An automatic segmentation method and system based on image

Abstract

The present application relates to an image-based automated segmentation method and system, the method comprising extracting a tooth feature model in a model image; creating a first contour line based on a sensing area and sequentially creating multiple second contour lines in a direction away from the first contour line; creating a second base surface reference contour line in a non-tooth feature model area on the model image; driving the second base surface reference contour line close to the first base surface reference contour line until the second base surface reference contour line and the first base surface reference contour line meet the requirements of coincidence; fusing the first base surface reference contour line and the second base surface reference contour line to obtain a base surface segmentation reference line. The image-based automated segmentation method and system disclosed in the present application uses an equal value division method to perform preliminary division, and uses edge fusion and local adjustment methods to obtain refined extracted edges to achieve accurate segmentation of teeth in the model image.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an image-based automatic segmentation method and system. Background Art

[0002] Automated processing of dental images can assist doctors in obtaining accurate tooth models during the orthodontic process. In the initialization stage of image processing, manual feature methods are generally used for processing, such as level sets, graph cuts, or template fitting. However, these low-level descriptors / features are sensitive to the complex appearance of the image and therefore require tedious manual intervention to initialize or correct.

[0003] There are also automated processing methods for identification, but they are limited to relatively healthy teeth. The main reason is that when factors such as inflammation appear around the teeth (periodontal area), it will cause local color changes. These small areas will cause errors in the fine processing of the edges. How to solve this problem requires further research. Summary of the invention

[0004] The present application provides an image-based automated segmentation method and system, which uses equal value division to perform preliminary segmentation, and uses edge fusion and local adjustment to obtain refined extracted edges, so as to achieve accurate segmentation of teeth in model images.

[0005] The above-mentioned purpose of the present application is achieved through the following technical solutions:

[0006] In a first aspect, the present application provides an image-based automated segmentation method, comprising:

[0007] In response to the acquired model image, extracting a tooth feature model in the model image;

[0008] determining a single tooth feature model in the tooth feature model;

[0009] Selecting at least one feature point on a single tooth feature model and establishing a sensing area belonging to the feature point based on the feature point;

[0010] Creating a first contour line based on the perception area and sequentially creating a plurality of second contour lines in a direction away from the first contour line;

[0011] Determine the last second contour line on the sequential sequence and use the last second contour line on the sequential sequence as the first base surface reference contour line;

[0012] Creating a second base surface reference contour line in a non-tooth feature model area on the model image;

[0013] Drive the second base surface reference contour line close to the first base surface reference contour line until the degree of coincidence between the second base surface reference contour line and the first base surface reference contour line meets the requirement;

[0014] The first base surface reference contour line and the second base surface reference contour line are merged to obtain the base surface segmentation reference line.

[0015] In a possible implementation manner of the first aspect, extracting the tooth feature model in the model image includes:

[0016] Perform grayscale processing on the model image to obtain a grayscale model image;

[0017] Randomly select multiple regions on the grayscale model image and calculate the mean value of the pixels in the region;

[0018] The regions are classified using the pixel mean to obtain first-category regions and second-category regions, wherein the first-category regions include tooth feature models and the second-category regions include non-tooth feature models;

[0019] Expand the edge of the first type of area to obtain the extraction area;

[0020] Use the extraction region to extract the tooth feature model from the model image.

[0021] In a possible implementation manner of the first aspect, classifying the regions by using the pixel mean value includes:

[0022] Connect the areas with the same or similar pixel mean values ​​to obtain a local grid;

[0023] Determine the boundaries of the local grid;

[0024] Determine the boundary between the first type of area and the second type of area on the boundary of the local grid;

[0025] Increase the density of the local grid at the junction to obtain the junction of the first type of area and the second type of area;

[0026] The regions are classified according to the boundary between the first and second category regions.

[0027] In a possible implementation manner of the first aspect, when expanding the edge of the first type of area, only a portion of the edge of the first type of area adjacent to the edge of the second type of area is expanded;

[0028] Driving the edge of the first type of area to move toward the direction close to the second type of area includes:

[0029] Cutting the edge of the first type of area to obtain multiple edge segments;

[0030] determining a midpoint of each edge segment and creating a movement direction based on the edge segment and the midpoint of the edge segment;

[0031] The edge segment is moved in the moving direction, and the blank area of ​​the moved edge segment is filled and the redundant part of the moved edge segment is removed.

