CN110991339A - Three-dimensional puckery palate identification method adopting circular spectrum - Google Patents
Three-dimensional puckery palate identification method adopting circular spectrum Download PDFInfo
- Publication number
- CN110991339A CN110991339A CN201911216905.1A CN201911216905A CN110991339A CN 110991339 A CN110991339 A CN 110991339A CN 201911216905 A CN201911216905 A CN 201911216905A CN 110991339 A CN110991339 A CN 110991339A
- Authority
- CN
- China
- Prior art keywords
- palate
- dimensional
- puckery
- ruffles
- wrinkle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the technical field of legal medical expert identification of human biological characteristics, and the specific technical scheme is as follows: a three-dimensional puckery palate recognition method adopting a cyclic frequency spectrum comprises the following specific steps: acquiring three-dimensional puckery palate information by adopting a portable color intraoral scanner; carrying out equidistant slicing on the acquired three-dimensional puckery palate image; corroding and binarizing the wrinkled palate slice image; constructing a palate wrinkle section database; extracting the circular spectrum characteristics of a palate wrinkle tangent line in the slice image; partitioning the extracted palatal wrinkle tangent cyclic frequency spectrum characteristic diagram by adopting a 4 multiplied by 4 uniform partitioning scheme, and further constructing a palatal wrinkle tangent cyclic frequency spectrum characteristic vector; finally, three-dimensional puckery palate recognition is carried out; the isometric slicing idea provided by the method simplifies the complexity of the three-dimensional puckery palate image recognition problem, can better reflect the essential characteristics of puckery palate tangent lines, and avoids the data disaster problem by blocking the cyclic spectrum characteristics.
Description
Technical Field
The invention belongs to the technical field of forensic identification of human body biological characteristics, and particularly relates to a three-dimensional pucker-palate recognition method for forensic identification by adopting equidistant slicing and cyclic spectrum characteristic extraction.
Background
The palate is an irregular, asymmetric ridge line located in the anterior third of the hard palate at the top of the human mouth, extending outward from the mastoid incisors and anterior to the median palatine suture. Biometric identification is a technique that utilizes a person's physiological or behavioral characteristics to authenticate the person. As a new biometric feature, pucker palate is equivalent to the role played by fingerprints, human faces, irises, and the like in biometric identification. The palate wrinkles are unique for each person, and the palate wrinkles keep the shape of the person unchanged in a life and only change the length and width of the person due to normal growth and development. The puckery palate meets the requirements of biological characteristics for inheritance, stability, collectability and uniqueness.
Biometric identification technology plays a very important role in some huge disasters and criminal cases (e.g. major aviation accidents, natural disasters, industrial explosions, etc.). Because the fingerprints, the human faces, the irises and the like are easily influenced by factors such as temperature, humidity, huge destructive power and the like, the fingerprints, the human faces, the irises and the like are difficult to store completely, and certain difficulty is brought to personal identification. The pucker palate is protected from damage in the presence of trauma and high temperatures by soft and hard tissues in the oral cavity (e.g., lips, cheeks, teeth, and bones). Research shows that the palate wrinkles can resist the damage of third degree burn, and within 7 days of death, the palate wrinkles have strong rancidity resistance and can still maintain a relatively complete shape. Thus, pucker palate can be used as a biometric for personal identification.
At present, researchers mainly focus on the fields of anthropology, genetics, forensic dentistry, stomatology, forensic dentistry, anatomy and the like, and mainly perform statistical classification research on the form, the number, the length, the sex, the race and the like of the puckery palate, but the research on the aspect of three-dimensional digital image recognition is less, and a set of complete theoretical system is not formed. In 1732, Winslow first proposed the anatomical concept of "pucker palate". In 1889, Allen et al first used pucker palate as a tool for identification in the study of pucker palate. In 1955, Lysell developed the first classification system of puckering palate. Subsequently, a large number of researchers classified the shape, number, length, direction, etc. of the wrinkles palate, and various types of systems for classifying the wrinkles palate were invented and used for identification of individuals, among which classification methods of Kapali and Thomas, etc. are most commonly used. In 2007, several researchers systematically summarized and organized the most commonly used classification methods, such as the Gamea classification method and the Martins dos Santos classification method, based on the classification of the wrinkles palate in the past. In 2010, Hamath M used software MS paint version5.1 to extract and match information from the pucker palate image. In 2018, Bernitz et al performed the same identification study on 1 forensic sample using the radiation conversion principle, and again emphasized the advantage of puckery palate as an identification tool when no DNA or the like could be obtained for the same identification study. Compared with the foreign research work on identifying the puckery palate, the domestic research on identifying the puckery palate starts later. In 2015, panfei et al discussed that pucker palate has three major characteristics of stability, variability and breadth that are legally medically recognized. In 2016, Wuxiduch et al established the same established digital system of rugose palate forensic medicine and evaluated its effectiveness. In 2017, jiayu discusses the importance of puckery palate in oral forensic medicine. In the same year, plum ice demonstrated the feasibility of forensic stomatology identity determination using digital images of pucker palate. In general, studies of two-dimensional wrinkle palate images by scholars at home and abroad mainly focus on extracting or counting the shape, number, length and fringe edge characteristics of wrinkle palate stripes.
To date, there has been relatively little research on the identification of three-dimensional digital images of ruffles. The traditional method collects three-dimensional puckery palate data by manufacturing a gypsum model, and has the problems of easy damage, time consumption in collection, difficult storage and the like. In 2015, Taneva E D makes the puckery palate into a plaster model, and then scans the model into a three-dimensional puckery palate digital image through a scanning instrument, and finally, the three-dimensional puckery palate digital image is used for feature matching of the image. In 2016, Batool Ali et al used Thamas and Kotze classification to evaluate the length and shape of the creases in the palate three-dimensional Gypsum model after orthodontic treatment in patients. In 2017, Daniele Gibelli et al verified the uniqueness of the three-dimensional model of pucker palate and applied it to identity determination. In the same year, Wu Xiao Xue uses digital camera to photograph the palate wrinkle plaster model, encodes the number of palate wrinkles and their duty ratio, and finally carries out feature matching research. In 2019, jiayu uses a three-dimensional digital puckery palate model to research the correlation between puckery palate type and gender and blood type, and explores the possibility of identifying gender by using the length and shape of puckery palate.
Disclosure of Invention
In order to solve the technical problem that the existing three-dimensional puckery palate digital image recognition technology is deficient, the invention provides a method for human body biological characteristic forensic identification, which can complete three-dimensional puckery palate recognition, and the method is accurate and rapid and has better recognition effect.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows: a three-dimensional puckery palate recognition method adopting a cyclic frequency spectrum specifically comprises the following steps:
step one, collecting three-dimensional puckery palate information. Specifically, the acquisition equipment is a portable color intraoral scanner (comprising a scanning gun, a USB key, a software U disk, a POD base, a color calibration head and a notebook computer), the dental chair is adjusted to be located at a position where the lower jaw of a volunteer is parallel to the ground and the upper jaw of the volunteer is perpendicular to the ground, the volunteer and the acquisition equipment are respectively disinfected, then the head of the scanning gun is extended into the oral cavity of the volunteer and moved along the teeth for scanning, whether each surface of the palate wrinkles is complete or not is checked while scanning, if the wrinkles exist or the scanning is incomplete, the collection equipment is subjected to supplementary scanning, and after the scanning is completed, data are guided into the computer to acquire three-dimensional palate wrinkle data.
And step two, carrying out equidistant slicing on the puckery palate image. The slicing principle is as follows: and intersecting the three-dimensional puckery palate model with a horizontal slice plane for slicing, wherein equidistant slicing refers to generating slices with constant intervals and fixed thickness. The slicing operation comprises the following specific steps: (1) firstly, opening a three-dimensional puckery palate image by using 3Shape software, measuring the length, width and height of the puckery palate image by using a rectangular frame of the software, and preparing uniform slices of the three-dimensional puckery palate image by using a circular tangent plane; (2) firstly, adjusting the position of a rectangular frame to ensure that the upper surface and the lower surface of a palate wrinkle image are overlapped (namely the upper surface and the lower surface of the rectangular frame are overlapped), connecting protruding points of three teeth from the left side and the right side of an incisor mastoid to form a cutting straight line in a plan view, then starting cutting from five teeth from the left side and the right side of the incisor mastoid by adopting a circular incisor line, continuously adjusting the positions of the circular incisor line and the rectangular frame before starting equidistant slicing to ensure that the circular incisor line is parallel to the cutting straight line, and finally carrying out equidistant slicing along the middle point of the circular incisor line and the direction of the incisor mastoid (note: a vertical line is the slicing direction, the length of the vertical line is between 39cm and 40cm from bottom to top, and the vertical line is vertical to; (3) the constant interval adopted by the invention is 1cm, equidistant slicing operation is carried out on three-dimensional pucker palate once every 1cm, the operation is carried out in sequence, 40 slicing operations are carried out totally, after equidistant slicing, 40 two-dimensional pucker palate slice images are obtained in sequence and stored in a jpg format, and the 40 slice images jointly form a pucker palate slice sample set.
Step three, constructing a palate wrinkle section database, establishing a palate wrinkle section image database according to unique characteristics of palate wrinkle images, acquiring 5 palate wrinkle samples for 91 persons, performing the equidistant section operation of the step two on each palate wrinkle sample, and further acquiring 40 palate wrinkle equidistant section curve images of each palate wrinkle sample, thereby obtaining 455 palate wrinkle samples and 18200 palate wrinkle section images of 91 persons. Where 5 samples of ruffles per person were three-dimensional ruffles images taken at different times, the same light-dark background, and with a slight rotation operation added.
And step four, corroding and binarizing the palate wrinkle slice image, wherein in order to eliminate irrelevant information in the image, recover useful real information, enhance the detectability of relevant information and simplify data to the maximum extent, so that the reliability of palate wrinkle characteristic extraction, palate wrinkle image segmentation, palate wrinkle matching and identification is improved, and the method corrodes and binarizes the palate wrinkle slice image in the database obtained in the step three.
And step five, extracting the circular spectrum characteristics of the tangent line of the ruffle in the slice image, constructing a ruffle characteristic vector by extracting the circular spectrum characteristics of the tangent line of the ruffle, specifically, firstly scanning the image of the tangent line of the ruffle to obtain one-dimensional components in the horizontal direction and the vertical direction, and further respectively calculating the spectrum correlation functions of the two one-dimensional components by adopting cumulative Fourier transform.
And step six, constructing circular spectrum characteristic vectors of the palate wrinkles tangent line, and in order to avoid dimension disasters and facilitate subsequent identification processing, the method adopts a 4 x 4 uniform partitioning scheme to partition the circular spectrum characteristic diagram of the palate wrinkles tangent line extracted in the step five. The blocking principle is as follows: and uniformly partitioning the blocks based on the positions of the pixel points, solving the mean value and the variance of each small block, and finally converting the mean value and the variance into one-dimensional row vectors in sequence.
And seventhly, identifying the three-dimensional puckery palate, wherein the three-dimensional puckery palate identification system adopted by the method comprises two components of a three-dimensional puckery palate image training process and a three-dimensional puckery palate image testing process, and each component comprises a three-dimensional puckery palate pretreatment module, a puckery palate tangent circulation frequency spectrum characteristic extraction module, a puckery palate tangent circulation frequency spectrum characteristic partitioning module and a puckery palate tangent characteristic vector classification module.
The isometric slicing idea provided by the method simplifies the complexity of the three-dimensional puckery palate image recognition problem, meanwhile, the circular spectrum features are adopted to construct the feature vectors, the essential features of the puckery palate tangent line can be better reflected, the blocking processing of the circular spectrum features avoids the occurrence of the data disaster problem, and in the actual operation process, the method can reduce the complexity of the three-dimensional puckery palate recognition problem calculation and accelerate the operation speed.
Drawings
Fig. 1 is a three-dimensional digital image of pucker palate.
Figure 2 is a schematic view of the circular cut lines of the pucker palate in the 3Shape software.
Fig. 3 is a schematic diagram of a standard top view of a three-dimensional crumpled-palate slice.
Fig. 4 is a schematic diagram of a three-dimensional palatoglossus equidistant slicing process.
Fig. 5 is a sample set of slices of a pucker palate.
Fig. 6 is a schematic diagram of the erosion and binarization processing of a palate fold tangent line image.
Fig. 7 is a schematic diagram of the extraction process of the palatoglossus tangent circulation spectrum characteristics.
Fig. 8 is a schematic diagram of a three-dimensional puckering recognition system.
Fig. 9 is a block diagram of a line-tangent circulation spectrum of puckery palate.
Fig. 10 is a schematic diagram of the construction process of the palatople fold tangent cycle spectrum feature vector.
Fig. 11 is a schematic diagram of a three-dimensional puckering recognition system.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
A three-dimensional puckery palate recognition method adopting a cyclic frequency spectrum comprises the following specific steps:
step one, collecting three-dimensional puckery palate information. Specifically, the acquisition equipment is a portable color intraoral scanner (comprising a scanning gun, a USB key, a software U disk, a POD base, a color calibration head and a notebook computer), the position of a dental chair is adjusted to be in a position where the lower jaw of a volunteer is parallel to the ground and the upper jaw of the volunteer is perpendicular to the ground, the volunteer and the acquisition equipment are respectively disinfected, then a head of the scanning gun is extended into the oral cavity of the volunteer to move and scan along teeth, whether each surface of the palate wrinkle is complete or not is checked while scanning, if the surface is loophole or incomplete, the collection equipment is subjected to complement scanning, after the scanning is finished, data is introduced into a computer to obtain three-dimensional palate wrinkle data, as shown in figure 1, compared with a traditional two-dimensional palate wrinkle image, the three-dimensional palate image can reflect the distribution of the palate wrinkles in a three-dimensional, intuitive and detailed manner, interference generated by lips, saliva, light rays and the like can be avoided, and the palate wrinkle image obtained, the method is more suitable for the real distribution condition of the wrinkled palate stripes in the oral cavity of the human body, the arrangement is orderly and not concentrated, each wrinkled palate stripe is clearer, and the detailed information of the wrinkled palate can be better kept.
And step two, carrying out equidistant slicing on the puckery palate image. Firstly, reducing the complexity of the problem, and carrying out equidistant slicing processing on the three-dimensional puckery palate image, wherein the slicing principle is as follows: and (3) intersecting the three-dimensional puckery palate model with horizontal slice planes for slicing, wherein each horizontal slice plane gives a piecewise linear contour of a slice, and equidistant slicing means that slices with constant intervals and fixed thickness are generated. The slicing operation comprises the following specific steps: (1) firstly, opening a three-dimensional crumple palate image by using 3Shape software, measuring the length, the width and the height of the crumple palate image by using a rectangular frame of the software, and preparing a uniform slice of the three-dimensional crumple palate image by using a circular cutting surface line (shown in figure 2); (2) as shown in fig. 3, the position of the rectangular frame is adjusted to make the upper and lower surfaces of the ruffle image coincide (i.e. the upper and lower surfaces of the rectangular frame coincide), in the top view, three teeth are respectively connected to form a cutting straight line from the left and right of the incisor mastoid, then the cutting is started from the five teeth of the left and right of the incisor mastoid by using the circular cutting line, before starting the equidistant slicing, the positions of the circular cutting line and the rectangular frame are continuously adjusted to make the circular cutting line parallel to the cutting straight line, and finally, the equidistant slicing is carried out along the middle point of the circular cutting line and the direction of the incisor mastoid (note: the vertical line is the slicing direction, from bottom to top, the length of the vertical line is about 39cm to 40cm, and the vertical line is vertical to the circular cutting; (3) as shown in fig. 4, the constant interval adopted by the present invention is 1cm, equidistant slicing operation is performed on three-dimensional pucker palate once every 1cm, the operations are sequentially performed for 40 times, after equidistant slicing, 40 two-dimensional pucker palate slice images are sequentially obtained and stored in a jpg format, and the 40 slice images together form a slice sample set of pucker palate, as shown in fig. 5.
And step three, constructing a palate wrinkle section database. According to the unique characteristics of the wrinkle palate images, a wrinkle palate section image database is established, 5 wrinkle palate samples are collected for 91 persons, the equidistant slicing operation in the second step is carried out on each wrinkle palate sample, and then 40 equal-distance wrinkle palate section curve images of each wrinkle palate sample can be obtained, so that 455 wrinkle palate samples and 18200 wrinkle palate section images of 91 persons can be obtained. Where 5 samples of ruffles per person were three-dimensional ruffles images taken at different times, the same light-dark background, and with a slight rotation operation added.
And step four, corroding and binarizing the wrinkled palate slice image. In order to eliminate irrelevant information in the image, recover useful real information, enhance the detectability of relevant information and simplify data to the maximum extent, thereby increasing the reliability of extraction of the palate wrinkle characteristic, segmentation of the palate wrinkle image, matching and identification of the palate wrinkles, the method carries out corrosion and binarization processing on the palate wrinkle slice image in the database obtained in the third step, as shown in fig. 6, it can be found that the original palate wrinkle curve image has lower brightness and grid lines, which can have certain influence on the identification result, and the change degree of the palate wrinkle curve bulge can be reflected by the change degree of the palate wrinkle stripes.
And step five, extracting the circular spectrum characteristics of the wrinkle palate tangent line in the slice image. The method comprises the steps of extracting circular spectrum features of a wrinkle palate tangent line to construct a wrinkle palate feature vector, specifically, scanning an wrinkle palate tangent line image to obtain one-dimensional components in the horizontal and vertical directions, and calculating spectrum correlation functions of the two one-dimensional components by adopting cumulative Fourier transform. Assuming a cyclic autocorrelation function of the discrete-time signal x (n) of
The spectral correlation function is:
in the formula (1), a is a cycle frequency, and f is a spectrum frequency. When a is 0, the signal has finite average power
The symmetry relationship and the discrete correlation periodicity of the spectral correlation function are shown in equations (3) and (4):
in the formula (4), m and n are arbitrary integers.
The cyclic spectrum is a complex spectrum containing phase information not contained in the power spectrum, and thus frequency information and phase information of the signal can be estimated by using a cyclic spectrum algorithm. Further, the spectral correlation function expression is as shown in formula (5):
in order to improve the calculation efficiency of the cyclic spectrum estimation of the signal and reduce the calculation amount, the spectrum correlation function is calculated by utilizing an efficient calculation algorithm accumulated Fourier transform, the idea of the accumulated Fourier transform algorithm is to divide the total observation time length of the signal into a plurality of small sections, and then, the fast Fourier transform algorithm is used for carrying out cyclic frequency movement on each small section. The spectral correlation function calculated by the cumulative fourier transform is:
in the formula (6), the reaction mixture is,
performing discrete digitization processing on equation (6), equation (6) becomes equation (8):
in formula (8):
for computational convenience, only the signal x (n) is considered as a real sequence. When discretized, the sequence has a frequency of
The flow of extraction of the tangent circulation spectrum of the pucker palate is shown in fig. 7, and taking an image of a tangent of the pucker palate sample No. 37 in the sample library as an example, the tangent circulation spectrum of the pucker palate is shown in fig. 8. And performing cyclic spectrum estimation on the one-dimensional signal components of the tangent line of the pucker palate in the horizontal direction and the vertical direction to obtain a spectrum correlation function. On the contour line schematic diagram and the spectrum correlation function three-dimensional schematic diagram, it can be seen that the partial feature extraction effect on the double frequency domain is very good.
And step six, constructing a circular spectrum characteristic vector of the palate wrinkle tangent line. The cyclic spectrum characteristics of the tangent to the pucker palate can be obtained by using an accumulated fourier transform algorithm, and how to construct a feature vector for identification is an important problem. If the dual-frequency plan constructed by using the cumulative Fourier transform algorithm has 64 cyclic frequency spectrum estimation feature blocks, the energy and standard deviation of each small block are calculated, and then a feature vector is constructed, wherein the dimension of the feature vector is 128(64 x 2), 40 palate wrinkle slice images are provided for each palate wrinkle sample in the method, and if the feature vector is constructed directly according to the method, the dimension of the feature vector is high, which is not beneficial to classification calculation on one hand, and dimension disasters can be caused on the other hand. Therefore, in order to avoid dimension disasters and facilitate subsequent identification processing, the method adopts a 4 × 4 uniform partitioning scheme to partition the circular spectrum characteristic diagram of the wrinkle palate tangent line extracted in the step five, as shown in fig. 9, the partitioning principle is as follows: the blocks are evenly partitioned based on the pixel point positions, then the mean value and the variance of each small block are solved, finally the mean value and the variance are sequentially converted into one-dimensional row vectors, and further, a schematic diagram of the construction process of the wrinkle palate tangent line circular spectrum feature vector is given in fig. 10. Compared with the traditional method for constructing the feature vector, the method based on the block can realize the purpose of feature dimension reduction while constructing the feature vector, and can reduce the complexity of calculation and accelerate the operation speed.
And seventhly, identifying three-dimensional puckery palate. The three-dimensional puckery palate recognition system adopted by the method comprises two components of a three-dimensional puckery palate image training process and a three-dimensional puckery palate image testing process, wherein each component also comprises a three-dimensional puckery palate preprocessing module, a puckery palate tangent line circulation frequency spectrum feature extraction module, a puckery palate tangent line circulation frequency spectrum feature partitioning module and a puckery palate tangent line feature vector classification module, and a schematic diagram of the three-dimensional puckery palate recognition system is given in fig. 11.
The three-dimensional puckery palate recognition method comprises the following specific steps:
firstly, dividing a three-dimensional puckery palate digital image into a training sample and a testing sample in a three-dimensional puckery palate preprocessing module, respectively sending the training sample and the testing sample into a training process and a testing process, respectively carrying out equidistant slicing processing on the three-dimensional puckery palate training sample and the testing sample in the two parts to obtain a three-dimensional puckery palate slice sample set, carrying out corrosion and binarization processing on a puckery palate slice image, and constructing a puckery palate slice database on the basis of the three-dimensional puckery palate slice sample set and the;
secondly, in a puckery palate tangent line cyclic spectrum feature extraction module, scanning training set and test set puckery palate slice images respectively to obtain one-dimensional signal components of puckery palate tangent lines in the horizontal direction and the vertical direction, performing cyclic spectrum analysis on the one-dimensional signal components in the two directions respectively by using an accumulative Fourier transform algorithm, and extracting cyclic spectrum features of the puckery palate slice images;
thirdly, in a palatine wrinkle tangent line circulation frequency spectrum characteristic blocking module, a 4 multiplied by 4 uniform blocking scheme is adopted to block the palatine wrinkle tangent line circulation frequency spectrum characteristic diagram which is extracted in the test set and the training set respectively, and the blocking principle is as follows: and uniformly partitioning the blocks based on the positions of the pixel points, solving the mean value and the variance of each small block, and finally converting the mean value and the variance into one-dimensional row vectors in sequence, thereby respectively constructing a feature vector dictionary of a training set and a test set.
And finally, adding a three-dimensional wrinkle palate sample label into a k nearest neighbor classifier in a wrinkle palate tangent characteristic vector classification module, and finally outputting a classification result by the classifier according to the provided information, wherein the classification result comprises a correct identification rate, an error rate and an error type label.
The concrete case is as follows:
selecting samples from 11 th person to 20 th person in a database, establishing a small three-dimensional crumple palate sample database, wherein each person has 5 three-dimensional samples, each sample has 40 crumple palate section images obtained through equidistant slicing, assuming that a crumple palate tangent line at the 40 th position is selected to construct the sample database, a crumple palate tangent line at the 39 th position is selected to construct the sample database, and the like so as to construct the crumple palate tangent database of the 40 samples. In order to verify how many of the 40 creases-palate tangents of each crease-palate sample can be successfully identified, the method selects the cyclic frequency spectrum characteristics of the horizontal signal components of the crease-palate tangents for classification, and the result is shown in table 1:
TABLE 1 Effect of different blocking schemes on recognition rates of pucker palate tangents at different positions
In table 1, if the 11 th test sample is taken as an example for analysis, the 40 palate fold tangents are completely and correctly identified, the identification rate is not affected by the blocking scheme, and the identification rate is kept at 100%; from the 12 th sample to the 17 th test sample, it can be seen that the number of recognized cuts of the palate is in an increasing trend as the number of the divided pieces increases. In fact, it can be seen from the table that the number of correctly identified cutting lines of the pucker palate is above 30 when the 4 × 8 block and 8 × 8 block schemes are adopted, which indicates that in some cases, the cutting lines of the pucker palate can be identified at different positions.
In the whole three-dimensional puckery palate sample bank, 91 individual puckery palate samples are collected, 5 three-dimensional puckery palate samples are collected for each person, the first 4 three-dimensional puckery palate samples are selected for each person to be used as training samples, the 5 th three-dimensional puckery palate sample is selected for each person to be used as a test sample, and the identification condition of the sample bank is shown in table 2:
TABLE 2 statistical table of different directional signal component identification of the cutting line of the pucker palate
In table 2, the number of identification errors of the vertical and horizontal direction signal components and the corresponding identification error class labels are given, and finally, the identification of both direction signal components achieves good identification effect.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principles of the present invention are intended to be included therein.
Claims (8)
1. A three-dimensional puckery palate recognition method adopting a cyclic spectrum is characterized by comprising the following specific steps:
step one, collecting three-dimensional puckery palate information;
step two, carrying out equidistant slicing on the three-dimensional puckery palate image;
thirdly, constructing a palate wrinkle section database;
step four, carrying out corrosion and binarization treatment on the wrinkled palate slice image;
step five, extracting the circular spectrum characteristics of the crease palate tangent line in the slice image;
step six, constructing a circular spectrum characteristic vector of a palate wrinkle tangent line;
and seventhly, identifying the three-dimensional puckery palate and outputting a result.
2. The method for identifying the three-dimensional ruffles using the cyclic spectrum according to claim 1, wherein in the first step, the position of the dental chair is adjusted to be in a position where the lower jaw of the volunteer is parallel to the ground and the upper jaw of the volunteer is perpendicular to the ground, the volunteer and the collection device are respectively sterilized, the head of the scanning gun is then extended into the oral cavity of the volunteer, the scanning gun is moved along the teeth to scan, whether each surface of the ruffles is complete or not is checked, if a leak exists or the scanning is incomplete, the scanning is performed, and after the scanning is completed, the scanning data is introduced into the computer to obtain the three-dimensional ruffles palatab.
3. The method for identifying ruffles in three dimensions using cyclic spectrum according to claim 2, wherein in step two, the principle of equidistant slicing is: intersecting the three-dimensional puckery palate model with a horizontal slice plane for slicing;
the slicing operation comprises the following specific steps: firstly, measuring the length, the width and the height of a palate wrinkle image by adopting image processing software, connecting protruding points of three teeth from the left side and the right side of an incisor mastoid to form a cutting straight line, then starting cutting from five teeth from the left side and the right side of the incisor mastoid by adopting a circular section line, adjusting the circular section line to be parallel to the cutting straight line, and carrying out equidistant slicing along the middle point of the circular section line and the direction of the incisor mastoid; and secondly, carrying out equidistant slicing operation on the three-dimensional pucker palate every 1cm, sequentially carrying out 40 slicing operations, sequentially obtaining 40 two-dimensional pucker palate slice images after equidistant slicing, storing the two-dimensional pucker palate slice images in a jpg format, and forming a pucker palate slice sample set by the 40 slice images together.
4. The method for identifying ruffles in three dimensions using a cyclic spectrum according to claim 1, wherein in step four, samples of ruffles in 91 individuals are collected, 5 samples of ruffles in each individual are collected, and the equidistant slicing operation of step two is performed on each ruffles sample, so that 40 equidistant sliced surface curve images of ruffles in each ruffle sample can be obtained, and thus 455 samples of ruffles in 91 individuals and 18200 sliced surface images of ruffles in each ruffle can be obtained.
5. The method of claim 4, wherein 5 samples of each person are obtained from the same light-dark background and the three-dimensional wrinkle images obtained by adding rotation at different times.
6. The method for identifying three-dimensional ruffles using cyclic spectrum according to claim 1, wherein in step five, the image of the tangent to ruffles is first scanned to obtain one-dimensional components in the horizontal direction and the vertical direction, and then the spectral correlation functions of the two one-dimensional components are calculated by using the cumulative fourier transform;
setting the cyclic autocorrelation function of the discrete-time signal x (n) to
The spectral correlation function is:
in the formula (1), a is a cycle frequency, and f is a spectrum frequency; when a is 0, the signal has a finite average power:
the symmetry relationship and the discrete correlation periodicity of the spectral correlation function are shown in equations (3) and (4):
in the formula (4), m and n are any integers;
the cyclic spectrum is a complex spectrum, frequency information and phase information of the signal are estimated by using a cyclic spectrum algorithm, and a spectrum correlation function expression is shown as a formula (5):
the spectral correlation function calculated by the cumulative fourier transform is:
in the formula (6), the reaction mixture is,
performing discrete digitization processing on equation (6), equation (6) becomes equation (8):
in formula (8):
the signal x (n) is a real sequence, which when discretized has a frequency of
7. The method for identifying the three-dimensional puckery palate through the circular spectrum according to claim 1, wherein in step six, the circular spectrum characteristic diagram of the puckery palate tangent line in step five is divided into 4 x 4 uniform blocks, and the principle of the block division is as follows: and uniformly partitioning the blocks based on the positions of the pixel points, solving the mean value and the variance of each small block, and sequentially converting the mean value and the variance into one-dimensional row vectors.
8. The method for identifying three-dimensional ruffles using cyclic spectrum according to claim 1, wherein in step seven, the three-dimensional ruffles identifying comprises two parts of a three-dimensional ruffles image training process and a three-dimensional ruffles image testing process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911216905.1A CN110991339B (en) | 2019-12-02 | 2019-12-02 | Three-dimensional palate wrinkle identification method adopting cyclic frequency spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911216905.1A CN110991339B (en) | 2019-12-02 | 2019-12-02 | Three-dimensional palate wrinkle identification method adopting cyclic frequency spectrum |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110991339A true CN110991339A (en) | 2020-04-10 |
| CN110991339B CN110991339B (en) | 2023-04-28 |
Family
ID=70089354
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911216905.1A Active CN110991339B (en) | 2019-12-02 | 2019-12-02 | Three-dimensional palate wrinkle identification method adopting cyclic frequency spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110991339B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111753709A (en) * | 2020-06-19 | 2020-10-09 | 太原科技大学 | Fractional Fourier transform-based three-dimensional pucker-palate recognition method |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1801213A (en) * | 2005-09-30 | 2006-07-12 | 铁岭市公安局213研究所 | Method and apparatus for three-dimensional cranium body source identification |
| CN103078816A (en) * | 2013-01-18 | 2013-05-01 | 河海大学 | Interference resisting method for 2400 MHz (megahertz) frequency band |
| CN103494653A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method for digitally measuring dental jaw growth and development laws |
| US20170039761A1 (en) * | 2014-05-14 | 2017-02-09 | Huawei Technologies Co., Ltd. | Image Processing Method And Apparatus |
| WO2017089953A1 (en) * | 2015-11-23 | 2017-06-01 | Politecnico Di Torino | Process for processing medical images of a face for recognition of facial dysmorphisms |
| CN106875409A (en) * | 2017-03-24 | 2017-06-20 | 云南大学 | A kind of light-type incisional hernia sticking patch three-dimensional ultrasound pattern feature extracting method |
| CN107169507A (en) * | 2017-01-06 | 2017-09-15 | 华南理工大学 | A kind of certificate photo exposure directions detection algorithm extracted based on grid search-engine |
| CN107563983A (en) * | 2017-09-28 | 2018-01-09 | 上海联影医疗科技有限公司 | Image processing method and medical imaging devices |
| CN109300486A (en) * | 2018-07-30 | 2019-02-01 | 四川大学 | Fricative automatic identifying method is swallowed based on the cleft palate speech that PICGTFs and SSMC enhances |
| CN109363787A (en) * | 2018-12-04 | 2019-02-22 | 北京大学口腔医学院 | A kind of malocclusion denture rescues rear three-dimensional configuration high efficiency digitlization preferred method |
| CN208659317U (en) * | 2017-10-26 | 2019-03-29 | 山西医科大学口腔医院 | Palate wrinkle image collecting device |
| CN109784335A (en) * | 2019-01-25 | 2019-05-21 | 太原科技大学 | A kind of E Zhou interesting image area boundary demarcation method based on least square fitting |
| CN109801222A (en) * | 2019-01-25 | 2019-05-24 | 太原科技大学 | A kind of palate wrinkle tilted image antidote based on direction vector |
| CN109802905A (en) * | 2018-12-27 | 2019-05-24 | 西安电子科技大学 | Digital signal Automatic Modulation Recognition method based on CNN convolutional neural networks |
| CN109829885A (en) * | 2018-12-24 | 2019-05-31 | 中山大学 | A kind of automatic identification nasopharyngeal carcinoma primary tumo(u)r method based on deep semantic segmentation network |
| CN109948400A (en) * | 2017-12-20 | 2019-06-28 | 宁波盈芯信息科技有限公司 | It is a kind of to be able to carry out the smart phone and its recognition methods that face characteristic 3D is identified |
| CN110033032A (en) * | 2019-03-29 | 2019-07-19 | 中国科学院西安光学精密机械研究所 | A kind of histotomy classification method based on micro- high light spectrum image-forming technology |
| CN110363803A (en) * | 2019-07-18 | 2019-10-22 | 深圳市思锐视科技有限公司 | A kind of object detection method and system of combination depth map slice and neural network |
-
2019
- 2019-12-02 CN CN201911216905.1A patent/CN110991339B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1801213A (en) * | 2005-09-30 | 2006-07-12 | 铁岭市公安局213研究所 | Method and apparatus for three-dimensional cranium body source identification |
| CN103078816A (en) * | 2013-01-18 | 2013-05-01 | 河海大学 | Interference resisting method for 2400 MHz (megahertz) frequency band |
| CN103494653A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method for digitally measuring dental jaw growth and development laws |
| US20170039761A1 (en) * | 2014-05-14 | 2017-02-09 | Huawei Technologies Co., Ltd. | Image Processing Method And Apparatus |
| WO2017089953A1 (en) * | 2015-11-23 | 2017-06-01 | Politecnico Di Torino | Process for processing medical images of a face for recognition of facial dysmorphisms |
| CN107169507A (en) * | 2017-01-06 | 2017-09-15 | 华南理工大学 | A kind of certificate photo exposure directions detection algorithm extracted based on grid search-engine |
| CN106875409A (en) * | 2017-03-24 | 2017-06-20 | 云南大学 | A kind of light-type incisional hernia sticking patch three-dimensional ultrasound pattern feature extracting method |
| CN107563983A (en) * | 2017-09-28 | 2018-01-09 | 上海联影医疗科技有限公司 | Image processing method and medical imaging devices |
| CN208659317U (en) * | 2017-10-26 | 2019-03-29 | 山西医科大学口腔医院 | Palate wrinkle image collecting device |
| CN109948400A (en) * | 2017-12-20 | 2019-06-28 | 宁波盈芯信息科技有限公司 | It is a kind of to be able to carry out the smart phone and its recognition methods that face characteristic 3D is identified |
| CN109300486A (en) * | 2018-07-30 | 2019-02-01 | 四川大学 | Fricative automatic identifying method is swallowed based on the cleft palate speech that PICGTFs and SSMC enhances |
| CN109363787A (en) * | 2018-12-04 | 2019-02-22 | 北京大学口腔医学院 | A kind of malocclusion denture rescues rear three-dimensional configuration high efficiency digitlization preferred method |
| CN109829885A (en) * | 2018-12-24 | 2019-05-31 | 中山大学 | A kind of automatic identification nasopharyngeal carcinoma primary tumo(u)r method based on deep semantic segmentation network |
| CN109802905A (en) * | 2018-12-27 | 2019-05-24 | 西安电子科技大学 | Digital signal Automatic Modulation Recognition method based on CNN convolutional neural networks |
| CN109784335A (en) * | 2019-01-25 | 2019-05-21 | 太原科技大学 | A kind of E Zhou interesting image area boundary demarcation method based on least square fitting |
| CN109801222A (en) * | 2019-01-25 | 2019-05-24 | 太原科技大学 | A kind of palate wrinkle tilted image antidote based on direction vector |
| CN110033032A (en) * | 2019-03-29 | 2019-07-19 | 中国科学院西安光学精密机械研究所 | A kind of histotomy classification method based on micro- high light spectrum image-forming technology |
| CN110363803A (en) * | 2019-07-18 | 2019-10-22 | 深圳市思锐视科技有限公司 | A kind of object detection method and system of combination depth map slice and neural network |
Non-Patent Citations (10)
| Title |
|---|
| SAMEH TALAAT等: "Validity and reliability of three-dimensional palatal superimposition of digital dental models", 《EUROPEAN JOURNAL OF ORTHODONTICS》 * |
| XIONG ZHANG等: "Three-dimensional palatal rugae recognition based on cyclic spectral analysis", 《BIOMEDICAL SIGNAL PROCESSING AND CONTROL》 * |
| 朱宸佑等: "锥形束电子计算机断层扫描在上颌前牙区牙槽骨的测量方法与分型标准中的应用", 《国际口腔医学杂志》 * |
| 李鹏等: "一种路面裂缝的循环频谱特征提取方法", 《安徽大学学报(自然科学版)》 * |
| 武小雪: "可视化腭皱识别技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 武秀萍等: "应用数字图像识别系统研究正畸治疗前后腭皱形态变化" * |
| 罗强: "三维腭皱生物特征识别方法研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
| 蔡志刚等: "18三体腭裂小鼠上颌牙弓发育的 比较形态测量学研究" * |
| 贾宇等: "腭皱法医学同一认定中倾斜图像矫正方法的研究", 《口腔医学研究》 * |
| 黄传波等: "基于视觉注意的彩色图像检索方法", 《光子学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111753709A (en) * | 2020-06-19 | 2020-10-09 | 太原科技大学 | Fractional Fourier transform-based three-dimensional pucker-palate recognition method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110991339B (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Harvati et al. | Evolution of middle-late Pleistocene human cranio-facial form: A 3-D approach | |
| US8842904B2 (en) | Method for tooth dissection in CBCT volume | |
| Braga et al. | Efficacy of diffeomorphic surface matching and 3D geometric morphometrics for taxonomic discrimination of Early Pleistocene hominin mandibular molars | |
| Baab et al. | Phylogenetic, ecological, and allometric correlates of cranial shape in Malagasy lemuriforms | |
| TW201224826A (en) | Systems and methods for automated extraction of measurement information in medical videos | |
| CN110189324B (en) | Medical image processing method and processing device | |
| Yang et al. | Skull sex estimation based on wavelet transform and Fourier transform | |
| CN111753709A (en) | Fractional Fourier transform-based three-dimensional pucker-palate recognition method | |
| Galantucci et al. | Is principal component analysis an effective tool to predict face attractiveness? A contribution based on real 3D faces of highly selected attractive women, scanned with stereophotogrammetry | |
| CN110991339A (en) | Three-dimensional puckery palate identification method adopting circular spectrum | |
| Harastani et al. | Methamphetamine drug abuse and addiction: Effects on face asymmetry | |
| Antolin et al. | Tooth condition classification for dental charting using convolutional neural network and image processing | |
| Tuan et al. | Shape prediction of nasal bones by digital 2D-photogrammetry of the nose based on convolution and back-propagation neural network | |
| Wilson et al. | Voyaging into the third dimension: A perspective on virtual methods and their application to studies of juvenile sex estimation and the ontogeny of sexual dimorphism | |
| Zhang et al. | Three-dimensional palatal rugae recognition based on cyclic spectral analysis | |
| CN113269160B (en) | Colonoscope operation predicament intelligent identification system based on eye movement characteristics | |
| ALbahbah et al. | A new optimized approach for detection of caries in panoramic images | |
| Li et al. | Trabecular texture analysis in dental cbct by multi-roi multi-feature fusion | |
| Jasso-Cuéllar et al. | Anterior dental arch shape and human identification: Kieser et al. method applied to 2D-3D dental models in Mexican population | |
| Hori et al. | A novel identification method using perceptual degree of concordance of occlusal surfaces calculated by a Python program | |
| Al-sherif | Novel Techniques for Automated Dental Identification | |
| Gupta et al. | Role of orthodontics in forensic facial reconstruction for human identification | |
| CN111222407B (en) | Palate wrinkle identification method adopting uniform slicing and inflection point characteristic extraction | |
| JP7226745B2 (en) | Apparatus, method, program, and system for determining three-dimensional facial morphology | |
| Hosntalab et al. | Automated dental recognition in MSCT images for human identification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |