CN116071355A - Auxiliary segmentation system and method for peripheral blood vessel image - Google Patents
Auxiliary segmentation system and method for peripheral blood vessel image Download PDFInfo
- Publication number
- CN116071355A CN116071355A CN202310203922.1A CN202310203922A CN116071355A CN 116071355 A CN116071355 A CN 116071355A CN 202310203922 A CN202310203922 A CN 202310203922A CN 116071355 A CN116071355 A CN 116071355A
- Authority
- CN
- China
- Prior art keywords
- blood vessel
- window
- suspected
- area
- region
- 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.)
- Withdrawn
Links
Images
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
- G06T7/12—Edge-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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to an auxiliary segmentation system and method for peripheral blood vessel images, wherein the method comprises the following steps: the method comprises the steps of obtaining suspected blood vessel pixel points and suspected blood vessel areas, obtaining window areas corresponding to different sizes and gray information uniformity of the suspected blood vessel areas corresponding to the window areas, obtaining extending trends of all boundary points in each window area, obtaining optimal window sizes and obtaining blood vessel areas. According to the invention, the optimal window size is obtained, so that the blood vessel region is obtained based on the optimal window size and by utilizing a local binarization algorithm, and the segmentation accuracy of the blood vessel region of the blood vessel image is improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an auxiliary segmentation system and method for peripheral blood vessel images.
Background
Peripheral blood vessels are other blood vessels except cardiovascular and cerebrovascular vessels, mainly trunk blood vessels, visceral blood vessels and limb blood vessels, and contain arteries, veins and capillaries therein. The blood vessel image is a two-dimensional expression of a three-dimensional blood vessel structure under a projection condition, and through knowing relevant information such as the shape, the distribution and the like of the blood vessels, the disease condition of a patient can be assisted to be analyzed, the operation planning is performed in advance, unnecessary damage to the patient caused by misjudgment on the disease condition or the damage to main blood vessels in the operation process is avoided, and the blood vessels in the peripheral blood vessel image are required to be segmented and extracted for clear analysis of the blood vessels.
The most commonly used peripheral blood vessel image segmentation method in the prior art is divided into two major types, namely a level set method based on region information and a level set method based on edge information, wherein the level set method based on the edge identifies blood vessels in an image through the edge information in the image, and can adapt to the condition that brightness at different positions in the image is inconsistent, but is excessively sensitive to noise and the edge information in the image, and the problem that edge leakage easily occurs under the condition that the intensity of the edge information of the blood vessels in the image is weak, and the edges of other non-blood vessel structures in the image are easily identified as blood vessel edges, so that the extraction precision of the blood vessels is not high.
Disclosure of Invention
The invention provides an auxiliary segmentation system and method for peripheral blood vessel images, which are used for solving the problem that the extraction accuracy of blood vessels is not high due to the influence of noise and edge information in the existing edge-based level set method.
The auxiliary segmentation method for the peripheral blood vessel image adopts the following technical scheme:
acquiring suspected blood vessel pixel points and suspected blood vessel areas in the peripheral blood vessel image;
acquiring corresponding window areas in different sizes along a suspected blood vessel boundary line of the suspected blood vessel area by taking a boundary point on the suspected blood vessel boundary line as a central point, wherein the edges of two adjacent window areas in the same size are adjacent; acquiring gray information uniformity of a suspected blood vessel region corresponding to each window region;
acquiring tangential directions at each boundary point on the boundary line of the suspected blood vessel in each window area, and acquiring extension trends of all boundary points in each window area according to the tangential directions at all boundary points in each window area;
constructing an objective function according to the extending trend difference value corresponding to each two adjacent window areas under each size and the gray information uniformity of the suspected blood vessel area corresponding to the window area, and taking the window size corresponding to the minimum objective function value in the objective function values corresponding to all the sizes as the optimal window size;
and acquiring an optimal window area corresponding to each suspected blood vessel pixel point according to the optimal window size, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point or not by utilizing a local binarization algorithm, and acquiring a blood vessel area according to all the blood vessel pixel points.
Preferably, constructing the objective function includes:
in the method, in the process of the invention,
representing the size as
The window area corresponds to an objective function value, wherein,
；
Preferably, the obtaining the gray information uniformity of the corresponding suspected blood vessel region in each window region includes:
taking the suspected blood vessel area in each window area as a characteristic area;
and acquiring the gray information uniformity of the corresponding suspected blood vessel region in each window region according to the number of the suspected blood vessel pixel points in the feature region, the total number of the suspected blood vessel pixel points in the window region in which the feature region is positioned and the gray variance in the feature region.
Preferably, the difference between the extending trends of the next window area and the previous window area in the two adjacent window areas under each size is obtained.
Preferably, the included angle between the tangent line at each boundary point on the boundary line of the suspected blood vessel in the window area and the horizontal direction is taken as the tangent line direction at each boundary point on the boundary line of the suspected blood vessel.
Preferably, the average value of the tangential directions corresponding to all the boundary points in the window area is used as the extending trend of all the boundary points in each window area.
Preferably, the determining whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point includes:
acquiring self-adaptive thresholds of suspected blood vessel pixel points corresponding to each optimal window area according to a local binarization algorithm;
and marking the suspected blood vessel pixel points with the pixel values larger than the corresponding self-adaptive threshold values as blood vessel pixel points.
Preferably, acquiring the suspected vascular region includes:
traversing adjacent continuous adjacent suspected blood vessel pixel points in the neighborhood of each suspected blood vessel pixel point;
and taking the region formed by all continuous and adjacent suspected blood vessel pixel points as a suspected blood vessel region.
The invention provides an auxiliary segmentation system for peripheral blood vessel images, which comprises the following components:
the image processing module is used for acquiring suspected blood vessel pixel points and suspected blood vessel areas in the peripheral blood vessel image;
the first parameter acquisition module is used for acquiring corresponding window areas under different sizes along a suspected blood vessel boundary line of the suspected blood vessel area by taking a boundary point on the suspected blood vessel boundary line as a center point, wherein the edges of two adjacent window areas under the same size are adjacent to each other, and the gray information uniformity of the corresponding suspected blood vessel area in each window area is acquired;
the second parameter calculation module is used for acquiring the tangential direction of each boundary point on the boundary line of the suspected blood vessel in each window area, and acquiring the extending trend of all boundary points in each window area according to the tangential directions of all boundary points in each window area;
the self-adaptive size acquisition module is used for constructing an objective function according to the extending trend difference value corresponding to each two adjacent window areas under each size and the gray information uniformity of the window areas, and taking the window size corresponding to the minimum objective function value in the objective function values corresponding to all the sizes as the optimal window size;
the image segmentation module is used for acquiring an optimal window area corresponding to each suspected blood vessel pixel point according to the optimal window size, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point or not by utilizing a local binarization algorithm, and acquiring a blood vessel area according to all the blood vessel pixel points.
The auxiliary segmentation system and the auxiliary segmentation method for the peripheral blood vessel image have the beneficial effects that:
the method comprises the steps of analyzing a suspected blood vessel region according to the characteristics of uniformity in the interior of the blood vessel region and smooth edge extension by taking the influence of the interfered region on the judgment of the blood vessel region into consideration by acquiring suspected blood vessel pixel points and the suspected blood vessel region in a peripheral blood vessel image, wherein the interior gray level of the interfered region is disordered and the edge trend is inconsistent, and firstly, corresponding window regions in different sizes are acquired along a suspected blood vessel boundary line of the suspected blood vessel region by taking boundary points on the suspected blood vessel boundary line as central points, wherein the edges of two adjacent window regions in the same size are adjacent; acquiring the gray information uniformity of the corresponding suspected blood vessel region in each window region, acquiring the extending direction of the boundary line of the suspected blood vessel region, namely acquiring the tangential direction of each boundary point on the boundary line of the suspected blood vessel in each window region, and acquiring the extending trend of all boundary points in each window region according to the tangential directions of all boundary points in each window region, wherein whether each two window regions are distributed on a continuous boundary of a section of trend is represented by an extending trend difference value, the smaller the extending trend difference value is, the more the uniformity and smoothness characteristics of the extending of the blood vessel edge are highlighted, the larger the gray information uniformity is, the more the uniformity is, the corresponding blood vessel boundary line is illustrated, so that an objective function is constructed according to the extending trend difference value corresponding to each two adjacent window regions in each size and the gray information uniformity of the suspected blood vessel region corresponding to the window region, taking the window size corresponding to the minimum objective function value in the objective function values under all sizes as the optimal window size, finally, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point by utilizing a local binarization algorithm, and obtaining a blood vessel area according to all the blood vessel pixel points, namely, the invention considers the characteristics of uniform inside and smooth edge extension of the blood vessel area and also considers the interference degree of environments of different boundary positions on the segmentation result, so that when the segmentation window is obtained based on the objective function, each section of blood vessel area with the same extension trend and similar pixel environment can uniformly use the dynamic threshold segmentation window with the same size, the interference area does not have uniformity and trend consistency, and can quickly converge when the applicability to the objective function is lower, the window sizes of the interference areas are not consistent almost every position, so that the segmentation effect of the blood vessel areas in the subsequent segmentation is smooth, and the interference areas still show disordered gray after segmentation, so that the segmentation accuracy of the blood vessel images is improved when the threshold segmentation is carried out according to the optimal window size.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of an embodiment of an assisted segmentation system and method for peripheral vessel imaging according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment of the auxiliary segmentation system and the auxiliary segmentation method for peripheral blood vessel images, disclosed by the invention, have the following applicable scenes: the method for assisting in segmenting the peripheral blood vessels in the peripheral blood vessel images with different brightness degrees at different positions specifically as shown in fig. 1 comprises the following steps:
s1, obtaining a suspected blood vessel pixel point and a suspected blood vessel region;
specifically, a suspected blood vessel pixel point and a suspected blood vessel region in the peripheral blood vessel image are obtained; the peripheral blood vessel image is an X-ray image obtained by injecting a contrast agent into a blood vessel and utilizing the property that the X-ray cannot penetrate the contrast agent, the X-ray image is the peripheral blood vessel image, and the peripheral blood vessel image is a gray scale image.