[0032] In a possible implementation manner of the first aspect, creating a second base surface reference isoline in a non-tooth feature model area on the model image includes:

[0033] A second base surface reference contour line is randomly created in the non-tooth feature model area on the model image;

[0034] driving the second base surface reference contour line to move toward the direction close to the first base surface reference contour line and recording the crossing segment on the second base surface reference contour line;

[0035] Adjust the crossing section according to the surrounding environment of the crossing section;

[0036] The second base surface reference contour line is repeatedly driven to move toward the direction close to the first base surface reference contour line and the crossing segment is repeatedly adjusted until the length of the crossing segment is less than or equal to the set length.

[0037] In a possible implementation manner of the first aspect, when the crossed segment is adjusted according to the surrounding environment of the crossed segment, the values ​​of the pixel points on the crossed segment are increased or decreased as a whole;

[0038] The overall increase or decrease of the pixel value includes value change and interval change.

[0039] In a possible implementation manner of the first aspect, when the first base surface reference contour line and the second base surface reference contour line are merged, the method includes:

[0040] Obtaining a non-overlapping area between the first base surface reference contour line and the second base surface reference contour line;

[0041] Create multiple moving reference lines in non-overlapping areas, and set adjacent moving reference lines in parallel.

[0042] Determine the demarcation points on the moving reference line;

[0043] The corresponding non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line is moved according to the dividing point, so that the non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line falls on the corresponding dividing point or in the area surrounded by the dividing point.

[0044] In a second aspect, the present application provides an image-based automatic segmentation device, comprising:

[0045] A model extraction unit, configured to extract a tooth feature model in the model image in response to the acquired model image;

[0046] A model determination unit, used for determining a single tooth feature model in the tooth feature model;

[0047] A perception region establishing unit, used for selecting at least one feature point on a single tooth feature model and establishing a perception region belonging to the feature point based on the feature point;

[0048] A first contour line establishing unit, used to create a first contour line based on the sensing area and sequentially create a plurality of second contour lines in a direction away from the first contour line;

[0049] A contour line selection unit, used for determining the last second contour line in the sequential sequence and using the last second contour line in the sequential sequence as the first base surface reference contour line;

[0050] A second contour line establishing unit is used to create a second base surface reference contour line in a non-tooth feature model area on the model image;

[0051] A contour processing unit, used for driving the second base surface reference contour line to approach the first base surface reference contour line until the coincidence degree between the second base surface reference contour line and the first base surface reference contour line meets the requirement;

[0052] The contour line fusion unit is used to fuse the first base surface reference contour line and the second base surface reference contour line to obtain the base surface segmentation reference line.

[0053] In a third aspect, the present application provides an image-based automated segmentation system, the system comprising:

[0054] one or more memories for storing instructions;

[0055] One or more processors, used to call and run the instructions from the memory to execute the method as described in the first aspect and any possible implementation of the first aspect.

[0056] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium comprises:

[0057] Program, when the program is executed by a processor, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0058] In a fifth aspect, the present application provides a computer program product, comprising program instructions. When the program instructions are executed by a computing device, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0059] In a sixth aspect, the present application provides a chip system, which includes a processor for implementing the functions involved in the above aspects, for example, generating, receiving, sending, or processing the data and / or information involved in the above methods.

[0060] The chip system may be composed of chips, or may include chips and other discrete devices.

[0061] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wired or wireless means, or the processor and the memory can also be coupled on the same device. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of an image-based automated segmentation method provided in this application.

[0063] Figure 2 This is a schematic diagram of the location of a feature point provided by this application.

[0064] Figure 3 This is a schematic diagram for creating a first contour line and a second contour line provided by the present application.

[0065] Figure 4 This is a schematic diagram provided by the present application of using a connection method to form a grid of regions, in which the arrows indicate the extension direction of the local grid.

