CN112017148B - Method and device for extracting single-segment skeleton contour - Google Patents
Method and device for extracting single-segment skeleton contour Download PDFInfo
- Publication number
- CN112017148B CN112017148B CN201910468788.1A CN201910468788A CN112017148B CN 112017148 B CN112017148 B CN 112017148B CN 201910468788 A CN201910468788 A CN 201910468788A CN 112017148 B CN112017148 B CN 112017148B
- Authority
- CN
- China
- Prior art keywords
- bone
- contour
- internode
- segment
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 210000000988 bone and bone Anatomy 0.000 claims abstract description 191
- 238000000605 extraction Methods 0.000 claims abstract description 22
- 230000002093 peripheral effect Effects 0.000 claims description 21
- 230000008859 change Effects 0.000 claims description 17
- 238000007621 cluster analysis Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—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
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- 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
- G06T2207/10116—X-ray image
- G06T2207/10124—Digitally reconstructed radiograph [DRR]
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
The application provides a method and a device for extracting a single-section skeleton contour; the extraction method may include: acquiring a foreground region containing the internode contour of any single bone in the target bone; projecting the foreground region along a preset projection direction to obtain pixel point distribution conditions of the foreground region in the preset projection direction, wherein the preset projection direction is used for representing the extension direction of the internode outline; determining a linear equation corresponding to the internode contour according to the pixel point distribution condition; and determining the edge profile of any single-section bone, and extracting the bone profile of any single-section bone according to a linear equation of the edge profile and the internode profile.
Description
Technical Field
The present disclosure relates to the field of image processing technologies, and in particular, to a method and an apparatus for extracting a single bone contour.
Background
When aligning an X-ray image and a CT image for the same bone, it is necessary to calculate the similarity of the X-ray image to a DRR (Digitally Reconstructured Radiograph, digitally reconstructed radiological image) image generated from the CT image.
In the related art, a similarity calculation mode based on features is generally adopted, and compared with a calculation method based on pixels, the method has higher robustness and stability and is suitable for an initial registration link. It can be seen that stable and reliable feature extraction is an important precondition for feature-based computing.
Disclosure of Invention
In view of this, the present application provides a method and apparatus for extracting a single bone contour.
In order to achieve the above purpose, the present application provides the following technical solutions:
according to a first aspect of the present application, a method for extracting a single-segment bone contour is provided, including:
acquiring a foreground region containing the internode contour of any single bone in the target bone;
projecting the foreground region along a preset projection direction to obtain pixel point distribution conditions of the foreground region in the preset projection direction, wherein the preset projection direction is used for representing the extension direction of the internode outline;
determining a linear equation corresponding to the internode contour according to the pixel point distribution condition;
and determining the edge profile of any single-section bone, and extracting the bone profile of any single-section bone according to a linear equation of the edge profile and the internode profile.
Optionally, the acquiring a foreground region containing an internode contour of any single segment of the target bone includes:
intercepting a single-segment bone image containing any single-segment bone from the image of the target bone, wherein the single-segment bone image does not contain other single-segment bones except the any single-segment bone;
intercepting a sub-region containing the internode contours from the single-segment bone image;
and carrying out binarization processing on the single-segment bone image to obtain the foreground region.
Optionally, the preset projection direction is perpendicular to a bone extension direction of the target bone.
Optionally, the determining a straight line equation corresponding to the internode contour according to the pixel point distribution condition includes:
determining the distribution position with the most pixel distribution in the preset projection direction in the foreground region;
and determining a linear equation corresponding to the internode contour according to the distribution position and the preset projection direction.
Optionally, the determining the edge profile of the any single bone includes:
identifying the edge contour based on pixel value differences of pixel points between the edge contour and a peripheral area; the peripheral region is a region of the any individual bone that is outside the edge contour.
Optionally, the identifying the edge contour based on the pixel value difference of the pixel points between the edge contour and the peripheral area includes:
traversing the pixel value of each pixel point along the inner side direction from the two side boundaries of the single-section bone image of any single-section bone, wherein the single-section bone image does not contain other single-section bones except any single-section bone;
adding all pixel points with the change rate between the first pixel value and the pixel value of the last pixel point in the inner side direction exceeding a preset change rate threshold value into a candidate edge contour set;
performing cluster analysis on the pixel points in the candidate edge contour set to delete discrete points in the candidate edge contour set;
and fitting the rest pixel points in the candidate edge contour set, and taking a fitting result as the edge contour.