Wherein, because the X-ray can not penetrate the contrast agent, the corresponding position of the blood vessel in the image is darker, namely the gray value corresponding to the pixel point in the area corresponding to the blood vessel is smaller, the suspected blood vessel pixel point which is possibly the blood vessel part in the image is divided according to the characteristic that the gray value corresponding to the pixel point in the area corresponding to the blood vessel is smaller, namely the maximum inter-class variance is used for the peripheral blood vessel image to obtain the divided threshold value, each pixel point with the gray value smaller than the threshold value in the image is selected, the pixel points with the gray value smaller than the threshold value are the suspected blood vessel pixel points,
because the adjacent distribution of each pixel point of the blood vessel position is darker and more uniform, that is, the gray value corresponding to the pixel point is smaller and the gray value near the pixel point is more uniform, the obtaining the suspected blood vessel region needs to exclude the isolated suspected blood vessel pixel points, that is, the obtaining the suspected blood vessel region includes: and traversing the continuous and adjacent suspicious blood vessel pixel points in the neighborhood of each suspicious blood vessel pixel point, and taking the region formed by all the continuous and adjacent suspicious blood vessel pixel points as a suspicious blood vessel region.
S2, acquiring corresponding window areas in different sizes and gray information uniformity of a suspected blood vessel area corresponding to the window areas;
since the dynamic threshold segmentation only occurs at the boundary portion, the window region is obtained according to the boundary point on the boundary line of the suspected blood vessel region as the center point, so that the boundary line of the suspected blood vessel region is taken as the center point along the boundary point on the boundary line of the suspected blood vessel, and the window regions corresponding to different sizes are obtained, wherein the edges of two adjacent window regions in the same size are adjacent.
It should be noted that, the edges of two adjacent window areas in the same size are adjacent to each other, i.e. the tile arrangement is made.
Specifically, when the pixel points of the tube correspond to the blood vessels, because the X-rays cannot penetrate the contrast agent, the pixel points of the blood vessel positions are distributed adjacently, and are darker and have more uniform gray level distribution, that is, the gray level value corresponding to the pixel points is smaller and the gray level value near the pixel points is also more uniform, that is, when the gray level value in the window area is more uniform, the probability that the pixel points correspond to the blood vessel positions is greater, and because the positions of the blood vessels where the pixel points are located are different, each pixel point in the window area acquired by taking the boundary point as the central point may be a suspected blood vessel pixel point, and may be other pixel points, so that uniformity analysis is only performed on the suspected blood vessel pixel points in the window area, and secondly, since the boundary point on the suspected blood vessel boundary line of the suspected blood vessel area may be a boundary point of the blood vessel may also be a boundary point of the interference area, the uniformity information of the corresponding suspected blood vessel area in each window area includes: according to the characteristic that each pixel point in the vascular area is tightly distributed, the suspected vascular area in each window area is used as a characteristic area, namely, adjacent continuous adjacent suspected vascular pixel points in the neighborhood of each suspected vascular pixel point are traversed; and taking the region formed by all continuous and adjacent suspected blood vessel pixel points as a suspected blood vessel region.
Since the discrete suspected blood vessel pixel points which are not adjacent to the central point of the window area do not have the characteristics about the central point, the discrete suspected blood vessel pixel points are excluded and not used as a part of the characteristic area, then when the boundary points are judged, the gray information of the characteristic area in the window area corresponding to the blood vessel pixel points serving as the centers is uniform, and the gray information uniformity of the characteristic area in the target window corresponding to the boundary points of the interference area is poor, so that the gray information uniformity of the corresponding suspected blood vessel area in each window area is obtained according to the number of the suspected blood vessel pixel points in the characteristic area, the total number of the suspected blood vessel pixel points in the window area where the characteristic area is located and the gray variance in the characteristic area.