[0066] Figure 5 This is a schematic diagram of expanding the edge of the first type of area provided by the present application, and the arrows in the figure indicate the moving direction of the edge segment.

[0067] Figure 6 It is a schematic diagram of a crossing segment provided by the present application.

[0068] Figure 7 This is a schematic diagram of adjusting the crossing segment provided by the present application.

[0069] Figure 8 It is a schematic diagram of a moving reference line and a dividing point provided in the present application.

[0070] Fig. 9 This is a schematic diagram provided by the present application for adjusting the edge of a non-overlapping area formed by a first base surface reference contour line and a second base surface reference contour line. The arrows in the figure indicate the adjustment directions of the first base surface reference contour line and the second base surface reference contour line. DETAILED DESCRIPTION

[0071] The technical solution in this application is further described in detail below in conjunction with the accompanying drawings.

[0072] This application discloses an automatic segmentation method based on images, please refer to Figure 1 In some examples, the present application discloses an image-based automated segmentation method including the following:

[0073] S101, in response to the acquired model image, extracting a tooth feature model in the model image;

[0074] S102, determining a single tooth feature model in the tooth feature model;

[0075] S103, selecting at least one feature point on a single tooth feature model and establishing a perception area belonging to the feature point based on the feature point;

[0076] S104, creating a first contour line based on the sensing area and sequentially creating a plurality of second contour lines in a direction away from the first contour line;

[0077] S105, determining the last second contour line in the sequential sequence and using the last second contour line in the sequential sequence as the first base surface reference contour line;

[0078] S106, creating a second base surface reference contour line in a non-tooth feature model area on the model image;

[0079] S107, driving the second base surface reference contour line close to the first base surface reference contour line until the degree of overlap between the second base surface reference contour line and the first base surface reference contour line meets the requirement;

[0080] S108, merging the first base surface reference contour line and the second base surface reference contour line to obtain a base surface segmentation reference line.

[0081] In general, in S101, after obtaining the model image, it is first necessary to extract the tooth feature model in the model image. The specific extraction method is to first use color to distinguish teeth and non-teeth, because the color of teeth and other tissues in the mouth are obvious. The specific technical details will be further introduced in the subsequent content.

[0082] Then, in S102, a single tooth feature model in the tooth feature model is determined. The tooth feature model is determined by edge recognition. Each tooth in the model image has an edge. The edges here can be divided into three categories: upper edge, side edge and bottom edge. There is no adjacent tissue around the upper edge, the side edge is surrounded by adjacent teeth, and the bottom edge is surrounded by adjacent soft tissue.

[0083] In S103, at least one feature point is selected on a single tooth feature model and a perception area belonging to the feature point is established based on the feature point.

[0084] Generally speaking, feature points are assigned manually. The advantage of manual assignment is that it is fast. This is because the manual recognition method can identify a single tooth feature model relatively quickly and accurately, and at the same time select appropriate feature points on the single tooth feature model.

[0085] In some possible implementations, the feature point is generally selected at the middle position of a single tooth feature model, such as Figure 2 This is because it is easy to establish a perception area when choosing this place. Figure 2 The square shapes on the middle teeth are brackets that are bonded to the teeth during the orthodontic process.

[0086] The perception area attributable to a feature point refers to an area existing around the feature point, in which the values ​​of the pixels in the area are the same as or close to the values ​​of the feature point. The closeness here means that the values ​​of the pixels in the area and the values ​​of the feature point are in the same area.

[0087] In S104, see Figure 3 , a first contour line will be created based on the perception area and multiple second contour lines will be created sequentially in the direction away from the first contour line. The first contour line is a part of the edge of the perception area. The reason why a part of the edge of the perception area is used here is because the segmentation involves the edge of the junction of the tooth and the soft tissue.

[0088] When creating multiple second contour lines, the reference direction is the direction of the first contour line, that is, the direction close to the junction of the tooth and soft tissue.

[0089] At the same time, it is also required that the edge of the first contour line and the edge of the second contour line both fall on the edge of the tooth.

