CN107437251A - Head mri image skull strip module - Google Patents
Head mri image skull strip module Download PDFInfo
- Publication number
- CN107437251A CN107437251A CN201710619840.XA CN201710619840A CN107437251A CN 107437251 A CN107437251 A CN 107437251A CN 201710619840 A CN201710619840 A CN 201710619840A CN 107437251 A CN107437251 A CN 107437251A
- Authority
- CN
- China
- Prior art keywords
- image
- otsu
- skull
- unit
- canny
- 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
- 210000003625 skull Anatomy 0.000 title claims abstract description 36
- 238000001914 filtration Methods 0.000 claims abstract description 20
- 238000003708 edge detection Methods 0.000 claims abstract description 15
- 238000009792 diffusion process Methods 0.000 claims abstract description 13
- 238000003709 image segmentation Methods 0.000 claims abstract description 10
- 238000010586 diagram Methods 0.000 claims abstract description 9
- 230000011218 segmentation Effects 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 abstract description 5
- 210000001519 tissue Anatomy 0.000 description 6
- 210000005013 brain tissue Anatomy 0.000 description 5
- 238000000034 method Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000002490 cerebral effect Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000686 essence Substances 0.000 description 1
- 230000005283 ground state Effects 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000005204 segregation Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Processing (AREA)
Abstract
The present invention discloses a kind of head mri image skull strip module, including anisotropic diffusion filtering unit, Otsu binarization units, Canny edge detection units and image segmentation unit, the anisotropic diffusion filtering unit is used to receive head mri image and carry out filtering smoothed image, and retains image border in smoothed image；The Otsu binarization units are calculated by Otsu and obtain binary picture to smoothed image progress binarization segmentation；The Canny edge detection units sketch the contours all contour generating profile diagrams in smoothed image by canny operator operations；The pixel of each bar contour line accounts for the ratio in region in binary picture in described image cutting unit calculating profile diagram, and the corresponding region in artwork whether is removed according to ratio in judgement, skull is peeled off from the MRI image of head, and remove small noise.In this way, improving the degree of accuracy of image procossing, the amount of calculation of image procossing is reduced.
Description
Technical field
The present invention relates to field of medical technology, particularly relates to a kind of head mri image skull strip module.
Background technology
MRI image-forming principle is that the relaxation time difference of ground state is returned again to after the hydrogen nuclei in different tissues is stimulated,
Inverse Projection is carried out according to the electromagnetic pulse detected to rebuild so as to obtain the image inside tissue.Due to skeletal tissue with
Its hetero-organization of inside of human body differs greatly, calculate gained CT values it is also larger with other histological differences, show on image just
It is that gray value and other histological differences are larger.
In the processing of various images (such as the skill such as Subcortex structural division, brain occupying lesion region recognition
Art), the identification that the presence of skull may be on cerebral tissue produce it is certain influence, and the presence of skull can increase various images
The operand of processing.Therefore, skull in MRI image is peeled off to can aid in and improves the various image recognition degrees of accuracy, reduce meter
Calculation amount.
The content of the invention
For problem present in background technology, it is an object of the invention to provide a kind of head mri image skull to peel off mould
Block, skull is peeled off from image during image preprocessing, improve the degree of accuracy of image procossing, reduce the meter of image procossing
Calculation amount.
The technical proposal of the invention is realized in this way：A kind of head mri image skull strip module, including anisotropy
Diffusing filter unit, Otsu binarization units, Canny edge detection units and image segmentation unit, the anisotropy expand
Dissipate filter unit to be connected with Otsu binarization units, the Otsu binarization units are connected with Canny edge detection units, described
Canny edge detection units are connected with image segmentation unit, wherein, the anisotropic diffusion filtering unit is used to receive head
MRI image simultaneously carries out filtering smoothed image, and retains image border in smoothed image；The Otsu binarization units pass through
Otsu calculates obtains binary picture to smoothed image progress binarization segmentation；The Canny edge detection units pass through canny
Operator operation sketches the contours all contour generating profile diagrams in smoothed image；Described image cutting unit calculates each bar wheel in profile diagram
The pixel of profile accounts for the ratio in region in binary picture, and the corresponding region in artwork whether is removed according to ratio in judgement, will
Skull is peeled off from the MRI image of head, and removes small noise.
In the above-mentioned technical solutions, the segmentation step of described image cutting unit is：
A. the pixel sum j occupied by by a contour line in the image after the processing of canny operators is calculated；
B. the contour line in a steps is mapped back into binary picture, finds the corresponding region in binary picture again, it is right to calculate this
Answer the pixel sum k shared by the gross area in region；
C. j and k ratio is calculated, if j/k value exceedes given threshold l, the corresponding region in binary picture is existed
Corresponding region in artwork removes, and otherwise remains.
Head mri image skull strip module of the present invention, including anisotropic diffusion filtering unit, Otsu binaryzation lists
Member, Canny edge detection units and image segmentation unit, anisotropic diffusion filtering unit are used to receive head mri image
And filtering smoothed image is carried out, and retain image border in smoothed image；Otsu binarization units are calculated to flat by Otsu
Sliding image carries out binarization segmentation and obtains binary picture；Canny edge detection units are sketched the contours smoothly by canny operator operations
All contour generating profile diagrams in image；The pixel that image segmentation unit calculates each bar contour line in profile diagram accounts for binaryzation
Whether the ratio in region in figure, the corresponding region in artwork is removed according to ratio in judgement, and skull is shelled from the MRI image of head
From, and remove small noise.
Brief description of the drawings
Fig. 1 is each unit flow chart in head mri image skull strip module of the present invention；
Fig. 2 is the operational flowchart of Canny operators in the present invention；
Fig. 3 is the design sketch that head mri image of the present invention carries out skull stripping.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.Based on this
Embodiment in invention, the every other reality that those of ordinary skill in the art are obtained under the premise of creative work is not made
Example is applied, belongs to the scope of protection of the invention.
A kind of head mri image skull strip module of the present invention, key point are that Canny operators are calculated with Otsu
The utilization of method, Canny operators are a kind of edge detection operator, are connected for rim detection and edge；And Otsu algorithms are used for
Image binaryzation.The ratio that the boundary pixel points that Canny operator operations are drawn account for former area pixel points is calculated, ratio is less than
Predetermined threshold value then removes the region, so as to have the function that skull and uncorrelated noise removing.
As shown in figure 1, the head mri image skull strip module includes anisotropic diffusion filtering unit, Otsu bis-
Value unit, Canny edge detection units and image segmentation unit, it is the detailed description to above-mentioned each unit below.
(1) anisotropic diffusion filtering unit：
The main purpose of anisotropic diffusion filtering unit is smoothed image, and this kind of filtering method overcomes gaussian filtering
It is fuzzy, remain image border in smoothed image.
Need to set three major parameters before anisotropic diffusion filtering is carried out：Iterations t, thermal conductivity factor are related
K values, λ value (span is [0,1/4]).
Used iterative equation is：
Divergence formula is that local derviation is sought on four direction, and four direction is respectively all around：
And cNX, y、cSX, y、cEX, y、cWX, yRepresent the thermal conductivity factor on four direction.Formula is as follows：
By iteration repeatedly, it becomes possible to filtered image is obtained, it is more smooth than original image.
(2) Otsu binarization units：
Otsu binarization units calculate for Otsu, are a kind of adaptive threshold image binaryzation dividing methods, also referred to as
Maximum variance between clusters, after carrying out binaryzation separation, the inter-class variance of prospect and background is maximum.Its computational methods is as follows：
The segmentation threshold of prospect and background is counted as T, belonging to the pixel number of prospect, to account for the ratio of entire image be ω0,
Average gray is μ0；And the ratio that the pixel number for belonging to background accounts for entire image is ω1, average gray μ1；The total of image puts down
Equal gray scale is counted as μ, and inter-class variance is designated as g.
The size of image is M × N, and pixel count of the gray value less than threshold value T is N0, the pixel count more than T is N1.Use above
Formula is expressed as：
ω0=N0/M×N；
ω1=N1/M×N；
G=ω0ω1(μ0-μ1)2；
T brings from 0 to maximum to the above is various into, t values are designated as adaptive threshold value T when g obtains maximum.By image
Middle gray scale less than threshold value T pixel put it is black, the pixel more than threshold value T be set to white, so i.e. obtain the image of binaryzation.
(3) Canny operators unit：The algorithm of Canny operators is as follows：
A. the convolution kernel example of Gaussian filter is as follows：
Convolution is carried out using each regional area in Gaussian filter convolution collecting image, is obtained smooth after gaussian filtering
Image, so as to eliminate noise.
B. the amplitude of gradient and direction are calculated with the finite difference of single order local derviation：
Used first differential convolution kernel has two, is respectively：
With
Convolution algorithm is carried out with the two convolution collecting images, while obtains the size and deflection of gradient magnitude, formula
It is as follows：
θ [x, y]=arctan (Gx(x,y)+Gy(x,y))
C. non-maxima suppression is carried out to gradient magnitude：The gradient for only obtaining the overall situation e insufficient to determine edge, therefore
To determine edge, for example retain the maximum point of partial gradient, and suppress non-maximum.Four fans by gradient angle straggling for circumference
One of area, to carry out suppression computing with 3 × 3 window.The numbering of four sectors is 0-3.
To a point, the center pixel M [x, y] of neighborhood is compared with two pixels along gradient line, if the pixel
Need not be big along two pixel gradients of gradient line, then it is 0 to make the center pixel.
D. edge is detected and connected with dual threashold value-based algorithm：Enter the statistics of column hisgram to image, threshold value is judged, need
Two threshold value th are set1And th2, both sides relation th1=0.4th2.Grad is less than th1The gray value of pixel be set to
0, obtain image 1；Grad is less than th2The gray value of pixel be set to 0, obtain image 2.
D.1 image 2 is scanned, when running into pixel p (x, y) of non-zero gray scale, tracks the wheel since the point
Profile, until the terminal q (x, y) of contour line.
D.2 eight adjacent domains around point s (x, y) corresponding to being put in image under consideration 1 with q (x, y) in image 2, if
Non-zero pixels be present in this eight regions, then include it in image 2, as r (x, y).Since r (x, y), repeat
D.1 step, untill it can not all be carried out again in image 1 and image 2.
D.3 after completing the connection to p (x, y) contour line, return to d.1, next contour line is found, until in image 2
New contour line can not be found again.
(4) image segmentation unit：
By processing of the canny operators to image, all contour lines are all sketched the contours of in image.
A. to a contour line in the image after the processing of canny operators, the pixel occupied by this contour line is calculated
Total j.
B. the contour line in a steps is mapped back into binary picture, finds the corresponding region in binary picture again.It is right to calculate this
Answer the pixel sum k shared by the gross area in region.
C. j and k ratio is calculated, if j/k value exceedes some given threshold l, by the corresponding area of this in binary picture
Corresponding region of the domain in artwork removes, and otherwise remains.
Split by image, skull is stripped out from the MRI image of head, while some small noises are also removed.
To sum up, a head mri image file is imported into the skull strip module, by anisotropic diffusion filtering, obtained
To a filtered image file.On the one hand the image file obtains a binary picture by Otsu operator operations, another
Aspect obtains a profile diagram by Canny operator operations.The pixel for calculating each bar contour line in profile diagram accounts for binary picture
The ratio in middle region, the corresponding region in artwork whether is removed according to ratio in judgement.Design sketch such as Fig. 3 institutes after skull stripping
Show.
Compared with prior art, head mri image skull strip module of the present invention, has the advantages that：
1. when carrying out rim detection, there are a variety of Image Edge-Detection operators available, used in this module
It is canny operators, the advantage of canny operators is to have used two threshold values, detects strong edge and weak edge, and strong side respectively
Edge can be connected by edge, and the weak edge on strong edge periphery is connected, so as to improve the integrality of profile.Strong and weak edge knot
Close and cause the profile detected by canny operators on the one hand to have higher signal to noise ratio, on the other hand can also detect weaker side
Edge, therefore ensure that peel results have higher fineness using canny operators in skull peels off computing.
2. if incorrect portion's MRI image carries out skull stripping, then on the one hand can be right when image processing is carried out
Recognition result impacts, and the pixel of another aspect redundancy can add unnecessary amount of calculation to calculating process, be unfavorable for
Successive image processing.The head mri image skull strip module has been used, skull can have been separated MRI image, has only retained hypencephalon
Tissue, the degree of accuracy of successive image identification is on the one hand improved, on the other hand decreases its amount of calculation.At the same time, use
The processing method of image segmentation also eliminates local small noise pixel point simultaneously, further increases skull and peels off effect
Fruit, eliminate picture noise caused by canny operators possibility.
3. compared with based on morphology skull stripping means, morphology skull stripping means easily by a small amount of brain tissue together
Remove, and easily remain some thicker bone tissues.And segregation method used in this module is pixel ratio in judgement,
As long as company belongs to the pixel of brain tissue all without being removed, the only higher skull of pixel ratio, and do not connect and belong to
The isolated point or pole pocket of brain tissue can be just removed.Due to the edge of brain tissue and bone tissue/tissue pixels point ratio
Value difference is not larger, and can set one by priori result can substantially distinguish both threshold values, therefore ensure that the skull stripping means
Higher specificity and sensitivity, it ensure that specificity removes skull and noise, retain brain tissue.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
God any modification, equivalent substitution and improvements made etc., should be included in the scope of the protection with principle.
Claims (2)
- A kind of 1. head mri image skull strip module, it is characterised in that：Including anisotropic diffusion filtering unit, Otsu bis- Value unit, Canny edge detection units and image segmentation unit, the anisotropic diffusion filtering unit and Otsu two-values Change unit connection, the Otsu binarization units are connected with Canny edge detection units, the Canny edge detection units and Image segmentation unit connects, wherein,The anisotropic diffusion filtering unit is used to receive head mri image and carry out filtering smoothed image, and is smoothly scheming Retain image border during picture；The Otsu binarization units are calculated by Otsu and obtain two to smoothed image progress binarization segmentation Value figure；The Canny edge detection units sketch the contours all contour generating profiles in smoothed image by canny operator operations Figure；The pixel of each bar contour line accounts for the ratio in region in binary picture in described image cutting unit calculating profile diagram, according to Whether ratio in judgement removes the corresponding region in artwork, skull is peeled off from the MRI image of head, and remove small noise.
- 2. head mri image skull strip module according to claim 1, it is characterised in that：Described image cutting unit Segmentation step be：A. the pixel sum j occupied by by a contour line in the image after the processing of canny operators is calculated；B. the contour line in a steps is mapped back into binary picture, finds the corresponding region in binary picture again, calculate the correspondence area Pixel sum k shared by the gross area in domain；C. j and k ratio is calculated, if j/k value exceedes given threshold l, by the corresponding region in binary picture in artwork In corresponding region remove, otherwise remain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710619840.XA CN107437251B (en) | 2017-07-26 | 2017-07-26 | Skull stripping module for head MRI (magnetic resonance imaging) image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710619840.XA CN107437251B (en) | 2017-07-26 | 2017-07-26 | Skull stripping module for head MRI (magnetic resonance imaging) image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107437251A true CN107437251A (en) | 2017-12-05 |
| CN107437251B CN107437251B (en) | 2020-11-13 |
Family
ID=60459819
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710619840.XA Expired - Fee Related CN107437251B (en) | 2017-07-26 | 2017-07-26 | Skull stripping module for head MRI (magnetic resonance imaging) image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107437251B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111084675A (en) * | 2019-10-12 | 2020-05-01 | 西安科智骨医疗器械有限公司 | Preparation method of personalized customized craniomaxillofacial bone surgical repair and reconstruction implant |
| CN111260673A (en) * | 2020-01-17 | 2020-06-09 | 中国人民解放军国防科学技术大学 | Visceral organ parenchyma segmentation method and device suitable for edge-breaking visceral organ radiography |
| CN111292348A (en) * | 2020-01-21 | 2020-06-16 | 滨州医学院 | MRA skull stripping method based on autonomous probe and three-dimensional labeling replacement and application thereof |
| CN111612793A (en) * | 2019-02-26 | 2020-09-01 | 中国科学院沈阳自动化研究所 | Automatic skull removing method for brain magnetic resonance image |
| CN114066922A (en) * | 2021-11-19 | 2022-02-18 | 数坤(北京)网络科技股份有限公司 | Medical image segmentation method and device, terminal equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102393963A (en) * | 2011-07-05 | 2012-03-28 | 电子科技大学 | Deformable surface model-based magnetic-resonance (MR) image skull stripping method |
| US20150206300A1 (en) * | 2014-01-21 | 2015-07-23 | Agency For Science, Technology And Research | Method and apparatus for extraction and quantification of hematoma from a brain scan such as computed tomography data |
-
2017
- 2017-07-26 CN CN201710619840.XA patent/CN107437251B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102393963A (en) * | 2011-07-05 | 2012-03-28 | 电子科技大学 | Deformable surface model-based magnetic-resonance (MR) image skull stripping method |
| US20150206300A1 (en) * | 2014-01-21 | 2015-07-23 | Agency For Science, Technology And Research | Method and apparatus for extraction and quantification of hematoma from a brain scan such as computed tomography data |
Non-Patent Citations (3)
| Title |
|---|
| CHADDAD A: ""Quantitative evaluation of robust skull stripping and tumor detection applied to axial MR images"", 《BRAIN INFORMATICS》 * |
| KALAVATHI P: ""Methods on Skull Stripping of MRI Head Scan Images—a Review"", 《JOURNAL OF DIGITAL IMAGING》 * |
| 王大溪: ""基于形态学的脑部MRI图像颅骨剥离算法"", 《计算机技术与发展》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111612793A (en) * | 2019-02-26 | 2020-09-01 | 中国科学院沈阳自动化研究所 | Automatic skull removing method for brain magnetic resonance image |
| CN111612793B (en) * | 2019-02-26 | 2023-07-25 | 中国科学院沈阳自动化研究所 | Automatic skull removing method for brain magnetic resonance image |
| CN111084675A (en) * | 2019-10-12 | 2020-05-01 | 西安科智骨医疗器械有限公司 | Preparation method of personalized customized craniomaxillofacial bone surgical repair and reconstruction implant |
| CN111260673A (en) * | 2020-01-17 | 2020-06-09 | 中国人民解放军国防科学技术大学 | Visceral organ parenchyma segmentation method and device suitable for edge-breaking visceral organ radiography |
| CN111292348A (en) * | 2020-01-21 | 2020-06-16 | 滨州医学院 | MRA skull stripping method based on autonomous probe and three-dimensional labeling replacement and application thereof |
| CN111292348B (en) * | 2020-01-21 | 2023-07-28 | 滨州医学院 | MRA skull peeling method based on autonomous probe and three-dimensional labeling displacement |
| CN114066922A (en) * | 2021-11-19 | 2022-02-18 | 数坤(北京)网络科技股份有限公司 | Medical image segmentation method and device, terminal equipment and storage medium |
| CN114066922B (en) * | 2021-11-19 | 2022-06-03 | 数坤(北京)网络科技股份有限公司 | Medical image segmentation method and device, terminal equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107437251B (en) | 2020-11-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107437251A (en) | Head mri image skull strip module | |
| Jung et al. | Robust watershed segmentation using wavelets | |
| Sreedevi et al. | A novel approach for removal of pectoral muscles in digital mammogram | |
| Hao et al. | Improved self-adaptive edge detection method based on Canny | |
| CN106023204A (en) | Method and system for removing mosquito noise based on edge detection algorithm | |
| CN101499164B (en) | Image interpolation reconstruction method based on single low-resolution image | |
| Choudhary et al. | Curvelet and fast marching method‐based technique for efficient artifact detection and removal in dermoscopic images | |
| CN104732530A (en) | Image edge detection method | |
| CN110930330B (en) | Image segmentation and region growth based salt and pepper noise reduction algorithm | |
| Mittal et al. | Brain tumour extraction using otsu based threshold segmentation | |
| Hansen et al. | Watershed-based maximum-homogeneity filtering | |
| Okuboyejo et al. | Unsupervised Restoration of Hair-Occluded Lesion in Dermoscopic Images. | |
| CN112785520B (en) | CT image artifact removing method | |
| Nain et al. | Dynamic thresholding based edge detection | |
| Tajudin et al. | An improved watershed segmentation technique for microbleeds detection in MRI images | |
| Yu et al. | Applications of edge preservation ratio in image processing | |
| Brannock et al. | A synopsis of recentwork in edge detection using the DWT | |
| CN105931192A (en) | Image texture filtering method based on weighted median filtering | |
| JP2005010842A (en) | Banded pattern extraction system and line pattern extraction method | |
| Bahaghighat et al. | Fingerprint image enhancement using GWT and DMF | |
| CN112614100A (en) | Ocean plankton image segmentation method | |
| CN105608702B (en) | A kind of background suppression method and system of MRA brain image datas | |
| Shetty et al. | Melanoma decision support system for dermatologist | |
| CN110047065A (en) | Kidney medical image contour line extraction system and method | |
| Chudasama et al. | Survey on Various Edge Detection Techniques on Noisy Images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201113 |