CN118135117A - Method, device, equipment and medium for generating region of three-dimensional model - Google Patents
Method, device, equipment and medium for generating region of three-dimensional model Download PDFInfo
- Publication number
- CN118135117A CN118135117A CN202410538639.9A CN202410538639A CN118135117A CN 118135117 A CN118135117 A CN 118135117A CN 202410538639 A CN202410538639 A CN 202410538639A CN 118135117 A CN118135117 A CN 118135117A
- Authority
- CN
- China
- Prior art keywords
- region
- interest
- dimensional model
- candidate
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 210000002455 dental arch Anatomy 0.000 claims description 54
- 238000010276 construction Methods 0.000 claims description 28
- 238000012545 processing Methods 0.000 claims description 17
- 210000001909 alveolar process Anatomy 0.000 claims description 16
- 238000004891 communication Methods 0.000 claims description 15
- 239000007943 implant Substances 0.000 claims description 9
- 238000002360 preparation method Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 8
- 238000005520 cutting process Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 238000013145 classification model Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 5
- 229940023487 dental product Drugs 0.000 description 5
- 210000000214 mouth Anatomy 0.000 description 5
- 238000013507 mapping Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000001680 brushing effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The present disclosure relates to a method, an apparatus, a device, and a medium for generating a region of a three-dimensional model, where the method includes: obtaining a three-dimensional model; and automatically identifying and determining a target region of interest on the three-dimensional model. The method and the device can automatically identify and determine the target region of interest, and improve the determination speed of the target region of interest.
Description
Technical Field
The disclosure relates to the technical field of oral cavity, in particular to a method, a device, equipment and a medium for generating a three-dimensional model.
Background
In the process of digital design of dental products, it is necessary to preserve the regions of interest of the doctor's teeth, teeth preparation, implant stems, alveolar ridges, etc., which are of most interest, and to delete other unwanted regions. At present, users select the region to be reserved or deleted under different visual angles through manual modes such as lasso, cutting planes, clicking, painting and brushing and the like for multiple interactions, and dental digital treatment is carried out on the reserved region of interest. This manual interaction has a number of problems, such as: the speed is low, different operator standards are different, a user may have multiple selection or fewer selection problems, a processing result is poor in a part of boundary areas, and the like. Therefore, how to quickly and accurately select a region of interest of a user is a problem to be solved.
Disclosure of Invention
In order to solve the technical problems, the present disclosure provides a method, an apparatus, a device, and a medium for generating a region of a three-dimensional model.
According to an aspect of the present disclosure, there is provided a region generating method of a three-dimensional model, including:
Obtaining a three-dimensional model;
and automatically identifying and determining a target region of interest on the three-dimensional model.
According to another aspect of the present disclosure, there is provided a region generating apparatus of a three-dimensional model, including:
the model acquisition module is used for acquiring a three-dimensional model;
and the region determining module is used for automatically identifying and determining the target region of interest on the three-dimensional model.
According to another aspect of the present disclosure, there is also provided an electronic apparatus including:
a processor;
a memory for storing the processor-executable instructions;
the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method described above.
According to another aspect of the present disclosure, there is also provided a computer-readable storage medium storing a computer program for executing the above method.
Compared with the prior art, the technical scheme provided by the embodiment of the disclosure has the following advantages:
Firstly, a three-dimensional model is acquired, and then a target region of interest is automatically identified and determined on the three-dimensional model. Compared with the prior art that a user manually brushes, lashes or cuts a plane to select the area to be reserved, the method of the embodiment can automatically identify and determine the target area of interest, and improves the determination speed of the target area of interest.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure.
In order to more clearly illustrate the embodiments of the present disclosure or the solutions in the prior art, the drawings that are required for the description of the embodiments or the prior art will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a flow chart of a method for generating a region of a three-dimensional model according to an embodiment of the present disclosure;
FIG. 2 is a schematic illustration of a three-dimensional model according to an embodiment of the present disclosure;
FIG. 3 is a schematic illustration of a projection view according to an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of a target region of interest according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of a second region of interest according to an embodiment of the present disclosure;
FIG. 6 is a schematic illustration of a target arch area according to an embodiment of the present disclosure;
FIG. 7 is a schematic diagram of a shell extraction model according to an embodiment of the disclosure;
FIG. 8 is a schematic view of a back cover model according to an embodiment of the present disclosure;
FIG. 9 is a schematic illustration of a dental product model according to an embodiment of the present disclosure;
FIG. 10 is a block diagram of a three-dimensional model region generating apparatus according to an embodiment of the present disclosure;
fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the disclosure.
Detailed Description
In order that the above objects, features and advantages of the present disclosure may be more clearly understood, a further description of aspects of the present disclosure will be provided below. It should be noted that, without conflict, the embodiments of the present disclosure and features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure, but the present disclosure may be practiced otherwise than as described herein; it will be apparent that the embodiments in the specification are only some, but not all, embodiments of the disclosure.