[0090] In S105, the last second contour line in the sequential sequence is determined and used as the first base surface reference contour line. The reason why the second contour line cannot be created wirelessly is that there is an obvious difference in color between the teeth and the soft tissue in the oral cavity, and the second contour line cannot be created on the soft tissue in the oral cavity.

[0091] The last second contour line obtained at this time is a reference segmentation line between the tooth and the adjacent soft tissue.

[0092] In S106, a second base surface reference contour line is created in the non-tooth feature model area on the model image, and then in S107, the second base surface reference contour line is driven close to the first base surface reference contour line until the second base surface reference contour line and the first base surface reference contour line meet the requirements of overlap.

[0093] In S106 and S107, the reference segmentation line created based on teeth and the reference segmentation line created based on oral soft tissue (second base surface reference contour line) are used respectively to obtain the base surface segmentation reference line. The specific processing method of the two reference lines is performed in S108.

[0094] In S108, the first base surface reference contour line and the second base surface reference contour line are fused to obtain the base surface segmentation reference line. The purpose of the fusion process is mainly to take into account that there are certain errors in the first base surface reference contour line and the second base surface reference contour line during the processing. The errors may cause part of the first base surface reference contour line and part of the second base surface reference contour line to be located in the wrong area. For example, part of the first base surface reference contour line and part of the second base surface reference contour line may not be close to the actual edge of the tooth and the soft tissue and may cross the actual edge.

[0095] These errors can be reduced or eliminated using fusion processing.

[0096] In some examples, the specific method of extracting the tooth feature model in the model image is as follows:

[0097] S201, grayscale processing is performed on the model image to obtain a grayscale model image;

[0098] S202, randomly selecting multiple regions on the grayscale model image and calculating the mean value of the pixels in the regions;

[0099] S203, using the pixel mean to classify the region, to obtain a first category of regions and a second category of regions, the first category of regions including tooth feature models, and the second category of regions including non-tooth feature models;

[0100] S204, expanding the edge of the first type of area to obtain an extraction area;

[0101] S205, using the extraction region to extract the tooth feature model in the model image.

[0102] In S201 to S205, the model image is firstly gray-scaled to obtain a gray-scale model image, and then a plurality of regions are randomly selected on the gray-scale model image and the pixel means of the regions are calculated. The plurality of regions selected here may be located in the tooth region or in the non-tooth region.

[0103] Then, the pixel mean is used to classify the regions, and the first and second regions are obtained. The first region includes the tooth feature model, and the second region includes the non-tooth feature model. At this point, the edge of the tooth can be preliminarily divided. Figure 4As can be seen in the figure, regions can be connected to form a grid, which can be used to quickly calibrate the range on the model image.

[0104] The reason why the regions can be classified is that the mean values ​​of pixels in the first and second types of regions are significantly different.

[0105] Combined with expanding the edge of the first type of area, the extraction area is obtained. The purpose of expanding the edge of the first type of area is to make part of the second type of area exist in the first type of area, so that the edge of the tooth can be completely retained and the problem of tooth edge loss caused by direct division can be avoided.

[0106] Finally, the tooth feature model in the model image is extracted using the extraction area.

[0107] The specific method of using pixel mean to classify regions is as follows:

[0108] Connect the areas with the same or similar pixel mean values ​​to obtain a local grid;

[0109] Determine the boundaries of the local grid;

[0110] Determine the boundary between the first type of area and the second type of area on the boundary of the local grid;

[0111] Increase the density of the local grid at the junction to obtain the junction of the first type of area and the second type of area;

[0112] The regions are classified according to the boundary between the first and second category regions.

[0113] See also Figure 4 In the above method, firstly, the areas with the same or similar pixel mean values ​​are connected to obtain local grids. At this time, the number of local grids is two, and the boundaries of the two local grids may partially overlap or not overlap at all.

[0114] Then, the boundary between the first and second types of areas is determined on the boundary of the local grid. The boundary refers to the potential overlap area between the first and second types of areas. At this time, the two local grids need to be extended towards each other at the same time (refer to Figure 4 ), then we will get the boundary between the first and second types of regions, where the boundary refers to one region.