Optionally, the target bone comprises a spinal column.
According to a second aspect of the present application, there is provided an extraction device of a single-segment bone profile, comprising:
an acquisition unit that acquires a foreground region containing an internode contour of any one single bone of a target bone;
the projection unit is used for projecting the foreground region along a preset projection direction to obtain the pixel point distribution condition of the foreground region in the preset projection direction, wherein the preset projection direction is used for representing the extension direction of the internode contour;
the determining unit is used for determining a linear equation corresponding to the internode outline according to the pixel point distribution condition;
and the extraction unit is used for determining the edge profile of any single-section bone and extracting the bone profile of any single-section bone according to a linear equation of the edge profile and the internode profile.
Optionally, the acquiring unit is specifically configured to:
intercepting a single-segment bone image containing any single-segment bone from the image of the target bone, wherein the single-segment bone image does not contain other single-segment bones except the any single-segment bone;
intercepting a sub-region containing the internode contours from the single-segment bone image;
and carrying out binarization processing on the subareas to obtain the foreground area.
Optionally, the preset projection direction is perpendicular to a bone extension direction of the target bone.
Optionally, the determining unit is specifically configured to:
determining the distribution position with the most pixel distribution in the preset projection direction in the foreground region;
and determining a linear equation corresponding to the internode contour according to the distribution position and the preset projection direction.
Optionally, the extracting unit is specifically configured to:
identifying the edge contour based on pixel value differences of pixel points between the edge contour and a peripheral area; the peripheral region is a region of the any individual bone that is outside the edge contour.
Optionally, the extracting unit is further configured to:
traversing the pixel value of each pixel point along the inner side direction from the two side boundaries of the single-section bone image of any single-section bone, wherein the single-section bone image does not contain other single-section bones except any single-section bone;
adding all pixel points with the change rate between the first pixel value and the pixel value of the last pixel point in the inner side direction exceeding a preset change rate threshold value into a candidate edge contour set;
performing cluster analysis on the pixel points in the candidate edge contour set to delete discrete points in the candidate edge contour set;
and fitting the rest pixel points in the candidate edge contour set, and taking a fitting result as the edge contour.
Optionally, the target bone comprises a spinal column.
According to a third aspect of the present application, there is provided an electronic device comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the method for extracting a single-segment bone contour according to any of the above embodiments by executing the executable instructions.
According to a fourth aspect of embodiments of the present disclosure, there is provided a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the method of extracting a single-segment bone contour as described in any of the above embodiments.
According to the technical scheme, the foreground region comprising the internode outline is projected along the extending direction of the internode outline of the single-section skeleton, so that the pixel point distribution condition of the foreground region in the extending direction can be obtained, and a linear equation representing the internode outline can be obtained based on the distribution condition.
Further, the present application utilizes geometric features of the target bone that have gray values on the edge contour (whether on the X-ray image or the DRR image) that are significantly lower than the pixel values of the peripheral region to identify the edge contour based on the pixel value differences of the pixels between the edge contour and the peripheral region.
Finally, a closed loop curve can be formed by a functional expression of the edge contour and the internode contour, thereby extracting the bone contour of the single-section bone.
Drawings
Fig. 1 is a flowchart illustrating a method for extracting a single bone contour according to an exemplary embodiment of the present application.
Fig. 2 is a flow chart illustrating an extraction of edge contours according to an exemplary embodiment of the present application.
Fig. 3A-3E are schematic illustrations of a fitted edge profile, as shown in an exemplary embodiment of the present application.
Fig. 4 is a flow chart illustrating an extraction of internode contours according to an exemplary embodiment of the present application.
Fig. 5A-5E are schematic illustrations of a fitting internode profile, as shown in an exemplary embodiment of the present application.
Fig. 6 is a schematic illustration of an extracted spinal profile, as shown in an exemplary embodiment of the present application.
Fig. 7 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present application.
Fig. 8 is a block diagram of an extraction device for a single-segment bone contour, as shown in an exemplary embodiment of the present application.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present application as detailed in the accompanying claims.
The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first message may also be referred to as a second message, and similarly, a second message may also be referred to as a first message, without departing from the scope of the present application. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for extracting a single bone contour according to an exemplary embodiment of the present application. As shown in fig. 1, the method may include the steps of:
step 102, a foreground region is acquired that contains the internode contours of any single segment of the target bone.
