Intelligent scanning method and system based on oral digital impression instrument

Abstract

The present invention relates to the field of image processing, and in particular to an intelligent scanning method and system based on an oral digital impression device. The method comprises the following steps: obtaining a current oral image at each angle and a historical oral image at each angle; performing matching processing on the current oral image at different angles, and obtaining suspicious pixels according to the pixel value difference between the matching pixels; obtaining several texture areas in the historical oral image and the current oral image, and fitting the area variation law and the interval variation law of the continuous texture area; calculating the probability that any suspicious pixel is an interference pixel according to the area variation law and the interval variation law; removing the interference pixels in the current oral image by comparing the probability with a preset probability threshold, so as to realize oral three-dimensional reconstruction. The accuracy of oral reconstruction is improved by excluding interference pixels.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to an intelligent scanning method and system based on an oral digital impression device. Background Art

[0002] In order to more intuitively display the state of teeth in the oral cavity and facilitate dental health diagnosis, it is necessary to use an oral digital impression device to scan the teeth and gums in the oral cavity, and then build a 3D model based on the scanned data. The scanned data is the data basis for building a 3D model. The accuracy of the scanned data will affect the construction effect of the 3D model. Therefore, in order to more accurately build a 3D model of teeth and gums, it is necessary to accurately obtain the scanned data.

[0003] In the process of scanning the teeth and gums in the oral cavity using an oral digital impression device, the tongue and saliva in the oral cavity occasionally cover the teeth, resulting in the presence of some tongue or saliva data in the scanned data, which interferes with the construction of the three-dimensional model. In order to accurately construct the three-dimensional model, the interference of the tongue or saliva data needs to be eliminated. Therefore, how to eliminate the interference data in the scanned data becomes the research focus of the present invention.

[0004] The patent application document with publication number CN114283236A discloses a method, device and storage medium for oral scanning using a smart phone. The patent application document mainly introduces how to scan data and how to use the scanned data to build a three-dimensional model. The patent application document does not involve the exclusion of interference data and other aspects. Therefore, the method in the patent application document cannot solve the technical problem in the present invention. Summary of the invention

[0005] In order to solve the problem of how to eliminate interference data in scan data, the present invention provides an intelligent scanning method and system based on an oral digital impression device.

[0006] In the first aspect, the present invention provides an intelligent scanning method based on an oral digital impression device, which adopts the following technical solutions:

[0007] The intelligent scanning method based on the oral digital impression instrument comprises the following steps:

[0008] Acquire the current oral cavity image at each angle and the historical oral cavity image at each angle;

[0009] Perform matching processing on the current oral cavity images at different angles, and obtain suspicious pixels according to the pixel value differences between the matching pixels;

[0010] Obtain several texture areas in historical oral images and current oral images, fit the area change law and interval change law of continuous texture areas; calculate the probability that any suspicious pixel is an interference pixel , taking a historical oral image at the same angle as the current oral image as a reference oral image, recording the current oral image containing the suspicious pixel as a backup oral image, obtaining the mean similarity between a preset window of the suspicious pixel in the backup oral image and the corresponding areas of all reference oral images as a similarity measurement value of the suspicious pixel in the backup oral image; represents the similarity measure value of the suspicious pixel in the i-th backup oral image, It represents the maximum similarity metric value of the suspicious pixel in all the backup oral images. Indicates the conformity of the texture interval distance of the texture area where the suspicious pixel is located in the i-th backup oral image to the interval variation law, Indicates the conformity of the area of ​​the texture region where the suspicious pixel is located in the i-th backup oral image to the area variation law, Indicates the preset anti-zero coefficient;

[0011] The interfering pixels in the current oral image are removed by comparing the probability with the preset probability threshold to achieve oral three-dimensional reconstruction.

[0012] Preferably, obtaining suspicious pixels according to pixel value differences between matching pixels includes:

[0013] Every two angles form an angle pair, and the absolute value of the difference in pixel values ​​of the corresponding matching pixels of the current oral image of the two angles in the angle pair under each channel is calculated and recorded as the pixel difference of each pixel in the angle pair under each channel, and the average of the pixel differences of each pixel in the angle pair under all channels is recorded as the comprehensive pixel difference of each pixel in the angle pair; if there is a pixel whose comprehensive pixel difference is greater than a preset difference threshold, then the pixel and the pixels matching the pixel in each angle are recorded as pixels to be analyzed;

[0014] The light interference degree of the pixel to be analyzed is calculated, and the pixel to be analyzed whose light interference degree is less than a preset degree threshold is recorded as a suspicious pixel.

[0015] Preferably, the calculating the light interference degree of the pixel to be analyzed includes:

[0016] The current oral image is segmented by a superpixel segmentation algorithm to obtain several superpixel blocks. The superpixel block where any pixel to be analyzed is located is obtained as a reference superpixel block. A Gaussian mixture model is fitted to the pixel values ​​of the reference superpixel blocks in the current oral image at each angle, and the KL divergence of the Gaussian mixture model of the reference superpixel blocks of the current oral image at every two angles is calculated. The normalized value of the mean of the KL divergence of the Gaussian mixture model of the reference superpixel blocks of the current oral image at all two angles is used as the light interference degree of the pixel to be analyzed.

[0017] Preferably, the acquiring of several texture regions in the historical oral image and the current oral image includes:

[0018] Perform edge detection on historical oral images to obtain edge pixels in the corresponding images;

[0019] After morphological processing of edge pixels, the closed area surrounded by edge pixels is obtained and recorded as texture area;

[0020] Get the texture area in the current oral image.

[0021] Preferably, the area variation law and interval variation law of the fitted continuous texture region include:

[0022] The texture area that is continuous and has no gaps is recorded as a continuous texture area;

[0023] According to the arrangement order of the continuous texture regions, the areas of all the texture regions in the continuous texture regions are fitted with polynomials to obtain the area variation law of the continuous texture regions;

[0024] Obtain the interval distance between the geometric center of each texture area and the previous texture area in the continuous texture area and record it as the texture interval distance of each texture area in the continuous texture area;

[0025] According to the arrangement order of the continuous texture regions, the texture interval distances of all the texture regions in the continuous texture region are fitted with a polynomial to obtain the interval variation rule of the continuous texture region.

[0026] Preferably, the method for obtaining the conformity of the texture spacing distance to the spacing variation rule includes:

[0027] Performing matching processing on the backup oral cavity image and the reference oral cavity image;

[0028] Obtain the texture region where the suspicious pixel is located, obtain the distance between the geometric center of the texture region where the suspicious pixel is located and the previous adjacent texture region as the texture interval distance of the texture region where the suspicious pixel is located, and obtain the corresponding region of the texture region where the suspicious pixel is located in the reference oral image according to the matching relationship;

[0029] The interval distance fitting value is fitted using the interval variation law of the corresponding area of ​​the texture area where the suspicious pixel is located in any reference oral image. The absolute value of the difference between the texture interval distance of the texture area where the suspicious pixel is located and the interval distance fitting value is calculated and recorded as the interval distance fitting deviation of the texture area where the suspicious pixel is located relative to the reference oral image; the inverse of the mean of the interval distance fitting deviation of the texture area where the suspicious pixel is located relative to all reference oral images is taken as the conformity of the texture interval distance to the interval variation law.

[0030] Preferably, the method for obtaining the conformity of the area to the area variation law includes:

[0031] The area fitting value is fitted using the area variation law of the corresponding area of ​​the texture area where the suspicious pixel is located in any reference oral image. The absolute value of the difference between the area of ​​the texture area where the suspicious pixel is located and the area fitting value is calculated and recorded as the area fitting deviation of the texture area where the suspicious pixel is located relative to the reference oral image. The inverse of the mean of the area fitting deviation of the texture area where the suspicious pixel is located relative to all reference oral images is taken as the conformity of the area to the area variation law.

[0032] In a second aspect, the present invention provides an intelligent scanning system based on an oral digital impression device, which adopts the following technical solutions:

[0033] The intelligent scanning system based on the oral digital impression instrument comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the intelligent scanning method based on the oral digital impression instrument is implemented.

[0034] By adopting the above technical solution, the above intelligent scanning method based on the oral digital impression device 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 according to the memory and the processor for easy use.