Fig. 1 is a flowchart of a method for generating a three-dimensional model according to an embodiment of the present disclosure, where the method may be suitable for identifying a region of interest such as a tooth, a planting bar, a tooth preparation, and an alveolar ridge in a mouth, that is, the embodiment of the present disclosure may provide a method for generating a three-dimensional model. The method may be performed by a region generating device configured to a three-dimensional model of the terminal, which may be implemented in software and/or hardware. As shown in fig. 1, the region generating method of the three-dimensional model may include the following steps.
S102, acquiring a three-dimensional model. As shown in fig. 2, the three-dimensional model may be a three-dimensional toothless mesh model.
S104, automatically identifying and determining the target region of interest on the three-dimensional model.
The present embodiment may include: first, candidate regions of interest are identified on the three-dimensional model based on the identification pattern. The specific implementation mode is as follows: carrying out multi-view projection on the three-dimensional model to obtain a plurality of projection views; identifying a first region of interest and a background region on each projection view through a preset classification and identification model; and projecting the first region of interest and the background region on each projection view back to the three-dimensional model to obtain candidate regions of interest on the three-dimensional model.
And secondly, performing geometric processing on the candidate region of interest to obtain a target region of interest.
According to the above embodiment, there is provided a region generating method of a three-dimensional model as follows, referring to steps 1 to 4 as follows.
Step 1, acquiring a three-dimensional model, and performing multi-view projection on the three-dimensional model to obtain a plurality of projection views. Wherein, as shown in fig. 2, the three-dimensional model may be a three-dimensional toothless mesh model. In one example, multiple viewing angles may be determined along the transverse and/or longitudinal axes and at preset angular intervals for the three-dimensional model; and performing multi-view projection on the three-dimensional model according to the view angles to obtain two-dimensional projection images under each view angle, wherein a plurality of projection images can be exemplified by referring to FIG. 3.
And 2, identifying a first region of interest and a background region on each projection view through a preset classification identification model. The first region of interest is such as an alveolar ridge region, and the alveolar ridge region includes: teeth, teeth preparation, implant stems, peripheral gums, and the like.
According to the embodiment, the three-dimensional model is converted into the two-dimensional projection image through projection, so that the classification recognition model can be used for recognizing the region of interest of the projection image, and the recognition difficulty is reduced. For any projection view, embodiments of identifying a first region of interest and a background region on the projection view by a classification recognition model may be referred to as follows.
Identifying a corresponding first category of each pixel on the projection view and a first probability value of the first category through a preset classification identification model; wherein the first category comprises: points of interest and background points. Specifically, a first class of a single pixel on the projection view is identified by the classification model, the first class being identified as two classes of interest points and background points, and each class corresponding to a respective first probability value. Alternatively, the classification model outputs a first probability value for each pixel of the projection view as the point of interest and a first probability value for the background point. If the first probability value of the pixel i as the interested point is identified as 0.8, the first probability value of the pixel i as the background point is identified as 0.2; in this case, a point of interest corresponding to a first class of pixel i being a higher first probability value may be determined.
Determining a region composed of a plurality of pixels with the first category as a point of interest as a first region of interest; and determining a region composed of a plurality of pixels of which the first category is background points as a background region.
In practical application, different marks such as colors, curves and the like can be used for distinguishing and marking the first interested area and the background area on the projection view, so that a user can more intuitively view the image.
It will be appreciated that the classification model in this embodiment is a pre-trained model, and the training process can be referred to as follows.
Marking the types of each triangular patch of the sample model one by one; carrying out multi-view projection on the sample model to obtain a plurality of sample images; according to the projection corresponding relation between the sample model and the sample image, determining the corresponding relation between each triangular patch in the sample model and each pixel in the sample image; assigning the types of the triangular patch marks to the pixels according to the corresponding relation, thereby obtaining sample images marked with the interesting points and the background points; training the classification recognition model to be trained by using the sample image marked with the interest points and the background points to obtain the classification recognition model which can be finally and directly used for carrying out classification recognition on the projection image.
Compared with the method of directly marking the class of the sample image, the method of marking the class of the triangular surface patch in the sample model and marking the class of the pixel in the sample image through the steps of projection, assignment and the like can improve the accuracy of marking the class of the sample image and the identification accuracy of the trained classification identification model, and can further strengthen the certainty of mapping between the two-dimensional image (namely the projection view) and the three-dimensional model, so that the classification identification model indirectly improves the accuracy of determining the target region of interest on the three-dimensional model on the basis of improving the accuracy of image identification.
Step 3, projecting the first interested region and the background region on each projection view back to the three-dimensional model to obtain candidate interested regions on the three-dimensional model; like the first region of interest, the candidate region of interest may also be an alveolar ridge region, and the alveolar ridge region includes: teeth, teeth preparation, and implant stems.
In this embodiment, for each projection view, the first class of the pixel and the first probability value thereof are projected onto a triangular patch corresponding to the pixel in the three-dimensional model, so as to obtain a target region of interest on the three-dimensional model. One specific implementation may refer to the following steps (1) to (3).
(1) Determining a triangular patch corresponding to each pixel on the three-dimensional model according to the coordinates of the pixel in each projection view; wherein one triangular patch corresponds to a pixel in at least one projection view.
Any one of these projection views will be described as an example. And finding out a triangular patch corresponding to each pixel on the three-dimensional model according to the coordinates of each pixel on the projection view and the mapping corresponding relation between the projection view and the three-dimensional model. It will be appreciated that during the mapping process, the same triangular patch may be mapped onto multiple projection views at different viewing angles, such that the same triangular patch corresponds to multiple different pixels on at least one projection view.
(2) Determining a second category of the current triangular patch according to the first category of the pixel corresponding to the current triangular patch and the first probability value thereof; wherein the current triangular patch is any one of a plurality of triangular patches.
In one embodiment, when the current triangular patch corresponds to only one pixel, the first class of the corresponding pixel and the first probability value thereof are directly assigned to the current triangular patch, and the second class of the current triangular patch is the same as the first class of the corresponding pixel.
In another embodiment, when the current triangular patch corresponds to a pixel in multiple projection views, the pixel in each projection view may be taken as a candidate pixel; accumulating the first probability values of the first categories corresponding to the candidate pixels to obtain a second probability value with the category as the interested point and a third probability value with the category as the background point; and determining the category corresponding to the larger one of the second probability value and the third probability value as the second category of the current triangular patch.
As an example, when a current triangular patch is observed by three perspectives, the current patch will correspond to three candidate pixels in three projection views at the three perspectives. In classifying pixels in a projection view, the classification of the pixels is divided into two categories, a point of interest and a background point, and a first probability value for the two first categories may be denoted as [ a i, bi ]. Where a i represents the first probability value for the point of interest for the first class of pixel i, and b i represents the first probability value for the background point for the first class of pixel i. Based on this, the first probability values of the three candidate pixels corresponding to the current triangular patch are, for example, [0.2, 0.8], [0.1, 0.9], [0.5, 0.5], respectively.
Accumulating the first probability values of the first categories corresponding to the candidate pixels to obtain second probability values of the categories as the interest points: 0.2+0.1+0.5=0.8, and a third probability value for the category background point is obtained: 0.8+0.9+0.5=2.2; and then, determining the category of the background point corresponding to the first probability value 2.2 with a larger value as the second category of the current triangular patch as the background.
According to the embodiment, aiming at the situation that a single triangular patch corresponds to a plurality of pixels, the second class of the triangular patch is comprehensively determined by fusing probability values of different classes of the plurality of pixels, so that the determination accuracy of the class of the triangular patch can be improved.
(3) And determining the regions formed by the triangular patches of which the second categories are the points of interest as candidate regions of interest.
So far, the first interested area and the background area on the projection view are projected back to the three-dimensional model, and in the process, the second category of the triangular patches is comprehensively determined by fusing probability values of a plurality of pixels belonging to the interested points or the background points, so that the second category determined for the triangular patches is more accurate, and the accuracy of the candidate interested areas is correspondingly improved.
In some approaches, candidate regions of interest may be determined for the triangular patches of interest according to the second category, and background regions may be determined for the triangular patches of background points according to the second category. Or after determining the candidate region of interest, determining all other regions of the three-dimensional model except the candidate region of interest as background regions. Or considering that the background area is not concerned in the digital design process of the dental product, in order to reduce the workload and improve the working efficiency, the embodiment can determine the candidate interested area for the triangular patches of the interested points according to the second category only, without determining the background area for the triangular patches of the background points according to the second category.
And 4, performing geometric processing on the candidate region of interest to obtain a target region of interest.
The present embodiment may include: removing noise and non-communication area small blocks in the candidate interested area according to preset filtering conditions to obtain a target interested area; wherein the filtering condition may comprise, for example, at least one of: the area of the candidate region of interest is less than a second area threshold, the volume of the bounding box corresponding to the candidate region of interest is less than a second volume threshold, and the number of triangular patches in the candidate region of interest is less than a second number threshold.
And performing geometric treatment on the candidate interested region according to the filtering condition to remove noise and non-communicated region small blocks, wherein the non-communicated region small blocks can be understood as island regions and triangular patches separated from the alveolar ridge regions, so that a target interested region with more accurate edge details is obtained. The target region of interest in this embodiment may be a part of a three-dimensional region in a three-dimensional model, as shown in fig. 4.
Compared with some deep learning schemes based on point clouds and grids, the embodiment for determining the target region of interest in the three-dimensional model is based on pixel-by-pixel identification and mapping, and is combined with a geometric processing mode, so that edge details are better processed, and edge accuracy of the target region of interest is improved.
