A method and system for collecting images of oral digital impression device

Abstract

The present invention discloses an image acquisition method and system for a digital oral impression device, comprising: downsampling an oral image acquired by a digital oral impression device, weighting the frequency of gray values ​​in each downsampled image according to the size ratio of each downsampled image to the oral image, performing edge detection on the oral image at gray levels of different gray scales, calculating the protrusion of the pixel point according to the gradient amplitude of the pixel point at different gray scales and the length of the edge to which it belongs, and correcting the weighted frequency of the gray value, obtaining a mapping interval of the gray value according to the corrected frequency, determining the mapping gray value of the pixel point from the mapping interval of the gray value of the pixel point according to the protrusion of the pixel point and the gray values ​​of other pixels in its neighborhood, and composing the mapping gray values ​​of all pixels in the oral image into an enhanced oral image, so as to reconstruct 3D information of teeth. The present invention improves the accuracy and authenticity of reconstructing 3D information.

Description

Technical Field

[0001] The present invention relates to the field of image acquisition technology, and more specifically, to a method and system for acquiring images of an oral digital impression device. Background Art

[0002] The oral digital impression device uses laser scanning, optical scanning and other technologies to obtain three-dimensional data of the oral structure, and converts traditional dental plaster models, silicone rubber impressions or intraoral dentition morphology into a three-dimensional visual digital model in the computer. The oral digital impression device has strong accuracy and operability and is an important tool in modern oral restoration and implantology.

[0003] The oral digital impression device reconstructs 3D information of teeth by collecting multiple 2D images of the oral cavity. In this process, the contrast of the collected 2D images is poor due to the influence of the oral environment, resulting in the loss of tooth surface details in the 2D images, especially in subtle structures such as tooth edges and textures. It is difficult to accurately capture data points on the tooth surface when reconstructing 3D information, which seriously affects the accuracy and authenticity of the reconstructed 3D information. Summary of the Invention

[0004] To solve the technical problem that when reconstructing 3D information of teeth by collecting multiple 2D images of the oral cavity, the poor contrast of the 2D images seriously affects the accuracy and authenticity of the reconstructed 3D information, the present invention provides solutions in the following aspects.

[0005] In a first aspect, the present invention provides an image acquisition method for an oral digital impression device, comprising: acquiring an oral image through an oral digital impression device; performing multiple downsampling on the oral image, and obtaining a weight of each downsampled image according to a size ratio of each downsampled image to the oral image; weighting the frequency of each grayscale value in each downsampled image according to the weight, and obtaining a weighted frequency of each grayscale value; dividing all grayscale values ​​into multiple grayscale levels at different grayscale scales, wherein the number of grayscale values ​​corresponding to the grayscale levels at different grayscale scales and the number of all grayscale levels obtained by the division are different; performing edge detection on the oral image at the grayscale levels at different grayscale scales, and obtaining the gradient amplitude of each pixel point at different grayscale scales and the corresponding grayscale value; the length of the edge; the protrusion of the pixel is calculated according to the gradient amplitude of the pixel at different grayscale scales and the length of the edge to which it belongs, and the weighted frequency of each grayscale value is corrected according to the protrusion of the pixel to obtain the corrected frequency of each grayscale value; the mapping interval of each grayscale value is obtained through histogram equalization according to the corrected frequencies of all grayscale values; for any pixel in the oral image, the mapping grayscale value of the pixel is determined from the mapping interval of the grayscale value of the pixel according to the protrusion of the pixel and the grayscale values ​​of other pixels in its neighborhood; the mapping grayscale values ​​of all pixels in the oral image are combined to form an enhanced oral image, so as to reconstruct the 3D information of the teeth through the enhanced oral image.

[0006] Preferably, the weight of each down-sampled image is obtained according to the size ratio of each down-sampled image to the oral image, including: The downsampled image obtained by downsampling The weight of The calculation formula is: ;in, , represents the number of times the oral image is downsampled, and the number of all downsampled images of different sizes obtained is equal to , downsampled image The size ratio of the oral image is equal to , downsampled image The size ratio of the oral image is equal to .

[0007] Preferably, the step of weighting the frequencies of the grayscale values ​​in the downsampled images according to the weights to obtain the weighted frequencies of each grayscale value comprises: wherein the grayscale value The weighted frequency The calculation formula is: ;in, , For downsampled images The weight of For downsampled images Medium gray value The frequency of all downsampled images of different sizes is equal to .