In this embodiment, the target bone includes a plurality of single bones, a single bone image including any one of the single bones (excluding any other single bones except the one of the single bones) may be first cut from the image of the target bone, then a sub-region including an internode contour may be cut from the single bone image, and finally the sub-region may be subjected to binarization processing to obtain the foreground region. Through the binarization processing, the foreground part (namely the foreground area, the pixel point corresponding to black) and the background part (the pixel point corresponding to white) in the subarea can display obvious black and white effects, so that the data volume in the image is greatly reduced, and the characteristics of the outline, the texture and the like of a single bone can be highlighted, thereby being beneficial to the follow-up identification of the outline.
And 104, projecting the foreground region along a preset projection direction to obtain the pixel point distribution condition of the foreground region in the preset projection direction, wherein the preset projection direction is used for representing the extension direction of the internode contour.
In this embodiment, the extension direction of the internode contour of each individual bone is generally perpendicular to the bone extension direction based on the structure in which the target bone is composed of a plurality of individual bones spliced together. Thus, the preset projection direction may be set to be perpendicular to the bone extension direction of the target bone.
And 106, determining a linear equation corresponding to the internode contour according to the pixel point distribution condition.
In this embodiment, based on a structure in which a target bone is composed of a plurality of single bones, the internode contour of each single bone is close to a straight line; such as the human spine, the internode contour of a single spinal column approximates a straight line. Thus, the internode contours can be fitted by straight lines.
Further, since the extension direction of the contour of other regions near the inter-node contour in the foreground region is more variable, and the extension direction of the inter-node contour is more fixed (which is similar to a straight line), the straight line equation corresponding to the inter-node contour can be determined by using the distribution of pixels projected from the extension direction of the projection direction which is the inter-node contour. Specifically, a distribution position with the greatest distribution of the pixel points in the projection direction in the foreground region can be determined, and a linear equation corresponding to the internode contour can be determined according to the distribution position and the projection direction.
Step 108, determining the edge profile of any single-section bone, and extracting the bone profile of any single-section bone according to the linear equation of the edge profile and the internode profile.
In this embodiment, the edge profile and the internode profile of the single-segment bone are based on fitting, and the edge profile and the internode profile together form a closed-loop curve, so that the bone profile of the single-segment bone can be extracted. The process of identifying the edge contour may be based on the difference in pixel values of the pixels between the edge contour and the peripheral region (the peripheral region being the region of the bone outside the edge contour).
Further, from the two side boundaries of the single-section bone image of any single-section bone (the single-section bone image does not contain other single-section bones except any single-section bone), traversing the pixel value of each pixel point along the medial direction, and adding the pixel points with the change rate between the pixel value of all first pixel values and the pixel value of the last pixel point in the medial direction exceeding a preset change rate threshold value into the candidate edge contour set; and performing cluster analysis on the pixel points in the candidate edge contour set to delete the discrete points in the candidate edge contour set, and finally performing fitting on the rest pixel points in the candidate edge contour set, thereby taking the fitting result as an edge contour.
According to the technical scheme, the foreground region comprising the internode outline is projected along the extending direction of the internode outline of the single-section skeleton, so that the pixel point distribution condition of the foreground region in the extending direction can be obtained, and a linear equation representing the internode outline can be obtained based on the distribution condition.
Further, the present application utilizes geometric features of the target bone that have gray values on the edge contour (whether on the X-ray image or the DRR image) that are significantly lower than the pixel values of the peripheral region to identify the edge contour based on the pixel value differences of the pixels between the edge contour and the peripheral region.
Finally, a closed loop curve can be formed by a functional expression of the edge contour and the internode contour, thereby extracting the bone contour of the single-section bone.
The extraction scheme of the single-segment bone profile of the application is divided into two stages: 1) Extracting an edge profile; 2) An internode contour is extracted. For ease of understanding, the two stages are described in detail below with reference to the drawings, taking as an example the extraction of the profile of the spine.
Referring to fig. 2, fig. 2 is a flowchart illustrating an edge contour extraction according to an exemplary embodiment of the present application. As shown in fig. 2, the method may include the steps of:
step 202, a gradient map in the vertical direction is calculated.
As shown in fig. 3A, the upper and lower side contours (i.e., edge contours) of the single-segment spine image 10 (which may be an X-ray image or a DRR image) have a significant change in gray scale relative to the background image. Thus, the upper and lower side profiles can be extracted by a gradient-based method.