According to the above embodiment, firstly, a three-dimensional model is projected into a plurality of two-dimensional projection views, and then a first region of interest and a background region on the projection views are identified through a classification identification model; according to the method, the two-dimensional projection view is identified by the classification identification model, and the first interested areas such as teeth, teeth preparation, implant rods, alveolar ridges and gums around the teeth, teeth preparation, implant rods, alveolar ridges and the like can be rapidly and accurately identified on the projection view. On the basis, the first interested region and the background region on each projection view are projected back to the three-dimensional model, and the process is carried out by projecting at the pixel level, and fusing the probabilities of the interested points and the background points to comprehensively determine the candidate interested regions, so that the accuracy of the candidate interested regions is improved, and missing or too many candidate interested regions can be well avoided. And then carrying out geometric treatment on the candidate region of interest to obtain a target region of interest on the three-dimensional model, wherein the geometric treatment can better treat edge details and improve the edge accuracy of the target region of interest. Therefore, compared with the previous method that the user manually brushes, lashes or cuts the plane to select the area to be reserved, the method of the embodiment can more accurately select the target area of interest of the user on the three-dimensional model.
According to the above embodiment, there may be provided another region generating method of a three-dimensional model, the method including the following steps.
Obtaining a three-dimensional model;
Carrying out multi-view projection on the three-dimensional model to obtain a plurality of projection views;
identifying a first region of interest and a background region on each projection view through a preset classification and identification model;
Projecting the first interested region and the background region on each projection view back to the three-dimensional model to obtain candidate interested regions on the three-dimensional model;
Performing geometric treatment on the candidate region of interest to obtain a target region of interest;
Fitting an dental arch line on the three-dimensional model, and connecting a target region of interest meeting preset communication conditions to a target dental arch region based on the dental arch line; wherein the communication condition includes: the area of the region is greater than a preset first area threshold, the bounding box volume is greater than a preset first volume threshold, and/or the number of triangular patches of the region is greater than a preset first number threshold.
In this embodiment, the target region of interest may be isolated connected regions, based on which an archwire may be fitted, and the target region of interest that is apparent to belong to the archwire is given in, thereby yielding a target arch region that can encompass the entire alveolar ridge region.
Wherein fitting the dental archwire may take place in existing ways, such as: extracting data according to the three-dimensional model to obtain key point cloud data; and performing morphological fitting according to the key point cloud data to obtain the dental archwire.
The implementation process of connecting a target region of interest satisfying a preset communication condition as a target dental arch region based on a dental arch wire may refer to the following embodiments.
And screening a second region of interest meeting preset communication conditions from the target region of interest. For example, a target region of interest having a region area greater than a preset first area threshold, a target region of interest having a bounding box volume corresponding to the region greater than the preset first volume threshold, and/or a target region of interest having a number of triangular patches greater than the preset first number threshold are used as the second region of interest. The second region of interest is for example each tooth region marked with a dark color in fig. 5.
The second region of interest is then connected by the archwire as the initial arch region. In specific implementation, the second region of interest is connected to a candidate dental arch region through a dental arch wire; identifying boundary features of the candidate arch region; and carrying out boundary expansion on the candidate dental arch region according to the boundary characteristics to obtain an initial dental arch region. The boundary expansion is carried out according to the boundary characteristics, so that the expanded boundary is smoother and more natural.
In fig. 5, the initial arch region and the second regions of interest are labeled with the same color, the initial arch region surrounding each of the second regions of interest. When the absence of teeth results in a discontinuity in the second region of interest, there is likewise an initial arch region for a single second region of interest. The initial arch region expands part of the space outwardly of the boundary of the second region of interest.
Then, constructing an initial dental arch region according to a preset region construction mode to obtain a target dental arch region; the region construction mode comprises the following steps: a boundary line-based construction method, a boundary line control point-based construction method, or a lowest cut plane-based construction method.
For ease of understanding, the region building approach is described by the following example.
In one embodiment, the region construction is based on the lowest cut plane; correspondingly, constructing the initial dental arch region according to a preset region construction mode, and obtaining the target dental arch region may include: traversing all triangular patches in the initial arch region to determine a lowest cut plane containing all triangular patches; and removing the area of the three-dimensional model below the lowest cutting plane to obtain a target dental arch area.
Specifically, for convenience of user's viewing, an initial arch region may be first selected in the three-dimensional model and a triangular patch in the initial arch region may be marked for display. Traversing the triangular patches to find a lowest cut plane capable of containing all the marked triangular patches; the area below the lowest cutting plane is an unnecessary area, so that the area of the three-dimensional model below the lowest cutting plane is removed, and the remaining area is used as a target dental arch area.
In one embodiment, the region construction mode is a construction mode based on a boundary line control point; correspondingly, constructing the initial dental arch region according to a preset region construction mode, and obtaining the target dental arch region may include: determining a boundary of an initial arch region in the three-dimensional model; performing spline fitting on the boundary of the initial dental arch region and calculating control points of the spline of the boundary after fitting; and adjusting the boundary spline according to the control operation of the control point to obtain the target dental arch region.
Specifically, an initial dental arch region is selected in the three-dimensional model, and triangular patches in the initial dental arch region are marked and displayed. The selected initial arch region and other unselected regions can be distinguished obviously through the marked display, so that boundaries between the initial arch region and the other unselected regions are determined in the three-dimensional model. Performing spline fitting on the boundary and calculating control points of the spline of the boundary after fitting; the user can execute control operations such as dragging on the control points, so that the boundary spline is adjusted according to the control operations on the control points, and a more accurate target dental arch region is obtained.