The method for obtaining the gray information uniformity of the suspected blood vessel region corresponding to each window region specifically comprises the following steps: carrying out negative correlation calculation on the gray variance in the characteristic region to obtain a target gray variance; acquiring the ratio of the number of the suspected blood vessel pixel points in the characteristic region to the number of the suspected blood vessel pixel points in the window region where the characteristic region is located; taking the product of the quantity ratio and the target gray variance as the gray information uniformity of the suspected blood vessel region corresponding to the window region, wherein the calculation formula of the gray information uniformity of the suspected blood vessel region corresponding to the window region is as follows:
in the method, in the process of the invention,
representing the first line on the boundary of a suspected blood vessel
Gray information uniformity of suspected blood vessel areas corresponding to the window areas;
it should be noted that the number of the substrates,
representing the gray variance, i.e., the gray information uniformity, within the feature region, with smaller values representing more uniform gray levels within the feature region, more toward the vessel region,
the ratio of the number of the suspected blood vessel pixel points in the characteristic area to the total number of the suspected blood vessel pixel points in the window area is represented, the initial gray information uniformity of the characteristic area is corrected by taking the ratio of the number of the suspected blood vessel pixel points in the characteristic area to the total number of the suspected blood vessel pixel points in the window area as an influence factor, when the ratio of the number of the suspected blood vessel pixel points in the characteristic area to the number of the suspected blood vessel pixel points in the window area is smaller, the interference degree is higher, otherwise, the interference degree is smaller when the ratio of the number of the suspected blood vessel pixel points in the characteristic area to the number of the suspected blood vessel pixel points in the window area is larger, in short, the threshold segmentation difficulty is higher when the number of the suspected blood vessel pixel points in the non-characteristic area is larger, and when the number of the suspected blood vessel pixel points in the window area is almost the characteristic area, the threshold segmentation is performedThe segmentation object is a suspected blood vessel pixel point and other pixel points, so that the segmentation difficulty is extremely low, and therefore, the influence factor is set to correct the gray information uniformity of the characteristic region, the larger the influence factor is, the higher the uniformity is, the higher the gray information uniformity is, and when a window with a certain size is utilized for segmentation, the higher the gray information uniformity is, namely the probability that the window belongs to the boundary of the blood vessel region is, and the smaller the threshold segmentation difficulty is.
S3, obtaining the extending trend of all boundary points in each window area;
because the gray level information uniformity of the suspected blood vessel region corresponding to the window region is obtained in the step S3, the gray level information uniformity reflects the position characteristics of the blood vessel pixel points, so that according to the interference of other tissues in the suspected blood vessel pixel points, but some pixel points are relatively uniform in a smaller range in the interference part of other tissues or the low-illumination region in the peripheral blood vessel image, in order to avoid that the pixel points are identified as blood vessel parts, the pixel point characteristics are required to be continuously analyzed according to the consistency and smoothness characteristics of the extending direction of the blood vessel in a certain range, and the possibility of misjudgment is reduced, so that the extending trend of all boundary points in each window region, namely the extending direction of the suspected blood vessel boundary line in the window region, is required to be obtained firstly, specifically, the extending trend of all boundary points in each window region is obtained according to the tangential direction of all boundary points in each window region.
The tangential direction at each boundary point on the boundary line of the suspected blood vessel in each window area is specifically obtained as follows: and taking the included angle between the tangent line at each boundary point on the boundary line of the suspected blood vessel in the window area and the horizontal direction as the tangent line direction at each boundary point on the boundary line of the suspected blood vessel.
The extending trend of all boundary points in each window area is specifically obtained as follows: and taking the average value of the tangential directions corresponding to all boundary points in the window area as the extending trend of all boundary points in each window area, namely solving the average value of included angles between the tangential lines of all boundary points on the suspected blood vessel boundary line in the window area and the horizontal direction, and taking the average value of the included angles as the extending trend of all boundary points in the window area, namely the extending direction of the suspected blood vessel boundary line in the window area.
S4, obtaining the optimal window size;
based on the gray information uniformity of the corresponding suspected blood vessel region in each window region obtained in the step S3 and the extending trend of all boundary points in the window region obtained in the step S4, obtaining the characteristics of consistency and smoothness of extending directions in a certain range based on the extending trend, and finally determining the optimal window size according to the gray information uniformity and the consistency and smoothness characteristics of the extending trend, specifically, constructing an objective function according to the extending trend difference value corresponding to every two adjacent window regions in each size and the gray information uniformity of the window region, and taking the window size corresponding to the minimum objective function value in the objective function values in all sizes as the optimal window size;
wherein, a specific formula of an objective function is constructed:
in the method, in the process of the invention,
representing the size as
The window area corresponds to an objective function value, wherein,
；
it should be noted that the number of the substrates,
representing the first along a boundary line of a suspected blood vessel
A window region and adjacent first window region along the suspected blood vessel boundary line
The difference value of the extending trend of each window area is smaller, and the smaller the difference value of the extending trend is, the size of the window area is represented as
When the size is
Is distributed in a section of window areaThe suspected blood vessel pixel point corresponding to the center of the window area on the boundary with continuous trend is at the size
The more the window region of the (B) is divided, the more the uniformity and smoothness characteristics of the vessel edge extension are emphasized, the more the window region is divided into the size of the window
Gray information uniformity corresponding to each window area
When the window area is larger, the suspected blood vessel pixel point corresponding to the center of the window area is more likely to be the blood vessel pixel point, so that the window area is formed by
The smaller the size, the more suitable the division is at the size of the window area, in
The smaller and
the smaller the window is, the more suitable the window with the size is segmented, so that two adjacent window areas in the selected continuous adjacent window areas are a group, the average value of the extending trend difference values corresponding to all groups is the average value of the gray information uniformity corresponding to the selected adjacent window areas, and the objective function is constructed.
And obtaining the corresponding objective function value under each size until the objective function is converged, then selecting the minimum objective function value in the objective function values corresponding to all sizes, and taking the window size corresponding to the minimum objective function value as the optimal window size.
S5, acquiring a blood vessel region;
and acquiring an optimal window area corresponding to each suspected blood vessel pixel point according to the optimal window size, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point or not by utilizing a local binarization algorithm, and acquiring a blood vessel area according to all the blood vessel pixel points.
Specifically, acquiring a self-adaptive threshold value of a suspected blood vessel pixel point corresponding to each optimal window area according to a local binarization algorithm (Sauvola algorithm); the suspected blood vessel pixel points with the pixel values larger than the corresponding self-adaptive threshold values are marked as blood vessel pixel points, wherein the Sauvola algorithm is the algorithm in the prior art, and the description of the embodiment is omitted; and then taking the region formed by all the vascular pixel points as a vascular region.
The invention provides an auxiliary segmentation system for peripheral blood vessel images, which comprises the following components: the device comprises an image processing module, a first parameter acquisition module, a second parameter calculation module, a self-adaptive size acquisition module and an image segmentation module, wherein the image processing module is used for acquiring suspected blood vessel pixel points and suspected blood vessel areas in peripheral blood vessel images; the first parameter acquisition module is used for acquiring corresponding window areas in different sizes along a suspected blood vessel boundary line of the suspected blood vessel area by taking a boundary point on the suspected blood vessel boundary line as a center point, wherein the edges of two adjacent window areas in the same size are adjacent; acquiring gray information uniformity of a suspected blood vessel region corresponding to each window region; the second parameter calculation module is used for obtaining the tangential direction of each boundary point on the suspected blood vessel boundary line in each window area, and obtaining the extending trend of all boundary points in each window area according to the tangential directions of all boundary points in each window area; the self-adaptive size acquisition module is used for constructing an objective function according to the extending trend difference value corresponding to each two adjacent window areas under each size and the gray information uniformity of the window areas, and taking the window size corresponding to the minimum objective function value in the objective function values corresponding to all the sizes as the optimal window size; the image segmentation module is used for obtaining an optimal window area corresponding to each suspected blood vessel pixel point according to the optimal window size, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point or not by utilizing a local binarization algorithm, and obtaining a blood vessel area according to all the blood vessel pixel points.
According to the auxiliary segmentation system and the auxiliary segmentation method for the peripheral blood vessel image, through acquiring the suspected blood vessel pixel points and the suspected blood vessel regions in the peripheral blood vessel image, the suspected blood vessel regions are analyzed according to the characteristics that the interiors of the blood vessel regions are uniform and the edge extensions are smooth in consideration of the influence of the interfered regions on the judgment of the blood vessel regions, and the suspected blood vessel boundary lines of the suspected blood vessel regions are firstly along the suspected blood vessel boundary lines of the suspected blood vessel regions by taking boundary points on the suspected blood vessel boundary lines as central points, wherein the edges of two adjacent window regions in the same size are adjacent; acquiring the gray information uniformity of the corresponding suspected blood vessel region in each window region, acquiring the extending direction of the boundary line of the suspected blood vessel region, namely acquiring the tangential direction of each boundary point on the boundary line of the suspected blood vessel in each window region, and acquiring the extending trend of all boundary points in each window region according to the tangential directions of all boundary points in each window region, wherein whether each two window regions are distributed on a continuous boundary of a section of trend is represented by an extending trend difference value, the smaller the extending trend difference value is, the more the uniformity and smoothness characteristics of the extending of the blood vessel edge are highlighted, the larger the gray information uniformity is, the more the uniformity is, the corresponding blood vessel boundary line is illustrated, so that an objective function is constructed according to the extending trend difference value corresponding to each two adjacent window regions in each size and the gray information uniformity of the suspected blood vessel region corresponding to the window region, taking the window size corresponding to the minimum objective function value in the objective function values under all sizes as the optimal window size, finally, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point by utilizing a local binarization algorithm, and obtaining a blood vessel area according to all the blood vessel pixel points, namely, the invention considers the characteristics of uniform inside and smooth edge extension of the blood vessel area and also considers the interference degree of environments of different boundary positions on the segmentation result, so that when the segmentation window is obtained based on the objective function, each section of blood vessel area with the same extension trend and similar pixel environment can uniformly use the dynamic threshold segmentation window with the same size, the interference area does not have uniformity and trend consistency, and can quickly converge when the applicability to the objective function is lower, the window sizes of the interference areas are not consistent almost every position, so that the segmentation effect of the blood vessel areas in the subsequent segmentation is smooth, and the interference areas still show disordered gray after segmentation, so that the segmentation accuracy of the blood vessel images is improved when the threshold segmentation is carried out according to the optimal window size.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (9)
1. An assisted segmentation method for peripheral vessel images, comprising:
acquiring suspected blood vessel pixel points and suspected blood vessel areas in the peripheral blood vessel image;
acquiring corresponding window areas in different sizes along a suspected blood vessel boundary line of the suspected blood vessel area by taking a boundary point on the suspected blood vessel boundary line as a central point, wherein the edges of two adjacent window areas in the same size are adjacent; acquiring gray information uniformity of a suspected blood vessel region corresponding to each window region;
acquiring tangential directions at each boundary point on the boundary line of the suspected blood vessel in each window area, and acquiring extension trends of all boundary points in each window area according to the tangential directions at all boundary points in each window area;
constructing an objective function according to the extending trend difference value corresponding to each two adjacent window areas under each size and the gray information uniformity of the suspected blood vessel area corresponding to the window area, and taking the window size corresponding to the minimum objective function value in the objective function values corresponding to all the sizes as the optimal window size;
and acquiring an optimal window area corresponding to each suspected blood vessel pixel point according to the optimal window size, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point or not by utilizing a local binarization algorithm, and acquiring a blood vessel area according to all the blood vessel pixel points.
2. The assisted segmentation method for peripheral vascular imaging according to claim 1, wherein constructing the objective function comprises:
in the method, in the process of the invention,
the expression size is +.>
An objective function value corresponding to the window region, wherein +.>
；
3. The assisted segmentation method for peripheral vascular imaging according to claim 1, wherein acquiring the gray level information uniformity of the corresponding suspected vascular region in each window region comprises:
taking the suspected blood vessel area in each window area as a characteristic area;
and acquiring the gray information uniformity of the corresponding suspected blood vessel region in each window region according to the number of the suspected blood vessel pixel points in the feature region, the total number of the suspected blood vessel pixel points in the window region in which the feature region is positioned and the gray variance in the feature region.
4. The assisted segmentation method for peripheral vascular image according to claim 1, wherein the difference between the extending trends of the next and previous window regions in the two adjacent window regions is obtained by differentiating the extending trends of the next and previous window regions in each size.
5. The method according to claim 1, wherein an angle between a tangent line at each boundary point on the boundary line of the suspected blood vessel in the window region and the horizontal direction is used as a tangent line direction at each boundary point on the boundary line of the suspected blood vessel.
6. The assisted segmentation method for peripheral vascular image according to claim 1, wherein the mean value of the tangential directions corresponding to all boundary points in the window area is used as the extending trend of all boundary points in each window area.
7. The assisted segmentation method for peripheral vascular image according to claim 1, wherein determining whether the suspected vascular pixel corresponding to each optimal window region is a vascular pixel comprises:
acquiring self-adaptive thresholds of suspected blood vessel pixel points corresponding to each optimal window area according to a local binarization algorithm;
and marking the suspected blood vessel pixel points with the pixel values larger than the corresponding self-adaptive threshold values as blood vessel pixel points.
8. The assisted segmentation method for peripheral vascular imaging according to claim 1, wherein acquiring the suspected vascular region comprises:
traversing adjacent continuous adjacent suspected blood vessel pixel points in the neighborhood of each suspected blood vessel pixel point;
and taking the region formed by all continuous and adjacent suspected blood vessel pixel points as a suspected blood vessel region.
9. An assisted segmentation system for peripheral vascular imaging, comprising:
the image processing module is used for acquiring suspected blood vessel pixel points and suspected blood vessel areas in the peripheral blood vessel image;
the first parameter acquisition module is used for acquiring corresponding window areas under different sizes along a suspected blood vessel boundary line of the suspected blood vessel area by taking a boundary point on the suspected blood vessel boundary line as a center point, wherein the edges of two adjacent window areas under the same size are adjacent to each other, and the gray information uniformity of the corresponding suspected blood vessel area in each window area is acquired;
the second parameter calculation module is used for acquiring the tangential direction of each boundary point on the boundary line of the suspected blood vessel in each window area, and acquiring the extending trend of all boundary points in each window area according to the tangential directions of all boundary points in each window area;
the self-adaptive size acquisition module is used for constructing an objective function according to the extending trend difference value corresponding to each two adjacent window areas under each size and the gray information uniformity of the window areas, and taking the window size corresponding to the minimum objective function value in the objective function values corresponding to all the sizes as the optimal window size;
the image segmentation module is used for acquiring an optimal window area corresponding to each suspected blood vessel pixel point according to the optimal window size, judging whether the suspected blood vessel pixel point corresponding to each optimal window area is a blood vessel pixel point or not by utilizing a local binarization algorithm, and acquiring a blood vessel area according to all the blood vessel pixel points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310203922.1A CN116071355A (en) | 2023-03-06 | 2023-03-06 | Auxiliary segmentation system and method for peripheral blood vessel image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310203922.1A CN116071355A (en) | 2023-03-06 | 2023-03-06 | Auxiliary segmentation system and method for peripheral blood vessel image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116071355A true CN116071355A (en) | 2023-05-05 |
Family
ID=86182097
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310203922.1A Withdrawn CN116071355A (en) | 2023-03-06 | 2023-03-06 | Auxiliary segmentation system and method for peripheral blood vessel image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116071355A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116542966A (en) * | 2023-06-28 | 2023-08-04 | 贵州医科大学附属医院 | Intelligent bone age analysis method for children endocrine abnormality detection |
| CN116993628A (en) * | 2023-09-27 | 2023-11-03 | 四川大学华西医院 | CT image enhancement system for tumor radio frequency ablation guidance |
| CN117252893A (en) * | 2023-11-17 | 2023-12-19 | 科普云医疗软件(深圳)有限公司 | Segmentation processing method for breast cancer pathological image |
| CN117422628A (en) * | 2023-12-18 | 2024-01-19 | 三亚中心医院(海南省第三人民医院、三亚中心医院医疗集团总院) | Optimized enhancement method for cardiac vascular ultrasonic examination data |
| CN117557460A (en) * | 2024-01-12 | 2024-02-13 | 济南科汛智能科技有限公司 | Angiography image enhancement method |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101127117A (en) * | 2007-09-11 | 2008-02-20 | 华中科技大学 | Method for segmenting blood vessel data using serial DSA image |
| CN101833757A (en) * | 2009-03-11 | 2010-09-15 | 深圳迈瑞生物医疗电子股份有限公司 | Method and system for detection edge of blood vessel graphic tissue structure and blood vessel endangium |
| CN107292890A (en) * | 2017-06-19 | 2017-10-24 | 北京理工大学 | A kind of medical image cutting method and device |
| US20180137612A1 (en) * | 2015-04-27 | 2018-05-17 | Wuhan Wuda Zoyon Science And Technology Co., Ltd. | A stepwise refinement detection method for pavement cracks |
| CN109544525A (en) * | 2018-11-15 | 2019-03-29 | 北京工业大学 | A kind of eyeground picture blood vessel recognition methods based on self-adapting window Model Matching |
| CN109658406A (en) * | 2018-12-25 | 2019-04-19 | 广州天鹏计算机科技有限公司 | Recognition methods, device, computer equipment and the storage medium of blood-vessel image |
| CN110276229A (en) * | 2018-03-14 | 2019-09-24 | 京东方科技集团股份有限公司 | Target object regional center localization method and device |
| CN112308846A (en) * | 2020-11-04 | 2021-02-02 | 赛诺威盛科技（北京）有限公司 | Blood vessel segmentation method and device and electronic equipment |
| CN113570570A (en) * | 2021-07-27 | 2021-10-29 | 蒋庆贺 | Device for segmenting pulmonary blood vessels from pulmonary mask image |
| WO2022062812A1 (en) * | 2020-09-28 | 2022-03-31 | 歌尔股份有限公司 | Screen defect detection method, apparatus, and electronic device |
| CN115661135A (en) * | 2022-12-09 | 2023-01-31 | 山东第一医科大学附属省立医院(山东省立医院) | Focus region segmentation method for cardio-cerebral angiography |
-
2023
- 2023-03-06 CN CN202310203922.1A patent/CN116071355A/en not_active Withdrawn
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101127117A (en) * | 2007-09-11 | 2008-02-20 | 华中科技大学 | Method for segmenting blood vessel data using serial DSA image |
| CN101833757A (en) * | 2009-03-11 | 2010-09-15 | 深圳迈瑞生物医疗电子股份有限公司 | Method and system for detection edge of blood vessel graphic tissue structure and blood vessel endangium |
| US20180137612A1 (en) * | 2015-04-27 | 2018-05-17 | Wuhan Wuda Zoyon Science And Technology Co., Ltd. | A stepwise refinement detection method for pavement cracks |
| CN107292890A (en) * | 2017-06-19 | 2017-10-24 | 北京理工大学 | A kind of medical image cutting method and device |
| CN110276229A (en) * | 2018-03-14 | 2019-09-24 | 京东方科技集团股份有限公司 | Target object regional center localization method and device |
| US20210334998A1 (en) * | 2018-03-14 | 2021-10-28 | Beijing Boe Optoelectronics Technology Co., Ltd. | Image processing method, apparatus, device and medium for locating center of target object region |
| CN109544525A (en) * | 2018-11-15 | 2019-03-29 | 北京工业大学 | A kind of eyeground picture blood vessel recognition methods based on self-adapting window Model Matching |
| CN109658406A (en) * | 2018-12-25 | 2019-04-19 | 广州天鹏计算机科技有限公司 | Recognition methods, device, computer equipment and the storage medium of blood-vessel image |
| WO2022062812A1 (en) * | 2020-09-28 | 2022-03-31 | 歌尔股份有限公司 | Screen defect detection method, apparatus, and electronic device |
| CN112308846A (en) * | 2020-11-04 | 2021-02-02 | 赛诺威盛科技（北京）有限公司 | Blood vessel segmentation method and device and electronic equipment |
| CN113570570A (en) * | 2021-07-27 | 2021-10-29 | 蒋庆贺 | Device for segmenting pulmonary blood vessels from pulmonary mask image |
| CN115661135A (en) * | 2022-12-09 | 2023-01-31 | 山东第一医科大学附属省立医院(山东省立医院) | Focus region segmentation method for cardio-cerebral angiography |
Non-Patent Citations (4)
| Title |
|---|
| BUKET TOPTAS¸ ET AL.: "Retinal blood vessel segmentation using pixel-based feature vector", 《BIOMEDICAL SIGNAL PROCESSING AND CONTROL》, pages 1 - 12 * |
| 刘帅 等: "自适应局部区域型水平集分割算法", 《计算机***应用》, vol. 26, no. 11, pages 145 - 151 * |
| 姜平: "眼底图像分割方法研究", 《中国博士学位论文全文数据库》, pages 1 - 105 * |
| 王卓英: "冠状动脉造影图像分割算法研究", 《中国优秀硕士学位论文全文数据库》, pages 1 - 70 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116542966A (en) * | 2023-06-28 | 2023-08-04 | 贵州医科大学附属医院 | Intelligent bone age analysis method for children endocrine abnormality detection |
| CN116542966B (en) * | 2023-06-28 | 2023-09-08 | 贵州医科大学附属医院 | Intelligent bone age analysis method for children endocrine abnormality detection |
| CN116993628A (en) * | 2023-09-27 | 2023-11-03 | 四川大学华西医院 | CT image enhancement system for tumor radio frequency ablation guidance |
| CN116993628B (en) * | 2023-09-27 | 2023-12-08 | 四川大学华西医院 | CT image enhancement system for tumor radio frequency ablation guidance |
| CN117252893A (en) * | 2023-11-17 | 2023-12-19 | 科普云医疗软件(深圳)有限公司 | Segmentation processing method for breast cancer pathological image |
| CN117252893B (en) * | 2023-11-17 | 2024-02-23 | 科普云医疗软件(深圳)有限公司 | Segmentation processing method for breast cancer pathological image |
| CN117422628A (en) * | 2023-12-18 | 2024-01-19 | 三亚中心医院(海南省第三人民医院、三亚中心医院医疗集团总院) | Optimized enhancement method for cardiac vascular ultrasonic examination data |
| CN117422628B (en) * | 2023-12-18 | 2024-03-05 | 三亚中心医院(海南省第三人民医院、三亚中心医院医疗集团总院) | Optimized enhancement method for cardiac vascular ultrasonic examination data |
| CN117557460A (en) * | 2024-01-12 | 2024-02-13 | 济南科汛智能科技有限公司 | Angiography image enhancement method |
| CN117557460B (en) * | 2024-01-12 | 2024-03-29 | 济南科汛智能科技有限公司 | Angiography image enhancement method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116071355A (en) | Auxiliary segmentation system and method for peripheral blood vessel image | |
| WO2021129323A1 (en) | Ultrasound image lesion describing method and apparatus, computer device, and storage medium | |
| CN109064476B (en) | CT chest radiography lung tissue image segmentation method based on level set | |
| EP2188779B1 (en) | Extraction method of tongue region using graph-based approach and geometric properties | |
| CN116109663B (en) | Stomach CT image segmentation method based on multi-threshold segmentation | |
| US5960102A (en) | X-ray image processing method and device for performing that method in which a portion corresponding to an x-ray absorption filter is selectively processed | |
| CN110610498A (en) | Mammary gland molybdenum target image processing method, system, storage medium and equipment | |
| CN117422628B (en) | Optimized enhancement method for cardiac vascular ultrasonic examination data | |
| CN116993628B (en) | CT image enhancement system for tumor radio frequency ablation guidance | |
| CN116152505A (en) | Bone target identification and segmentation method based on X-ray data | |
| CN111127373B (en) | Blood vessel image extraction method and device based on local section analysis | |
| CN111105427B (en) | Lung image segmentation method and system based on connected region analysis | |
| CN116485814A (en) | Intracranial hematoma region segmentation method based on CT image | |
| CN114820663A (en) | Assistant positioning method for determining radio frequency ablation therapy | |
| CN107564021A (en) | Detection method, device and the digital mammographic system of highly attenuating tissue | |
| CN112529918B (en) | Method, device and equipment for segmenting brain room area in brain CT image | |
| CN116993764B (en) | Stomach CT intelligent segmentation extraction method | |
| US8160336B2 (en) | Reducing false positives for automatic computerized detection of objects | |
| CN116596810B (en) | Automatic enhancement method for spine endoscope image | |
| CN111127404B (en) | Medical image contour rapid extraction method | |
| CN111861984A (en) | Method and device for determining lung region, computer equipment and storage medium | |
| CN112634280B (en) | MRI image brain tumor segmentation method based on energy functional | |
| CN111161285B (en) | Pericardial area positioning method, device and system based on feature analysis | |
| CN111402284A (en) | Image threshold value determination method and device based on three-dimensional connectivity | |
| EP0635804A1 (en) | Image processing method and device for performing that 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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230505 |