CN109191456A - Lung CT image processing method and system based on two-dimentional S-transformation - Google Patents
Lung CT image processing method and system based on two-dimentional S-transformation Download PDFInfo
- Publication number
- CN109191456A CN109191456A CN201811098822.2A CN201811098822A CN109191456A CN 109191456 A CN109191456 A CN 109191456A CN 201811098822 A CN201811098822 A CN 201811098822A CN 109191456 A CN109191456 A CN 109191456A
- Authority
- CN
- China
- Prior art keywords
- vertical direction
- horizontal direction
- lung
- image
- transformation
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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
- 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]
-
- 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
- G06T2207/30061—Lung
Abstract
The invention discloses a kind of lung CT image processing method and system based on two-dimentional S-transformation, the lung CT image processing method, the following steps are included: carrying out the two-dimentional S-transformation on horizontal direction and vertical direction to the original lung CT image of input, horizontal direction transform domain and vertical direction transform domain are obtained;The extraction that energy properties are carried out to the horizontal direction transform domain and vertical direction transform domain, obtains horizontal direction energy properties figure and vertical direction energy properties figure;The horizontal direction energy properties figure and vertical direction energy properties figure are subjected to Threshold segmentation, obtain the lung images with frequency domain character.The present invention is based on two-dimentional S-transformations, and the extraction of energy properties is carried out to obtained transform domain, Threshold segmentation is then carried out, finally obtains processing image.By calculating energy properties, a kind of feature of new type, i.e. frequency domain character are provided for the processing of lung CT image, to obtain the lung images with frequency domain character.
Description
Technical field
The invention belongs to lung images process field, in particular to a kind of lung CT image processing based on two-dimentional S-transformation
Method and system.
Background technique
The processing of lung CT image, the main feature high using extraction discrimination is come normal and different in area's lung images
Often, the discrimination of extracted feature directly influences processing result.
Currently, common feature can be roughly divided into three classes: gray feature, morphological feature, textural characteristics.Gray feature is normal
There are mean value, standard deviation and statistical moment;There are commonly eccentricities, circularity, the degree of packing, maximum radius, two-dimensional for shape feature
Area, three-dimensional volume etc., there are also some shape descriptors based on Hessian matrix, description based on HOG feature;Line
Reason Feature Descriptor has very much, usually using description based on GLCM, description of LBP, description such as comentropy.But these
Spatial information and statistical information is only utilized in feature, and there is no utilize frequency information.
Summary of the invention
It is an object of the invention to: it is extracted for the processing of existing lung CT image in the method for feature and space is only utilized
Information and statistical information provide a kind of lung CT image processing based on two-dimentional S-transformation there is no frequency information is utilized
Method and system provides a kind of feature of new type, i.e. frequency using the energy feature in S-transformation domain for lung CT image processing
Characteristic of field, to obtain the lung images with frequency domain character.
The technical solution adopted by the invention is as follows:
A kind of lung CT image processing method based on two-dimentional S-transformation, comprising the following steps:
Step 1: the two-dimentional S-transformation on horizontal direction and vertical direction being carried out to the original lung CT image of input, is obtained
Horizontal direction transform domain and vertical direction transform domain;
Step 2: carrying out the extraction of energy properties to the horizontal direction transform domain and vertical direction transform domain, obtain level
Oriented energy attributed graph and vertical direction energy properties figure;
Step 3: the horizontal direction energy properties figure and vertical direction energy properties figure being subjected to Threshold segmentation, had
There are the lung images of frequency domain character.
Further, step 1 specifically comprises the following steps:
Step 1.1: two-dimensional Fourier transform being carried out to the original lung CT image of input, obtains its spectrogram;
Step 1.2: the value of initialization horizontal direction frequency variable and vertical direction frequency variable;
Step 1.3: the frequency according to the value of the horizontal direction frequency variable and vertical direction frequency variable, after calculating translation
Spectrogram;
Step 1.4: according to the value of the horizontal direction frequency variable and vertical direction frequency variable, calculating dimensional Gaussian window
Spectrum of function;
Step 1.5: the product of spectrogram and dimensional Gaussian window function frequency spectrum after calculating the translation;
Step 1.6: two-dimentional inversefouriertransform being carried out to the result of step 1.5, respectively obtains horizontal direction and vertical side
To S-transformation result;
Step 1.7: after executing step 1.6, whether the value of determined level direction frequency variable and vertical direction frequency variable
All values have been looped through, if not having, the horizontal direction frequency variable and vertical direction frequency variable of all values will not traversed
From after increasing, step 1.3 is continued to execute to step 1.6, otherwise jumps out circulation.
Further, in step 1.1, the formula of two-dimensional Fourier transform is as follows:
Wherein, F (α, β) indicates that obtained spectrogram, f (x, y) indicate that original lung CT image, α indicate that horizontal direction is empty
Between frequency, β indicate vertical direction spatial frequency.
Further, step 1.2 specifically:
When calculating horizontal direction transform domain, fixed vertical direction frequency variable ky=0, initialize horizontal direction frequency variable
kx=1;
When calculating vertical direction transform domain, fixed horizontal direction frequency variable kx=0, initialize vertical direction frequency variable
ky=1.
Further, in step 1.4, the formula for calculating dimensional Gaussian window function frequency spectrum is as follows:
Wherein, W (kx, ky) indicate calculated dimensional Gaussian window function frequency spectrum, kxAnd kyIndicate the frequency domain number of S-transformation
Value, α and β indicate the frequency domain frequency variable of dimensional Gaussian window function frequency spectrum.
Further, in step 1.6, the calculation formula for carrying out inversefouriertransform to the result of step 1.5 is as follows:
Wherein, S (x, y, kx, ky) it is S-transformation as a result, M (α, β) is the resulting result of product of step 1.5, α and β indicate anti-
The frequency domain frequency variable of Fourier transformation, x indicate the spatial position of horizontal direction, and y indicates the spatial position of vertical direction.
Further, in step 2, the extraction of energy properties is carried out to horizontal direction transform domain and vertical direction transform domain,
Obtain the formula of energy properties figure are as follows:
Wherein, S (x, y, kx, ky) indicate original lung CT image f (x, y) frequency variable k in the horizontal directionxWith vertical side
To frequency variable kyRespectively certain value when S-transformation as a result, Re expression takes real part, Im expression takes imaginary part, and m and n respectively indicate defeated
The height and width of the original lung CT image f (x, y) entered.
Further, step 3 specifically:
Step 3.1: the horizontal direction energy properties figure and vertical direction energy properties figure acquire to step 2 carries out threshold value point
It cuts, respectively obtains horizontal direction binary image and vertical direction binary image;
Step 3.2: horizontal direction direction binary image and vertical direction binary image being carried out and operated, is obtained
To the lung images with frequency domain character.
Further, in step 3.1, the calculation formula of the threshold value are as follows: T=m+ ε σ;
Wherein, T indicate threshold value, m indicate energy properties figure mean value, σ indicate energy properties figure variance, ε be control because
Son, value range are 0.3~0.5.
A kind of lung CT image processing system based on two-dimentional S-transformation, comprising:
Image input units, for inputting original lung CT image;
Image processing unit obtains final processing image for handling the original lung CT image of input;
Image output unit, for exporting the final processing image;
The original lung CT image of described image processing unit processes input, the step of obtaining final processing image
Include:
Step 1: the two-dimentional S-transformation on horizontal direction and vertical direction being carried out to the original lung CT image of input, is obtained
Horizontal direction transform domain and vertical direction transform domain;
Step 2: carrying out the extraction of energy properties to the horizontal direction transform domain and vertical direction transform domain, obtain level
Oriented energy attributed graph and vertical direction energy properties figure;
Step 3: the horizontal direction energy properties figure and vertical direction energy properties figure being subjected to Threshold segmentation, had
There are the lung images of frequency domain character.
In conclusion by adopting the above-described technical solution, the beneficial effects of the present invention are:
The present invention is based on two-dimentional S-transformations, and the extraction of energy properties is carried out to obtained transform domain, then carry out threshold value point
It cuts, finally obtains processing image;By calculating energy properties, a kind of spy of new type is provided for the processing of lung CT image
Sign, i.e. frequency domain character, to obtain the lung images with frequency domain character.S-transformation compared to other Time-Frequency Analysis Methods, have compared with
Good time-frequency focusing, the interference for item of also not reporting to the leadship after accomplishing a task, frequency-domain information obtained have better discrimination.Threshold process rank
The adaptive thresholding of section has preferable robustness to clinical complicated lung images.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is that the present invention is based on the flow charts of the lung CT image processing method of two-dimentional S-transformation;
Fig. 2 is that the present invention is based on the detail flowcharts of the lung CT image processing method of two-dimentional S-transformation;
Fig. 3 is the original lung CT image that the present invention inputs;
Fig. 4 is the horizontal direction energy properties figure that the present invention obtains;
Fig. 5 is the vertical direction energy properties figure that the present invention obtains;
Fig. 6 is the horizontal direction binary image that the present invention obtains;
Fig. 7 is the vertical direction binary image that the present invention obtains;
Fig. 8 is the lung images with frequency domain character that the present invention obtains;
Fig. 9 is that the present invention is based on the structural schematic diagrams of the lung CT image processing system of two-dimentional S-transformation.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention, i.e., described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is logical
The component for the embodiment of the present invention being often described and illustrated herein in the accompanying drawings can be arranged and be designed with a variety of different configurations.
Therefore, the detailed description of the embodiment of the present invention provided in the accompanying drawings is not intended to limit below claimed
The scope of the present invention, but be merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art
Member's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
It should be noted that the terms "include", "comprise" or its any other variant are intended to the packet of nonexcludability
Contain, so that the process, method, article or equipment for including a series of elements not only includes those elements, but also including
Other elements that are not explicitly listed, or further include for elements inherent to such a process, method, article, or device.
In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including the element
Process, method, article or equipment in there is also other identical elements.
Feature and performance of the invention are described in further detail with reference to embodiments.
Embodiment 1
The present embodiment provides a kind of lung CT image processing methods based on two-dimentional S-transformation, as shown in Figure 1, including following
Step:
Step 1: the two-dimentional S-transformation on horizontal direction and vertical direction being carried out to the original lung CT image of input, is obtained
Horizontal direction transform domain and vertical direction transform domain;
Step 2: carrying out the extraction of energy properties to the horizontal direction transform domain and vertical direction transform domain, obtain level
Oriented energy attributed graph and vertical direction energy properties figure;
Step 3: the horizontal direction energy properties figure and vertical direction energy properties figure being subjected to Threshold segmentation, had
There are the lung images of frequency domain character.
Specifically, as shown in Fig. 2,
Step 1 includes the following steps:
Step 1.1: two-dimensional Fourier transform being carried out to the original lung CT image as shown in Figure 3 of input, obtains its frequency
Spectrogram F (α, β);The formula of two-dimensional Fourier transform is as follows:
Wherein, f (x, y) indicates that original lung CT image, α indicate horizontal direction spatial frequency, and β indicates vertical direction space
Frequency:
Step 1.2: initialization horizontal direction frequency variable kxWith vertical direction frequency variable kyValue;Specifically:
When calculating horizontal direction transform domain, fixed vertical direction frequency variable ky=0, initialize horizontal direction frequency variable
kx=1;
When calculating vertical direction transform domain, fixed horizontal direction frequency variable kx=0, initialize vertical direction frequency variable
ky=1;
Step 1.3: according to the horizontal direction frequency variable kxWith vertical direction frequency variable kyValue, calculate translation after
Spectrogram;Specifically:
In fixed vertical direction frequency variable kySpectrogram F (α+k when=0, after calculating horizontal direction translationx, β);
In fixed horizontal direction frequency variable kxSpectrogram F (α, β+k when=0, after calculating vertical direction translationy);
Step 1.4: according to the horizontal direction frequency variable kxWith vertical direction frequency variable kyValue, it is high to calculate two dimension
This window function frequency spectrum W (kx, ky);Calculation formula is as follows:
Wherein, W (kx, ky) indicate calculated dimensional Gaussian window function frequency spectrum, kxAnd kyIndicate the frequency domain number of S-transformation
Value, α and β indicate the frequency domain frequency variable of dimensional Gaussian window function frequency spectrum;
Specifically:
In fixed vertical direction frequency variable kyWhen=0, the Gauss function frequency spectrum of horizontal direction is calculated
In fixed horizontal direction frequency variable kxWhen=0, the Gauss function frequency spectrum of vertical direction is calculated
Step 1.5: the product M (α, β) of spectrogram and dimensional Gaussian window function frequency spectrum after calculating the translation;Specifically
Ground:
In fixed vertical direction frequency variable kySpectrogram and horizontal direction when=0, after calculating horizontal direction translation
The product Mh (α, β) of dimensional Gaussian window function frequency spectrum=F (α+kx, β) and Wh(kx, ky);
In fixed horizontal direction frequency variable kxSpectrogram and vertical direction when=0, after calculating vertical direction translation
The product M of dimensional Gaussian window function frequency spectrumv(α, β)=F (α, β+ky)·Wv(kx, ky);
Step 1.6: two-dimentional inversefouriertransform being carried out to the result of step 1.5, respectively obtains horizontal direction and vertical side
To S-transformation as a result, calculation formula is as follows:
Wherein, S (x, y, kx, ky) it is S-transformation as a result, M (α, β) is the resulting result of product of step 1.5, α and β indicate anti-
The frequency domain frequency variable of Fourier transformation, x indicate the spatial position of horizontal direction, and y indicates the spatial position of vertical direction;
Specifically,
In fixed vertical direction frequency variable kyWhen=0, to two of spectrogram and horizontal direction after horizontal direction translation
The product Mh (α, β) for tieing up Gauss function frequency spectrum carries out inversefouriertransform, obtains horizontal direction transform domain Sh(x, y, kx, ky);
In fixed horizontal direction frequency variable kxWhen=0, to two of spectrogram and vertical direction after vertical direction translation
Tie up the product M of Gauss function frequency spectrumv(α, β) carries out inversefouriertransform, obtains vertical direction transform domain Sv(x, y, kx, ky);
Step 1.7: after executing step 1.6, judging the horizontal direction frequency variable kxWith vertical direction frequency variable ky's
Whether value has looped through all values, if not having, will not traverse the horizontal direction frequency variable k of all valuesxAnd vertical direction
Frequency variable kyFrom after increasing, step 1.3 is continued to execute to step 1.6, otherwise jumps out circulation;
Specifically:
In fixed vertical direction frequency variable ky=0, initialize horizontal direction frequency variable kxWhen=1, horizontal direction is obtained
S-transformation as a result, then determined level direction frequency variable kxWhether all values, k have been looped throughxValue range 0~m it
Between, indicate the height of original lung CT image.If not having, by horizontal direction frequency variable kxFrom increasing 1, i.e. execution kxAfter+1,
Step 1.3 is continued to execute to step 1.6, otherwise jumps out circulation;
In fixed horizontal direction frequency variable kx=0, initialize vertical direction frequency variable kyWhen=1, vertical direction is obtained
S-transformation as a result, then judging vertical direction frequency variable kyWhether all values, k have been looped throughyValue range 0~n it
Between, indicate the width of original lung CT image.If no, by horizontal direction frequency variable kyFrom increasing 1, i.e. execution kyAfter+1, after
The continuous step 1.3 that executes arrives step 1.6, otherwise jumps out circulation.
Step 2 is specifically, to horizontal direction transform domain Sh(x, y, kx, ky) and vertical direction transform domain Sv(x, y, kx, ky)
The extraction for carrying out energy properties, obtains the formula of energy properties figure are as follows:
Wherein, S (x, y, kx, ky) indicate original lung CT image f (x, y) frequency variable k in the horizontal directionxWith vertical side
To frequency variable kyRespectively certain value when S-transformation as a result, Re expression takes real part, Im expression takes imaginary part, and m and n respectively indicate defeated
The height and width of the original lung CT image f (x, y) entered;
Specifically, to the horizontal direction transform domain Sh(x, y, kx, ky) carry out energy properties extraction, obtain such as Fig. 4 institute
The horizontal direction energy properties figure E shownh(x, y);
To the vertical direction transform domain Sv(x, y, kx, ky) extraction that carries out energy properties, it obtains as shown in Figure 5 perpendicular
Histogram is to energy properties figure Ev(x, y).
Step 3 includes the following steps:
Step 3.1: the horizontal direction energy properties figure E that step 2 is acquiredh(x, y) and vertical direction energy properties figure Ev
(x, y) carries out Threshold segmentation, obtains horizontal direction binary image B as shown in FIG. 6h(x, y) and vertical side as shown in Figure 7
To binary image Bv(x, y);
Further, the calculation formula of threshold value T are as follows: T=m+ ε σ;
Wherein, m indicates the mean value of energy properties figure E (x, y), and σ indicates the variance of energy properties figure E (x, y), and ε is control
The factor, value range are 0.3~0.5;
Step 3.2: to the horizontal direction direction binary image Bh(x, y) and vertical direction binary image Bv(x, y)
It carries out and operates, obtain the lung images O (x, y) with frequency domain character as shown in Figure 8, calculation formula are as follows: O (x, y)=Bh
(x, y) ∩ Bv(x, y).
Embodiment 2
The present embodiment provides a kind of lung CT image processing systems based on two-dimentional S-transformation, as shown in Figure 9, comprising:
Image input units, for inputting original lung CT image;
Image processing unit obtains final processing image for handling the original lung CT image of input;
Image output unit, for exporting the final processing image;
The original lung CT image of described image processing unit processes input, the step of obtaining final processing image
Include:
Step 1: the two-dimentional S-transformation on horizontal direction and vertical direction being carried out to the original lung CT image of input, is obtained
Horizontal direction transform domain and vertical direction transform domain;
Step 2: carrying out the extraction of energy properties to the horizontal direction transform domain and vertical direction transform domain, obtain level
Oriented energy attributed graph and vertical direction energy properties figure;
Step 3: the horizontal direction energy properties figure and vertical direction energy properties figure being subjected to Threshold segmentation, had
There are the lung images of frequency domain character.
It is apparent to those skilled in the art that the convenience and letter for description are bought, foregoing description based on
The specific work process of the lung CT image processing system of two-dimentional S-transformation and its each functional unit can refer to aforementioned implementation
The corresponding process in intelligent control method in example 1, details are not described herein.
Above-mentioned bright each functional unit can integrate in one processing unit, is also possible to the independent physics of each unit and deposits
It can also be integrated in one unit with two or more units.Above-mentioned integrated unit can both use the shape of hardware
Formula is realized, can also be realized in the form of software functional units.
If integrated each functional unit is realized in the form of SFU software functional unit and sells as independent product
Or it in use, can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention
Substantially all or part of the part that contributes to existing technology or the technical solution can be with software product in other words
Form embody, which is stored in a storage medium, including some instructions use so that one
Computer equipment (can be smart phone, tablet computer, personal computer, server or the network equipment etc.) executes this hair
The all or part of the steps of the bright lung CT image processing method based on two-dimentional S-transformation.And storage medium above-mentioned includes:
USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random
Access Memory), the various media that can store program code such as magnetic or disk.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of lung CT image processing method based on two-dimentional S-transformation, which comprises the following steps:
Step 1: the two-dimentional S-transformation on horizontal direction and vertical direction being carried out to the original lung CT image of input, obtains level
Direction transformation domain and vertical direction transform domain;
Step 2: carrying out the extraction of energy properties to the horizontal direction transform domain and vertical direction transform domain, obtain horizontal direction
Energy properties figure and vertical direction energy properties figure;
Step 3: the horizontal direction energy properties figure and vertical direction energy properties figure being subjected to Threshold segmentation, obtain that there is frequency
The lung images of characteristic of field.
2. as described in claim 1 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that step 1 is specific
Include the following steps:
Step 1.1: two-dimensional Fourier transform being carried out to the original lung CT image of input, obtains its spectrogram;
Step 1.2: the value of initialization horizontal direction frequency variable and vertical direction frequency variable;
Step 1.3: the frequency spectrum according to the value of the horizontal direction frequency variable and vertical direction frequency variable, after calculating translation
Figure;
Step 1.4: according to the value of the horizontal direction frequency variable and vertical direction frequency variable, calculating dimensional Gaussian window function
Frequency spectrum;
Step 1.5: the product of spectrogram and dimensional Gaussian window function frequency spectrum after calculating the translation;
Step 1.6: two-dimentional inversefouriertransform being carried out to the result of step 1.5, respectively obtains the S of horizontal direction and vertical direction
Transformation results;
Step 1.7: after executing step 1.6, whether the value of determined level direction frequency variable and vertical direction frequency variable is recycled
All values have been traversed, if not having, the horizontal direction frequency variable for not traversing all values and vertical direction frequency variable have been increased certainly
Afterwards, step 1.3 is continued to execute to step 1.6, otherwise jumps out circulation.
3. as claimed in claim 2 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that step 1.1
In, the formula of two-dimensional Fourier transform is as follows:
Wherein, F (α, β) indicates that obtained spectrogram, f (x, y) indicate that original lung CT image, α indicate horizontal direction space frequency
Rate, β indicate vertical direction spatial frequency.
4. as claimed in claim 2 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that step 1.2 tool
Body are as follows:
When calculating horizontal direction transform domain, fixed vertical direction frequency variable ky=0, initialize horizontal direction frequency variable kx=
1;
When calculating vertical direction transform domain, fixed horizontal direction frequency variable kx=0, initialize vertical direction frequency variable ky=
1。
5. as claimed in claim 2 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that step 1.4
In, the formula for calculating dimensional Gaussian window function frequency spectrum is as follows:
Wherein, W (kx,ky) indicate calculated dimensional Gaussian window function frequency spectrum, kxAnd kyIndicate S-transformation frequency domain numerical value, α and
The frequency domain frequency variable of β expression dimensional Gaussian window function frequency spectrum.
6. as claimed in claim 2 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that step 1.6
In, the calculation formula for carrying out inversefouriertransform to the result of step 1.5 is as follows:
Wherein, S (x, y, kx,ky) it is S-transformation as a result, M (α, β) is the resulting result of product of step 1.5, α and β are indicated in anti-Fu
The frequency domain frequency variable of leaf transformation, x indicate the spatial position of horizontal direction, and y indicates the spatial position of vertical direction.
7. as described in claim 1 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that in step 2,
The extraction that energy properties are carried out to horizontal direction transform domain and vertical direction transform domain, obtains the formula of energy properties figure are as follows:
Wherein, S (x, y, kx,ky) indicate original lung CT image f (x, y) frequency variable k in the horizontal directionxWith vertical direction frequency
Rate variable kyRespectively certain value when S-transformation as a result, Re expression takes real part, Im expression takes imaginary part, and m and n respectively indicate input
The height and width of original lung CT image f (x, y).
8. as described in claim 1 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that step 3 is specific
Are as follows:
Step 3.1: the horizontal direction energy properties figure and vertical direction energy properties figure acquire to step 2 carries out Threshold segmentation,
Respectively obtain horizontal direction binary image and vertical direction binary image;
Step 3.2: horizontal direction direction binary image and vertical direction binary image being carried out and operated, is had
There are the lung images of frequency domain character.
9. as claimed in claim 8 based on the lung CT image processing method of two-dimentional S-transformation, which is characterized in that step 3.1
In, the calculation formula of the threshold value are as follows: T=m+ ε σ;
Wherein, T indicates threshold value, and m indicates the mean value of energy properties figure, and σ indicates the variance of energy properties figure, and ε is controlling elements, is taken
Being worth range is 0.3~0.5.
10. a kind of lung CT image processing system based on two-dimentional S-transformation characterized by comprising
Image input units, for inputting original lung CT image;
Image processing unit obtains final processing image for handling the original lung CT image of input;
Image output unit, for exporting the final processing image;
The original lung CT image of described image processing unit processes input, the step of obtaining final processing image packet
It includes:
Step 1: the two-dimentional S-transformation on horizontal direction and vertical direction being carried out to the original lung CT image of input, obtains level
Direction transformation domain and vertical direction transform domain;
Step 2: carrying out the extraction of energy properties to the horizontal direction transform domain and vertical direction transform domain, obtain horizontal direction
Energy properties figure and vertical direction energy properties figure;
Step 3: the horizontal direction energy properties figure and vertical direction energy properties figure being subjected to Threshold segmentation, obtain that there is frequency
The lung images of characteristic of field.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811098822.2A CN109191456A (en) | 2018-09-19 | 2018-09-19 | Lung CT image processing method and system based on two-dimentional S-transformation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811098822.2A CN109191456A (en) | 2018-09-19 | 2018-09-19 | Lung CT image processing method and system based on two-dimentional S-transformation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109191456A true CN109191456A (en) | 2019-01-11 |
Family
ID=64908696
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811098822.2A Pending CN109191456A (en) | 2018-09-19 | 2018-09-19 | Lung CT image processing method and system based on two-dimentional S-transformation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109191456A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109829902A (en) * | 2019-01-23 | 2019-05-31 | 电子科技大学 | A kind of lung CT image tubercle screening technique based on generalized S-transform and Teager attribute |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009067680A1 (en) * | 2007-11-23 | 2009-05-28 | Mercury Computer Systems, Inc. | Automatic image segmentation methods and apparartus |
CN101493934A (en) * | 2008-11-27 | 2009-07-29 | 电子科技大学 | Weak target detecting method based on generalized S-transform |
CN102156984A (en) * | 2011-04-06 | 2011-08-17 | 南京大学 | Method for determining optimal mark image by adaptive threshold segmentation |
CN105424709A (en) * | 2015-11-20 | 2016-03-23 | 陕西科技大学 | Fruit surface defect detection method based on image marking |
CN107330885A (en) * | 2017-07-07 | 2017-11-07 | 广西大学 | A kind of multi-operator image reorientation method of holding important content region the ratio of width to height |
-
2018
- 2018-09-19 CN CN201811098822.2A patent/CN109191456A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009067680A1 (en) * | 2007-11-23 | 2009-05-28 | Mercury Computer Systems, Inc. | Automatic image segmentation methods and apparartus |
CN101493934A (en) * | 2008-11-27 | 2009-07-29 | 电子科技大学 | Weak target detecting method based on generalized S-transform |
CN102156984A (en) * | 2011-04-06 | 2011-08-17 | 南京大学 | Method for determining optimal mark image by adaptive threshold segmentation |
CN105424709A (en) * | 2015-11-20 | 2016-03-23 | 陕西科技大学 | Fruit surface defect detection method based on image marking |
CN107330885A (en) * | 2017-07-07 | 2017-11-07 | 广西大学 | A kind of multi-operator image reorientation method of holding important content region the ratio of width to height |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109829902A (en) * | 2019-01-23 | 2019-05-31 | 电子科技大学 | A kind of lung CT image tubercle screening technique based on generalized S-transform and Teager attribute |
CN109829902B (en) * | 2019-01-23 | 2022-04-12 | 电子科技大学 | Lung CT image nodule screening method based on generalized S transformation and Teager attribute |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zafari et al. | Segmentation of overlapping elliptical objects in silhouette images | |
Singh et al. | Svm-bdt pnn and fourier moment technique for classification of leaf shape | |
Tomita et al. | Computer analysis of visual textures | |
Ramalho et al. | Rotation-invariant feature extraction using a structural co-occurrence matrix | |
CN104834922B (en) | Gesture identification method based on hybrid neural networks | |
Awrangjeb et al. | An improved curvature scale-space corner detector and a robust corner matching approach for transformed image identification | |
Lahdenoja et al. | Towards understanding the formation of uniform local binary patterns | |
Kumar et al. | Colour image segmentation with histogram and homogeneity histogram difference using evolutionary algorithms | |
Guan et al. | Blood cell image segmentation based on the Hough transform and fuzzy curve tracing | |
Chaibou et al. | Adaptive strategy for superpixel-based region-growing image segmentation | |
Narwade et al. | Offline handwritten signature verification using cylindrical shape context | |
Azad et al. | Real-time human face detection in noisy images based on skin color fusion model and eye detection | |
CN109191456A (en) | Lung CT image processing method and system based on two-dimentional S-transformation | |
Carreira-Perpiñán | Clustering methods based on kernel density estimators: mean-shift algorithms | |
Desai et al. | Performance evaluation of image retrieval systems using shape feature based on wavelet transform | |
Lu et al. | Particle filter vehicle tracking based on surf feature matching | |
Hammouda et al. | Pattern recognition based on compound complex shape-invariant Radon transform | |
Desai et al. | Image retrieval using wavelet based shape features | |
Florindo et al. | Fractal descriptors of texture images based on the triangular prism dimension | |
El-Sayed et al. | Novel approach of edges detection for digital images based on hybrid types of entropy | |
Angadi et al. | Iris recognition: a symbolic data modeling approach using Savitzky-Golay filter energy features | |
Akhtar et al. | Big data mining based on computational intelligence and fuzzy clustering | |
Bustos et al. | 3D shape matching for retrieval and recognition | |
Gheorghe et al. | Key learning features as means for terrain classification | |
Rahimzadeh Arashloo | Sparse binarised statistical dynamic features for spatio-temporal texture analysis |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190111 |
|
RJ01 | Rejection of invention patent application after publication |