Target arch area as shown in fig. 6, it can be seen that the initial arch area does not enclose a region lacking teeth, and the constructed target arch area can predict and enclose a region lacking teeth, and at the same time, the target arch area enlarges the boundary range compared to the initial arch area.
The present embodiment connects all larger connected regions in a three-dimensional model together by fitting an archwire and connecting the target region of interest satisfying a preset connected condition as a target archwire based on the archwire, and smoothly expands a boundary region. According to the method, on the basis of identifying the target region of interest, the whole region of interest of the user is expanded by using a geometric method, so that applicable scenes are more various.
In practical application, the three-dimensional model marked with the target dental arch region can be manufactured by using a shell extraction model (fig. 7), a back cover model (fig. 8) and other working models, so that the dental product model shown in fig. 9 is finally obtained, and the dental product model is used for being installed in the oral cavity of a user, so that the user experiences the adaptation degree of the dental product model and the oral cavity of the user, and the three-dimensional model is adjusted in uncomfortable places.
In summary, according to the embodiment of the disclosure, the three-dimensional model is projected into a plurality of two-dimensional projection views, and then the classification recognition model is used for performing classification recognition on the projection views, so that the classification recognition model is used for recognizing the two-dimensional projection views, and the first interested areas such as teeth, teeth preparation, implant stems, alveolar ridges and the like can be rapidly and accurately recognized on the projection views. And then, the first interested region and the background region on each projection view are projected back to the three-dimensional model, and the process integrates the probability of the interested point and the background point through the projection of the pixel level so as to comprehensively determine the candidate interested region, thereby improving the accuracy of the candidate interested region and well avoiding missing or too many determined candidate interested regions. And then, carrying out geometric treatment on the candidate region of interest to obtain a target region of interest on the three-dimensional model, wherein the geometric treatment can better treat edge details, and the identification accuracy of the region of interest in the three-dimensional model is improved. On the basis, the dental archwire is fitted on the three-dimensional model, and a communicated target dental archwire area is generated on the three-dimensional model based on the dental archwire and the target area of interest, so that the dental archwire can be used for generating and manufacturing a working model, and the practicability is improved.
Fig. 10 is a block diagram of a three-dimensional model region generating device according to an embodiment of the present disclosure, where the device may be suitable for identifying regions of interest such as teeth, implant stems, standby teeth, and alveolar ridges in the mouth, and is used to implement the above-described three-dimensional model region generating method. As shown in fig. 10, the region generating apparatus of the three-dimensional model may include the following modules.
A model acquisition module 210, configured to acquire a three-dimensional model;
The region determining module 220 is configured to automatically identify and determine a target region of interest on the three-dimensional model.
In one embodiment, the region determination module 220 includes:
a candidate region determination unit for identifying and determining a candidate region of interest on the three-dimensional model based on an identification pattern;
and the geometric processing unit is used for carrying out geometric processing on the candidate region of interest to obtain a target region of interest.
In one embodiment, the candidate region determination unit includes:
The first projection subunit is used for performing multi-view projection on the three-dimensional model to obtain a plurality of projection views;
The identification subunit is used for identifying a first region of interest and a background region on each projection view through a preset classification identification model;
And the second projection subunit is used for projecting the first interested region and the background region on each projection view back to the three-dimensional model to obtain candidate interested regions on the three-dimensional model.
In one embodiment, the identification subunit is specifically configured to:
Identifying a corresponding first category of each pixel on the projection view and a first probability value of the first category through a preset classification identification model; wherein the first category comprises: interest points and background points;
Determining a plurality of areas consisting of the pixels with the first category as the interest points as first interest areas;
And determining a plurality of areas consisting of the pixels with the first category as background points as background areas.
In one embodiment, the second projection subunit is specifically configured to:
determining triangular patches corresponding to the pixels on the three-dimensional model according to the coordinates of the pixels in the projection views; wherein one of the triangular patches corresponds to the pixel in at least one of the projection views;
Determining a second category of the current triangular patch according to the first category of the pixel corresponding to the current triangular patch and the first probability value thereof; wherein the current triangular patch is any one of a plurality of triangular patches;
And determining the areas formed by the triangular patches of which the second categories are the points of interest as candidate areas of interest.
In one embodiment, the second projection subunit is specifically configured to:
When the current triangular patch corresponds to the pixels in a plurality of projection views, taking the pixels in each projection view as candidate pixels;
Accumulating the first probability values of the first categories corresponding to the candidate pixels to obtain a second probability value with the category as the interest point and a third probability value with the category as the background point;
And determining the category corresponding to the larger one of the second probability value and the third probability value as the second category of the current triangular patch.
In one embodiment, the geometry processing unit is specifically configured to:
Removing noise and non-communication area small blocks in the candidate interested area according to preset filtering conditions to obtain a target interested area; wherein the filtering condition comprises at least one of the following: the area of the candidate interested region is smaller than a second area threshold, the volume of the bounding box corresponding to the candidate interested region is smaller than a second volume threshold, and the number of triangular patches in the candidate interested region is smaller than a second number threshold.
In one embodiment, the apparatus further comprises an arch processing module for:
Fitting an dental arch line on the three-dimensional model, and connecting the target region of interest meeting preset communication conditions into a target dental arch region based on the dental arch line; wherein the communication condition includes: the area of the region is greater than a preset first area threshold, the bounding box volume is greater than a preset first volume threshold, and/or the number of triangular patches of the region is greater than a preset first number threshold.
In one embodiment, the dental arch processing module is specifically configured to:
Screening a second region of interest meeting preset communication conditions from the target region of interest;
Connecting the second region of interest to an initial arch region through the dental archwire;
constructing the initial dental arch region according to a preset region construction mode to obtain a target dental arch region; the region construction mode comprises the following steps: a boundary line-based construction method, a boundary line control point-based construction method, or a lowest cut plane-based construction method.
In one embodiment, the dental arch processing module is specifically configured to:
connecting the second region of interest to a candidate dental arch region through the dental archwire;
Identifying boundary features of the candidate arch region;
and carrying out boundary expansion on the candidate dental arch region according to the boundary characteristics to obtain an initial dental arch region.
In one embodiment, the region construction mode is the construction mode based on the lowest cutting plane; the dental arch processing module is specifically used for:
traversing all triangular patches in the initial arch region to determine a lowest cut plane containing the all triangular patches;
and removing the area of the three-dimensional model below the lowest cutting plane to obtain a target dental arch area.
In one embodiment, the area construction mode is the construction mode based on the boundary line control point; the dental arch processing module is specifically used for:
determining a boundary of the initial arch region in the three-dimensional model;
performing spline fitting on the boundary of the initial dental arch region and calculating control points of the spline of the boundary after fitting;
And adjusting the boundary spline according to the control operation of the control point to obtain a target dental arch region.
In one embodiment, the target region of interest is an alveolar ridge region, and the alveolar ridge region comprises: teeth, teeth preparation and implant stems.
The device provided in this embodiment has the same implementation principle and technical effects as those of the foregoing method embodiment, and for brevity, reference may be made to the corresponding content of the foregoing method embodiment where the device embodiment is not mentioned.
Fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the disclosure. As shown in fig. 11, the electronic device 300 includes one or more processors 301 and memory 302.
The processor 301 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities and may control other components in the electronic device 300 to perform desired functions.
Memory 302 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that may be executed by the processor 301 to implement the region generation method and/or other desired functions of the three-dimensional model of the embodiments of the present disclosure described above. Various contents such as an input signal, a signal component, a noise component, and the like may also be stored in the computer-readable storage medium.
In one example, the electronic device 300 may further include: an input device 303, and an output device 304, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).
In addition, the input device 303 may also include, for example, a keyboard, a mouse, and the like.
The output device 304 may output various information to the outside, including the determined distance information, direction information, and the like. The output device 304 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 300 relevant to the present disclosure are shown in fig. 11 for simplicity, components such as buses, input/output interfaces, and the like are omitted. In addition, the electronic device 300 may include any other suitable components depending on the particular application.
Further, the present embodiment also provides a computer-readable storage medium storing a computer program for executing the region generating method of the three-dimensional model described above.
The embodiment of the present disclosure provides a method, an apparatus, an electronic device, and a computer program product of a medium for generating a three-dimensional model, including a computer readable storage medium storing program codes, where the program codes include instructions for executing the method described in the foregoing method embodiment, and specific implementation may refer to the method embodiment and will not be described herein.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is merely a specific embodiment of the disclosure to enable one skilled in the art to understand or practice the disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown and described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (16)
1. A method for generating a region of a three-dimensional model, comprising:
Obtaining a three-dimensional model;
and automatically identifying and determining a target region of interest on the three-dimensional model.
2. The method of claim 1, wherein automatically identifying the determination target region of interest on the three-dimensional model comprises:
Identifying and determining candidate regions of interest on the three-dimensional model based on the identification pattern;
And performing geometric processing on the candidate region of interest to obtain a target region of interest.
3. The method of claim 1, wherein identifying the determined candidate region of interest on the three-dimensional model based on the identification pattern comprises:
carrying out multi-view projection on the three-dimensional model to obtain a plurality of projection views;
Identifying a first region of interest and a background region on each projection view through a preset classification and identification model;
and projecting the first region of interest and the background region on each projection view back to the three-dimensional model to obtain candidate regions of interest on the three-dimensional model.
4. A method according to claim 3, wherein the identifying the first region of interest and the background region on each of the projection views by a predetermined classification model comprises:
Identifying a corresponding first category of each pixel on the projection view and a first probability value of the first category through a preset classification identification model; wherein the first category comprises: interest points and background points;
Determining a plurality of areas consisting of the pixels with the first category as the interest points as first interest areas;
And determining a plurality of areas consisting of the pixels with the first category as background points as background areas.
5. The method of claim 4, wherein projecting the first region of interest and the background region on each of the projection views back to the three-dimensional model results in candidate regions of interest on the three-dimensional model, comprising:
determining triangular patches corresponding to the pixels on the three-dimensional model according to the coordinates of the pixels in the projection views; wherein one of the triangular patches corresponds to the pixel in at least one of the projection views;
Determining a second category of the current triangular patch according to the first category of the pixel corresponding to the current triangular patch and the first probability value thereof; wherein the current triangular patch is any one of a plurality of triangular patches;
And determining the areas formed by the triangular patches of which the second categories are the points of interest as candidate areas of interest.
6. The method of claim 5, wherein the determining the second class of the current triangular patch based on the first class of the pixel corresponding to the current triangular patch and the first probability value thereof comprises:
When the current triangular patch corresponds to the pixels in a plurality of projection views, taking the pixels in each projection view as candidate pixels;
Accumulating the first probability values of the first categories corresponding to the candidate pixels to obtain a second probability value with the category as the interest point and a third probability value with the category as the background point;
And determining the category corresponding to the larger one of the second probability value and the third probability value as the second category of the current triangular patch.
7. The method according to claim 2, wherein geometrically processing the candidate region of interest to obtain a target region of interest comprises:
Removing noise and non-communication area small blocks in the candidate interested area according to preset filtering conditions to obtain a target interested area; wherein the filtering condition comprises at least one of the following: the area of the candidate interested region is smaller than a second area threshold, the volume of the bounding box corresponding to the candidate interested region is smaller than a second volume threshold, and the number of triangular patches in the candidate interested region is smaller than a second number threshold.
8. The method according to claim 1, wherein the method further comprises:
Fitting an dental arch line on the three-dimensional model, and connecting the target region of interest meeting preset communication conditions into a target dental arch region based on the dental arch line; wherein the communication condition includes: the area of the region is greater than a preset first area threshold, the bounding box volume is greater than a preset first volume threshold, and/or the number of triangular patches of the region is greater than a preset first number threshold.
9. The method of claim 8, wherein the connecting the target region of interest satisfying a preset communication condition based on the dental archwire as a target dental arch region comprises:
Screening a second region of interest meeting preset communication conditions from the target region of interest;
Connecting the second region of interest to an initial arch region through the dental archwire;
constructing the initial dental arch region according to a preset region construction mode to obtain a target dental arch region; the region construction mode comprises the following steps: a boundary line-based construction method, a boundary line control point-based construction method, or a lowest cut plane-based construction method.
10. The method of claim 9, wherein the connecting the second region of interest by the dental archwire as an initial arch region comprises:
connecting the second region of interest to a candidate dental arch region through the dental archwire;
Identifying boundary features of the candidate arch region;
and carrying out boundary expansion on the candidate dental arch region according to the boundary characteristics to obtain an initial dental arch region.
11. The method of claim 9, wherein the region construction is the lowest cut plane-based construction; the constructing the initial dental arch region according to a preset region constructing mode to obtain a target dental arch region comprises the following steps:
traversing all triangular patches in the initial arch region to determine a lowest cut plane containing the all triangular patches;
and removing the area of the three-dimensional model below the lowest cutting plane to obtain a target dental arch area.
12. The method according to claim 9, wherein the region construction method is the boundary line control point-based construction method; the constructing the initial dental arch region according to a preset region constructing mode to obtain a target dental arch region comprises the following steps:
determining a boundary of the initial arch region in the three-dimensional model;
performing spline fitting on the boundary of the initial dental arch region and calculating control points of the spline of the boundary after fitting;
And adjusting the boundary spline according to the control operation of the control point to obtain a target dental arch region.
13. The method of any one of claims 1-12, wherein the target region of interest is an alveolar ridge region, and the alveolar ridge region comprises: teeth, teeth preparation and implant stems.
14. An area generating apparatus of a three-dimensional model, comprising:
the model acquisition module is used for acquiring a three-dimensional model;
and the region determining module is used for automatically identifying and determining the target region of interest on the three-dimensional model.
15. An electronic device, the electronic device comprising:
a processor;
a memory for storing the processor-executable instructions;
The processor being configured to read the executable instructions from the memory and execute the instructions to implement the method of any of the preceding claims 1-13.
16. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein instructions, which when run on a terminal device, cause the terminal device to implement the method according to any of claims 1-13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410538639.9A CN118135117A (en) | 2024-04-30 | 2024-04-30 | Method, device, equipment and medium for generating region of three-dimensional model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410538639.9A CN118135117A (en) | 2024-04-30 | 2024-04-30 | Method, device, equipment and medium for generating region of three-dimensional model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118135117A true CN118135117A (en) | 2024-06-04 |
Family
ID=91230388
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410538639.9A Pending CN118135117A (en) | 2024-04-30 | 2024-04-30 | Method, device, equipment and medium for generating region of three-dimensional model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118135117A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6396940B1 (en) * | 1999-05-27 | 2002-05-28 | Litton Systems, Inc. | Optical correlator based automated pathologic region of interest selector for integrated 3D ultrasound and digital mammography |
| US20100322514A1 (en) * | 2007-01-04 | 2010-12-23 | Koninklijke Philips Electronics N. V. | Apparatus, method and computer program for producing a corrected image of a region of interest from acquired projection data |
| CN109598752A (en) * | 2017-10-03 | 2019-04-09 | 佳能株式会社 | Image processing apparatus and its control method, computer readable storage medium |
| CN111461230A (en) * | 2020-04-02 | 2020-07-28 | 南京佳和牙科技术有限公司 | Dental model classification and identification method based on dental stl model and two-dimensional image registration |
| CN111932520A (en) * | 2018-08-31 | 2020-11-13 | 上海联影智能医疗科技有限公司 | Medical image display method, viewing device and computer device |
| CN117745939A (en) * | 2023-12-15 | 2024-03-22 | 先临三维科技股份有限公司 | Acquisition method, device, equipment and storage medium of three-dimensional digital model |
-
2024
- 2024-04-30 CN CN202410538639.9A patent/CN118135117A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6396940B1 (en) * | 1999-05-27 | 2002-05-28 | Litton Systems, Inc. | Optical correlator based automated pathologic region of interest selector for integrated 3D ultrasound and digital mammography |
| US20100322514A1 (en) * | 2007-01-04 | 2010-12-23 | Koninklijke Philips Electronics N. V. | Apparatus, method and computer program for producing a corrected image of a region of interest from acquired projection data |
| CN109598752A (en) * | 2017-10-03 | 2019-04-09 | 佳能株式会社 | Image processing apparatus and its control method, computer readable storage medium |
| CN111932520A (en) * | 2018-08-31 | 2020-11-13 | 上海联影智能医疗科技有限公司 | Medical image display method, viewing device and computer device |
| CN111461230A (en) * | 2020-04-02 | 2020-07-28 | 南京佳和牙科技术有限公司 | Dental model classification and identification method based on dental stl model and two-dimensional image registration |
| CN117745939A (en) * | 2023-12-15 | 2024-03-22 | 先临三维科技股份有限公司 | Acquisition method, device, equipment and storage medium of three-dimensional digital model |
Non-Patent Citations (2)
| Title |
|---|
| 张慧滔;陈明;张朋;: "一种新的针对感兴趣区域的CT扫描模式及其重建公式", 自然科学进展, no. 11, 15 November 2007 (2007-11-15), pages 131 - 136 * |
| 章新友: "医学成像及处理技术", 31 July 2011, 中国铁道出版社, pages: 158 - 159 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110163198B (en) | Table identification reconstruction method and device and storage medium | |
| US7889193B2 (en) | Method of and system for determining a data model designed for being superposed with an image of a real object in an object tracking process | |
| US20100315410A1 (en) | System and method for recovering three-dimensional particle systems from two-dimensional images | |
| KR20210068077A (en) | 3D model creation method, device, device and storage medium | |
| JP2000187731A (en) | Picture feature extraction method and recording medium which records program for making computer execute respective processes of the same and which computer can read | |
| CN112515787B (en) | Three-dimensional dental data analysis method | |
| WO2014070145A1 (en) | Object segmentation | |
| CN107272899B (en) | VR (virtual reality) interaction method and device based on dynamic gestures and electronic equipment | |
| EP4224419A1 (en) | Scanning result processing method and apparatus, and processor and scanning system | |
| KR20240032962A (en) | Three-dimensional dynamic tracking method, device, electronic device, and storage medium | |
| CN110084885A (en) | A kind of cloud and image optimization method, device, equipment and storage medium | |
| JP3647407B2 (en) | Image processing device | |
| CN110007764B (en) | Gesture skeleton recognition method, device and system and storage medium | |
| JP7362265B2 (en) | Information processing device, information processing method and program | |
| CN117745939A (en) | Acquisition method, device, equipment and storage medium of three-dimensional digital model | |
| CN110648336A (en) | Method and device for dividing tongue texture and tongue coating | |
| CN118135117A (en) | Method, device, equipment and medium for generating region of three-dimensional model | |
| CN111860515A (en) | Interactive intelligent 2D semantic segmentation system, method, storage medium and device | |
| EP3018626A1 (en) | Apparatus and method for image segmentation | |
| US20230071291A1 (en) | System and method for a precise semantic segmentation | |
| JP2005353101A (en) | Image-feature extraction method and computer-readable recording medium stored with program for executing on computer each step of the method | |
| CN112364693B (en) | Binocular vision-based obstacle recognition method, device, equipment and storage medium | |
| CN114708237A (en) | Detection algorithm for hair health condition | |
| CN110276290B (en) | Quick face model acquisition method and quick face model acquisition device based on TOF module | |
| CN106127679B (en) | Fingerprint image conversion 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 |