Denture model generation system based on image processing
Technical Field
The invention relates to the technical field of image processing, in particular to a false tooth model generating system based on image processing.
Background
Because of the factors such as congenital factors, acquired living habits and accidents, etc., the current many people need to carry out medical treatment activities of dental correction and denture installation, in order to improve the working efficiency and quality of the dental correction and denture installation work, the current medical staff is often required to firstly establish a corresponding model for the oral cavity structure of a patient before diagnosis and treatment, and then carry out corresponding denture modeling, installation and dental correction work according to the established model.
Referring to Chinese patent, the patent name is: an implant modeling method and system based on image recognition technology (patent publication No. CN115984470A, patent publication No. 2023.04.18), the method comprises: determining a target tooth position and attribute data thereof according to the first image data, and determining second image data according to the attribute data; according to the method, the curve data corresponding to the target tooth position is determined according to the second image data, the implant modeling is carried out on the target tooth position according to the curve data, the image recognition technology is utilized to extract the image of the oral cavity of the user, the position of the tooth to be implanted and attribute data thereof are automatically recognized, and then the curve data related to the position of the tooth to be implanted are extracted, so that the implant model with higher adaptation to the peripheral teeth can be determined, the adaptation of the implant is greatly improved, and more personalized and more comfortable tooth experience is brought to the user.
In the existing denture model generation process, the denture is generally manufactured by adopting a 3D printing technology, but the following problems still exist:
at present, a denture needs to be matched and customized for a customer, the denture is manufactured through 3D printing, the user needs to be sampled before customization, an image recognition processing is carried out through a scanner to automatically generate a denture model, then the denture model is optimized to obtain an exact model, and on the basis that after image recognition, the obtained image definition has defects in the image processing process, more shadow parts exist in the image, and whether the shadow parts are protruding or recessed entities can not be accurately judged through shadow part comparison, so that the formed denture model has errors in accuracy.
To this end, the present invention provides a denture model generation system based on image processing.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a false tooth model generation system based on image processing, which solves the problem that in the existing 3D false tooth generation operation, the obtained image definition has defects in the process of processing the image, more shadow parts exist in the image, and whether the shadow parts are raised or recessed entities cannot be accurately judged through the shadow part comparison, so that the formed false tooth model precision has errors.
In order to achieve the above purpose, the invention is realized by the following technical scheme: a denture model generation system based on image processing, comprising a 3D scanner, an image forming system, a denture model building module and a denture generating device, wherein the 3D scanner is used for scanning the oral cavity dental site of a custom person and forming image data, the denture model building module is used for combining the data processed by the image forming system to form a denture model, the denture generating device is used for printing the denture model to generate a entity of the denture model, and the 3D scanner, the image forming system, the denture model building module and the denture generating device sequentially realize bidirectional data interaction operation, the image forming system comprises:
the image receiving module is used for receiving the pictures formed by scanning and transmitting the pictures to the image processing module for image processing operation through classification;
the image processing module is used for adjusting the definition of the image, improving the resolution of the image and combining the image with the improved definition to form a new picture;
the digital conversion module is used for converting the image signals of the plurality of new photos into digital signals, inputting the digital signals into the denture model building module, and forming the 3D denture model through the combination of the digital signals.
Preferably, the image processing module includes:
an image segmentation unit for preprocessing the image and segmenting the image according to the gray threshold valueIs divided into a plurality of image blocksRemoving background image blocks by adopting blocks in the segmentation process, and obtaining dental image blocks;
the image enhancement unit is used for expanding channels in the feature images of the low-resolution dental image blocks by inputting the dental image blocks into the scale-expanding model, compensating the expanded errors and obtaining the high-resolution dental image blocks;
the image analysis unit is used for judging the specific information of the dental image block and transmitting the specific information to the next unit by extracting the characteristic part of the high-resolution dental image block and performing threshold comparison operation after the characteristic part is subjected to gray threshold segmentation operation;
and the data summarizing unit is used for forming an obtained amplified image by splicing and combining the amplified high-resolution dental image blocks, and obtaining a 3D false tooth image by mapping the characteristics on each image.