[0115] Then, by increasing the density of the local grid at the junction, the junction of the first type of area and the second type of area will be obtained. The purpose of increasing the density is to improve the accuracy of the junction. After determining the junction, the area is classified according to the junction of the first type of area and the second type of area.

[0116] In some possible implementations, when the edge of the first type of area is expanded, only the portion of the edge of the first type of area adjacent to the edge of the second type of area is expanded, and the expansion is implemented by moving, and the specific process is as follows:

[0117] Cutting the edge of the first type of area to obtain multiple edge segments;

[0118] determining a midpoint of each edge segment and creating a movement direction based on the edge segment and the midpoint of the edge segment;

[0119] The edge segment is moved in the moving direction, and the blank area of ​​the moved edge segment is filled and the redundant part of the moved edge segment is removed.

[0120] In the above-mentioned moving process, the edge of the first type of area is segmented, and then each edge segment is moved, and finally the blank area of ​​the moved edge segment is filled and the redundant part of the moved edge segment is removed.

[0121] See also Figure 5 , the implementation method of segmentation (cutting) is to examine the straightness of the edge of the first type of area. The specific implementation method is to set two parallel straight lines ( Figure 5 The two parallel line segments are then applied to the edge of the first type of area and rotated so that the portion of the edge of the first type of area between the two parallel line segments is as long as possible.

[0122] Creating a moving direction based on an edge segment and a midpoint of the edge segment refers to creating a line segment on the midpoint of the edge segment, and the angle between the line segment and the edge segment is as close to 90 degrees as possible.

[0123] In some examples, the specific method of creating the second base surface reference contour line in the non-tooth feature model area on the model image is:

[0124] S301, randomly creating a second base surface reference contour line in a non-tooth feature model area on the model image;

[0125] S302, driving the second base surface reference contour line to move in a direction close to the first base surface reference contour line and recording the crossing segment on the second base surface reference contour line;

[0126] S303, adjusting the crossing segment according to the surrounding environment of the crossing segment;

[0127] S304, repeatedly driving the second base surface reference contour line to move toward the first base surface reference contour line and repeatedly adjusting the crossing segment until the length of the crossing segment is less than or equal to the set length.

[0128] In S301 to S304, a second base surface reference contour line is first randomly created, and then the second base surface reference contour line is driven to move in a direction close to the first base surface reference contour line and the crossing segment on the second base surface reference contour line is recorded. Figure 6 As shown, at this time, at least one crossing segment will be generated on the second base surface reference contour line, and the crossing segment refers to the portion of the second base surface reference contour line that passes through the first base surface reference contour line.

[0129] Then, the crossing section is adjusted according to the surrounding environment of the crossing section until the length of the crossing section is less than or equal to the set length. In this way, the shape of the second base surface reference contour line can be more consistent with the actual shape, but at this time, the shapes of the first base surface reference contour line and the second base surface reference contour line are not exactly the same.

[0130] See also Figure 6 and Figure 7 In some possible implementations, when the crossing segment is adjusted according to the surrounding environment of the crossing segment, the values ​​of the pixels on the crossing segment are increased or decreased as a whole. That is, when the crossing segment is adjusted, the values ​​of the pixels on the crossing segment are allowed to change. This is because the color change will appear locally near the root of the tooth on the soft tissue.

[0131] The overall increase or decrease of the pixel value includes value change and interval change. The value change refers to that the change value of the pixel is a fixed value, and the interval change refers to that the change value of the pixel is selected within a given curve.

[0132] Obviously, the interval change method is more advantageous because it can minimize the total amount of change in the crossing segment. Of course, restrictions need to be imposed at this time. The specific method of restriction is to assign the change value of the pixel point in proportion to the distance between the crossing segment and the adjacent position on the first base surface reference contour line.

[0133] That is, the value change of the pixel point is directly proportional to the distance between the pixel point and the corresponding position on the first base surface reference contour line.

