the scanning; Carry out above-mentioned processing; And then based on image calculation diffusion weighted images or ADC value after handling, what just can reduce to a great extent that diffuse images gradient eddy current causes is image blurring, makes that the calculating of ADC value is more accurate.
Describe the image distortion correction device 400 of the embodiment of the invention below with reference to Fig. 4, shown in the structural representation of Fig. 4, image distortion correction device 400 comprises: extraction unit 402, the image outline of extraction source image; Acquiring unit 404 obtains distortion information according to image outline; And correcting unit 406, adopt distortion information that source images is proofreaied and correct, with the target image registration.
According to the embodiment of the invention, acquiring unit 404 compares the image outline of source images and the image outline of target image, and displacement calculating coefficient and coefficient of dilatation are as distortion information.
According to the embodiment of the invention, this device 400 can also comprise equalization unit (not shown), source images is carried out histogram equalization handle.Extraction unit 402 extracts the image outline of the source images after the equalization cell processing.
According to the embodiment of the invention, to all pixel columns of source images image outline, acquiring unit 404 is displacement calculating coefficient and coefficient of dilatation line by line, to obtain the pixel-shift figure of whole source images, as distortion information.
According to the embodiment of the invention, to the one part of pixel row of source images image outline, acquiring unit 404 is displacement calculating coefficient and coefficient of dilatation line by line.In this case, device 400 also comprises extrapolation unit 408, and the displacement coefficient and the coefficient of dilatation of the one part of pixel row that acquiring unit 404 is calculated carry out match and be extrapolated to other pixel columns, to obtain the pixel-shift figure of whole source images, as distortion information.
Below with reference to the image distortion correction method 100 of the bright embodiment of the invention of as an exampleBSEMGVR takeN-PSVSEMOBJ of Fig. 5 and 6 and the calibration result of device 400.
Fig. 5 shows the example of the correcting result of the image distortion correction method that adopts the embodiment of the invention.In this experimental example, the object of testing at body is the healthy volunteer, and imaging parameters is: TE/TR=105/3400msec; The visual field: 230 * 230mm; Imaging array: 160 * 160; Bandwidth: 895Hz/Pixel; Average time: 3; The b value that diffusion imaging uses is respectively: 0,500,1000.Fig. 5 has shown the result who the image of the image of b=500 and b=1000 is carried out distortion correction, and wherein white dashed line is represented the outline line that extracts.Fig. 5 a and Fig. 5 d are the distortionless target images of b=0, and outline line and image boundary are fine identically; Fig. 5 b and Fig. 5 c are without the b=500 of overcorrect and the source images of b=1000, can find out, in the zone of white arrow indication, outline line and image boundary have bigger difference; Fig. 5 e and Fig. 5 f are the images that adopts corrected b=500 of image distortion correction method of the present invention and b=1000, and the outline line of extraction and image boundary are fine identically, and be also fine identically with the outline line of the image of b=0.It is thus clear that Fig. 5 example description image distortion correction method of the present invention has effectively been proofreaied and correct the pattern distortion that cause the vortex field.
Fig. 6 shows the ADC figure that the image calculation after the image distortion correction method of utilizing the embodiment of the invention is proofreaied and correct obtains.The ADC figure that Fig. 6 a is to use not calibrated diffusion-weighted figure to calculate, the part of white arrow indication has manifest error, and this is because the distortion of the diffusion-weighted figure of b=500 and b=1000 causes.The ADC figure that Fig. 6 b is to use the corrected diffusion-weighted figure of bearing calibration of the present invention to calculate, picture quality has had obvious improvement.Fig. 6 example has shown the improvement effect of the image distortion correction method of the embodiment of the invention to ADC figure.
Specifically describe in conjunction with above, image distortion correction method and apparatus of the present invention obtains distortion information based on image outline, is used for proofreading and correct, and does not need image that similar contrast is arranged, and can be used in multiple different b value, the particularly bigger situation of b value; In addition, do not need the change of sequence and extra reference scan, can not increase sweep time.Image after the correction can be used to calculate isotropic DWI image and ADC figure, has eliminated the pseudo-shadow in these images.It is thus clear that image distortion correction method and apparatus of the present invention can be easily and fast and the distortion and the pseudo-shadow that cause of corrected for dispersion gradient exactly, is applicable to polytype diffusion-weighted imaging technique.