Preferably, the image blocks in the image segmentation unitThe specific operation mode of gray threshold segmentation is as follows:
s1, dividing each image block through background gray level distributionRemoving the background image blocks by the combination of the background image blocks and the dental image blocks, and leaving the dental image blocks;
s2, calculating variances of the dental image blocks, and arranging the variances of the dental image blocks into an ordered sequence from small to largeThen the dental image blocks with the variance of 50% of the ordered sequences from small to large are removed, and the rest of the ordered sequences are removedThe tooth image block of the tooth part is marked;
s3, calculating an average threshold value T of the dental image blocks, and performing sequential sequenceIn (3) a pixelPerforming S-shaped traversal operation, and traversing the obtained pixel valuesAnd comparing the image characteristics of the dental image blocks with the average threshold T.
Preferably, the variance calculation formula in S2 is:;
the image mean value calculation formula of the dental image block is as follows:
where a is the threshold value of image block X and n is the number of image blocks X.
Preferably, the comparison method of the pixel value in S3 and the average threshold T is as follows:
if the pixel value isThe following condition is satisfied, then the pixels in the dental image block of the ordered sequence GMarked as dental concave profile:;
if the pixel value isThe following condition is satisfied, then the pixels in the dental image block of the ordered sequence GMarked as tooth convex profile:。
preferably, the image enhancement unit enlarges the low-resolution dental image block to a target size by:
z1, according to the 50% ordered sequence from large to smallDispersing the low-resolution dental image blocks, extracting the characteristic parts of the marks of the images, and simultaneously establishing an image scale-up model for amplifying the images;
z2, inputting the extracted characteristic part into a scale-up model, and expanding a channel in a characteristic diagram of the low-resolution dental image block;
and Z3, calculating and compensating errors of the feature map obtained after expansion, and finally obtaining the high-resolution dental image block.
Preferably, the error calculation formula of the feature map obtained after expansion in Z3 is:
where i represents the number of measurements, q represents the number of output channels, and k is the basic constant of the feature map.
Preferably, the resolution calculation formula of the high-resolution dental image block is as follows:
wherein U is the length pixel number; v is the number of width pixels; w is the screen size, i.e. diagonal length.
Preferably, the feature mapping specific mode of the data summarizing unit is as follows: each layer of learning based on the neural network model is oneMapping of (2) instead ofI.e. the initial input is added at the end of each layer, where the input is d and the output isThenThe value of (2) tends to be 0 and as the number of layers increases, the errorAnd the high-resolution mapping image block is obtained without change.
Preferably, the data summarizing unit performs stitching and combining on the high-resolution mapping image blocks with the same resolution obtained through calculation according to the initial sequence of the image blocks to form a plurality of new high-resolution images, and links the high-resolution images to form a three-dimensional high-resolution denture image.
The invention provides a false tooth model generating system based on image processing. Compared with the prior art, the method has the following beneficial effects:
(1) The denture model generating system based on image processing is provided with an image forming system, an image receiving module is used for receiving and classifying images, the image processing module is used for adjusting the definition of the images and improving the resolution of the images, the images with improved definition are combined to form a new image, the digital conversion module is used for converting high-resolution image signals into digital signals, the digital signals are input into a denture model building module, and a 3D denture model is formed through the combination of the digital signals, so that the definition and the resolution of the images are improved in the image processing process, and the generated 3D denture model is more accurate.
(2) The denture model generating system based on image processing comprises an image dividing unit for preprocessing an image, dividing the image according to gray threshold valueThe method comprises the steps of dividing the sequence of the tooth image block into a plurality of image blocks, removing background image blocks by adopting blocks in the dividing process, obtaining tooth image blocks, and realizing the condition of primarily judging the image characteristics of the tooth image blocks, so as to judge whether the image elements in the tooth image blocks are concave surfaces or convex surfaces, realize the marking of uncertain image characteristics, improve the information accuracy in image processing, reduce the processing workload and perform optimization operation on key tooth images.
(3) According to the denture model generating system based on image processing, the image enhancement unit is arranged, the channels in the feature images of the low-resolution dental image blocks are enlarged through inputting the dental image blocks into the scale-enlarging model, error calculation and compensation are carried out on the feature images obtained after enlargement, the high-resolution dental image blocks are obtained, the sharpness of the dental image blocks with larger errors is improved, feature identification is carried out after error compensation is completed, and the high-resolution dental image blocks with accurate information are judged again, so that the follow-up denture models are more attached.
Drawings
Fig. 1 is a schematic block diagram of a denture model generating system according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the present invention provides three technical solutions:
an embodiment is a denture model generating system based on image processing, including a 3D scanner, an image forming system, a denture model building module and a denture generating device, wherein the 3D scanner is used for scanning oral cavity teeth of a customized person and forming image data, the denture model building module is used for combining the data processed by the image forming system to form a denture model, the denture generating device is used for printing the denture model to generate a entity of the denture model, and the 3D scanner, the image forming system, the denture model building module and the denture generating device sequentially realize bidirectional data interaction operation, the image forming system includes:
the image receiving module is used for receiving the pictures formed by scanning and transmitting the pictures to the image processing module for image processing operation through classification;
the image processing module is used for adjusting the definition of the image, improving the resolution of the image and combining the image with the improved definition to form a new picture;
the digital conversion module is used for converting the image signals of the plurality of new photos into digital signals, inputting the digital signals into the denture model building module, and forming the 3D denture model through the combination of the digital signals.
The 3D scanner and the denture generating equipment are matched equipment, are in the prior art and are electrically connected with an external power supply, and finish model generating operation after scanning and subsequent data importing.
The image forming system is arranged, the image receiving module is used for receiving and classifying images, the image processing module is used for adjusting the definition of the images and improving the resolution of the images, the images with the improved definition are combined to form a new image, the digital conversion module is used for converting high-resolution image signals into digital signals, the digital signals are input into the denture model building module, and the digital signals are combined to form a 3D denture model, so that the definition and the resolution of the images are improved in the image processing process, and the generated 3D denture model is more accurate.
In an embodiment of the present invention, an image processing module includes:
an image segmentation unit for preprocessing the image and segmenting the image according to the gray threshold valueIs divided into a plurality of image blocksRemoving background image blocks by adopting blocks in the segmentation process, and obtaining dental image blocks;
the image enhancement unit is used for expanding channels in the feature images of the low-resolution dental image blocks by inputting the dental image blocks into the scale-expanding model, compensating the expanded errors and obtaining the high-resolution dental image blocks;
the image analysis unit is used for judging the specific information of the dental image block and transmitting the specific information to the next unit by extracting the characteristic part of the high-resolution dental image block and performing threshold comparison operation after the characteristic part is subjected to gray threshold segmentation operation;
and the data summarizing unit is used for forming an obtained amplified image by splicing and combining the amplified high-resolution dental image blocks, and obtaining a 3D false tooth image by mapping the characteristics on each image.
The output end of the image segmentation unit is connected with the input end of the image enhancement unit, the output end of the image enhancement unit is connected with the input end of the image analysis unit, and the output end of the image analysis unit is connected with the input end of the data summarization unit, so that the image segmentation unit is used for firstly segmenting a plurality of image blocks, removing background image blocks, leaving dental image blocks with small variance, and then traversing pixel values in the image blocksComparing with the average threshold value T to judge the image characteristics of the dental image blocks, knowing whether the mark is a concave surface or a convex surface, dispersing the dental image blocks with low resolution and large variance, extracting the characteristic parts of the marks of the images, and simultaneously establishing a graphThe image scale-up model is used for amplifying the image, compensating the resolution of the feature map after amplifying the resolution, obtaining a high-resolution dental image block, carrying out segmentation operation again based on the high-resolution dental image block, outputting the image block with specific information, transmitting the image block to the data summarizing unit, reducing the error after resolution conversion through the mapping learning of the neural network model, finally splicing and combining the processed high-resolution mapping image blocks to form a plurality of new high-resolution images, and forming a three-dimensional high-resolution denture image by linking the high-resolution images.
In the embodiment of the invention, the image blocks in the image segmentation unitThe specific operation mode of gray threshold segmentation is as follows:
s1, dividing each image block through background gray level distributionRemoving the background image blocks by the combination of the background image blocks and the dental image blocks, and leaving the dental image blocks;
s2, calculating variances of the dental image blocks, and arranging the variances of the dental image blocks into an ordered sequence from small to largeThen the dental image blocks with the variance of 50% of the ordered sequences from small to large are removed, and the rest of the ordered sequences are removedThe tooth image block of the tooth part is marked;
s3, calculating an average threshold value T of the dental image blocks, and performing sequential sequenceIn (3) a pixelPerforming S-shaped traversal operation, and traversing the obtained pixel valuesAnd comparing the image characteristics of the dental image blocks with the average threshold T.
In the embodiment of the present invention, the variance calculation formula in S2 is:;
the image mean value calculation formula of the dental image block is as follows:
where a is the threshold value of image block X and n is the number of image blocks X.
In the embodiment of the invention, the comparison method of the pixel value in S3 and the average threshold T is as follows:
if the pixel value isThe following condition is satisfied, then the pixels in the dental image block of the ordered sequence GMarked as dental concave profile:;
if the pixel value isThe following condition is satisfied, then the pixels in the dental image block of the ordered sequence GMarked as tooth convex profile:。
wherein, the image segmentation unit is arranged to preprocess the image and then segment the image according to the gray threshold valueIs divided into a plurality of sequencesThe image blocks are used for eliminating background image blocks by adopting blocks in the segmentation process, so that dental image blocks are obtained, the condition of primarily judging the image characteristics of the dental image blocks is realized, and therefore, the condition that the image elements in the dental image blocks are concave surfaces or convex surfaces is judged, and the marking of uncertain image characteristics is realized, so that the information accuracy in image processing is improved, the processing workload is reduced, and the optimization operation is performed on key dental images.
The difference between the second embodiment and the first embodiment is that:
in the embodiment of the invention, the operation steps of amplifying the low-resolution dental image block to the target size in the image enhancement unit are as follows:
z1, according to the 50% ordered sequence from large to smallDispersing the low-resolution dental image blocks, extracting the characteristic parts of the marks of the images, and simultaneously establishing an image scale-up model for amplifying the images;
z2, inputting the extracted characteristic part into a scale-up model, and expanding a channel in a characteristic diagram of the low-resolution dental image block;
and Z3, calculating and compensating errors of the feature map obtained after expansion, and finally obtaining the high-resolution dental image block.
In the embodiment of the invention, the error calculation formula of the feature map obtained after expansion in Z3 is as follows:
where i represents the number of measurements, q represents the number of output channels, and k is the basic constant of the feature map.
In the embodiment of the invention, the resolution calculation formula of the high-resolution dental image block is as follows:
wherein U is the length pixel number; v is the number of width pixels; w is screen rulerI.e. diagonal length.
Through being provided with the image enhancement unit, through inputting the tooth image piece into the scale-up model, enlarge the passageway in the feature map of low resolution tooth image piece, carry out the calculation and the compensation of error to the feature map that obtain after the enlargement to obtain high resolution tooth image piece, with this improvement to the great tooth image piece of error carries out the definition in the completion, thereby carry out the feature identification again after accomplishing the compensation to the error, judge again and obtain the high resolution tooth image piece of accurate information, make subsequent artificial tooth model more laminate.
The difference between the third embodiment and the first embodiment is that:
in the embodiment of the invention, the characteristic mapping specific mode of the data summarizing unit is as follows: each layer of learning based on the neural network model is oneMapping of (2) instead ofI.e. the initial input is added at the end of each layer, where the input is d and the output isThenThe value of (2) tends to be 0 and as the number of layers increases, the errorAnd the high-resolution mapping image block is obtained without change.
In the embodiment of the invention, the data summarizing unit splices and combines the high-resolution mapping image blocks with the same resolution obtained by calculation according to the initial sequence of the image blocks to form a plurality of new high-resolution images, and links the high-resolution images to form a three-dimensional high-resolution denture image.
And all that is not described in detail in this specification is well known to those skilled in the art.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.