[0134] In some examples, when the first base surface reference contour line and the second base surface reference contour line are merged, the following processing is used:

[0135] S401, obtaining a non-overlapping area of ​​a first base surface reference contour line and a second base surface reference contour line;

[0136] S402, creating moving reference lines in the non-overlapping area, where the number of moving reference lines is multiple and adjacent moving reference lines are arranged in parallel;

[0137] S403, determining a dividing point on the moving reference line;

[0138] S404, moving the corresponding non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line according to the dividing point, so that the non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line falls on the corresponding dividing point or the area surrounded by the dividing point.

[0139] When moving the first base surface reference contour line and the second base surface reference contour line, first move the second base surface reference contour line and stop after a contact point appears with the first base surface reference contour line, and then move the first base surface reference contour line and the second base surface reference contour line at the same time. At this time, the first base surface reference contour line and the second base surface reference contour line move towards each other.

[0140] During the movement, non-overlapping areas will appear on the first base surface reference contour line and the second base surface reference contour line. For the investigation of the non-overlapping areas, the area of ​​the non-overlapping area located on the left side (one side) of the first base surface reference contour line is given a positive value, and the area of ​​the non-overlapping area located on the right side (the other side) of the first base surface reference contour line is given a negative value.

[0141] The absolute value of the cumulative value of the area of ​​the non-overlapping region needs to be minimized.

[0142] See also Figure 8 and Fig. 9 Then create moving reference lines in the non-overlapping area. The requirements for moving reference lines are: there are multiple moving reference lines and adjacent moving reference lines are set in parallel. Then determine the dividing point on the moving reference line. The dividing point here is obtained by using the numerical values ​​of the pixel points on the moving reference line to form a sequence, and then calculating the quadratic difference sequence of this sequence.

[0143] The position corresponding to the maximum value in the quadratic difference series is the dividing point on the moving reference line.

[0144] Finally, the corresponding non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line is moved according to the dividing point, so that the non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line falls on the corresponding dividing point or the area surrounded by the dividing point.

[0145] After the non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line falls on the corresponding demarcation point or the area surrounded by the demarcation point, one of the edges of the non-overlapping area (a part of the first base surface reference contour line and a part of the second base surface reference contour line) is selected and retained.

[0146] The present application also provides an image-based automatic segmentation device, comprising:

[0147] A model extraction unit, configured to extract a tooth feature model in the model image in response to the acquired model image;

[0148] A model determination unit, used for determining a single tooth feature model in the tooth feature model;

[0149] A perception region establishing unit, used for selecting at least one feature point on a single tooth feature model and establishing a perception region belonging to the feature point based on the feature point;

[0150] A first contour line establishing unit, used to create a first contour line based on the sensing area and sequentially create a plurality of second contour lines in a direction away from the first contour line;

[0151] A contour line selection unit, used for determining the last second contour line in the sequential sequence and using the last second contour line in the sequential sequence as the first base surface reference contour line;

[0152] A second contour line establishing unit is used to create a second base surface reference contour line in a non-tooth feature model area on the model image;

[0153] A contour processing unit, used for driving the second base surface reference contour line to approach the first base surface reference contour line until the coincidence degree between the second base surface reference contour line and the first base surface reference contour line meets the requirement;

[0154] The contour line fusion unit is used to fuse the first base surface reference contour line and the second base surface reference contour line to obtain the base surface segmentation reference line.

[0155] Furthermore, extracting the tooth feature model in the model image includes:

[0156] Perform grayscale processing on the model image to obtain a grayscale model image;

[0157] Randomly select multiple regions on the grayscale model image and calculate the mean value of the pixels in the region;

[0158] The regions are classified using the pixel mean to obtain first-category regions and second-category regions, wherein the first-category regions include tooth feature models and the second-category regions include non-tooth feature models;

[0159] Expand the edge of the first type of area to obtain the extraction area;

[0160] Use the extraction region to extract the tooth feature model from the model image.

[0161] Furthermore, using the pixel mean to classify the region includes:

[0162] Connect the areas with the same or similar pixel mean values ​​to obtain a local grid;

[0163] Determine the boundaries of the local grid;

[0164] Determine the boundary between the first type of area and the second type of area on the boundary of the local grid;

[0165] Increase the density of the local grid at the junction to obtain the junction of the first type of area and the second type of area;

[0166] The regions are classified according to the boundary between the first and second category regions.

[0167] Furthermore, when expanding the edge of the first type of region, only the portion of the edge of the first type of region adjacent to the edge of the second type of region is expanded;

[0168] Driving the edge of the first type of area to move toward the direction close to the second type of area includes:

[0169] Cutting the edge of the first type of area to obtain multiple edge segments;

[0170] determining a midpoint of each edge segment and creating a movement direction based on the edge segment and the midpoint of the edge segment;

[0171] The edge segment is moved in the moving direction, and the blank area of ​​the moved edge segment is filled and the redundant part of the moved edge segment is removed.

[0172] Further, creating a second base surface reference contour line in the non-tooth feature model area on the model image includes:

[0173] A second base surface reference contour line is randomly created in the non-tooth feature model area on the model image;

[0174] driving the second base surface reference contour line to move toward the direction close to the first base surface reference contour line and recording the crossing segment on the second base surface reference contour line;

[0175] Adjust the crossing section according to the surrounding environment of the crossing section;

[0176] The second base surface reference contour line is repeatedly driven to move toward the direction close to the first base surface reference contour line and the crossing segment is repeatedly adjusted until the length of the crossing segment is less than or equal to the set length.

[0177] Furthermore, when the crossing segment is adjusted according to the surrounding environment of the crossing segment, the values ​​of the pixels on the crossing segment are increased or decreased as a whole;

[0178] The overall increase or decrease of the pixel value includes value change and interval change.

[0179] Furthermore, when the first base surface reference contour line and the second base surface reference contour line are merged, it includes:

[0180] Obtaining a non-overlapping area between the first base surface reference contour line and the second base surface reference contour line;

[0181] Create multiple moving reference lines in non-overlapping areas, and set adjacent moving reference lines in parallel.

[0182] Determine the demarcation points on the moving reference line;

[0183] The corresponding non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line is moved according to the dividing point, so that the non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line falls on the corresponding dividing point or in the area surrounded by the dividing point.

[0184] In one example, the unit in any of the above devices can be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASICs), or, one or more digital signal processors (DSPs), or, one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0185] For another example, when the units in the device can be implemented in the form of a processing element scheduling program, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0186] Various objects such as various messages / information / equipment / network elements / systems / devices / actions / operations / processes / concepts that may appear in this application are named. It can be understood that these specific names do not constitute a limitation on the relevant objects. The names assigned may change with factors such as scenarios, contexts or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from the functions and technical effects embodied / executed in the technical scheme.

[0187] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0188] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0189] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0190] Those of ordinary skill in the art will appreciate that the units and algorithms of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0191] It should also be understood that in various embodiments of the present application, the first, second, etc. are only used to indicate that multiple objects are different. For example, the first time window and the second time window are only used to indicate different time windows. They should not have any impact on the time window itself, and the first, second, etc. mentioned above should not impose any limitations on the embodiments of the present application.

[0192] It should also be understood that in the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other, and the technical features in different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0193] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a computer-readable storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned computer-readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0194] The present application also provides an image-based automated segmentation system, the system comprising:

[0195] one or more memories for storing instructions;

[0196] One or more processors are used to call and run the instructions from the memory to execute the method as described above.

[0197] The present application also provides a computer program product, which includes instructions. When the instructions are executed, the terminal device and the network device perform operations of the terminal device and the network device corresponding to the above method.

[0198] The present application also provides a chip system, which includes a processor for implementing the functions involved in the above content, such as generating, receiving, sending, or processing the data and / or information involved in the above method.

[0199] The chip system may be composed of chips, or may include chips and other discrete devices.

[0200] The processor mentioned in any of the above places can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for executing programs for controlling the above-mentioned feedback information transmission method.

[0201] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wire or wireless means to support the chip system to implement various functions in the above embodiments. Alternatively, the processor and the memory can also be coupled on the same device.

[0202] Optionally, the computer instructions are stored in a memory.

[0203] Optionally, the memory is a storage unit within the chip, such as a register, a cache, etc. The memory can also be a storage unit within the terminal located outside the chip, such as a ROM or other types of static storage devices that can store static information and instructions, RAM, etc.

[0204] It can be understood that the memory in the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0205] The non-volatile memory may be a ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory.

[0206] The volatile memory may be a RAM, which is used as an external cache. There are many different types of RAM, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct memory bus RAM.

[0207] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. An image-based automated segmentation method, characterized in that: include: In response to the acquired model image, extracting a tooth feature model in the model image; determining a single tooth feature model in the tooth feature model; Selecting at least one feature point on a single tooth feature model and establishing a sensing area belonging to the feature point based on the feature point; Creating a first contour line based on the perception area and sequentially creating a plurality of second contour lines in a direction away from the first contour line; Determine the last second contour line on the sequential sequence and use the last second contour line on the sequential sequence as the first base surface reference contour line; Creating a second base surface reference contour line in a non-tooth feature model area on the model image; Drive the second base surface reference contour line close to the first base surface reference contour line until the degree of coincidence between the second base surface reference contour line and the first base surface reference contour line meets the requirement; The first base surface reference contour line and the second base surface reference contour line are merged to obtain a base surface segmentation reference line; Creating the second base surface reference contour line in the non-tooth feature model area on the model image includes: A second base surface reference contour line is randomly created in the non-tooth feature model area on the model image; driving the second base surface reference contour line to move toward the direction close to the first base surface reference contour line and recording the crossing segment on the second base surface reference contour line; Adjust the crossing section according to the surrounding environment of the crossing section; Repeatedly drive the second base surface reference contour line to move in a direction close to the first base surface reference contour line and repeatedly adjust the crossing segment until the length of the crossing segment is less than or equal to the set length; When the first base surface reference contour line and the second base surface reference contour line are merged, it includes: Obtaining a non-overlapping area between the first base surface reference contour line and the second base surface reference contour line; Create multiple moving reference lines in non-overlapping areas, and set adjacent moving reference lines in parallel. Determine the demarcation points on the moving reference line; The corresponding non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line is moved according to the dividing point, so that the non-overlapping area of ​​the first base surface reference contour line and the second base surface reference contour line falls on the corresponding dividing point or in the area surrounded by the dividing point.

2. The image-based automated segmentation method according to claim 1, characterized in that: The tooth feature models extracted from the model image include: Perform grayscale processing on the model image to obtain a grayscale model image; Randomly select multiple regions on the grayscale model image and calculate the mean value of the pixels in the region; The regions are classified using the pixel mean to obtain first-category regions and second-category regions, wherein the first-category regions include tooth feature models and the second-category regions include non-tooth feature models; Expand the edge of the first type of area to obtain the extraction area; Use the extraction region to extract the tooth feature model from the model image.

3. The image-based automated segmentation method according to claim 2, characterized in that: Using pixel mean to classify regions includes: Connect the areas with the same or similar pixel mean values ​​to obtain a local grid; Determine the boundaries of the local grid; Determine the boundary between the first type of area and the second type of area on the boundary of the local grid; Increase the density of the local grid at the junction to obtain the junction of the first type of area and the second type of area; The regions are classified according to the boundary between the first and second category regions.

4. The image-based automated segmentation method according to claim 2, characterized in that: When expanding the edge of the first type of area, only the portion of the edge of the first type of area adjacent to the edge of the second type of area is expanded; Driving the edge of the first type of area to move toward the direction close to the second type of area includes: Cutting the edge of the first type of area to obtain multiple edge segments; determining a midpoint of each edge segment and creating a movement direction based on the edge segment and the midpoint of the edge segment; The edge segment is moved in the moving direction, and the blank area of ​​the moved edge segment is filled and the redundant part of the moved edge segment is removed.

5. The image-based automated segmentation method according to claim 1, characterized in that: When the crossing segment is adjusted according to the surrounding environment of the crossing segment, the values ​​of the pixels on the crossing segment are increased or decreased as a whole; The overall increase or decrease of the pixel value includes value change and interval change.

6. An image-based automated segmentation system, characterized in that: The system comprises: one or more memories for storing instructions; One or more processors, configured to call and execute the instructions from the memory to perform the method according to any one of claims 1 to 5.

Images