As shown in fig. 3B, a gradient map G of the change in pixel values of the spine image in the vertical direction (i.e., the vertical direction in the figure) is calculated. The size of the gradient map G is consistent with the size of the single-segment spine image 10, and the brightness of a pixel point in the gradient map G represents the vertical gradient value of the pixel point. For example, a gradient map in the vertical direction may be calculated by:
G y ＝|f(x,y)-f(x,y-1)|；
where the f (x, y) function represents the gray value of the pixel point at the coordinate (x, y) location.
Step 204, a set of candidate edge contours is generated.
In connection with the above example, a threshold t may be set for measuring the rate of change of gray values at the edge profile. Specifically, for each column of pixel points in the gradient map G, the pixel point with the first vertical gradient value from top to bottom being greater than a threshold t is marked as an upper candidate contour point of the spine; similarly, the pixel point with the first vertical gradient value larger than the threshold value t from bottom to top is marked as the candidate contour point of the lower side of the spine; the upper candidate contour point and the lower candidate contour point together constitute a candidate edge contour set. The threshold t may be flexibly set according to practical situations, for example, the threshold t may be set to 10.
As shown in fig. 3C, the upper candidate contour point is located in the upper broken line frame in the single-segment spine image 10; similarly, the inferior candidate contour point is located in the dashed box at the lower side in the single-segment spine image 10.
Step 206, deleting the discrete points.
As shown in FIG. 3D, a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) spatial Density clustering algorithm can be used to perform cluster analysis on pixel points in the candidate edge contour set so as to delete discrete pixel points and improve the accuracy of subsequent fitting of edge contours. Of course, the application is not limited to the algorithm specifically employed for cluster analysis.
Step 208, fitting an edge profile.
As shown in fig. 3E, after deleting the discrete pixel points, fitting is performed on the remaining pixel points in the candidate edge contour set, and the fitting result is used as an edge contour. Specifically, fitting the upper candidate contour points remained in the candidate edge contour set, so as to obtain a curve a as an upper contour; and fitting the rest lower candidate contour points in the candidate edge contour set, so as to obtain a curve b as a lower contour.
For example, a cubic curve model may be employed under the random sample consensus (RANSAC) framework to fit the remaining set of pixel points P in the candidate edge contour set.
1) Randomly selecting four points p1, p2, p3 and p4; wherein, P1 epsilon P, P2 epsilon P, P3 epsilon P and P4 epsilon P;
2) Fitting a cubic curve l=polyfit (p 1, p2, p3, p 4) with the four selected points;
3) Traversing all points in the set P, and counting the number of points conforming to the curve model under the condition of meeting a certain tolerance error delta dWherein->dis (pi, l) is a function of the distance from the calculated point pi to the curve l, and the point conforming to the model becomes an inner point;
4) If N is more than 0.95 XN, N is the number of all points in the set P, directly taking the estimation result as the final profile of two sides of the spine; otherwise, the current curve is re-estimated by using all the inner points, the estimated result is used as the final profile of the two sides of the spine, and the profile extraction is finished.
5) When the circulation times reach the maximum circulation times, all internal points under the condition of n maximum are taken to re-fit the curve, and the estimated result is taken as the final profile of the two sides of the spine.
Referring to fig. 4, fig. 4 is a flowchart illustrating an internode contour extraction according to an exemplary embodiment of the present application. As shown in fig. 4, the method may include the steps of:
step 402, a foreground region is acquired that includes an internode contour.
In this embodiment, since the internode contour of the spine is closer to a straight line, a straight line is used to fit the internode contour. Wherein first a foreground region containing the internode contours needs to be acquired. Specifically, a single-segment spine image (i.e., the single-segment spine image 10 in fig. 3A) may be first cut from the spine image, then the left and right boundaries of the single-segment spine image are used to cut the sub-regions including the internode contours on the left and right sides, and finally the sub-regions are subjected to binarization processing to obtain the foreground region.
For example, as shown in fig. 5A, after a single-segment spine image 10 is manually captured by a person, the single-segment spine image 10 may be input to a device for performing contour extraction operation, or any other related art method may be used to extract the single-segment spine image; the sub-region 21 of corresponding width (e.g., 3/11 of the length of the spine) is then truncated from the boundaries of the single-segment spine image 10 according to a predefined scale (the single-segment spine image 10 includes two boundaries, one of which is exemplified). Further, as shown in fig. 5B, the sub-region 21 is binarized to obtain a foreground region 22. The foreground portion (black portion in the binary image 22) and the background portion (white portion in the binary image 22) in the sub-region 21 (i.e., the binary image 22 in fig. 5B) after the binarization process show a significant black-and-white effect. In other words, the foreground region 22 corresponds to a pixel point for representing black (hereinafter simply referred to as a black pixel point; for example, a pixel value of a black pixel point after binarization is normally 0) in fig. 5B. It should be noted that, the binarization algorithm used in the present embodiment may refer to the description in the related art, and will not be described herein. For example, a large gold method, a double peak method, a P parameter method, an iterative method, an OTSU method, and the like can be employed; of course, this application is not limited thereto.
Step 404, projection along the extension of the internode contour (perpendicular to the extension of the spine).
In this embodiment, the spine direction (i.e., the extending direction of the spine) may be set manually, i.e., the apparatus performing the contour extraction operation may acquire the spine direction by receiving an instruction to set the spine direction. Of course, any other technical means for determining the direction of the spine in the related art may be used, and the present application is not limited thereto.
In step 406, the distribution position where the pixel point is most distributed is determined.
In this embodiment, since the foreground region (including the inter-node contour) is projected along the extending direction of the inter-node contour, the obtained projection distribution map shows the distribution of all the pixels in the foreground region in the extending direction (i.e., the projection direction), and the straight line located in the projection direction and passing through the distribution position with the most distribution of the pixels is most suitable for representing the inter-node contour.
As shown in fig. 5C, the position of the projection maximum point (the distribution position where the pixel points are most distributed) in the projection profile is the position of the point a in the map, that is, the point a is located on the internode contour.
In step 408, a linear equation of the internode contour is calculated.
For example, a coordinate system as shown in FIG. 5D may be established: the ordinate is parallel to the projection direction, i.e. the coordinate system in the same direction as in fig. 5C. Then, with the slope (i.e., projection direction) and the position of point a known, an equation expression for line d in that coordinate system can be calculated.
As shown in FIG. 5E, a straight line d is used to represent the internode contour
As shown in fig. 6, a closed loop curve may be formed based on the edge profiles a, b obtained from the embodiments shown in fig. 3A-3E and the internode profiles d, c obtained from the embodiments shown in fig. 5A-5E, so that the spinal profile of a single spinal column may be further extracted.
Fig. 7 shows a schematic block diagram of a master-side based electronic device according to an exemplary embodiment of the present application. Referring to fig. 7, at a hardware level, the electronic device includes a processor 702, an internal bus 704, a network interface 707, a memory 708, and a nonvolatile memory 710, although other hardware required by other services is also possible. The processor 702 reads the corresponding computer program from the non-volatile memory 710 into the memory 708 and then runs to form the extraction means for the single-segment bone contours at the logic level. Of course, other implementations, such as logic devices or combinations of hardware and software, are not excluded from the present application, that is, the execution subject of the following processing flows is not limited to each logic unit, but may be hardware or logic devices.
Referring to fig. 8, in a software embodiment, the extraction device for a single bone contour may include:
an acquisition unit 81 that acquires a foreground region containing an internode contour of any single bone of the target bones;
the projection unit 82 is configured to project the foreground region along a preset projection direction, so as to obtain a distribution situation of pixels of the foreground region in the preset projection direction, where the preset projection direction is used to represent an extension direction of the internode contour;
a determining unit 83 that determines a straight line equation corresponding to the internode contour according to the pixel point distribution situation;
an extraction unit 84 determines an edge profile of the any one-segment bone, and extracts a bone profile of the any one-segment bone according to a linear equation of the edge profile and the internode profile.
Optionally, the acquiring unit 81 is specifically configured to:
intercepting a single-segment bone image containing any single-segment bone from the image of the target bone, wherein the single-segment bone image does not contain other single-segment bones except the any single-segment bone;
intercepting a sub-region containing the internode contours from the single-segment bone image;
and carrying out binarization processing on the subareas to obtain the foreground area.
Optionally, the preset projection direction is perpendicular to a bone extension direction of the target bone.
Optionally, the determining unit 83 is specifically configured to:
determining the distribution position with the most pixel distribution in the preset projection direction in the foreground region;
and determining a linear equation corresponding to the internode contour according to the distribution position and the preset projection direction.
Optionally, the extracting unit 84 is specifically configured to:
identifying the edge contour based on pixel value differences of pixel points between the edge contour and a peripheral area; the peripheral region is a region of the any individual bone that is outside the edge contour.
Optionally, the extracting unit 84 is further configured to:
traversing the pixel value of each pixel point along the inner side direction from the two side boundaries of the single-section bone image of any single-section bone, wherein the single-section bone image does not contain other single-section bones except any single-section bone;
adding all pixel points with the change rate between the first pixel value and the pixel value of the last pixel point in the inner side direction exceeding a preset change rate threshold value into a candidate edge contour set;
performing cluster analysis on the pixel points in the candidate edge contour set to delete discrete points in the candidate edge contour set;
and fitting the rest pixel points in the candidate edge contour set, and taking a fitting result as the edge contour.
Optionally, the target bone comprises a spinal column.
The implementation process of the functions and roles of each unit in the above device is specifically shown in the implementation process of the corresponding steps in the above method, and will not be described herein again.
For the device embodiments, reference is made to the description of the method embodiments for the relevant points, since they essentially correspond to the method embodiments. The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purposes of the present application. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, e.g. a memory, comprising instructions executable by a processor of the extraction device of a single bone contour as described above to implement a method as described in any of the above embodiments, e.g. the method may comprise: acquiring a foreground region containing the internode contour of any single bone in the target bone; projecting the foreground region along a preset projection direction to obtain pixel point distribution conditions of the foreground region in the preset projection direction, wherein the preset projection direction is used for representing the extension direction of the internode outline; determining a linear equation corresponding to the internode contour according to the pixel point distribution condition; and determining the edge profile of any single-section bone, and extracting the bone profile of any single-section bone according to a linear equation of the edge profile and the internode profile.
Optionally, the acquiring a foreground region containing an internode contour of any single segment of the target bone includes:
intercepting a single-segment bone image containing any single-segment bone from the image of the target bone, wherein the single-segment bone image does not contain other single-segment bones except the any single-segment bone;
intercepting a sub-region containing the internode contours from the single-segment bone image;
and carrying out binarization processing on the subareas to obtain the foreground area.
Optionally, the preset projection direction is perpendicular to a bone extension direction of the target bone.
Optionally, the determining a straight line equation corresponding to the internode contour according to the pixel point distribution condition includes:
determining the distribution position with the most pixel distribution in the preset projection direction in the foreground region;
and determining a linear equation corresponding to the internode contour according to the distribution position and the preset projection direction.
Optionally, the determining the edge profile of the any single bone includes:
identifying the edge contour based on pixel value differences of pixel points between the edge contour and a peripheral area; the peripheral region is a region of the any individual bone that is outside the edge contour.
Optionally, the identifying the edge contour based on the pixel value difference of the pixel points between the edge contour and the peripheral area includes:
traversing the pixel value of each pixel point along the inner side direction from the two side boundaries of the single-section bone image of any single-section bone, wherein the single-section bone image does not contain other single-section bones except any single-section bone;
adding all pixel points with the change rate between the first pixel value and the pixel value of the last pixel point in the inner side direction exceeding a preset change rate threshold value into a candidate edge contour set;
performing cluster analysis on the pixel points in the candidate edge contour set to delete discrete points in the candidate edge contour set;
and fitting the rest pixel points in the candidate edge contour set, and taking a fitting result as the edge contour.
Optionally, the target bone comprises a spinal column.
Wherein the non-transitory computer readable storage medium may be a ROM, random-access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc., which is not limited in this application.
The foregoing description of the preferred embodiments of the present invention is not intended to limit the invention to the precise form disclosed, and any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A method for extracting a single bone contour, comprising:
acquiring a foreground region containing the internode contour of any single bone in the target bone;
projecting the foreground region along a preset projection direction to obtain pixel point distribution conditions of the foreground region in the preset projection direction, wherein the preset projection direction is used for representing the extension direction of the internode outline;
determining a linear equation corresponding to the internode contour according to the pixel point distribution condition;
determining the edge profile of any single-section bone, and extracting the bone profile of any single-section bone according to a linear equation of the edge profile and the internode profile;
the determining a straight line equation corresponding to the internode contour according to the pixel point distribution condition comprises the following steps:
determining the distribution position with the most pixel distribution in the preset projection direction in the foreground region;
determining a linear equation corresponding to the internode contour according to the distribution position and the preset projection direction;
the determining the edge profile of the any single segment of bone comprises:
identifying the edge contour based on pixel value differences of pixel points between the edge contour and a peripheral area; the peripheral area is an area of the any single bone located outside the edge contour;
the identifying the edge contour based on the pixel value difference of the pixel points between the edge contour and the peripheral area comprises the following steps:
traversing the pixel value of each pixel point along the inner side direction from the two side boundaries of the single-section bone image of any single-section bone, wherein the single-section bone image does not contain other single-section bones except any single-section bone;
adding all pixel points with the change rate between the first pixel value and the pixel value of the last pixel point in the inner side direction exceeding a preset change rate threshold value into a candidate edge contour set;
performing cluster analysis on the pixel points in the candidate edge contour set to delete discrete points in the candidate edge contour set;
and fitting the rest pixel points in the candidate edge contour set, and taking a fitting result as the edge contour.
2. The method of claim 1, wherein the acquiring a foreground region containing an internode contour of any single segment of the target bone comprises:
intercepting a single-segment bone image containing any single-segment bone from the image of the target bone, wherein the single-segment bone image does not contain other single-segment bones except the any single-segment bone;
intercepting a sub-region containing the internode contours from the single-segment bone image;
and carrying out binarization processing on the subareas to obtain the foreground area.
3. The method of claim 1, wherein the predetermined projection direction is perpendicular to a bone extension direction of the target bone.
4. The method of claim 1, wherein the target bone comprises a spinal column.
5. An extraction device for a single bone contour, comprising:
an acquisition unit that acquires a foreground region containing an internode contour of any one single bone of a target bone;
the projection unit is used for projecting the foreground region along a preset projection direction to obtain the pixel point distribution condition of the foreground region in the preset projection direction, wherein the preset projection direction is used for representing the extension direction of the internode contour;
the determining unit is used for determining a linear equation corresponding to the internode outline according to the pixel point distribution condition;
the extraction unit is used for determining the edge profile of any single-section bone and extracting the bone profile of any single-section bone according to a linear equation of the edge profile and the internode profile;
the determining unit is specifically configured to:
determining the distribution position with the most pixel distribution in the preset projection direction in the foreground region;
determining a linear equation corresponding to the internode contour according to the distribution position and the preset projection direction;
the extraction unit is specifically used for:
identifying the edge contour based on pixel value differences of pixel points between the edge contour and a peripheral area; the peripheral area is an area of the any single bone located outside the edge contour;
the extraction unit is further configured to:
traversing the pixel value of each pixel point along the inner side direction from the two side boundaries of the single-section bone image of any single-section bone, wherein the single-section bone image does not contain other single-section bones except any single-section bone;
adding all pixel points with the change rate between the first pixel value and the pixel value of the last pixel point in the inner side direction exceeding a preset change rate threshold value into a candidate edge contour set;
performing cluster analysis on the pixel points in the candidate edge contour set to delete discrete points in the candidate edge contour set;
and fitting the rest pixel points in the candidate edge contour set, and taking a fitting result as the edge contour.
6. The apparatus according to claim 5, wherein the acquisition unit is specifically configured to:
intercepting a single-segment bone image containing any single-segment bone from the image of the target bone, wherein the single-segment bone image does not contain other single-segment bones except the any single-segment bone;
intercepting a sub-region containing the internode contours from the single-segment bone image;
and carrying out binarization processing on the subareas to obtain the foreground area.
7. The device of claim 5, wherein the predetermined projection direction is perpendicular to a bone extension direction of the target bone.
8. The device of claim 5, wherein the target bone comprises a spinal column.
9. An electronic device, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the method of any of claims 1-4 by executing the executable instructions.
10. A computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the method according to any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910468788.1A CN112017148B (en) | 2019-05-31 | 2019-05-31 | Method and device for extracting single-segment skeleton contour |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910468788.1A CN112017148B (en) | 2019-05-31 | 2019-05-31 | Method and device for extracting single-segment skeleton contour |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112017148A CN112017148A (en) | 2020-12-01 |
| CN112017148B true CN112017148B (en) | 2024-03-22 |
Family
ID=73502013
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910468788.1A Active CN112017148B (en) | 2019-05-31 | 2019-05-31 | Method and device for extracting single-segment skeleton contour |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112017148B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114638828B (en) * | 2022-05-18 | 2022-09-23 | 数聚(山东)医疗科技有限公司 | Radiological image intelligent segmentation method based on computer vision |
| CN115131301B (en) * | 2022-06-15 | 2023-04-18 | 北京长木谷医疗科技有限公司 | Method and system for intelligently identifying osteoarthritis based on deep learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0626656A2 (en) * | 1993-05-28 | 1994-11-30 | South Birmingham District Health Authority | Image processing system and method for automatic feature extraction |
| CN103729614A (en) * | 2012-10-16 | 2014-04-16 | 上海唐里信息技术有限公司 | People recognition method and device based on video images |
| CN106663309A (en) * | 2014-07-03 | 2017-05-10 | 西门子产品生命周期管理软件公司 | User-guided shape morphing in bone segmentation for medical imaging |
| CN106683090A (en) * | 2016-12-31 | 2017-05-17 | 上海联影医疗科技有限公司 | Rib positioning method in medical image and system thereof |
| CN107862699A (en) * | 2017-09-22 | 2018-03-30 | 中国科学院深圳先进技术研究院 | Bone edges extracting method, device, equipment and the storage medium of Bone CT image |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4437333B2 (en) * | 2007-09-28 | 2010-03-24 | ジーイー・メディカル・システムズ・グローバル・テクノロジー・カンパニー・エルエルシー | Image processing method, image processing apparatus, and program |
| WO2016046143A1 (en) * | 2014-09-28 | 2016-03-31 | Koninklijke Philips N.V. | Image processing apparatus and method |
-
2019
- 2019-05-31 CN CN201910468788.1A patent/CN112017148B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0626656A2 (en) * | 1993-05-28 | 1994-11-30 | South Birmingham District Health Authority | Image processing system and method for automatic feature extraction |
| CN103729614A (en) * | 2012-10-16 | 2014-04-16 | 上海唐里信息技术有限公司 | People recognition method and device based on video images |
| CN106663309A (en) * | 2014-07-03 | 2017-05-10 | 西门子产品生命周期管理软件公司 | User-guided shape morphing in bone segmentation for medical imaging |
| CN106683090A (en) * | 2016-12-31 | 2017-05-17 | 上海联影医疗科技有限公司 | Rib positioning method in medical image and system thereof |
| CN107862699A (en) * | 2017-09-22 | 2018-03-30 | 中国科学院深圳先进技术研究院 | Bone edges extracting method, device, equipment and the storage medium of Bone CT image |
Non-Patent Citations (1)
| Title |
|---|
| 基于改进活动轮廓模型的手掌骨的提取;孙振国;电子世界(2013, (04));77-78 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112017148A (en) | 2020-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112633144A (en) | Face occlusion detection method, system, device and storage medium | |
| JP5713790B2 (en) | Image processing apparatus, image processing method, and program | |
| CN112967236B (en) | Image registration method, device, computer equipment and storage medium | |
| EP1465109A2 (en) | Method for automated analysis of digital chest radiographs | |
| CN108230376B (en) | Remote sensing image processing method and device and electronic equipment | |
| CN110675407B (en) | Image instance segmentation method and device, electronic equipment and storage medium | |
| CN112017148B (en) | Method and device for extracting single-segment skeleton contour | |
| EP1771821A1 (en) | Projection views and orientation of chest radiographs | |
| US20090060332A1 (en) | Object segmentation using dynamic programming | |
| CN111932545A (en) | Image processing method, target counting method and related device thereof | |
| CN111310710A (en) | Face detection method and system | |
| CN113723399A (en) | License plate image correction method, license plate image correction device and storage medium | |
| US11842275B2 (en) | Improving segmentations of a deep neural network | |
| CN113240661A (en) | Deep learning-based lumbar vertebra analysis method, device, equipment and storage medium | |
| US11704807B2 (en) | Image processing apparatus and non-transitory computer readable medium storing program | |
| US9672600B2 (en) | Clavicle suppression in radiographic images | |
| CN111209908A (en) | Method and device for updating label box, storage medium and computer equipment | |
| CN111723836A (en) | Image similarity calculation method and device, electronic equipment and storage medium | |
| US9454814B2 (en) | PACS viewer and a method for identifying patient orientation | |
| JP2004188202A (en) | Automatic analysis method of digital radiograph of chest part | |
| CN111753723B (en) | Fingerprint identification method and device based on density calibration | |
| CN115205113A (en) | Image splicing method, device, equipment and storage medium | |
| CN113962953A (en) | Image processing method and system | |
| CN113592876A (en) | Training method and device for split network, computer equipment and storage medium | |
| CN112017147B (en) | Bone image extraction method and device |
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 |