Instructions of the invention described above and embodiment only are illustrated the image distortion correction method and apparatus of the embodiment of the invention in an exemplary fashion, and are not used in the scope of the present invention that limits.It all is possible changing and revise for disclosed embodiment, other feasible selection property embodiment and can be understood by those skilled in the art the equivalent variations of element among the embodiment.Other variations of disclosed embodiment of this invention and modification do not exceed spirit of the present invention and protection domain.
Claims (12)
1. an image distortion correction method comprises the steps:
A. the image outline of extraction source image;
B. obtain distortion information according to image outline; And
C. adopt distortion information that source images is proofreaied and correct, with the target image registration.
2. method according to claim 1, wherein, step b comprises: the image outline of source images and the image outline of target image are compared, and displacement calculating coefficient and coefficient of dilatation are as distortion information.
3. method according to claim 1 and 2 also comprised before step a:
D. source images being carried out histogram equalization handles;
Wherein, in step a, extract the image outline of the source images after handling.
4. method according to claim 1 and 2, wherein, in step a, the usage level diversity method comes the image outline of extraction source image.
5. method according to claim 2, wherein, in step b, to all pixel columns of image outline, displacement calculating coefficient and coefficient of dilatation line by line are to obtain the pixel-shift figure of whole source images, as distortion information.
6. method according to claim 2, wherein, in step b, to the one part of pixel row of image outline, displacement calculating coefficient and coefficient of dilatation line by line;
Wherein, said method also comprises step e between step b and step c: the displacement coefficient and the coefficient of dilatation of said one part of pixel row are carried out match and be extrapolated to other pixel columns, to obtain the pixel-shift figure of whole source images, as distortion information.
7. method according to claim 1, wherein, source images is image or the image sequence that diffusion-weighted imaging forms.
8. image distortion correction device comprises:
Extraction unit, the image outline of extraction source image;
Acquiring unit obtains distortion information according to image outline; And
Correcting unit adopts distortion information that source images is proofreaied and correct, with the target image registration.
9. device according to claim 8, wherein, acquiring unit compares the image outline of source images and the image outline of target image, and displacement calculating coefficient and coefficient of dilatation are as distortion information.
10. also comprise according to Claim 8 or 9 described devices:
The equalization unit carries out histogram equalization to source images and handles;
Wherein, the image outline of the source images after the extraction unit extraction equalization cell processing.
11. device according to claim 9, wherein, to all pixel columns of image outline, acquiring unit is displacement calculating coefficient and coefficient of dilatation line by line, to obtain the pixel-shift figure of whole source images, as distortion information.
12. device according to claim 9, wherein, to the one part of pixel row of image outline, acquiring unit is displacement calculating coefficient and coefficient of dilatation line by line;
Wherein, said device also comprises: the extrapolation unit, the displacement coefficient and the coefficient of dilatation of the said one part of pixel row that acquiring unit is calculated carry out match and are extrapolated to other pixel columns, to obtain the pixel-shift figure of whole source images, as distortion information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010618547.XA CN102542534B (en) | 2010-12-31 | 2010-12-31 | Image distortion correcting method and device based on image contour |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010618547.XA CN102542534B (en) | 2010-12-31 | 2010-12-31 | Image distortion correcting method and device based on image contour |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102542534A true CN102542534A (en) | 2012-07-04 |
| CN102542534B CN102542534B (en) | 2015-01-14 |
Family
ID=46349359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010618547.XA Expired - Fee Related CN102542534B (en) | 2010-12-31 | 2010-12-31 | Image distortion correcting method and device based on image contour |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102542534B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103376433A (en) * | 2012-04-27 | 2013-10-30 | 西门子(深圳)磁共振有限公司 | Method and system for correcting image distortion and magnetic resonance imaging device |
| CN105096247A (en) * | 2014-05-09 | 2015-11-25 | 北京大学 | Image interpolation method and device |
| CN105122298A (en) * | 2013-03-29 | 2015-12-02 | 皇家飞利浦有限公司 | Image registration |
| CN105989596A (en) * | 2015-02-12 | 2016-10-05 | 东芝医疗***株式会社 | Device and method of processing medial image and magnetic resonance imaging device |
| CN106873162A (en) * | 2017-03-14 | 2017-06-20 | 上海天马微电子有限公司 | The pixel arrangement method of display device, display device and near-eye display device |
| CN107368806A (en) * | 2017-07-18 | 2017-11-21 | 广东欧珀移动通信有限公司 | Image correction method, device, computer-readable recording medium and computer equipment |
| CN108426702A (en) * | 2018-01-19 | 2018-08-21 | 华勤通讯技术有限公司 | The dispersion measurement device and method of augmented reality equipment |
| CN108885246A (en) * | 2015-12-09 | 2018-11-23 | 皇家飞利浦有限公司 | For generating the diffusion MRI method of the synthesis diffusion image at high b value |
| CN109143134A (en) * | 2017-06-13 | 2019-01-04 | 西门子(深圳)磁共振有限公司 | A kind of segmentation reads Diffusion-Weighted MR Imaging method, apparatus and can storage medium |
| CN109480884A (en) * | 2018-12-06 | 2019-03-19 | 宁波耀通管阀科技有限公司 | Profile thickness analysis institution |
| CN110942064A (en) * | 2019-11-25 | 2020-03-31 | 维沃移动通信有限公司 | Image processing method and device and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6061476A (en) * | 1997-11-24 | 2000-05-09 | Cognex Corporation | Method and apparatus using image subtraction and dynamic thresholding |
| US20080055654A1 (en) * | 2006-09-01 | 2008-03-06 | Canon Kabushiki Kaisha | Image processing apparatus and its method |
| CN101354785A (en) * | 2008-09-04 | 2009-01-28 | 湖南大学 | Method for accurately positioning vision in cleaning robot of condenser |
| WO2010017524A2 (en) * | 2008-08-07 | 2010-02-11 | New York University | System, method and computer accessible medium for providing real-time diffusional kurtosis imaging |
| CN101916445A (en) * | 2010-08-25 | 2010-12-15 | 天津大学 | Affine parameter estimation-based image registration method |
-
2010
- 2010-12-31 CN CN201010618547.XA patent/CN102542534B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6061476A (en) * | 1997-11-24 | 2000-05-09 | Cognex Corporation | Method and apparatus using image subtraction and dynamic thresholding |
| US20080055654A1 (en) * | 2006-09-01 | 2008-03-06 | Canon Kabushiki Kaisha | Image processing apparatus and its method |
| WO2010017524A2 (en) * | 2008-08-07 | 2010-02-11 | New York University | System, method and computer accessible medium for providing real-time diffusional kurtosis imaging |
| CN101354785A (en) * | 2008-09-04 | 2009-01-28 | 湖南大学 | Method for accurately positioning vision in cleaning robot of condenser |
| CN101916445A (en) * | 2010-08-25 | 2010-12-15 | 天津大学 | Affine parameter estimation-based image registration method |
Non-Patent Citations (2)
| Title |
|---|
| 张森: "数字图像几何畸变自动校正算法的研究与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 03, 31 March 2009 (2009-03-31), pages 138 - 479 * |
| 韦春荣 等: "基于轮廓提取的医学图像配准方法", 《广西师范大学学报(自然科学版)》, vol. 21, no. 2, 30 June 2003 (2003-06-30), pages 33 - 36 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103376433A (en) * | 2012-04-27 | 2013-10-30 | 西门子(深圳)磁共振有限公司 | Method and system for correcting image distortion and magnetic resonance imaging device |
| CN105122298A (en) * | 2013-03-29 | 2015-12-02 | 皇家飞利浦有限公司 | Image registration |
| CN105096247A (en) * | 2014-05-09 | 2015-11-25 | 北京大学 | Image interpolation method and device |
| CN105096247B (en) * | 2014-05-09 | 2018-02-06 | 北京大学 | Image interpolation method and device |
| CN105989596A (en) * | 2015-02-12 | 2016-10-05 | 东芝医疗***株式会社 | Device and method of processing medial image and magnetic resonance imaging device |
| CN105989596B (en) * | 2015-02-12 | 2019-09-10 | 东芝医疗***株式会社 | Medical image processing devices and method and MR imaging apparatus |
| CN108885246A (en) * | 2015-12-09 | 2018-11-23 | 皇家飞利浦有限公司 | For generating the diffusion MRI method of the synthesis diffusion image at high b value |