[0008] Preferably, there are 4 different grayscales, namely 1, 5, 9 and 13. Grayscale , ; For any grayscale , in grayscale There are two methods for dividing all gray values ​​into multiple gray levels, and the specific implementation processes of the two methods are as follows: Under the first method, the gray scale can be Divide all grayscale values ​​into Gray levels, The gray value within is taken as the Gray levels, , Indicates rounding up. Indicates the minimum function; Under the second division method, it can be divided according to the grayscale Divide all grayscale values ​​into Gray levels, The gray value within is taken as the Gray levels, , .

[0009] Preferably, the edge detection is performed on the oral image at different grayscale levels to obtain the gradient amplitude of each pixel at different grayscale levels and the length of the edge to which it belongs, including: for any grayscale level , according to the gray value of each pixel in the oral image, Determine the grayscale level of each pixel in all grayscale levels under the image; the image composed of the grayscale levels of all pixels in the oral image is recorded as the grayscale scale Grayscale transformation image under the grayscale scale There are two grayscale transformation images under Perform edge detection on each grayscale transformed image under the The gradient amplitude of each pixel in each grayscale transformation image and the length of the edge to which it belongs are converted into grayscale. The mean of the gradient amplitude of each pixel in the two grayscale transformation images under the condition of , and the mean of the length of the edge to which it belongs, is used as the grayscale scale. The gradient amplitude and edge length of each pixel in the grayscale transformation image under The coordinates of the grayscale transformed image under The gradient amplitude of the pixel point and the length of the edge to which it belongs are used as the coordinates in the oral image. The pixel point is in grayscale The gradient magnitude and the length of the edge under is the horizontal coordinate of the pixel point, is the vertical coordinate of the pixel.

[0010] Preferably, the calculation of the prominence of a pixel point according to its gradient amplitude at different grayscales and the length of its edge includes: wherein the coordinates are The prominence of the pixel The calculation formula is: ;in, The coordinates are The gradient amplitude of the pixel point, The coordinates are The length of the edge where the pixel point is located, Indicates the Grayscale, , is the length of the oral image, is the horizontal coordinate of the pixel point, is the vertical coordinate of the pixel point, Represents the inverse tangent function.

[0011] Preferably, the method of correcting the weighted frequency of each gray value according to the protrusion degree of the pixel point to obtain the corrected frequency of each gray value includes: , the gray value in the oral image is equal to The pixel point is recorded as the target pixel point; the prominence of all target pixels is sorted in order from small to large, and the second quartile is obtained as the grayscale value Correction coefficient of grayscale value; calculate grayscale value Correction coefficient and gray value The product of the weighted frequencies of The corrected frequency of all gray values ​​is recorded as the cumulative value; the gray value The ratio of the corrected frequency to the accumulated value is used as the grayscale value The frequency of correction.

[0012] Preferably, the method of obtaining the mapping interval of each gray value by histogram equalization according to the correction frequency of all gray values ​​includes: ,when When the gray value The left and right boundaries of the mapping interval are equal , at this time the gray value There is only one grayscale value in the mapping interval; when When the gray value The left boundary of the mapping interval , right boundary ,in, Grayscale value The correction frequency, , Grayscale value The correction frequency, .

[0013] Preferably, determining the mapped grayscale value of the pixel from a mapping interval of the grayscale value of the pixel according to the protrusion degree of the pixel and the grayscale values ​​of other pixels in its neighborhood includes: ;in, is the mapped gray value of the pixel, is the grayscale value of the pixel, Grayscale value The left boundary of the mapping interval, Grayscale value The right boundary of the mapping interval, is the prominence of the pixel, is the mean of the grayscale values ​​of other pixels in the neighborhood of the pixel. To round down, the neighborhood of the pixel is centered on the pixel and has a size equal to area.

[0014] In a second aspect, the present invention provides an oral digital impression image acquisition system, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned oral digital impression image acquisition method is implemented.

[0015] By adopting the above technical solution, the above-mentioned oral digital impression device image acquisition method is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that a terminal device is made based on the memory and the processor for easy use.

[0016] The beneficial effects of the present invention are:

[0017] The present invention analyzes downsampled images of different sizes and understands the content of oral images at multiple scales. The weighted frequency of each grayscale value obtained in this way can more comprehensively reflect the more representative features in the oral image; all grayscale values ​​are divided into multiple grayscale levels by different grayscale scales, and the edge detection of the oral image is performed at the grayscale levels of different grayscale scales, which helps to capture features of different sizes in the oral image and can more finely identify the edge details in the oral image; the protrusion degree of the pixel point is calculated according to the gradient amplitude of the pixel point at different grayscale scales and the length of the edge to which it belongs, and the weighted frequency of each grayscale value is corrected according to the protrusion degree of the pixel point to obtain the weighted frequency of each grayscale value. correction frequency, thereby protecting important edge information in the oral image from being over-smoothed, which helps to maintain the sharpness and clarity of the edges in the oral image; according to the protrusion of the pixel point and the grayscale values ​​of other pixels in its neighborhood, the mapping grayscale value of the pixel point is determined from the mapping interval of the grayscale value of the pixel point, thereby enhancing the local contrast of the edge in the oral image, making the local details of the oral image clearer; in summary, the present invention retains and enhances the local details of the edge in the oral image while enhancing the contrast of the oral image, thereby ensuring the clarity of the tooth edge and texture, and then accurately capturing the data points on the tooth surface when reconstructing 3D information, thereby improving the accuracy and authenticity of the reconstructed 3D information. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0019] Figure 1 is a flow chart schematically illustrating an image acquisition method of an oral digital impression device in the present invention;

[0020] Figure 2 is a schematic diagram schematically illustrating an occlusal surface scan;

[0021] Figure 3 is a diagram schematically illustrating a buccal scan;

[0022] Figure 4 is a diagram schematically illustrating a lingual scan;

[0023] Figure 5 is a schematic diagram schematically showing the scanning of the adjacent surfaces. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0025] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0026] The embodiment of the present invention discloses a method for collecting images of an oral digital impression device, referring to Figure 1 , including steps S1 to S5:

[0027] S1. Collect oral images using an oral digital impression device.

[0028] Specifically, the oral digital impression device scans the occlusal surface, buccal side, lingual side and adjacent surfaces to take multiple oral images, which are 2D pictures, that is, two-dimensional images.

[0029] Before starting the scan, the teeth need to be blown dry. During the entire scanning process, the distance between the oral digital impression device's radiation window and the prepared tooth surface must be maintained between 0 and 15 mm (millimeters), with the optimal distance being 5 mm. When scanning the occlusal surface, the camera must not be located on the teeth or gums, as shown in the diagram. Figure 2 As shown; when performing buccal scanning, place the scanning head on the adjacent tooth near the prepared tooth, rotate the oral digital impression instrument 45 degrees to 90 degrees (maximum angle) toward the buccal side, and move the oral digital impression instrument in the distal direction across the prepared tooth, passing through the entire buccal side, as shown in the schematic diagram Figure 3 As shown; when performing lingual scanning, place the scanning head on the adjacent tooth near the prepared tooth, rotate the oral digital impression instrument from the buccal side 90 degrees to the other side to the lingual side 45 degrees to a maximum of 90 degrees, and move the oral digital impression instrument in the mesial direction across the prepared tooth and through the entire lingual side, as shown in the schematic diagram. Figure 4 As shown; when scanning the proximal surface, move the oral digital impression instrument toward the prepared tooth on the occlusal surface, and take images of the proximal surface along the distal and mesial directions, as shown in the schematic diagram. Figure 5 shown.

[0030] Furthermore, the size of the collected oral image is recorded as , is the length of the oral image, is the width of the oral image; take the pixel point in the lower left corner of the oral image as the origin, the horizontal left direction of the origin as the positive direction of the horizontal axis, and the vertical upward direction of the origin as the positive direction of the vertical axis to construct a rectangular coordinate system, then the coordinates of the pixel points in the oral image are marked as , is the horizontal coordinate of the pixel point, is the vertical coordinate of the pixel point, and , .

[0031] It should be noted that in the process of collecting multiple 2D images of the oral cavity by the oral digital impression device, the contrast of the collected 2D images is poor due to the influence of the oral environment, resulting in the loss of tooth surface details in the 2D images, especially in fine structures such as tooth edges and textures. It is difficult to accurately capture data points on the tooth surface when reconstructing 3D information, which seriously affects the accuracy and authenticity of the reconstructed 3D information. In order to obtain accurate 3D reconstruction results, it is necessary to ensure that the collected 2D images have sufficient contrast so that the structure and characteristics of the teeth can be clearly captured. The present invention improves the image contrast by using image enhancement technology, thereby improving the accuracy and reliability of 3D reconstruction.

[0032] It should be further explained that histogram equalization is an existing image enhancement technology. This method mainly adjusts the grayscale distribution of the image to make the grayscale value range of the entire image wider, thereby enhancing the details and contrast of the image. The implementation steps of histogram equalization are: (1) Calculate the grayscale histogram. First, calculate the grayscale histogram of the input image, that is, count the proportion of the number of times each grayscale value appears in the image, that is, the frequency; (2) Calculate the cumulative distribution function (CDF), which is a monotonically increasing function used to maintain the relative size relationship of the grayscale values, and the value range is between 0 and 1, meeting the requirements of grayscale value mapping; (3) Map the grayscale value: Use CDF to map the original grayscale value to the new grayscale value, so that the grayscale histogram of the output image is as evenly distributed as possible, thereby achieving the purpose of enhancing the image contrast; because histogram equalization is based only on the frequency of grayscale values, the enhancement effect on fine structures such as tooth edges and textures is not significant.

[0033] Therefore, the present invention weights the frequency of each grayscale value in each downsampled image according to the weight of each downsampled image to obtain the weighted frequency of each grayscale value, performs edge detection on the oral image at grayscale levels of different grayscales, calculates the protrusion of the pixel point according to the gradient amplitude of the pixel point at different grayscale scales and the length of the edge to which it belongs, corrects the weighted frequency of each grayscale value according to the protrusion of the pixel point, and obtains the mapping interval of each grayscale value through histogram equalization based on the corrected frequencies of all grayscale values. While enhancing the contrast of the oral image, the local details of the edge in the oral image are retained and enhanced, ensuring the clarity of the tooth edge and texture, and thus being able to accurately capture the data points on the tooth surface when reconstructing 3D information, thereby improving the accuracy and authenticity of the reconstructed 3D information.

[0034] S2. Downsample the oral image, obtain the weight of each downsampled image based on the size ratio of each downsampled image to the oral image, and weight the frequency of each grayscale value in each downsampled image according to the weight to obtain the weighted frequency of each grayscale value.

[0035] It should be noted that by analyzing downsampled images of different sizes and understanding the content of oral images at multiple scales, a more comprehensive understanding of the structure and characteristics of oral images can be achieved.

[0036] Specifically, the oral image is downsampled to obtain downsampled images of different sizes, including: in one embodiment, the oral image is decomposed by Haar wavelet transform, and the wavelet transform mainly performs low-pass filtering and high-pass filtering on the oral image in both horizontal and vertical directions, and obtains a downsampled image each time of decomposition; in other embodiments, the oral image is downsampled by nearest neighbor interpolation, and obtains a downsampled image each time of downsampling; for the two-dimensional image, Haar transform is no longer filtered in one direction, but low-pass and high-pass filtering are performed in both horizontal and vertical directions; nearest neighbor interpolation is a simple downsampling method; and Haar wavelet transform and nearest neighbor interpolation are both well-known technologies and will not be described in detail here.

[0037] The size of the downsampled image obtained by each downsampling is the size of the downsampled image obtained by the previous downsampling. , therefore, the The downsampled image obtained by downsampling is denoted as , then the downsampled image The size of , , represents the number of times the oral image is downsampled, and the number of all downsampled images of different sizes obtained is equal to , the number of times the oral image is downsampled The specific value of can be set according to the actual application scenario and requirements, and The value of is , the present invention will Set to 5.

[0038] It should be noted that the features in oral images have different representations at different scales. Assigning weights to downsampled images of different sizes can help extract more representative features in oral images.

[0039] Specifically, the weight of each down-sampled image is obtained according to the size ratio of each down-sampled image to the oral image. The downsampled image obtained by downsampling The weight of The calculation formula is:

[0040] ;

[0041] in, For downsampled images The weight of , represents the number of times the oral image is downsampled, and the number of all downsampled images of different sizes obtained is equal to , the size of the oral image is , downsampled image The size of , then the downsampled image The size ratio of the oral image is equal to , No. The downsampled image obtained by downsampling The size of , , then the downsampled image The size ratio of the oral image is equal to .

[0042] It should be noted that, the larger the size ratio of the downsampled image to the oral image, the greater the weight of the downsampled image.

[0043] Furthermore, the frequency statistics of the grayscale values ​​of each downsampled image are performed to obtain the frequency of each grayscale value in each downsampled image, wherein the downsampled image Medium gray value Frequency Equal to the downsampled image The mid-gray value is equal to The number of all pixels and the downsampled image The ratio of the total number of all pixels in , where , downsampled image The total number of all pixels in .

[0044] Furthermore, the frequency of each gray value in each downsampled image is weighted according to the weight to obtain the weighted frequency of each gray value, then the gray value The weighted frequency The calculation formula is:

[0045] ;

[0046] in, Grayscale value The weighted frequency of For downsampled images The weight of , For downsampled images Medium gray value The frequency of all downsampled images of different sizes is equal to .

[0047] It should be noted that weighting the frequency of each grayscale value in each downsampled image by the weight of each downsampled image can help extract more representative grayscale features in the oral image.

[0048] S3. Divide all grayscale values ​​into multiple gray levels at different grayscales, perform edge detection on the oral image at the grayscales at different grayscales, and obtain the gradient amplitude and the length of the edge of each pixel at different grayscales.

[0049] Specifically, all grayscale values ​​contain All integers in the range, therefore, the number of types of all gray values ​​is equal to 256; all gray values ​​are divided into multiple gray levels under different gray scales, wherein the number of gray values ​​corresponding to the gray levels under different gray scales and the number of all gray levels obtained by division are different; in this embodiment, there are 4 different gray scales, namely 1, 5, 9 and 13, that is, Grayscale , In other embodiments, the number of types of different grayscales and the specific value of each grayscale can be set according to actual application scenarios and requirements.

[0050] It should be noted that for Grayscale , the number of all gray levels obtained by division is equal to 256, that is, for the A grayscale, each gray value is regarded as a gray level.

[0051] It should be noted that by dividing all grayscale values ​​into multiple gray levels through different grayscales, edge detection is performed on oral images at gray levels of different grayscales, which provides multi-scale edge detection capabilities, helps to capture structures and features of different sizes in oral images, and can more finely identify edge details in oral images, especially when the edge intensity and width vary greatly, which helps to distinguish between real edges and noise. At the same time, by setting a coarser grayscale division, the influence of noise is reduced, thereby improving the stability and reliability of edge detection.

[0052] Furthermore, for any grayscale , in grayscale There are two methods for dividing all gray values ​​into multiple gray levels, and the specific implementation processes of the two methods are:

[0053] 1. Under the first division method, it can be divided into two groups according to the grayscale Divide all grayscale values ​​into Gray levels, specifically, The gray value within is taken as the grayscale, that is, Grayscale correspondence The gray value within , Indicates rounding up. Represents the minimum function;

[0054] 2. Under the second division method, it can be divided into two groups according to the grayscale Divide all grayscale values ​​into Gray levels, specifically, The gray value within is taken as the grayscale, that is, Grayscale correspondence The gray value within , Indicates rounding up. represents the maximum value function, Represents the minimum function.

[0055] For example, when the grayscale When the first division method is used, the grayscale Divide all grayscale values ​​into gray levels, that is, 、 、 、 、…、 The grayscale values ​​within are respectively used as Grayscale to grayscale, 255 is the first Grayscale; Under the second division method, it can be divided into Divide all grayscale values ​​into gray levels, that is, 、 、 、 、…、 、 The grayscale values ​​within are respectively used as Grayscale to Gray levels.

[0056] It should be noted that the two division methods can make the grayscale division results more diverse, which helps to capture structures and features of different sizes in oral images.

[0057] Furthermore, at different grayscale levels, edge detection is performed on the oral image to obtain the gradient amplitude and the length of the edge of each pixel at different grayscale levels, including: for any grayscale level , according to the gray value of each pixel in the oral image, Determine the grayscale level of each pixel in all grayscale levels under the image; the image composed of the grayscale levels of all pixels in the oral image is recorded as the grayscale scale Grayscale transformation image under grayscale There are two ways to divide all gray values ​​into multiple gray levels. Therefore, the gray scale There are two grayscale transformation images under Perform edge detection on each grayscale transformed image under the The gradient amplitude of each pixel in each grayscale transformation image and the length of the edge to which it belongs are converted into grayscale. The mean of the gradient amplitude of each pixel in the two grayscale transformation images under the condition of , and the mean of the length of the edge to which it belongs, is used as the grayscale scale. The gradient amplitude and edge length of each pixel in the grayscale transformation image under The coordinates of the grayscale transformed image under The gradient amplitude of the pixel point and the length of the edge to which it belongs are used as the coordinates in the oral image. The pixel point is in grayscale The gradient magnitude and the length of the edge under is the horizontal coordinate of the pixel point, is the vertical coordinate of the pixel.

[0058] Among them, the algorithm used for edge detection is the Canny edge detection algorithm, and the algorithm steps include: (1) smoothing the image with a Gaussian filter; (2) calculating the gradient amplitude and gradient direction with first-order partial derivative finite difference; (3) performing non-maximum suppression on the gradient amplitude; (4) detecting and connecting edges with a dual threshold algorithm; The Canny edge detection algorithm is a reliable and flexible edge detection method that extracts edges in an image by applying non-maximum suppression and hysteresis threshold processing, and the algorithm has good resistance to noise, and the operators used by the Canny edge detection algorithm include but are not limited to the Sobel operator, Prewitt operator, Laplacian operator and Canny operator; The Canny edge detection algorithm, Sobel operator, Prewitt operator, Laplacian operator and Canny operator are all well-known technologies and will not be described in detail here.

[0059] It should be noted that the present invention divides all grayscale values ​​into multiple gray levels through different grayscale scales, and performs edge detection on oral images at gray levels of different grayscale scales, which helps to capture features of different sizes in oral images and can more finely identify edge details in oral images.

[0060] S4. Calculate the prominence of the pixel point based on its gradient amplitude at different grayscales and the length of its edge, and use it to correct the weighted frequency of each grayscale value to obtain the corrected frequency of each grayscale value.

[0061] Specifically, the prominence of the pixel is calculated based on the gradient amplitude of the pixel at different grayscales and the length of the edge to which it belongs. The coordinates are The prominence of the pixel The calculation formula is:

[0062] ;

[0063] in, The coordinates are The prominence of the pixel point, The coordinates are The gradient amplitude of the pixel point, The coordinates are The length of the edge where the pixel point is located, Indicates the Grayscale, , is the length of the oral image, is the horizontal coordinate of the pixel point, is the vertical coordinate of the pixel point, Represents the inverse tangent function.

[0064] Specifically, according to the prominence of the pixel point, the weighted frequency of each gray value is corrected to obtain the corrected frequency of each gray value, including: , the gray value in the oral image is equal to The pixel point is recorded as the target pixel point; the prominence of all target pixels is sorted in order from small to large, and the second quartile, that is, the median, is obtained as the grayscale value Correction coefficient of grayscale value; calculate grayscale value Correction coefficient and gray value The product of the weighted frequencies of The corrected frequency of all gray values ​​is recorded as the cumulative value; the gray value The ratio of the corrected frequency to the accumulated value is used as the grayscale value The correction frequency, the gray value The ratio of the corrected frequency to the accumulated value is the grayscale value The result of the normalization process of the corrected frequency.

[0065] It should be noted that the present invention calculates the protrusion of the pixel point based on the gradient amplitude of the pixel point at different grayscale scales and the length of the edge to which it belongs, and corrects the weighted frequency of each grayscale value according to the protrusion of the pixel point to obtain the corrected frequency of each grayscale value, thereby protecting important edge information in the oral image from being over-smoothed, which helps to maintain the sharpness and clarity of the edges in the oral image.

[0066] S5. Obtain the mapping interval of each grayscale value based on the correction frequency of all grayscale values. Determine the mapping grayscale value of the pixel from the mapping interval of its grayscale value based on the protrusion of the pixel and the grayscale values ​​of other pixels in its neighborhood. Reconstruct the 3D information of the teeth through the enhanced oral image composed of the mapping grayscale values ​​of all pixels in the oral image.

[0067] Specifically, according to the correction frequency of all gray values, the mapping interval of each gray value is obtained by histogram equalization. ,when When the gray value The left and right boundaries of the mapping interval are equal , at this time the gray value There is only one grayscale value in the mapping interval. When the gray value The left boundary of the mapping interval , right boundary ,in, Grayscale value The correction frequency, , Grayscale value The correction frequency, .

[0068] Furthermore, for any pixel in the oral image, the mapping grayscale value of the pixel is determined from the mapping interval of the pixel's grayscale value according to the protrusion of the pixel and the grayscale values ​​of other pixels in its neighborhood. The calculation formula for the mapping grayscale value of the pixel is:

[0069] ;

[0070] in, is the mapped gray value of the pixel, is the grayscale value of the pixel, Grayscale value The left boundary of the mapping interval, Grayscale value The right boundary of the mapping interval, is the prominence of the pixel, is the mean of the grayscale values ​​of other pixels in the neighborhood of the pixel. is rounded down; in this embodiment, the neighborhood of the pixel point is centered on the pixel point and is equal to In other embodiments, the size of the neighborhood of the pixel point can be set according to actual application scenarios and requirements.

[0071] It should be noted that when the gray value of the pixel Less than the mean grayscale value of other pixels in the neighborhood of this pixel When the gray value Select a smaller grayscale value in the mapping interval, and the greater the prominence of the pixel, the smaller the grayscale value selected, so as to expand the grayscale difference between the pixel and other pixels in its neighborhood, thereby retaining the neighborhood characteristics of the pixel; when the grayscale value of the pixel is Greater than or equal to the mean of the grayscale values ​​of other pixels in the neighborhood of this pixel When the gray value A larger grayscale value is selected in the mapping interval, and the greater the prominence of the pixel, the larger the grayscale value is selected, so as to expand the grayscale difference between the pixel and other pixels in its neighborhood, thereby retaining the neighborhood characteristics of the pixel.

[0072] Furthermore, an image composed of the mapped grayscale values ​​of all pixels in the oral image is used as the enhanced oral image; so that the 3D information of the teeth can be reconstructed through the enhanced oral image, and the 3D information of the teeth includes the occlusal relationship between the upper and lower jaws and the upper and lower jaws.

[0073] It should be noted that the present invention analyzes and understands the content of oral images at multiple scales by analyzing downsampled images of different sizes. The weighted frequency of each grayscale value obtained in this way can more comprehensively reflect the more representative features in the oral image. All grayscale values ​​are divided into multiple grayscale levels by different grayscale scales. Edge detection of oral images is performed at grayscale levels of different grayscale scales, which helps to capture features of different sizes in the oral image and can more finely identify edge details in the oral image. The protrusion degree of the pixel point is calculated according to the gradient amplitude of the pixel point at different grayscale scales and the length of the edge to which it belongs. The weighted frequency of each grayscale value is corrected according to the protrusion degree of the pixel point to obtain the weighted frequency of each grayscale value. The correction frequency of the grayscale value protects important edge information in the oral image from being over-smoothed, which helps to maintain the sharpness and clarity of the edges in the oral image; based on the protrusion of the pixel point and the grayscale values ​​of other pixels in its neighborhood, the mapping grayscale value of the pixel point is determined from the mapping interval of the grayscale value of the pixel point, so as to enhance the local contrast of the edge in the oral image and make the local details of the oral image clearer; in summary, the present invention retains and enhances the local details of the edge in the oral image while enhancing the contrast of the oral image, thereby ensuring the clarity of the tooth edge and texture, and then accurately captures the data points on the tooth surface when reconstructing 3D information, thereby improving the accuracy and authenticity of the reconstructed 3D information.

[0074] An embodiment of the present invention further discloses an oral digital impression image acquisition system, comprising a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, an oral digital impression image acquisition method according to the present invention is implemented.

[0075] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0076] In the description of this specification, "multiple" and "several" mean at least two, such as two, three or more, unless otherwise clearly defined.

[0077] While this specification has shown and described several embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that in practicing the present invention, alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for collecting images of an oral digital impression device, characterized in that: include: Collect oral images through oral digital impression device; Downsampling the oral image multiple times, and obtaining the weight of each downsampled image according to the size ratio of each downsampled image to the oral image; Weighting the frequency of each gray value in each downsampled image according to the weight to obtain the weighted frequency of each gray value; At different grayscales, all grayscale values ​​are divided into multiple grayscale levels, wherein the number of grayscale values ​​corresponding to the grayscale levels at different grayscales and the number of all grayscale levels obtained by the division are different; at different grayscales, edge detection is performed on the oral image to obtain the gradient amplitude of each pixel point at different grayscales and the length of the edge to which it belongs; According to the gradient amplitude of the pixel at different grayscales and the length of the edge to which it belongs, the prominence of the pixel is calculated, and the weighted frequency of each grayscale value is corrected according to the prominence of the pixel to obtain the corrected frequency of each grayscale value, including: , the gray value in the oral image is equal to The pixel point is recorded as the target pixel point; the prominence of all target pixels is sorted in order from small to large, and the second quartile is obtained as the gray value Correction coefficient of gray value; calculate gray value The product of the correction coefficient and the weighted frequency is recorded as the gray value The corrected frequency of all gray values ​​is recorded as the cumulative value; the gray value The ratio of the corrected frequency to the accumulated value is used as the gray value The frequency of correction; According to the correction frequency of all grayscale values, the mapping interval of each grayscale value is obtained through histogram equalization; for any pixel in the oral image, the mapping grayscale value of the pixel is determined from the mapping interval of the grayscale value of the pixel according to the protrusion of the pixel and the grayscale values ​​of other pixels in its neighborhood; the mapping grayscale values ​​of all pixels in the oral image are combined into an enhanced oral image, so as to reconstruct the 3D information of the teeth through the enhanced oral image.

2. The method for acquiring an image of an oral digital impression device according to claim 1, characterized in that: The weight of each down-sampled image is obtained according to the size ratio of each down-sampled image to the oral image. include: Among them, The downsampled image obtained by downsampling The weight of The calculation formula is: ； in, , represents the number of times the oral image is downsampled, and the number of all downsampled images of different sizes obtained is equal to , downsampled image The size ratio of the oral image is equal to , downsampled image The size ratio of the oral image is equal to .

3. The method for acquiring an image of an oral digital impression device according to claim 1, characterized in that: The frequency of each gray value in each down-sampled image is weighted according to the weight to obtain the weighted frequency of each gray value, include: Among them, the gray value The weighted frequency The calculation formula is: ； in, , For downsampled images The weight of For downsampled images Medium gray value The frequency of all downsampled images of different sizes is equal to .

4. The method for acquiring an image of an oral digital impression device according to claim 1, characterized in that: There are 4 different grayscales, namely 1, 5, 9 and 13. Grayscale , ; For any grayscale , in grayscale There are two division methods to divide all gray values ​​into multiple gray levels, and the specific implementation processes of the two division methods are: In the first division method, it is possible to divide the Divide all gray values ​​into grayscale The gray value within is taken as the Gray levels, , Indicates rounding up. Represents the minimum value function; In the second division method, it is possible to divide the Divide all gray values ​​into grayscale The gray value within is taken as the Gray levels, , Represents the maximum value function.

5. The method for acquiring an image of an oral digital impression device according to claim 4, characterized in that: The method of performing edge detection on the oral image at different grayscale levels to obtain the gradient amplitude of each pixel at different grayscale levels and the length of the edge to which it belongs includes: For any grayscale According to the gray value of each pixel in the oral image, Determine the grayscale of each pixel from all grayscales under the image; the image composed of the grayscales of all pixels in the oral image is recorded as the grayscale scale The grayscale transformation image under There are two grayscale transformation images under Perform edge detection on each grayscale transformed image under the condition to obtain the grayscale scale The gradient amplitude of each pixel in each grayscale transformation image and the length of the edge to which it belongs are converted into grayscale The mean value of the gradient amplitude of each pixel in the two grayscale transformation images under the condition of , and the mean value of the length of the edge to which it belongs, are used as the grayscale scale The gradient amplitude of each pixel in the grayscale transformation image and the length of the edge to which it belongs; the grayscale scale The coordinates of the grayscale transformation image are The gradient amplitude of the pixel point and the length of the edge to which it belongs are taken as the coordinates in the oral image. The pixel point is in gray scale The gradient magnitude and the length of the edge under is the horizontal coordinate of the pixel point, is the vertical coordinate of the pixel.

6. The method for acquiring an image of an oral digital impression device according to claim 1, characterized in that: The prominence of the pixel is calculated according to the gradient amplitude of the pixel at different grayscales and the length of the edge to which it belongs. include: The coordinates are The prominence of the pixels The calculation formula is: ； in, The coordinates are The gradient amplitude of the pixel point, The coordinates are The length of the edge where the pixel point is located, Indicates Gray scale, , is the length of the oral image, is the horizontal coordinate of the pixel point, is the vertical coordinate of the pixel, Represents the inverse tangent function.

7. The method for acquiring an image of an oral digital impression device according to claim 1, characterized in that: The step of obtaining a mapping interval of each gray value by histogram equalization according to the correction frequency of all gray values ​​includes: For gray value ,when When the gray value The left and right boundaries of the mapping interval are equal , at this time the gray value There is only one grayscale value in the mapping interval; when When the gray value The left boundary of the mapping interval , right border ,in, Gray value The correction frequency, , Gray value The correction frequency, .

8. The method for acquiring an image of an oral digital impression device according to claim 1, characterized in that: The step of determining the mapping grayscale value of the pixel from the mapping interval of the grayscale value of the pixel according to the prominence of the pixel and the grayscale values ​​of other pixels in its neighborhood includes: ； in, is the mapped gray value of the pixel, is the gray value of the pixel, Gray value The left boundary of the mapping interval, Gray value The right boundary of the mapping interval, is the prominence of the pixel, is the mean grayscale value of other pixels in the neighborhood of the pixel. To round down, the neighborhood of the pixel is centered on the pixel and is equal to area.

9. An oral digital impression device image acquisition system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an image acquisition method for an oral digital impression device according to any one of claims 1 to 8 is implemented.

Images