[0035] The present invention has the following technical effects:

[0036] The present invention analyzes the interference information in the scanned image and accurately removes the interference information in the image, thereby providing a data basis for the subsequent accurate reconstruction of the three-dimensional model of the oral cavity; further, when analyzing the situation where each pixel in the image is interference information, considering that the grayscale value of the interference information at different angles is different, possible interference pixels are screened out by analyzing the pixel value difference; further, considering that the interference information at different viewing angles is different, when analyzing the situation where each pixel in the image is interference information, the situation where each possible interference pixel is an interference pixel is accurately measured by introducing the similarity analysis of the area where each possible interference pixel is located at different viewing angles; further, considering that non-interference information such as teeth has a certain distribution law, therefore when analyzing the situation where each pixel in the image is interference information, the distribution law is used to accurately measure the situation where each possible interference pixel is an interference pixel. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0038] Figure 1 It is a flow chart of a method in an intelligent scanning method based on an oral digital impression device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0040] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0041] The embodiment of the present invention discloses an intelligent scanning method based on an oral digital impression instrument, referring to Figure 1 , comprising steps S1 to S4:

[0042] S1: Acquire the current oral cavity image at each angle and the historical oral cavity image at each angle.

[0043] Specifically, the teeth and gums are scanned at each angle of the oral cavity using an oral data impression device to obtain a current oral cavity image at each angle;

[0044] The teeth and gum images at each angle historically acquired by the oral data impression device are recorded as historical oral images at each angle.

[0045] S2: Perform matching processing on the current oral cavity images at different angles, and obtain suspicious pixels according to the pixel value differences between the matching pixels.

[0046] It should be noted that the positions of the tongue and saliva will change during the scanning process, so the tongue or saliva will sometimes cover the teeth and sometimes not. Therefore, in the images collected at different angles, the tooth areas of some images are disturbed by the presence of the tongue or saliva, and the tooth areas of some images are not disturbed by the tongue or saliva. There is a difference between the pixel values ​​of the disturbed pixels and the pixel values ​​of the undisturbed pixels, so the possible disturbed pixels can be extracted by the pixel value difference.

[0047] S20: performing matching processing on the current oral cavity images at different angles.

[0048] Preferably, as an example, matching processing is performed on the current oral cavity images at different angles, including:

[0049] Matching processing is performed on the current oral cavity images at every two different angles to obtain the matching relationship between pixels.

[0050] S21: Obtain suspicious pixels according to the pixel value differences between the matching pixels.

[0051] It should be noted that light changes can also cause pixel value differences. In order to prevent light changes from causing interference in pixel value differences, it is necessary to analyze the light interference of pixels after filtering out pixels with large pixel value differences.

[0052] Preferably, as an example, obtaining suspicious pixels according to the pixel value difference between matching pixels includes:

[0053] Every two angles form an angle pair, and the absolute value of the difference in pixel values ​​of the corresponding matching pixels of the current oral image of the two angles in the angle pair under each channel is calculated and recorded as the pixel difference of each pixel in the angle pair under each channel, and the average of the pixel differences of each pixel in the angle pair under all channels is recorded as the comprehensive pixel difference of each pixel in the angle pair; if there is a pixel whose comprehensive pixel difference is greater than a preset difference threshold, then the pixel and the pixels matching the pixel in each angle are recorded as pixels to be analyzed; in this embodiment, the lower quartile of the distribution histogram of the comprehensive pixel differences of all pixels in all angle pairs is used as the preset difference threshold, and other embodiments may set other values, which are not specifically limited in this embodiment.

[0054] The current oral image is segmented using a superpixel segmentation algorithm to obtain several superpixel blocks, and the superpixel block where any pixel to be analyzed is located is obtained and recorded as a reference superpixel block. A Gaussian mixture model is fitted to the pixel values ​​of the reference superpixel blocks in the current oral image at each angle, and the KL divergence of the Gaussian mixture model of the reference superpixel blocks of the current oral image at every two angles is calculated. The normalized value of the mean of the KL divergence of the Gaussian mixture model of the reference superpixel blocks of the current oral image at all two angles is used as the light interference degree of the pixel to be analyzed; the pixel to be analyzed whose light interference degree is less than a preset degree threshold is recorded as a suspicious pixel. This embodiment is described by taking the preset degree threshold of 0.65 as an example, and other embodiments may take other values, and this embodiment does not make specific restrictions.

[0055] It can be understood that the Gaussian mixture model of the reference superpixel block to which the pixel to be analyzed belongs reflects the light distribution characteristics of the reference superpixel block to which the pixel to be analyzed belongs, and the KL divergence of the Gaussian mixture model of the reference superpixel block at different angles reflects the difference in light distribution characteristics. The larger the value, the greater the difference in light distribution at the pixels to be analyzed at different angles. Therefore, by eliminating the pixel value differences caused by the light distribution differences, the accuracy of screening out suspicious pixels is improved.

[0056] It should be added that the Gaussian mixture model is fitted to the pixel values ​​of the reference superpixel block in the current oral image at each angle, including:

[0057] Obtaining the pixel value of each pixel in the reference pixel block in each channel, and recording the average of the pixel values ​​of each pixel in the reference pixel block in all channels as the comprehensive pixel value of each pixel in the reference pixel block;

[0058] A Gaussian mixture model is fitted to the integrated pixel values ​​of the reference superpixel blocks in the current oral image at each angle.

[0059] S3: Obtain several texture regions in historical oral images and current oral images, fit the area variation law and interval variation law of continuous texture regions; calculate the probability that any suspicious pixel is an interference pixel.

[0060] It should be noted that the suspicious pixels screened out above are interference pixels in images at some angles, but normal pixels in images at other angles. Therefore, it is necessary to analyze whether the suspicious pixels are interference pixels in images at each angle, so as to more accurately screen out the interference pixels.

[0061] S30: Acquire a plurality of texture regions in a historical oral cavity image and a current oral cavity image.

[0062] It should be further explained that if the tooth area is disturbed by the tongue and saliva in some angle images, and not in some angle images, the similarity of the tooth area in the images at different angles will decrease. Therefore, the difference between the similarity of the suspicious pixels in the images at different angles and the maximum similarity can be used to determine whether each suspicious pixel in the images at each angle is an interference pixel. At the same time, if the tooth area is similarly disturbed, the similarity of the suspicious pixels in the images at different angles is still large, and the interference pixels cannot be distinguished by the similarity alone. Since the shape and arrangement of people's teeth have certain regularities, the interference pixels can be further distinguished based on the shape of people's teeth and the exclusion rules.

[0063] Preferably, as an example, several texture areas in the historical oral image and the current oral image are obtained, including:

[0064] Perform edge detection on historical oral images to obtain edge pixels in the corresponding images;

[0065] After morphological processing of edge pixels, the closed area surrounded by edge pixels is obtained and recorded as texture area;

[0066] Get the texture area in the current oral image.

[0067] S31: Fitting the area variation law and interval variation law of continuous texture regions.

[0068] Preferably, as an example, fitting the area variation law and interval variation law of the continuous texture region includes:

[0069] The texture area that is continuous and has no gaps is recorded as a continuous texture area;

[0070] According to the arrangement order of the continuous texture regions, the areas of all the texture regions in the continuous texture regions are fitted with polynomials to obtain the area variation law of the continuous texture regions;

[0071] Obtain the distance between the geometric center of each texture region and the previous texture region in the continuous texture region and record it as the texture interval distance of each texture region in the continuous texture region;

[0072] According to the arrangement order of the continuous texture regions, the texture interval distances of all the texture regions in the continuous texture region are fitted with a polynomial to obtain the interval variation rule of the continuous texture region.

[0073] S32: Calculate the probability that any suspicious pixel is an interference pixel.

[0074] Preferably, as an example, calculating the probability that any suspicious pixel is an interference pixel includes:

[0075]

[0076] The current oral image containing the suspicious pixel is recorded as a backup oral image, and the historical oral image at the same angle as the backup oral image is recorded as a reference oral image. The mean similarity between the preset window of the suspicious pixel in the backup oral image and the corresponding areas of all reference oral images is obtained and recorded as the similarity measurement value of the suspicious pixel in the backup oral image; represents the similarity measure value of the suspicious pixel in the i-th backup oral image, It represents the maximum similarity metric value of the suspicious pixel in all the backup oral images. Indicates the conformity of the texture interval distance of the texture area where the suspicious pixel is located in the i-th backup oral image to the interval variation law, Indicates the conformity of the area of ​​the texture region where the suspicious pixel is located in the i-th backup oral image to the area variation law, represents the probability that the suspicious pixel in the i-th backup oral image is an interference pixel. The area is used as the preset window. Other embodiments may set other areas, and this embodiment does not specifically limit it. Represents a preset anti-zero coefficient, which is used to prevent the denominator from being zero. 0.001 is taken as an example for description, and other embodiments may take other values, which are not specifically limited in this embodiment.

[0077] It can be understood that in some angle images, the tongue or saliva interferes with the tooth area; compared with the similarity between the tooth areas in different angle images without interference, the similarity between the tooth areas in the angle image with interference and the tooth areas in the angle image without interference is lower. It reflects the similarity between tooth regions in images from different angles without interference. It reflects the similarity between the tooth area in the angle image with interference and the tooth area in the angle image without interference. It reflects the decrease in similarity. The larger the value, the more likely the suspicious pixel is a tongue or saliva interference pixel in the i-th backup oral image. The area change law reflects the distribution law of people's tooth shapes, and the interval change law reflects the arrangement law of people's teeth. It reflects the compliance of the suspicious pixel with the tooth shape distribution law in the i-th backup oral image. The interference of tongue or saliva will reduce the tooth shape distribution law. The larger the value is, the smaller the probability that the suspicious pixel is an interference pixel in the i-th backup oral image. It reflects the compliance of the suspicious pixel with the tooth arrangement rule in the i-th backup oral image. The interference of tongue or saliva will reduce the tooth arrangement rule. The larger the value, the smaller the probability that the suspicious pixel is an interference pixel in the i-th backup oral image.

[0078] The above embodiments involve similarity measurement values, conformity of texture spacing distance to spacing variation rules, and conformity of texture spacing distance to spacing variation rules. The following describes a method for determining similarity measurement values, conformity of texture spacing distance to spacing variation rules, and conformity of texture spacing distance to spacing variation rules.

[0079] First, the method of obtaining similarity measurement values ​​is introduced.

[0080] Preferably, as an example, a method for obtaining a similarity measurement value includes:

[0081] The comprehensive pixel values ​​of all pixels in the preset window of the suspicious pixel in the backup oral image are statistically obtained to obtain the occurrence frequency of each comprehensive pixel value, and the vector formed by the occurrence frequency of all kinds of comprehensive pixel values ​​in the preset window of the suspicious pixel in the backup oral image is recorded as the comprehensive pixel value distribution vector, and the cosine similarity of the comprehensive pixel value distribution vector of the preset window of the suspicious pixel in the backup oral image and the corresponding area of ​​the reference oral image is calculated and recorded as the similarity between the preset window of the suspicious pixel in the backup oral image and the corresponding area of ​​the reference oral image, and the average similarity between the preset window of the suspicious pixel in the backup oral image and the corresponding areas of all reference oral images is recorded as the similarity measure value of the suspicious pixel in the backup oral image.

[0082] Then, the method of obtaining the conformity of texture interval distance to the interval variation law is introduced.

[0083] Preferably, as an example, a method for obtaining the conformity of the texture interval distance to the interval variation rule includes:

[0084] Performing matching processing on the backup oral cavity image and the reference oral cavity image;

[0085] Obtain the texture region where the suspicious pixel is located, obtain the distance between the geometric center of the texture region where the suspicious pixel is located and the previous adjacent texture region as the texture interval distance of the texture region where the suspicious pixel is located, and obtain the corresponding region of the texture region where the suspicious pixel is located in the reference oral image according to the matching relationship;

[0086] The interval distance fitting value is fitted using the interval variation law of the corresponding area of ​​the texture area where the suspicious pixel is located in any reference oral image. The absolute value of the difference between the texture interval distance of the texture area where the suspicious pixel is located and the interval distance fitting value is calculated and recorded as the interval distance fitting deviation of the texture area where the suspicious pixel is located relative to the reference oral image; the inverse of the mean of the interval distance fitting deviation of the texture area where the suspicious pixel is located relative to all reference oral images is taken as the conformity of the texture interval distance to the interval variation law.

[0087] In particular, if the suspicious pixel does not exist in the texture area or the texture of the suspicious pixel does not have an area regularity or spacing regularity, it means that the suspicious pixel is not a pixel in the tooth area, and the pixels in the non-tooth area have little effect on the three-dimensional construction. In other words, the loss of a part of the pixel data in the non-tooth area has little effect on the three-dimensional construction. Therefore, if the suspicious pixel does not exist in the texture area, the conformity of the texture interval distance to the interval variation law is taken as 0.

[0088] Then we will introduce the method of obtaining the conformity of area to the law of area change.

[0089] Preferably, as an example, the conformity of the area to the area variation law includes:

[0090] The area fitting value is fitted using the area variation law of the corresponding area of ​​the texture area where the suspicious pixel is located in any reference oral image. The absolute value of the difference between the area of ​​the texture area where the suspicious pixel is located and the area fitting value is calculated and recorded as the area fitting deviation of the texture area where the suspicious pixel is located relative to the reference oral image. The inverse of the mean of the area fitting deviation of the texture area where the suspicious pixel is located relative to all reference oral images is taken as the conformity of the area to the area variation law.

[0091] In particular, if the suspicious pixel does not exist in the texture area, it means that the suspicious pixel is not a pixel in the tooth area, and the pixels in the non-tooth area have little effect on the three-dimensional construction. In other words, the loss of a part of the pixel data in the non-tooth area has little effect on the three-dimensional construction. Therefore, if the suspicious pixel does not exist in the texture area, the conformity of the area to area change law is taken as 0.

[0092] S4: removing interfering pixels in the current oral image by comparing the probability with a preset probability threshold, so as to achieve three-dimensional reconstruction of the oral cavity.

[0093] Preferably, as an example, removing interfering pixels in the current oral image by comparing the probability with a preset probability threshold to achieve oral three-dimensional reconstruction includes:

[0094] The suspicious pixels in the current oral image at each angle whose probability is greater than the preset probability threshold are removed as interference pixels. This embodiment is described by taking the preset probability threshold of 0.85 as an example, and other embodiments may take other values, which are not specifically limited in this embodiment.

[0095] The remaining pixel information is used to reconstruct the oral cavity three-dimensional model according to the method, device and storage medium for oral cavity scanning using a smart phone disclosed in the patent application document with publication number CN114283236A.

[0096] The embodiment of the present invention also discloses an intelligent scanning system based on an oral digital impression device, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the intelligent scanning method based on an oral digital impression device according to the present invention is implemented.

[0097] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0098] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory, a dynamic random access memory, a static random access memory, an enhanced dynamic random access memory, a high bandwidth memory, a hybrid storage cube, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0099] Although this specification has shown and described a number of 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. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0100] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent scanning method based on an oral digital impression device, characterized in that: Includes steps: Acquire the current oral cavity image at each angle and the historical oral cavity image at each angle; Perform matching processing on the current oral cavity images at different angles, and obtain suspicious pixels according to the pixel value differences between the matching pixels; Obtain several texture areas in historical oral images and current oral images, fit the area change law and interval change law of continuous texture areas; calculate the probability that any suspicious pixel is an interference pixel , taking a historical oral image at the same angle as the current oral image as a reference oral image, recording the current oral image containing the suspicious pixel as a backup oral image, obtaining the mean similarity between a preset window of the suspicious pixel in the backup oral image and the corresponding areas of all reference oral images as a similarity measurement value of the suspicious pixel in the backup oral image; represents the similarity measure value of the suspicious pixel in the i-th backup oral image, It represents the maximum similarity metric value of the suspicious pixel in all the backup oral images. Indicates the conformity of the texture interval distance of the texture area where the suspicious pixel is located in the i-th backup oral image to the interval variation law, Indicates the conformity of the area of ​​the texture region where the suspicious pixel is located in the i-th backup oral image to the area variation law, Indicates the preset anti-zero coefficient; The method for obtaining the conformity of the texture interval distance to the interval variation law includes: Performing matching processing on the backup oral cavity image and the reference oral cavity image; Obtain the texture region where the suspicious pixel is located, obtain the distance between the geometric center of the texture region where the suspicious pixel is located and the previous adjacent texture region as the texture interval distance of the texture region where the suspicious pixel is located, and obtain the corresponding region of the texture region where the suspicious pixel is located in the reference oral image according to the matching relationship; The interval distance fitting value is fitted using the interval variation law of the corresponding area of ​​the texture area where the suspicious pixel is located in any reference oral image, and the absolute value of the difference between the texture interval distance of the texture area where the suspicious pixel is located and the interval distance fitting value is calculated and recorded as the interval distance fitting deviation of the texture area where the suspicious pixel is located relative to the reference oral image; the inverse of the mean of the interval distance fitting deviation of the texture area where the suspicious pixel is located relative to all reference oral images is taken as the conformity of the texture interval distance to the interval variation law; The method for obtaining the conformity of the area to the area variation law includes: The area fitting value is fitted using the area variation law of the corresponding area of ​​the texture area where the suspicious pixel is located in any reference oral image, and the absolute value of the difference between the area of ​​the texture area where the suspicious pixel is located and the area fitting value is calculated and recorded as the area fitting deviation of the texture area where the suspicious pixel is located relative to the reference oral image; the reciprocal of the mean of the area fitting deviation of the texture area where the suspicious pixel is located relative to all reference oral images is taken as the conformity of the area to the area variation law; The interfering pixels in the current oral image are removed by comparing the probability with the preset probability threshold to achieve oral three-dimensional reconstruction.

2. The intelligent scanning method based on oral digital impression instrument according to claim 1, characterized in that: The obtaining of suspicious pixels according to the pixel value differences between the matching pixels includes: Every two angles form an angle pair, and the absolute value of the difference in pixel values ​​of the corresponding matching pixels of the current oral image of the two angles in the angle pair under each channel is calculated and recorded as the pixel difference of each pixel in the angle pair under each channel, and the average of the pixel differences of each pixel in the angle pair under all channels is recorded as the comprehensive pixel difference of each pixel in the angle pair; if there is a pixel whose comprehensive pixel difference is greater than a preset difference threshold, then the pixel and the pixels matching the pixel in each angle are recorded as pixels to be analyzed; The light interference degree of the pixel to be analyzed is calculated, and the pixel to be analyzed whose light interference degree is less than a preset degree threshold is recorded as a suspicious pixel.

3. The intelligent scanning method based on oral digital impression instrument according to claim 2, characterized in that: The calculating the light interference degree of the pixel to be analyzed includes: The current oral image is segmented by a superpixel segmentation algorithm to obtain several superpixel blocks. The superpixel block where any pixel to be analyzed is located is obtained as a reference superpixel block. A Gaussian mixture model is fitted to the pixel values ​​of the reference superpixel blocks in the current oral image at each angle, and the KL divergence of the Gaussian mixture model of the reference superpixel blocks of the current oral image at every two angles is calculated. The normalized value of the mean of the KL divergence of the Gaussian mixture model of the reference superpixel blocks of the current oral image at all two angles is used as the light interference degree of the pixel to be analyzed.

4. The intelligent scanning method based on oral digital impression instrument according to claim 1, characterized in that: The obtaining of several texture regions in the historical oral image and the current oral image includes: Perform edge detection on historical oral images to obtain edge pixels in the corresponding images; After morphological processing of edge pixels, the closed area surrounded by edge pixels is obtained and recorded as texture area; Get the texture area in the current oral image.

5. The intelligent scanning method based on oral digital impression instrument according to claim 1, characterized in that: The area variation law and interval variation law of the fitted continuous texture region include: The texture area that is continuous and has no gaps is recorded as a continuous texture area; According to the arrangement order of the continuous texture regions, the areas of all the texture regions in the continuous texture regions are fitted with polynomials to obtain the area variation law of the continuous texture regions; Obtain the interval distance between the geometric center of each texture area and the previous texture area in the continuous texture area and record it as the texture interval distance of each texture area in the continuous texture area; According to the arrangement order of the continuous texture regions, the texture interval distances of all the texture regions in the continuous texture region are fitted with a polynomial to obtain the interval variation rule of the continuous texture region.

6. Intelligent scanning system based on oral digital impression instrument, 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, the intelligent scanning method based on the oral digital impression instrument according to any one of claims 1 to 5 is implemented.

Images