| CN108885246B (en) * | 2015-12-09 | 2021-04-09 | 皇家飞利浦有限公司 | Diffusion MRI method for generating synthetic diffusion images at high b-values |
| CN106873162A (en) * | 2017-03-14 | 2017-06-20 | 上海天马微电子有限公司 | The pixel arrangement method of display device, display device and near-eye display device |
| CN109143134A (en) * | 2017-06-13 | 2019-01-04 | 西门子(深圳)磁共振有限公司 | A kind of segmentation reads Diffusion-Weighted MR Imaging method, apparatus and can storage medium |
| US10928476B2 (en) | 2017-06-13 | 2021-02-23 | Siemens Healthcare Gmbh | Readout-segmented diffusion-weighted imaging method, apparatus, and storage medium |
| CN107368806B (en) * | 2017-07-18 | 2020-01-10 | Oppo广东移动通信有限公司 | Image rectification method, image rectification device, computer-readable storage medium and computer equipment |
| CN107368806A (en) * | 2017-07-18 | 2017-11-21 | 广东欧珀移动通信有限公司 | Image correction method, device, computer-readable recording medium and computer equipment |
| CN108426702A (en) * | 2018-01-19 | 2018-08-21 | 华勤通讯技术有限公司 | The dispersion measurement device and method of augmented reality equipment |
| CN108426702B (en) * | 2018-01-19 | 2020-06-02 | 华勤通讯技术有限公司 | Dispersion measurement device and method of augmented reality equipment |
| CN109480884A (en) * | 2018-12-06 | 2019-03-19 | 宁波耀通管阀科技有限公司 | Profile thickness analysis institution |
| CN110942064A (en) * | 2019-11-25 | 2020-03-31 | 维沃移动通信有限公司 | Image processing method and device and electronic equipment |
| CN110942064B (en) * | 2019-11-25 | 2023-05-09 | 维沃移动通信有限公司 | Image processing method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102542534B (en) | 2015-01-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102542534B (en) | Image distortion correcting method and device based on image contour | |
| Lutti et al. | Using high-resolution quantitative mapping of R1 as an index of cortical myelination | |
| JP5789196B2 (en) | Means and methods for providing high resolution MRI | |
| Zhuang et al. | Correction of eddy‐current distortions in diffusion tensor images using the known directions and strengths of diffusion gradients | |
| CN105308469B (en) | A kind of MR imaging method, MR equipment and relevant data carrier | |
| US10162037B2 (en) | Navigator-based data correction for simultaneous multislice MR imaging | |
| CN104337516B (en) | The determination of magnetic resonance control sequence | |
| JP2014158731A (en) | Magnetic resonance imaging system and method therefor | |
| EP3123192B1 (en) | Epi ghost correction involving sense | |
| US6815952B1 (en) | Magnetic resonance diffusion imaging with eddy-current compensation | |
| US10429476B2 (en) | Algebraic reconstruction method for off-resonance and eddy-current correction in functional and diffusion weighted magnetic resonance imaging | |
| CN104688226A (en) | Medical image processing device and magnetic resonance imaging device | |
| Keating et al. | Real‐time dynamic frequency and shim correction for single‐voxel magnetic resonance spectroscopy | |
| Vos et al. | Trade-off between angular and spatial resolutions in in vivo fiber tractography | |
| Bastin et al. | On the use of water phantom images to calibrate and correct eddy current induced artefacts in MR diffusion tensor imaging | |
| Chung et al. | An improved PSF mapping method for EPI distortion correction in human brain at ultra high field (7T) | |
| CN103018694A (en) | Distortion correction method, magnetic resonance apparatus, program product and data medium | |
| US7689015B2 (en) | Magnetic resonance imaging apparatus and image correction estimating method | |
| Chen et al. | Application of k-space energy spectrum analysis to susceptibility field mapping and distortion correction in gradient-echo EPI | |
| US7042220B2 (en) | Magnetic resonance imaging system and method | |
| Corbin et al. | Imperfect spoiling in variable flip angle T1 mapping at 7T: Quantifying and minimizing impact | |
| Lee et al. | A comprehensive approach for correcting voxel‐wise b‐value errors in diffusion MRI | |
| Schwarz et al. | GRAPPA reconstructed wave‐CAIPI MP‐RAGE at 7 tesla | |
| EP3589197A1 (en) | System and method for improved diffusion-weighted imaging | |
| US10746832B2 (en) | DTI with correction of motion-induced diffusion gradient inconsistency |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150114 Termination date: 20181231 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |