CN107862699A - Bone edges extracting method, device, equipment and the storage medium of Bone CT image - Google Patents
Bone edges extracting method, device, equipment and the storage medium of Bone CT image Download PDFInfo
- Publication number
- CN107862699A CN107862699A CN201710863598.0A CN201710863598A CN107862699A CN 107862699 A CN107862699 A CN 107862699A CN 201710863598 A CN201710863598 A CN 201710863598A CN 107862699 A CN107862699 A CN 107862699A
- Authority
- CN
- China
- Prior art keywords
- bone
- image
- edges
- data
- fitting function
- 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
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
- G06T7/13—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
- 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
- G06T2207/30008—Bone
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
Abstract
The applicable field of computer technology of the present invention, there is provided a kind of bone edges extracting method, device, equipment and the storage medium of Bone CT image, this method include：Receive the Bone CT image bone edges extraction request of user's input, bone edges extraction request includes corresponding Bone CT image, extracted and asked according to bone edges, the gradation data of Bone CT image is fitted using default data fitting function, obtain the data matched curve of Bone CT image, according to the data matched curve of Bone CT image, calculate the gray average in bone edges region in Bone CT image, according to the gray average being calculated, the high threshold parameter and Low threshold parameter of default bone edges detection algorithm are configured, edge extracting is carried out to Bone CT image using bone edges detection algorithm, so as to improve the accuracy and speed of the extraction of Bone CT image bone edges, and then improve the efficiency of bone edges extraction.
Description
Technical field
The invention belongs to field of computer technology, more particularly to a kind of bone edges extracting method of Bone CT image, dress
Put, equipment and storage medium.
Background technology
Backbone is the important support of human body, is made up of vertebra, rumpbone, coccyx and relevant connection.Minimally-invasive treatment interverbebral disc is dashed forward
The nucleus pulposus that the key gone out is safety, thoroughly, effectively eliminates backbone leaks part.In order to ensure the accuracy and safety of operation
Property, it is necessary in the preoperative shoot disc herniation patient x-ray tomography image (CT).The edge extracting of Bone CT two dimensional image is
The first step of three-dimensional reconstruction, while be also the key of 2 d-to-3 d image registration in art.Therefore, it is accurate quickly to obtain CT figures
As marginal information has great importance in the diagnosis of area of computer aided backbone and minimal invasive operation.
Edge in image is the set that point jumpy occurs for the Local Extremum of gray scale or gray scale.Rim detection
Conventional thought be that border is asked for according to the first derivative and second dervative of image, more classical edge detection operator has：
Sobel, Prewitt, Roberts, Laplacian, Canny operator etc..Wherein, Canny operators are because it has high s/n ratio, height
Positioning precision and single edges response etc. premium properties and be used widely.The basic thought of Canny operators is high using two dimension
First directional derivative on any direction of this function is noise filter, by being filtered with image convolution, then to filter
Image after ripple finds partial gradient maximum, and image border is determined with this.Then, occur some new edge inspections again
Method of determining and calculating, such as Nonlinear Wavelet Transform method, neural network, fuzzy technology, Mathematical Morphology method.
Mainly include bone, marrow, muscle and body fluid noise in Bone CT image, because each several part composition is different, therefore
Different gray values is shown as in imaging, edge is primarily present in each composition connection.Calculated according to traditional rim detection
Method such as Canny operators, then will detect that the edge of various other tissues in addition to bone edges.And new rim detection
Operator is generally required largely to calculate, and the bone edges of this human body that Various Complex composition be present of backbone bone are carried
Taking often has certain limitation.
The content of the invention
It is an object of the invention to provide a kind of bone edges extracting method, device, equipment and the storage of Bone CT image
Medium, it is intended to solve, because existing bone edges extracting method is computationally intensive, bone edges extraction accuracy is relatively low, to cause bone
The problem of bone edge extracting efficiency is low.
On the one hand, the invention provides a kind of bone edges extracting method of Bone CT image, methods described to include following
Step：
Receive the Bone CT image bone edges extraction request of user's input, bone edges extraction request includes pair
The Bone CT image answered；
Extracted and asked according to the bone edges, the gray scale using default data fitting function to the Bone CT image
Data are fitted, and obtain the data matched curve of the Bone CT image；
According to the data matched curve of the Bone CT image, bone edges region in the Bone CT image is calculated
Gray average；
According to the gray average being calculated, high threshold parameter and low threshold to default bone edges detection algorithm
Value parameter is configured, and edge extracting is carried out to the Bone CT image using the bone edges detection algorithm.
On the other hand, the invention provides a kind of bone edges extraction element of Bone CT image, described device to include：
Request reception unit, the Bone CT image bone edges for receiving user's input extract request, the bone side
Edge extraction request includes corresponding Bone CT image；
Data fitting unit, asked for being extracted according to the bone edges, using default data fitting function to institute
The gradation data for stating Bone CT image is fitted, and obtains the data matched curve of the Bone CT image；
Average calculation unit, for the data matched curve according to the Bone CT image, calculate the Bone CT
The gray average in bone edges region in image；And
Edge acquiring unit, for the gray average being calculated according to, to default bone edges detection algorithm
High threshold parameter and Low threshold parameter be configured, the Bone CT image is carried out using the bone edges detection algorithm
Edge extracting.
On the other hand, present invention also offers a kind of Medical Devices, including memory, processor and it is stored in described deposit
In reservoir and the computer program that can run on the processor, realized such as during computer program described in the computing device
The step of bone edges extracting method of the Bone CT image.
On the other hand, present invention also offers a kind of computer-readable recording medium, the computer-readable recording medium
Computer program is stored with, realizes that the bone edges such as the Bone CT image carry when the computer program is executed by processor
The step of taking method.
The present invention receives the Bone CT image bone edges extraction request of user's input, is wrapped in bone edges extraction request
Bone CT image corresponding to including, extracts according to bone edges and asks, using default data fitting function to Bone CT image
Gradation data is fitted, and obtains the data matched curve of Bone CT image, according to the data matched curve of Bone CT image, meter
The gray average in bone edges region in Bone CT image is calculated, according to the gray average being calculated, to default bone edges
The high threshold parameter and Low threshold parameter of detection algorithm are configured, and Bone CT image is carried out using bone edges detection algorithm
Edge extracting, so as to improve the accuracy and speed of the extraction of Bone CT image bone edges, and then improve bone edges and carry
The efficiency taken.
Brief description of the drawings
Fig. 1 is the implementation process figure of the bone edges extracting method for the Bone CT image that the embodiment of the present invention one provides；
Fig. 2 is the schematic diagram of the Bone CT Image Edge detection process of the embodiment of the present invention one；
Fig. 3 is pair for having threshold skirt testing result that the edge detection results of no threshold value provide with the embodiment of the present invention one
Than figure；
Fig. 4 is the structural representation of the bone edges extraction element for the Bone CT image that the embodiment of the present invention two provides；
Fig. 5 is the structural representation of the bone edges extraction element for the Bone CT image that the embodiment of the present invention three provides；With
And
Fig. 6 is the structural representation for the Medical Devices that the embodiment of the present invention four provides.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
It is described in detail below in conjunction with specific implementation of the specific embodiment to the present invention：
Embodiment one：
Fig. 1 shows the implementation process of the bone edges extracting method for the Bone CT image that the embodiment of the present invention one provides,
For convenience of description, the part related to the embodiment of the present invention is illustrate only, details are as follows：
In step S101, the Bone CT image bone edges extraction request of user's input, bone edges extraction are received
Request includes corresponding Bone CT image.
The embodiment of the present invention is applied to Medical Devices, and the bone edges of Bone CT image in Medical Devices that are particularly suitable for use in carry
Take.In embodiments of the present invention, corresponding bone edges to be extracted can be included in the bone edges extraction request received
Bone CT image, Bone CT image can also be received individually.
In step s 102, extracted and asked according to bone edges, using default data fitting function to Bone CT image
Gradation data be fitted, obtain the data matched curve of Bone CT image.
In embodiments of the present invention, due to main bone, marrow, muscle including human body or animal body in Bone CT image
And multiple different pieces such as body fluid noise, and these different pieces possess different compositions, cause these different pieces being imaged
To show different gray values during Bone CT image, and the gray value of these different pieces shows different distribution shapes
State, so as to need to use different fitting functions to be fitted gradation data of these different pieces in Bone CT image.Cause
This, it is preferable that default data fitting function is the blended data fitting function of multiple Gaussian functions composition, to improve Bone CT
The fitting effect of the data matched curve of image.
Because the distribution of the marrow area data in bone, muscle region data and background noise data is different, because
This, it is preferable that when being fitted using default data fitting function to the gradation data of Bone CT image, use mixed number
According to fitting functionTo the bone edges area grayscale data in Bone CT image, marrow area grayscale number
It is fitted according to, muscle region gradation data and ambient noise area grayscale data.Wherein, wGlFor in blended data fitting function
Ratio shared by each function, l=1,2,3,4, and wGlMeetfGl(x) represent in the blended data fitting function
Each function, it is preferable that each function can beSo, the mixing is passed through
Data fitting function can effectively improve the accuracy of Bone CT view data fitting, and fitting effect refers to Fig. 2.Fig. 2 (a) is shown
Bone CT image, Fig. 2 (b) is shown is fitted obtained design sketch using the blended data fitting function of the embodiment of the present invention.
Preferably, it is first before being fitted using default data fitting function to the gradation data of Bone CT image
Initial estimation first is carried out to the parameter in data fitting function, then the parameter that initial estimation obtains optimized, so as to carry
The high availability of data fitting functions, and then improve the accuracy of data fitting.Wherein, these parameters are fitted including data
The weight of the parameter of each Gaussian function and each Gaussian function in data fitting function in function.
It is further preferred that when the parameter in data fitting function carries out initial estimation, entered using k means clustering algorithms
Row initial estimation, obtain the initial value of these parameters in data fitting function
So as to further increase the availability of data fitting function.
It is further preferred that when being optimized to the parameter that initial estimation obtains, first by the initial value of these parameters
As the initial value of EM algorithm (Expectation Maximization Algorithm, abbreviation EM algorithm) iteration,
Then the initial value using EM algorithms to these parametersIt is iterated
Optimization operation, the parameter after being optimizedSo as to further increase number
According to the availability of fitting function, and then improve the accuracy of data fitting.
In step s 103, according to the data matched curve of Bone CT image, bone edges area in Bone CT image is calculated
The gray average in domain.
In embodiments of the present invention, after fitting obtains the data matched curve of Bone CT image, first according to gray value
Size, curved section corresponding to bone edges regions is obtained from the data matched curve of Bone CT image, then according to acquisition
Curved section corresponding to data, calculate Bone CT image in bone edges region gray average.
In step S104, according to the gray average being calculated, to the high threshold of default bone edges detection algorithm
Parameter and Low threshold parameter are configured, and edge extracting is carried out to Bone CT image using bone edges detection algorithm.
In embodiments of the present invention, first according to the gray average being calculated, to default bone edges detection algorithm
High threshold parameter be configured, then according to the relations between Low threshold and high threshold, default bone edges are detected and calculated
The Low threshold parameter of method is configured, so that high threshold parameter and Low threshold parameter by limiting bone edge detection algorithm,
The influence at other edges outside bone edges is eliminated, and then improves the effect of bone edges extraction.Fig. 2 and Fig. 3 are refer to, its
In, Fig. 2 (a) and Fig. 3 (a) they are that Bone CT image, Fig. 2 (c) and Fig. 3 (c) have threshold skirt inspection to be provided in an embodiment of the present invention
Result schematic diagram is surveyed, Fig. 3 (b) is the edge detection results schematic diagram without threshold value, from Fig. 3 (b) and Fig. 3 (c) it can be seen that passing through
The high threshold parameter and Low threshold parameter of bone edge detection algorithm are limited, eliminates the shadow at other edges outside bone edges
Ring, improve the effect of bone edges extraction.Afterwards, edge is carried out to Bone CT image using the bone edges detection algorithm to carry
Take, so as to improve the accuracy and speed of the extraction of Bone CT image bone edges, and then improve the effect of bone edges extraction
Rate.
Preferably, bone edges detection algorithm for Canny edge detection algorithms, so as to improve the effect of bone edges extraction
Rate.
Because the edge of different piece is primarily present in the junction of each several part, it is therefore preferred that using bone edges
When detection algorithm carries out edge extracting to Bone CT image, smoothing denoising is carried out to Bone CT image using default filtering algorithm
Processing, the amplitude of the shade of gray of each pixel and direction in Bone CT image are calculated after denoising, shade of gray amplitude is entered
Row non-maxima suppression operates, the shade of gray amplitude after being operated according to non-maxima suppression, by the pixel of Bone CT image
It is divided into strong edge point, weak marginal point and non-edge point.Afterwards, from weak marginal point, acquisition is connected weak with strong edge point
Marginal point, the weak marginal point being connected with strong edge point of acquisition and strong edge point are arranged to the marginal point of Bone CT image,
Attended operation is performed to marginal point, the edge of Bone CT image is obtained, so as to further increase Bone CT image bone edges
The accuracy and speed of extraction, and then improve the efficiency of bone edges extraction.
Specifically, the shade of gray amplitude after being operated according to non-maxima suppression, by the pixel dot-dash of Bone CT image
When being divided into strong edge point, weak marginal point and non-edge point, the high threshold and Low threshold pre-set is obtained first, then to non-pole
Big value suppresses the shade of gray amplitude after operation compared with high threshold and Low threshold, and shade of gray amplitude is in into gray scale ladder
The pixel that degree amplitude is more than high threshold is divided into strong edge point, and shade of gray amplitude is in into shade of gray amplitude is less than low threshold
The pixel of value is divided into weak marginal point, and shade of gray amplitude is in into shade of gray amplitude is between high threshold and Low threshold
Pixel be divided into non-edge point.
Embodiment two：
Fig. 4 shows the structure of the bone edges extraction element for the Bone CT image that the embodiment of the present invention two provides, in order to
It is easy to illustrate, illustrate only the part related to the embodiment of the present invention, including：
Request reception unit 41, the Bone CT image bone edges for receiving user's input extract request, bone edges
Extraction request includes corresponding Bone CT image.
Data fitting unit 42, asked for being extracted according to bone edges, using default data fitting function to bone
The gradation data of CT images is fitted, and obtains the data matched curve of Bone CT image.
Average calculation unit 43, for the data matched curve according to Bone CT image, calculate bone in Bone CT image
The gray average of fringe region.
Edge acquiring unit 44, the gray average being calculated for basis, to default bone edges detection algorithm
High threshold parameter and Low threshold parameter are configured, and edge extracting is carried out to Bone CT image using bone edges detection algorithm.
In embodiments of the present invention, request reception unit 41 receives the Bone CT image bone edges extraction that user inputs and asked
Ask, bone edges extraction request includes corresponding Bone CT image, and data fitting unit 42 please according to bone edges extraction
Ask, the gradation data of Bone CT image is fitted using default data fitting function, obtains the data of Bone CT image
Matched curve, average calculation unit 43 calculate bone edges in Bone CT image according to the data matched curve of Bone CT image
The gray average in region, edge acquiring unit 44 is according to the gray average being calculated, to default bone edges detection algorithm
High threshold parameter and Low threshold parameter be configured, and using bone edges detection algorithm to Bone CT image carry out edge carry
Take, so as to improve the accuracy and speed of the extraction of Bone CT image bone edges, and then improve the effect of bone edges extraction
Rate.
In embodiments of the present invention, each unit of the bone edges extraction element of Bone CT image can by corresponding hardware or
Software unit realizes that each unit can be independent soft and hardware unit, can also be integrated into a soft and hardware unit, herein not
To limit the present invention.The embodiment of each unit refers to the description of embodiment one, will not be repeated here.
Embodiment three：
Fig. 5 shows the structure of the bone edges extraction element for the Bone CT image that the embodiment of the present invention three provides, in order to
It is easy to illustrate, illustrate only the part related to the embodiment of the present invention, including：
Request reception unit 51, the Bone CT image bone edges for receiving user's input extract request, bone edges
Extraction request includes corresponding Bone CT image.
In embodiments of the present invention, can include in the bone edges extraction request that request reception unit 51 receives corresponding
Bone edges to be extracted Bone CT image, certainly, do not include corresponding bone side to be extracted in bone edges extraction request
During the Bone CT image of edge, Bone CT image can also be received individually.
Parameter optimization unit 52, for carrying out initial estimation to the parameter in data fitting function, and initial estimation is obtained
To parameter optimize.
In embodiments of the present invention, parameter optimization unit 52 is initially estimated to the parameter in data fitting function first
Meter, then optimizes to the parameter that initial estimation obtains, so as to improve the availability of data fitting function, and then improves
The accuracy of data fitting.Wherein, these parameters include in data fitting function the parameter of each Gaussian function and each high
Weight of this function in data fitting function.
Preferably, when the parameter in data fitting function carries out initial estimation, carried out just using k means clustering algorithms
Begin to estimate, obtain the initial value of these parameters in data fitting function
So as to further increase the availability of data fitting function.
Preferably, when being optimized to the parameter that initial estimation obtains, first using the initial value of these parameters as greatest hope
The initial value of algorithm EM algorithm iterations, the then initial value using EM algorithms to these parameters
It is iterated optimization operation, the parameter after being optimizedSo as to further
The availability of data fitting function is improved, and then improves the accuracy of data fitting.
Data fitting unit 53, asked for being extracted according to bone edges, using default data fitting function to bone
The gradation data of CT images is fitted, and obtains the data matched curve of Bone CT image.
In embodiments of the present invention, due to main bone, marrow, muscle including human body or animal body in Bone CT image
And multiple different pieces such as body fluid noise, and these different pieces possess different compositions, cause these different pieces being imaged
To show different gray values during Bone CT image, and the gray value of these different pieces shows different distribution shapes
State so that data fitting unit 53 need to use different fitting functions to these different pieces Bone CT image ash
Degrees of data is fitted, it is therefore preferred that default data fitting function is the blended data fitting of multiple Gaussian functions composition
Function, to improve the fitting effect of the data matched curve of Bone CT image.
Because the distribution of the marrow area data in bone, muscle region data and background noise data is different, because
This, it is preferable that when being fitted using default data fitting function to the gradation data of Bone CT image, use mixed number
According to fitting functionTo the bone edges area grayscale data in Bone CT image, marrow area grayscale number
It is fitted according to, muscle region gradation data and ambient noise area grayscale data.Wherein, wGlFor in blended data fitting function
Ratio shared by each function, l=1,2,3,4, and wGlMeetfGl(x) represent in the blended data fitting function
Each function, it is preferable that each function can beSo, the mixing is passed through
Data fitting function can effectively improve the accuracy of Bone CT view data fitting.
Average calculation unit 54, for the data matched curve according to Bone CT image, calculate bone in Bone CT image
The gray average of fringe region.
In embodiments of the present invention, after fitting obtains the data matched curve of Bone CT image, average calculation unit 54
First according to the size of gray value, curve corresponding to bone edges region is obtained from the data matched curve of Bone CT image
Section, then data according to corresponding to the curved section of acquisition, calculate the gray average in bone edges region in Bone CT image.
Edge acquiring unit 55, the gray average being calculated for basis, to default bone edges detection algorithm
High threshold parameter and Low threshold parameter are configured, and edge extracting is carried out to Bone CT image using bone edges detection algorithm.
In embodiments of the present invention, edge acquiring unit 55 is first according to the gray average being calculated, to default bone
The high threshold parameter of bone edge detection algorithm is configured, then according to the relation between Low threshold and high threshold, to default
The Low threshold parameter of bone edges detection algorithm is configured, so as to the high threshold parameter by limiting bone edge detection algorithm
With Low threshold parameter, the influence at other edges outside bone edges is eliminated, and then improve the effect of bone edges extraction.It
Afterwards, edge extracting is carried out to Bone CT image using the bone edges detection algorithm, so as to improve Bone CT image bone side
The accuracy and speed of edge extraction, and then improve the efficiency of bone edges extraction.
Preferably, bone edges detection algorithm for Canny edge detection algorithms, so as to improve the effect of bone edges extraction
Rate.
Because the edge of different piece is primarily present in the junction of each several part, it is therefore preferred that using bone edges
When detection algorithm carries out edge extracting to Bone CT image, smoothing denoising is carried out to Bone CT image using default filtering algorithm
Processing, the amplitude of the shade of gray of each pixel and direction in Bone CT image are calculated after denoising, shade of gray amplitude is entered
Row non-maxima suppression operates, the shade of gray amplitude after being operated according to non-maxima suppression, by the pixel of Bone CT image
It is divided into strong edge point, weak marginal point and non-edge point.Afterwards, from weak marginal point, acquisition is connected weak with strong edge point
Marginal point, the weak marginal point being connected with strong edge point of acquisition and strong edge point are arranged to the marginal point of Bone CT image,
Attended operation is performed to marginal point, the edge of Bone CT image is obtained, so as to further increase Bone CT image bone edges
The accuracy and speed of extraction, and then improve the efficiency of bone edges extraction.
Specifically, the shade of gray amplitude after being operated according to non-maxima suppression, by the pixel dot-dash of Bone CT image
When being divided into strong edge point, weak marginal point and non-edge point, the high threshold and Low threshold pre-set is obtained first, then to non-pole
Big value suppresses the shade of gray amplitude after operation compared with high threshold and Low threshold, and shade of gray amplitude is in into gray scale ladder
The pixel that degree amplitude is more than high threshold is divided into strong edge point, and shade of gray amplitude is in into shade of gray amplitude is less than low threshold
The pixel of value is divided into weak marginal point, and shade of gray amplitude is in into shade of gray amplitude is between high threshold and Low threshold
Pixel be divided into non-edge point.
It is therefore preferred that the data fitting unit 53 includes：
Data are fitted subelement 531, for using blended data fitting functionTo Bone CT figure
Bone edges area grayscale data, marrow area grayscale data, muscle region gradation data and ambient noise region ash as in
Degrees of data is fitted；
Wherein, wGlFor the ratio shared by each function in blended data fitting function, l=1,2,3,4, and wGlMeetfGl(x) each function in blended data fitting function is represented；
Preferably, the edge acquiring unit 55 includes：
Image denoising unit 551, for carrying out smoothing denoising processing to Bone CT image using default filtering algorithm；
Data processing unit 552, for calculating the amplitude of the shade of gray of each pixel in Bone CT image after denoising
And direction, non-maxima suppression operation is carried out to shade of gray amplitude；
Pixel taxon 553, for the shade of gray amplitude after being operated according to non-maxima suppression, by Bone CT figure
The pixel of picture is divided into strong edge point, weak marginal point and non-edge point；
Marginal point extraction unit 554, for obtaining the weak marginal point being connected with strong edge point from weak marginal point, it will obtain
The weak marginal point being connected with strong edge point and strong edge point that take are arranged to the marginal point of Bone CT image；And
Marginal point connection unit 555, for performing attended operation to marginal point, obtain the edge of Bone CT image.
In embodiments of the present invention, each unit of the bone edges extraction element of Bone CT image can by corresponding hardware or
Software unit realizes that each unit can be independent soft and hardware unit, can also be integrated into a soft and hardware unit, herein not
To limit the present invention.
Example IV：
Fig. 6 shows the structure for the Medical Devices that the embodiment of the present invention four provides, and for convenience of description, illustrate only and this
The related part of inventive embodiments.
The Medical Devices 6 of the embodiment of the present invention include processor 60, memory 61 and are stored in memory 61 and can
The computer program 62 run on processor 60.The processor 60 realizes above-mentioned each Bone CT when performing computer program 62
Step in the bone edges extracting method embodiment of image, for example, the step S101 to S104 shown in Fig. 1.Or processor
Realize the function of each unit in above-mentioned each device embodiment during 60 execution computer program 62, for example, unit 41 shown in Fig. 4 to
44th, the function of unit 51 to 55 shown in Fig. 5.
In embodiments of the present invention, above-mentioned each Bone CT image is realized when the processor 60 performs computer program 62
During step in bone edges extracting method embodiment, the Bone CT image bone edges extraction request of user's input is received, should
Bone edges extraction request includes corresponding Bone CT image, is extracted and asked according to bone edges, intended using default data
The gradation data for closing function pair Bone CT image is fitted, and the data matched curve of Bone CT image is obtained, according to Bone CT
The data matched curve of image, the gray average in bone edges region in Bone CT image is calculated, according to the gray scale being calculated
Average, the high threshold parameter and Low threshold parameter of default bone edges detection algorithm are configured, examined using bone edges
Method of determining and calculating carries out edge extracting to Bone CT image, so as to improve the accuracy and speed of the extraction of Bone CT image bone edges
Degree, and then improve the efficiency of bone edges extraction.
The step of processor 60 is realized when performing computer program 62 in the Medical Devices 6 specifically refers to embodiment one
The description of middle method, will not be repeated here.
Embodiment five：
In embodiments of the present invention, there is provided a kind of computer-readable recording medium, the computer-readable recording medium are deposited
Computer program is contained, the computer program realizes the bone edges extraction of above-mentioned each Bone CT image when being executed by processor
Step in embodiment of the method, for example, the step S101 to S104 shown in Fig. 1.Or the computer program is executed by processor
The function of each unit in the above-mentioned each device embodiments of Shi Shixian, for example, unit 41 to 44 shown in Fig. 4, unit 51 to 55 shown in Fig. 5
Function.
In embodiments of the present invention, the Bone CT image bone edges extraction request of user's input, the bone edges are received
Extraction request includes corresponding Bone CT image, is extracted and asked according to bone edges, uses default data fitting function pair
The gradation data of Bone CT image is fitted, and the data matched curve of Bone CT image is obtained, according to the number of Bone CT image
According to matched curve, the gray average in bone edges region in Bone CT image is calculated, according to the gray average being calculated, to pre-
If bone edges detection algorithm high threshold parameter and Low threshold parameter be configured, using bone edges detection algorithm to bone
Bone CT images carry out edge extracting, so as to improve the accuracy and speed of the extraction of Bone CT image bone edges, and then improve
The efficiency of bone edges extraction.
The bone edges extracting method for the Bone CT image that the computer program is realized when being executed by processor further may be used
With reference to the description of step in preceding method embodiment, will not be repeated here.
The computer-readable recording medium of the embodiment of the present invention can include that any of computer program code can be carried
Entity or device, recording medium, for example, the memory such as ROM/RAM, disk, CD, flash memory.
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
All any modification, equivalent and improvement made within refreshing and principle etc., should be included in the scope of the protection.
Claims (10)
1. a kind of bone edges extracting method of Bone CT image, it is characterised in that methods described comprises the steps：
The Bone CT image bone edges extraction request that user inputs is received, the bone edges extraction request includes corresponding
Bone CT image；
Extracted and asked according to the bone edges, the gradation data using default data fitting function to the Bone CT image
It is fitted, obtains the data matched curve of the Bone CT image；
According to the data matched curve of the Bone CT image, the ash in bone edges region in the Bone CT image is calculated
Spend average；
According to the gray average being calculated, high threshold parameter and Low threshold ginseng to default bone edges detection algorithm
Number is configured, and edge extracting is carried out to the Bone CT image using the bone edges detection algorithm.
2. the method as described in claim 1, it is characterised in that using default data fitting function to the Bone CT image
Gradation data the step of being fitted, including：
Use blended data fitting functionTo the bone edges area grayscale in the Bone CT image
Data, marrow area grayscale data, muscle region gradation data and ambient noise area grayscale data are fitted；
Wherein, the wGlFor the ratio shared by each function in blended data fitting function, l=1,2,3,4, and the wGlMeetThe fGl(x) each function in the blended data fitting function is represented.
3. method as claimed in claim 2, it is characterised in that using default data fitting function to the Bone CT image
Gradation data the step of being fitted before, methods described also includes：
Initial estimation is carried out to the parameter in the data fitting function, and the parameter progress obtained to the initial estimation is excellent
Change, parameter and each Gaussian function of the parameter including each Gaussian function in the data fitting function are described
Weight in data fitting function.
4. the method as described in claim 1, it is characterised in that using the bone edges detection algorithm to the Bone CT figure
The step of as carrying out edge extracting, including：
Smoothing denoising processing is carried out to the Bone CT image using default filtering algorithm；
The amplitude of the shade of gray of each pixel and direction in Bone CT image are calculated after the denoising, to the shade of gray
Amplitude carries out non-maxima suppression operation；
Shade of gray amplitude after being operated according to the non-maxima suppression, the pixel of the Bone CT image is divided into by force
Marginal point, weak marginal point and non-edge point；
The weak marginal point being connected with the strong edge point is obtained from the weak marginal point, by the acquisition and the strong side
Weak marginal point and strong edge point that edge point is connected are arranged to the marginal point of the Bone CT image；
Attended operation is performed to the marginal point, obtains the edge of the Bone CT image.
5. a kind of bone edges extraction element of Bone CT image, it is characterised in that described device includes：
Request reception unit, the Bone CT image bone edges for receiving user's input extract request, and the bone edges carry
Request is taken to include corresponding Bone CT image；
Data fitting unit, asked for being extracted according to the bone edges, using default data fitting function to the bone
The gradation data of bone CT images is fitted, and obtains the data matched curve of the Bone CT image；
Average calculation unit, for the data matched curve according to the Bone CT image, calculate the Bone CT image
The gray average in middle bone edges region；And
Edge acquiring unit, for the gray average being calculated according to, to the height of default bone edges detection algorithm
Threshold parameter and Low threshold parameter are configured, and edge is carried out to the Bone CT image using the bone edges detection algorithm
Extraction.
6. device as claimed in claim 5, it is characterised in that the data fitting unit includes：
Data are fitted subelement, for using blended data fitting functionTo in the Bone CT image
Bone edges area grayscale data, marrow area grayscale data, muscle region gradation data and ambient noise area grayscale number
According to being fitted；
Wherein, the wGlFor the ratio shared by each function in blended data fitting function, the l=1,2,3,4, and the wGl
MeetThe fGl(x) each function in the blended data fitting function is represented.
7. device as claimed in claim 6, it is characterised in that described device also includes：
Parameter optimization unit, for carrying out initial estimation to the parameter in the data fitting function, and to the initial estimation
Obtained parameter optimizes, and the parameter includes in the data fitting function parameter of each Gaussian function and described each
Weight of the individual Gaussian function in the data fitting function.
8. device as claimed in claim 5, it is characterised in that the edge acquiring unit includes：
Image denoising unit, for carrying out smoothing denoising processing to the Bone CT image using default filtering algorithm；
Data processing unit, for calculating after the denoising amplitude of the shade of gray of each pixel and side in Bone CT image
To shade of gray amplitude progress non-maxima suppression operation；
Pixel taxon, for the shade of gray amplitude after being operated according to the non-maxima suppression, by the Bone CT
The pixel of image is divided into strong edge point, weak marginal point and non-edge point；
Marginal point extraction unit, will for obtaining the weak marginal point being connected with the strong edge point from the weak marginal point
The weak marginal point being connected with the strong edge point and strong edge point of the acquisition are arranged to the Bone CT image
Marginal point；And
Marginal point connection unit, for performing attended operation to the marginal point, obtain the edge of the Bone CT image.
9. a kind of Medical Devices, including memory, processor and it is stored in the memory and can be on the processor
The computer program of operation, it is characterised in that realize such as Claims 1-4 described in the computing device during computer program
The step of any one methods described.
10. a kind of computer-readable recording medium, the computer-readable recording medium storage has computer program, and its feature exists
In when the computer program is executed by processor the step of realization such as any one of Claims 1-4 methods described.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710863598.0A CN107862699B (en) | 2017-09-22 | 2017-09-22 | Bone edge extraction method, device, equipment and storage medium of bone CT image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710863598.0A CN107862699B (en) | 2017-09-22 | 2017-09-22 | Bone edge extraction method, device, equipment and storage medium of bone CT image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107862699A true CN107862699A (en) | 2018-03-30 |
| CN107862699B CN107862699B (en) | 2020-11-10 |
Family
ID=61699557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710863598.0A Active CN107862699B (en) | 2017-09-22 | 2017-09-22 | Bone edge extraction method, device, equipment and storage medium of bone CT image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107862699B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112017147A (en) * | 2019-05-31 | 2020-12-01 | 杭州三坛医疗科技有限公司 | Skeleton image extraction method and device |
| CN112017146A (en) * | 2019-05-31 | 2020-12-01 | 杭州三坛医疗科技有限公司 | Skeleton segmentation method and device |
| CN112017148A (en) * | 2019-05-31 | 2020-12-01 | 杭州三坛医疗科技有限公司 | Method and device for extracting single-joint skeleton contour |
| CN114037719A (en) * | 2022-01-11 | 2022-02-11 | 北京威高智慧科技有限公司 | Bone region determination method and device, electronic equipment and storage medium |
| CN114897703A (en) * | 2022-06-08 | 2022-08-12 | 西安交通大学 | Automatic splicing system and method for orthopedic full-length X-ray films, computer equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103714538A (en) * | 2013-12-20 | 2014-04-09 | 中联重科股份有限公司 | Road edge detection method and device and vehicle |
| US8913817B2 (en) * | 2011-10-28 | 2014-12-16 | Carestream Health, Inc. | Rib suppression in radiographic images |
| CN104574361A (en) * | 2014-11-27 | 2015-04-29 | 沈阳东软医疗系统有限公司 | Image processing method and device for mammary peripheral tissue equalization |
| CN104700421A (en) * | 2015-03-27 | 2015-06-10 | 中国科学院光电技术研究所 | Edge detection algorithm based on canny self-adaptive threshold value |
| US9269139B2 (en) * | 2011-10-28 | 2016-02-23 | Carestream Health, Inc. | Rib suppression in radiographic images |
| CN105551041A (en) * | 2015-12-15 | 2016-05-04 | 中国科学院深圳先进技术研究院 | Universal blood vessel segmentation method and system |
| CN106683085A (en) * | 2016-12-23 | 2017-05-17 | 浙江大学 | CT image spine and spinal dura mater automation detection method |
-
2017
- 2017-09-22 CN CN201710863598.0A patent/CN107862699B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8913817B2 (en) * | 2011-10-28 | 2014-12-16 | Carestream Health, Inc. | Rib suppression in radiographic images |
| US9269139B2 (en) * | 2011-10-28 | 2016-02-23 | Carestream Health, Inc. | Rib suppression in radiographic images |
| CN103714538A (en) * | 2013-12-20 | 2014-04-09 | 中联重科股份有限公司 | Road edge detection method and device and vehicle |
| CN104574361A (en) * | 2014-11-27 | 2015-04-29 | 沈阳东软医疗系统有限公司 | Image processing method and device for mammary peripheral tissue equalization |
| CN104700421A (en) * | 2015-03-27 | 2015-06-10 | 中国科学院光电技术研究所 | Edge detection algorithm based on canny self-adaptive threshold value |
| CN105551041A (en) * | 2015-12-15 | 2016-05-04 | 中国科学院深圳先进技术研究院 | Universal blood vessel segmentation method and system |
| CN106683085A (en) * | 2016-12-23 | 2017-05-17 | 浙江大学 | CT image spine and spinal dura mater automation detection method |
Non-Patent Citations (4)
| Title |
|---|
| KUI FANG 等: "3D MODEL-BASED METHOD FOR VESSEL SEGMENTATION IN TOF-MRA", 《PROCEEDINGS OF THE 2011 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS》 * |
| PEI LU 等: "a vessel segmentation method for MRA ata based on probabilistic mixture model", 《《2015 IET INTERNATIONAL CONFERENCE ON BIOMEDICAL IMAGE AND SIGNAL PROCESSING（ICBISP 2015）》》 * |
| SEMA CANDEMIR 等: "Atlas-based rib-bone detection in chest X-rays", 《COMPUTERIZED MEDICAL IMAGING AND GRAPHICS》 * |
| 朱志松 等: "基于数学形态变换的骨骼CT图像边缘提取", 《微计算机信息》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112017147A (en) * | 2019-05-31 | 2020-12-01 | 杭州三坛医疗科技有限公司 | Skeleton image extraction method and device |
| CN112017146A (en) * | 2019-05-31 | 2020-12-01 | 杭州三坛医疗科技有限公司 | Skeleton segmentation method and device |
| CN112017148A (en) * | 2019-05-31 | 2020-12-01 | 杭州三坛医疗科技有限公司 | Method and device for extracting single-joint skeleton contour |
| CN112017147B (en) * | 2019-05-31 | 2024-03-22 | 杭州三坛医疗科技有限公司 | Bone image extraction method and device |
| CN112017146B (en) * | 2019-05-31 | 2024-03-22 | 杭州三坛医疗科技有限公司 | Bone segmentation method and device |
| CN112017148B (en) * | 2019-05-31 | 2024-03-22 | 杭州三坛医疗科技有限公司 | Method and device for extracting single-segment skeleton contour |
| CN114037719A (en) * | 2022-01-11 | 2022-02-11 | 北京威高智慧科技有限公司 | Bone region determination method and device, electronic equipment and storage medium |
| CN114897703A (en) * | 2022-06-08 | 2022-08-12 | 西安交通大学 | Automatic splicing system and method for orthopedic full-length X-ray films, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107862699B (en) | 2020-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Duarte-Salazar et al. | Speckle noise reduction in ultrasound images for improving the metrological evaluation of biomedical applications: an overview | |
| CN107862699A (en) | Bone edges extracting method, device, equipment and the storage medium of Bone CT image | |
| Tan et al. | Segmentation of lung lesions on CT scans using watershed, active contours, and Markov random field | |
| Kaur et al. | A comprehensive review of denoising techniques for abdominal CT images | |
| JP2019516460A (en) | System and method for noise control in multi-energy CT images based on spatial and spectral information | |
| EP2869261B1 (en) | Method for processing image data representing a three-dimensional volume | |
| KR101312459B1 (en) | Method for denoising of medical image | |
| Zhang et al. | Low-dose lung CT image restoration using adaptive prior features from full-dose training database | |
| Deng et al. | Speckle reduction of ultrasound images based on Rayleigh-trimmed anisotropic diffusion filter | |
| Prasath | Quantum noise removal in X-ray images with adaptive total variation regularization | |
| CN108885786B (en) | Medical image processing | |
| CN115456890A (en) | Method for generating anti-medical CT image denoising based on multi-scale dual-domain discriminator | |
| Raj et al. | Denoising of magnetic resonance and x-ray images using variance stabilization and patch based algorithms | |
| Belgherbi et al. | Morphological segmentation of the kidneys from abdominal ct images | |
| Raj et al. | Denoising of medical images using total variational method | |
| Hsieh et al. | A limit on dose reduction possible with CT reconstruction algorithms without prior knowledge of the scan subject | |
| Menon et al. | Enhancement of dental digital x-ray images based on the image quality | |
| EP3658031B1 (en) | Motion compensated cardiac valve reconstruction | |
| WO2017102467A1 (en) | Method of data processing for computed tomography | |
| Myint et al. | Effective kidney segmentation using gradient based approach in abdominal CT images | |
| Shafer et al. | Validation of a 3D hidden-Markov model for breast tissue segmentation and density estimation from MR and tomosynthesis images | |
| JP5632920B2 (en) | System and method for determining blur characteristics in a blurred image | |
| Kumar et al. | Semiautomatic method for segmenting pedicles in vertebral radiographs | |
| Shahina et al. | Renal stone detection and analysis by contour based algorithm | |
| Devi et al. | CT image reconstruction using NLMfuzzyCD regularization method |
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 |