CN112471643B - 3D child protection mask construction method based on head-face modeling - Google Patents
3D child protection mask construction method based on head-face modeling Download PDFInfo
- Publication number
- CN112471643B CN112471643B CN202011441258.7A CN202011441258A CN112471643B CN 112471643 B CN112471643 B CN 112471643B CN 202011441258 A CN202011441258 A CN 202011441258A CN 112471643 B CN112471643 B CN 112471643B
- Authority
- CN
- China
- Prior art keywords
- face
- head
- mask
- curved surface
- data
- 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.)
- Active
Links
- 238000010276 construction Methods 0.000 title description 3
- 238000000034 method Methods 0.000 claims abstract description 26
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 230000000877 morphologic effect Effects 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 23
- 230000001681 protective effect Effects 0.000 claims abstract description 13
- 238000013461 design Methods 0.000 claims abstract description 10
- 238000007621 cluster analysis Methods 0.000 claims abstract description 9
- 230000001815 facial effect Effects 0.000 claims description 15
- 238000005259 measurement Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 241000746998 Tragus Species 0.000 claims description 9
- 241000083547 Columella Species 0.000 claims description 7
- 230000000593 degrading effect Effects 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 claims description 4
- 238000011282 treatment Methods 0.000 claims description 4
- 238000013499 data model Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 230000003287 optical effect Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims description 3
- 230000003213 activating effect Effects 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims 2
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000006872 improvement Effects 0.000 abstract description 2
- 238000011160 research Methods 0.000 description 5
- 239000006185 dispersion Substances 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 238000000556 factor analysis Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 241001070941 Castanea Species 0.000 description 1
- 235000014036 Castanea Nutrition 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 239000013618 particulate matter Substances 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000036558 skin tension Effects 0.000 description 1
- 210000003625 skull Anatomy 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010971 suitability test Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A41—WEARING APPAREL
- A41D—OUTERWEAR; PROTECTIVE GARMENTS; ACCESSORIES
- A41D13/00—Professional, industrial or sporting protective garments, e.g. surgeons' gowns or garments protecting against blows or punches
- A41D13/05—Professional, industrial or sporting protective garments, e.g. surgeons' gowns or garments protecting against blows or punches protecting only a particular body part
- A41D13/11—Protective face masks, e.g. for surgical use, or for use in foul atmospheres
- A41D13/1107—Protective face masks, e.g. for surgical use, or for use in foul atmospheres characterised by their shape
- A41D13/113—Protective face masks, e.g. for surgical use, or for use in foul atmospheres characterised by their shape with a vertical fold or weld
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/12—Geometric CAD characterised by design entry means specially adapted for CAD, e.g. graphical user interfaces [GUI] specially adapted for CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- 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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2210/00—Indexing scheme for image generation or computer graphics
- G06T2210/16—Cloth
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The invention discloses a 3D child protection mask forming method based on head-face modeling, which comprises the following steps: step 1: selecting an experimental object and acquiring experimental data; step 2: performing cluster analysis on the head and face characteristic data to obtain the characteristic data of the standard head and face; step 3: performing discriminant analysis on the head-face characteristic data, and verifying the accuracy of head-face clustering analysis and the feasibility and accuracy of classification rules; step 4: performing reverse modeling on the head and the face, processing data, establishing a database and obtaining a mean characteristic curve; step 5: and carrying out mask modeling design according to the characteristic curve. The invention provides a child head and face feature classification method combining front and side features, which can provide more complete control part information for the design of a protective mask by analyzing morphological features and section angles of the head and face under the front and side angles, and is more beneficial to the improvement of the three-dimensional fit of the mask.
Description
Technical Field
The invention belongs to the field of mask design, and particularly relates to a 3D child protection mask construction method based on head-face modeling.
Background
Because of the great difference of physical characteristics of children, the accumulation of medical data and experimental data, which influence on the growth of children, of respiratory resistance is still to be perfected, so that the standard does not specially specify the mask for children, but explicitly indicates that the mask for children is not suitable for children. The Zhejiang market supervision bureau of 3 months and 13 days in 2020 issued the child mask group standard at home first, and standardized means were used to standardize the production and supply of child masks. However, in the standard, only the types of masks are classified according to the shapes of the small children (3-6 years), the middle children (7-12 years), the large children (13-14 years) and the constituent shapes, namely, the plane and the three-dimensional, and a specific method for solving the problem of mask fitting is still lacking.
How to accurately find out the index representing the head characteristics of the human body and the parameters of the facial form and the face are key to the head-face part classification research. A great deal of researches are carried out on human body type classification for improving the fit and comfort of clothes by students at home and abroad, and experimental researches on quantitative body type classification are respectively developed by using different methods such as human body size information, human body surface angle, human body transverse and longitudinal section characteristics, different age groups and the like. Chen Shourong and the like, by measuring and observing morphological characteristics of 27 mark points of the head and the face of 15 ethnic groups 3182 adults in China, compared with those of Han nationality male and female adults and researching , the measured values and the facial morphological characteristics of the different ethnic group facial mark points are found to be different, and the eye characteristics, the nose characteristics and the facial characteristics of each ethnic group are proposed. The quality of the chestnut and the like perform cluster analysis on the head and face sizes of 30 people in China so as to analyze the head shape characteristics and differences of each ethnic group. The minority constitution research group represented by Xu Biao performs microcomputer measurement and morphological observation research on 15 minority head and face organs of the Dai nationality, yi nationality, lisu nationality and the like in Yunnan province. Zhang Xueyan and the like detect the FF of 4 kinds of particulate matter protective masks by using quantitative suitability test and human head-face size measurement methods, and discuss the influence of head-face size (head-face size data such as morphological face length, face width, intertragus width, intermandibular angle width, nasal height, nose length, nose width, and intertragus point submandibular arc length of mask wearer), BMI, sex, mask model and the like on the FF.
In summary, these studies are mostly based on the ratio of the transverse vector of the head and the face or the classification of the head and the face combined with a single angle value, and the analysis is not fully performed by combining the outline morphological characteristics of the front and the side surfaces of the human body and the curve morphology of the characteristic section, so that the considered influencing factors are not comprehensive enough.
Disclosure of Invention
The invention aims to provide a 3D child protection mask forming method based on head-face modeling, and provides a child head-face feature classification method combining front and side features aiming at the defects in the prior art.
In order to solve the technical problems, the following technical scheme is adopted:
the 3D child protection mask forming method based on head-face modeling is characterized by comprising the following steps of:
step 1: selecting an experimental object and acquiring experimental data;
step 2: performing cluster analysis on the head and face characteristic data to obtain the characteristic data of the standard head and face;
step 3: performing discriminant analysis on the head-face characteristic data, and verifying the accuracy of head-face clustering analysis and the feasibility and accuracy of classification rules;
step 4: performing reverse modeling on the head and the face, processing data, establishing a database and obtaining a mean characteristic curve;
step 5: and carrying out mask modeling design according to the characteristic curve.
Preferably, in the step 1, the samples of the subjects are divided into four experimental groups of 6-9 years old, 9-12 years old, 12-15 years old and 15-18 years old according to the age specification, and the four experimental groups are respectively marked as S, M, L, XL, and the sample capacity of each group is 100 persons; carrying out face three-dimensional scanning on each experimental sample by using an optical three-dimensional scanner to obtain point cloud data, and obtaining four groups of experimental sample data in total;
the experimental data comprise 7 head-face morphological characteristics of morphological face length, morphological face width, width between two tragus, width between two mandibular angles, nose length, nose width and nose depth selected under the front view angle of human face; the facial side view of the human body is selected with morphological characteristics of 4 parts of nasal height, submandibular arc length between two tragus points, side nasal tip and nasal columella section angle, and nasal back and facial middle section included angle, and is combined with the transverse vector diameter ratio of the head and the face of the human body, and finally 12 morphological parameters related to the mask fitting degree are extracted through a direct measurement or calculation mode.
And preferably, when the cross section angle between the nose tip and the nose columella at the side, the cross section included angle between the nose back and the face and the transverse vector diameter ratio of the head and the face of the human body are obtained, using Geomagic Studio software to carry out secondary measurement calculation on three-dimensional point cloud data of the human body.
Preferably, the point cloud data is processed, the polygonal object is converted into an accurate curved surface, and the rest part of the accurate curved surface options are activated to generate a curved surface which is completely matched with the created object, and the specific operations comprise:
contour line processing, namely placing contour lines in a curvature area, observing cross confusion of the contour lines if the contour lines are detected after the curvature is detected, degrading all orange contour lines into black patch lines by degrading all the contour lines, and upgrading black lines at high curvature to obtain contour lines;
the curved surface sheet processing, namely constructing a curved surface sheet through a more regular curved surface sheet structure generated by a contour line and a boundary line, arranging the curved surface sheets in the panel and filling the blank panel with the curved surface sheets; defining vertexes and paths of the curved surface sheets, and ensuring that the curved surface sheets can be divided uniformly; the same method processes the curved surface piece on the other side and rearranges the curved surface piece by 'free detection';
grid processing, constructing grids, automatically parameterizing curved surfaces, dividing each curved surface piece into 20 x 20 small grids, manually adjusting resolution according to requirements, and repairing the intersected grids if imperfect curved surface pieces exist;
NURBS curved surface processing, namely, the NURBS curved surface obtained by fitting the curved surface and combining the curved surfaces can be output as IGES/STEP files and input into any CAD/CAM system; the auxiliary line and the face are constructed, and the nose height, the submandibular arc length between two tragus points, the side nose tip and the section angle of the nose columella and the section angle between the nose back and the face are obtained through a measuring tool.
Preferably, the specific step of cluster analysis of the head and face feature data in the step 2 includes:
(1) Selecting a clustering variable: in order to reduce inconvenience of high-dimensional data to subsequent processing, it is necessary to perform dimension reduction processing before cluster analysis; first, carrying out principal component analysis on 12 variables to obtain factor analysis initial solutions; and then selecting a representative index which contains the most abundant information of the category from the two factors of the front face, the side face and the head face by adopting a correlation index maximum value method, wherein the calculation formula of the correlation index is as follows:
calculating the correlation index of each index to other indexes with the same factor according to the formula (1), and then selecting the index with the highest correlation index result as a representative clustering index;
(2) Determining the optimal classification number by using mixed F statistic to make F Mixed ；F Mixed The intra-class tightness and the inter-class dispersion of all the variables are comprehensively reflected, the larger the value of the intra-class tightness and the inter-class dispersion of all the variables is, the more compact the intra-class linkage of all the variables is, and otherwise, the more the inter-class linkage is dispersed; when F Mixed The corresponding classification number is the optimal classification number when the value is maximum, and the calculation formula is as follows:
wherein, P in the formula (2) is the variable number of the cluster; f (k) is the F value of the kth cluster variable, and can be calculated by the following formula (3):
(3) Comparison analysis of head and face characteristic factors of each class: and counting the characteristic indexes of each face according to the classification of the face of the experimental sample, the clustering center value of the clustering index and the corresponding frequency, and acquiring the characteristic data of the standard head faces of the 4 experimental groups S, M, L, XL from the class analysis of the front and the side according to the factor load comparison analysis after the rotation of the principal component factors.
Preferably, in the step 3, in order to verify the accuracy of the head-face clustering analysis and the feasibility and accuracy of the classification rule, the initial 400 head-face data and the newly added 50 head-face feature data are subjected to discriminant analysis; the feasibility and the accuracy of the classification rule are verified through the accuracy of the classification result of the percentage discriminant analysis and the new sample, and the percentage discriminant indexes comprise ungrouped observation values, estimated categories and discriminant categories.
Preferably, in the step 4, the steps of performing reverse modeling of the head and the face, processing the data and establishing the database include:
(1) Converting the point cloud data into grid data: the point cloud data are subjected to preprocessing of splicing, fitting and noise reduction, converted into grid data, and a grid data model is exported in a stl format;
(2) Characterizing head-face data: the grid data is imported into the Rhino software, the upper edge of the mask is drawn by using a curve tool in the front view, the side edge of the mask is drawn on the side view, the bottom edge of the mask is drawn on the bottom view, and the bottom edge of the mask is respectively projected on a grid model of the head and the face, so that characteristic multiple straight lines are obtained;
(3) Multiple lines were converted into nurbs curves: and converting the obtained multiple sections of straight lines into a nurbs curve through reconstruction, and simultaneously ensuring that the maximum error value of the reconstructed curve is controlled within 0.1. Finally, a 6 th-order 7-point nurbs curve was obtained, designated as S (CV 1 ,CV 2 ,…,CV 7 ) The method comprises the steps of carrying out a first treatment on the surface of the The above steps are repeated to obtain four sets of characteristic curves, each set of 100 (S 1 ,S 2 ,…,S 100 ) Thus, four groups of facial form databases with different models are obtained, and then the coordinates of the corresponding control points of each group of curves are subjected to mean value calculation to finally obtain four groups of S, M, L, XL mean value characteristic curves.
Preferably, in the step 5, mask modeling design is performed according to the characteristic curve:
(1) A structural midline required by a mask structural model is built, namely the midline of the mask is designed according to the mean characteristic curve and the human head and face model, and a 6-order 7-point fairing curve is obtained through adjustment;
(2) Establishing a curved surface of a cover surface main body: establishing a curved surface of the mask, carrying out mixed reconstruction on a joint surface in the middle of the mask main body, and adjusting UVN coordinates of the curved surface; creating a cover surface on the solid model of the head and the face through a section line; the three-dimensional X, Y, Z coordinate axis shape angle rotation deformation is changed into a cross section, and the cross section line strings generated by the cross section and the entity are required to intersect within a set tolerance range and are mutually perpendicular; the mask is based on a facial curve, belongs to a complex curved surface, and needs to finish a main surface or a large surface firstly, then smooth the connecting curved surface, and finally edit and modify;
(3) Perfecting a mask three-dimensional model: constructing a three-dimensional model of the mask structure according to the ear positions and the cheek shapes, and providing a final structure model of the 3D protective mask;
(4) Extracting a plane paper pattern: using forward engineering design software to expand the 3D protective mask model from a three-dimensional model to a two-dimensional plane; selecting a laminar surface to be unfolded, after determining an unfolding point and a primary and secondary azimuth, unfolding by taking four corners as starting points, comparing the unfolded model size obtained by UG software calculation with the original head-face measured size, and taking a two-dimensional unfolding diagram with the minimum error point and the minimum size error as a final 3D protective mask paper sample;
repeating the steps to obtain the 3D child protection mask paper patterns with four different specifications S, M, L, XL.
Due to the adoption of the technical scheme, the method has the following beneficial effects:
the invention provides a child head and face feature classification method combining front and side features, which can provide more complete control part information for the design of a protective mask by analyzing morphological features and section angles of the head and face under the front and side angles, and is more beneficial to the improvement of the three-dimensional fit of the protective mask. Clustering and multi-curved surface modeling analysis are carried out on relevant indexes of the head and the face of a human body, so that a basis is provided for solving the problem of mask tightness, and a new idea is provided for solving the problem of contradiction between clothing and the human body. Meanwhile, theoretical guidance is provided for enriching the product types of the child protection mask, improving the functionality and comfort of the product and forward design production and development of the head and face protection product, so that the health and rapid development of the whole industry can be promoted.
Drawings
The invention is further described below with reference to the accompanying drawings:
FIG. 1 is a 3D scanned face type grid data diagram in the present invention;
FIG. 2 is a front, side, bottom curve projection of the present invention;
FIG. 3 is a profile plot of the present invention;
FIG. 4 is a three-view of a mask model in accordance with the present invention;
Detailed Description
The invention is further described with reference to specific embodiments, as shown in fig. 1 to 4:
A3D child protection mask forming method based on head-face modeling comprises the following steps:
step 1: selecting an experimental object to obtain experimental data
Firstly, a study object is established, sample data are collected, and a face data collection range is determined. The invention takes Hangzhou area as an example to collect facial feature data. According to international standard ISO 15535:2012 general requirement for establishing anthropometric database, 6-18 year old children are selected as data acquisition objects, in order to eliminate the influence of skin factors, experimental samples with better skin tension and parallel to bones are selected, the three-dimensional shape of the skin surface can embody the morphological characteristics of regional skull, 400 samples are selected according to a random sampling method, the object samples are divided into four specification groups of S (6-9 years old), M (9-12 years old), L (12-15 years old) and XL (15-18 years old) according to the age specification, and the capacity of each category sample is 100 persons. And carrying out face three-dimensional scanning on each experimental sample by using an optical three-dimensional scanner to obtain point cloud data, and obtaining four groups of experimental sample data. In making manual contact face measurements, the tools mainly used are vernier calipers (sliding calipers) and pin gauges (spreading caliper) in Martin gauges. According to the related measurement study of scholars at home and abroad, a subject is required to calm for 3min before measurement in manual measurement, a measured person needs to sit at an end and look at the front, the measured person marks the position of a measuring point on the head and the face of the measured person by using a marker pen according to related regulations in national standard GB/T5703-2010 'basic project for human body measurement for technical design', then the measured person takes an average value after 3 times of measurement, and the measurement result requires that the stability error does not exceed a specified range.
The experimental data comprise 7 head-face morphological characteristics of morphological face length, morphological face width, width between two tragus, width between two mandibular angles, nose length, nose width and nose depth selected under the front view angle of human face; the facial side view of the human body is selected with morphological characteristics of 4 parts of nasal height, submandibular arc length between two tragus points, side nasal tip and nasal columella section angle, and nasal back and facial middle section included angle, and is combined with the transverse vector diameter ratio of the head and the face of the human body, and finally 12 morphological parameters related to the mask fitting degree are extracted through a direct measurement or calculation mode. Wherein the transverse vector diameter ratio of the head and the face of the human body is the ratio of the form surface width (width) under the front view angle to the form surface width (thickness) under the side view angle.
When the cross section angle between the nose tip and the small nose column of the front side, the cross section included angle between the nose back and the face and the transverse vector diameter ratio of the head and the face of the human body are obtained, geomagic Studio software is used for carrying out secondary measurement calculation on three-dimensional point cloud data of the human body.
Because the surface quality of the scanning point cloud (STL) is poor and excessive, the surface quality is poor and cannot be applied, so that a NURBS curved surface with better quality and fewer surfaces needs to be obtained through shape stage processing. The ideal curved surface model is obtained through a series of technical treatments after the transformation from the polygonal stage.
Firstly, converting a polygonal object into a precise curved surface by a precise curved surface, and activating the rest part of the precise curved surface options to generate a curved surface which is completely matched with the created object, wherein the specific operation comprises the following steps:
contour line processing, namely placing contour lines in a curvature area, observing cross confusion of the contour lines if the contour lines are detected after the curvature is detected, degrading all orange contour lines into black patch lines by degrading all the contour lines, and upgrading black lines at high curvature to obtain contour lines;
the curved surface sheet processing, namely constructing a curved surface sheet through a more regular curved surface sheet structure generated by a contour line and a boundary line, arranging the curved surface sheets in the panel and filling the blank panel with the curved surface sheets; defining vertexes and paths of the curved surface sheets, and ensuring that the curved surface sheets can be divided uniformly; the same method processes the curved surface piece on the other side and rearranges the curved surface piece by 'free detection';
grid processing, constructing grids, automatically parameterizing curved surfaces, dividing each curved surface piece into 20 x 20 small grids, manually adjusting resolution according to requirements, and repairing the intersected grids if imperfect curved surface pieces exist;
NURBS curved surface processing, namely, the NURBS curved surface obtained by fitting the curved surface and combining the curved surfaces can be output as IGES/STEP files and input into any CAD/CAM system; the auxiliary line and the face are constructed, and the nose height, the submandibular arc length between two tragus points, the side nose tip and the section angle of the nose columella and the section angle between the nose back and the face are obtained through a measuring tool.
Step 2: clustering analysis of head-to-face feature data
(1) Selecting a clustering variable: in order to reduce inconvenience of high-dimensional data to subsequent processing, it is necessary to perform dimension reduction processing before cluster analysis; first, carrying out principal component analysis on 12 variables to obtain factor analysis initial solutions; and then selecting a representative index which contains the most abundant information of the category from the two factors of the front face, the side face and the head face by adopting a correlation index maximum value method, wherein the calculation formula of the correlation index is as follows:
calculating the correlation index of each index to other indexes with the same factor according to the formula (1), and then selecting the index with the highest correlation index result as a representative clustering index;
(2) Determining the optimal classification number by using mixed F statistic to make F Mixed ；F Mixed The intra-class tightness and the inter-class dispersion of all the variables are comprehensively reflected, the larger the value of the intra-class tightness and the inter-class dispersion of all the variables is, the more compact the intra-class linkage of all the variables is, and otherwise, the more the inter-class linkage is dispersed; when F Mixed The corresponding classification number is the optimal classification number when the value is maximum, and the calculation formula is as follows:
wherein, P in the formula (2) is the variable number of the cluster; f (k) is the F value of the kth cluster variable, and can be calculated by the following formula (3):
(3) Comparison analysis of head and face characteristic factors of each class: and counting the characteristic indexes of each face according to the classification of the face of the experimental sample, the clustering center value of the clustering index and the corresponding frequency, and obtaining the characteristic data of the standard head faces of 4 experimental groups of S (6-9 years old), M (9-12 years old), L (12-15 years old) and XL (15-18 years old) from the class analysis of the front and the side according to the factor load comparison analysis after the rotation of the principal component factors.
Step 3: performing discriminant analysis on the head and face characteristic data, and verifying accuracy of head and face cluster analysis and feasibility and accuracy of classification rules
To verify the accuracy of the head-face cluster analysis and the feasibility and accuracy of the classification rule, the discrimination analysis is performed on the initial 400 head-face data and the newly added 50 head-face feature data. The feasibility and the accuracy of the classification rule are verified through the accuracy of the classification result of the percentage discriminant analysis and the new sample, and the percentage discriminant indexes comprise ungrouped observation values, estimated categories and discriminant categories.
Step 4: reverse modeling of head and face, processing data and establishing database
(1) Converting the point cloud data into grid data: the point cloud data are subjected to preprocessing of splicing, fitting and noise reduction, converted into grid data, and a grid data model is exported in a stl format;
(2) Characterizing head-face data: the grid data is imported into the Rhino software, the upper edge of the mask is drawn by using a curve tool in the front view, the side edge of the mask is drawn on the side view, the bottom edge of the mask is drawn on the bottom view, and the bottom edge of the mask is respectively projected on a grid model of the head and the face, so that characteristic multiple straight lines are obtained;
(3) Multiple lines were converted into nurbs curves: and converting the obtained multiple sections of straight lines into a nurbs curve through reconstruction, and simultaneously ensuring that the maximum error value of the reconstructed curve is controlled within 0.1. Finally, a 6 th-order 7-point nurbs curve was obtained, designated as S (CV 1 ,CV 2 ,…,CV 7 ) The method comprises the steps of carrying out a first treatment on the surface of the The above steps are repeated to obtain four sets of characteristic curves, each set of 100 (S 1 ,S 2 ,…,S 100 ) Thus, four groups of facial form databases with different models are obtained, and then the corresponding control point coordinates of each group of curves are subjected to mean value calculation to finally obtain four groups of mean characteristic curves of S (6-9 years old), M (9-12 years old), L (12-15 years old) and XL (15-18 years old).
Step 5: mask modeling design according to characteristic curve
(1) Building a structural center line required by a mask structural model: and designing a mask center line according to the mean value curve and the human head and face model, and obtaining a 6-order 7-point fairing curve through adjustment.
(2) Establishing a curved surface of a cover surface main body: and establishing a curved surface of the mask, and reconstructing the joint surface in the middle of the mask body in a mixed mode and adjusting UVN coordinates of the curved surface to ensure attractive appearance and fit. Over the solid model of the head and face, a cover is created by the section lines. The three-dimensional X, Y, Z coordinate axis is deformed by angular rotation into a cross section, and the cross section line strings (main line string and cross line string) generated by the cross section and the entity need to intersect within a set tolerance range and are mutually perpendicular. The mask is based on a facial curve, belongs to a complex curved surface, and needs to finish a main surface or a large surface, then smooth the connecting curved surface, and finally edit and modify.
(3) Perfecting a mask three-dimensional model: and constructing a three-dimensional model of the mask structure according to the ear positions and the cheek shapes, and providing a final structural model of the 3D protective mask.
(4) Extracting a plane paper pattern: and (3) unfolding the 3D protective mask model from the three-dimensional model to a two-dimensional plane by using forward engineering design software. After the unfolding points and the primary and secondary directions are determined, the sheet surfaces to be unfolded are unfolded by taking four corners as starting points, the unfolded model size obtained through UG software calculation is compared with the original head-face measured size, and a two-dimensional unfolding diagram with the minimum error points and the minimum size error is used as a final 3D protective mask paper pattern.
The steps are repeated to obtain the 3D child protection mask paper patterns with four different specifications of S (6-9 years old), M (9-12 years old), L (12-15 years old) and XL (15-18 years old).
The above is only a specific embodiment of the present invention, but the technical features of the present invention are not limited thereto. Any simple changes, equivalent substitutions or modifications made on the basis of the present invention to solve the substantially same technical problems and achieve the substantially same technical effects are encompassed within the scope of the present invention.
Claims (3)
1. The 3D child protection mask forming method based on head-face modeling is characterized by comprising the following steps of:
step 1: selecting an experimental object and acquiring experimental data:
dividing the test object samples into four test groups of 6-9 years old, 9-12 years old, 12-15 years old and 15-18 years old according to the age specification, respectively marking as S, M, L, XL, wherein the sample capacity of each group is 100 persons; carrying out face three-dimensional scanning on each experimental sample by using an optical three-dimensional scanner to obtain point cloud data, and obtaining four groups of experimental sample data in total;
the experimental data comprise 7 head-face morphological characteristics of morphological face length, morphological face width, width between two tragus, width between two mandibular angles, nose length, nose width and nose depth selected under the front view angle of human face; the method comprises the steps that the morphological characteristics of 4 parts of the nasal height, the submandibular arc length between two tragus points, the cross section angle of the lateral nasal tip and the nasal columella, and the cross section included angle between the nasal back and the middle face are selected on a human face side view, and 12 morphological parameters related to the fitting degree of the mask are finally extracted through a direct measurement or calculation mode by combining the transverse vector diameter ratio of the head and the face of the human body;
processing point cloud data, converting a polygonal object into a precise curved surface, and activating the rest part of the precise curved surface options to generate a curved surface which is completely matched with the created object, wherein the specific operation comprises the following steps:
contour line processing, namely placing contour lines in a curvature area, observing cross confusion of the contour lines if the contour lines are detected after the curvature is detected, degrading all orange contour lines into black patch lines by degrading all the contour lines, and upgrading black lines at high curvature to obtain contour lines;
the curved surface sheet processing, namely constructing a curved surface sheet through a more regular curved surface sheet structure generated by a contour line and a boundary line, arranging the curved surface sheets in the panel and filling the blank panel with the curved surface sheets; defining vertexes and paths of the curved surface sheets, and ensuring that the curved surface sheets can be divided uniformly; the same method processes the curved surface piece on the other side and rearranges the curved surface piece by 'free detection';
grid processing, constructing grids, automatically parameterizing curved surfaces, dividing each curved surface piece into 20 x 20 small grids, manually adjusting resolution according to requirements, and repairing the intersected grids if imperfect curved surface pieces exist;
NURBS curved surface processing, namely, the NURBS curved surface obtained by fitting the curved surface and combining the curved surfaces can be output as IGES/STEP files and input into any CAD/CAM system; constructing auxiliary lines and planes, and acquiring the height of the nose, the length of the submandibular arc between two tragus points, the angles of the sections of the lateral nasal tips and the nasal columella, and the angles of the sections of the nasal backs and the middle planes through a measuring tool;
step 2: performing cluster analysis on the head and face characteristic data to obtain the characteristic data of the standard head and face;
step 3: performing discriminant analysis on the head-face characteristic data, and verifying the accuracy of head-face clustering analysis and the feasibility and accuracy of classification rules;
step 4: performing reverse modeling on the head and the face, processing data, establishing a database and obtaining a mean characteristic curve:
(1) Converting the point cloud data into grid data: the point cloud data are subjected to preprocessing of splicing, fitting and noise reduction, converted into grid data, and a grid data model is exported in a stl format;
(2) Characterizing head-face data: the grid data is imported into the Rhino software, the upper edge of the mask is drawn by using a curve tool in the front view, the side edge of the mask is drawn on the side view, the bottom edge of the mask is drawn on the bottom view, and the bottom edge of the mask is respectively projected on a grid model of the head and the face, so that characteristic multiple straight lines are obtained;
(3) Multiple lines were converted into nurbs curves: converting the obtained multi-section straight line into a nurbs curve through reconstruction, and simultaneously ensuring that the maximum error value of the reconstructed curve is controlled within 0.1; finally, a 6 th-order 7-point nurbs curve was obtained, designated as S (CV 1 ,CV 2 ,…,CV 7 ) The method comprises the steps of carrying out a first treatment on the surface of the The above steps are repeated to obtain four sets of characteristic curves, each set of 100 (S 1 ,S 2 ,…,S 100 ) Thus obtaining four groups of facial form databases with different models, and then carrying out mean value calculation on the coordinates of the corresponding control points of each group of curves to finally obtain four groups of S, M, L, XL mean value characteristic curves;
step 5: carrying out mask modeling design according to the mean characteristic curve:
(1) A structural midline required by a mask structural model is built, namely the midline of the mask is designed according to the mean characteristic curve and the human head and face model, and a 6-order 7-point fairing curve is obtained through adjustment;
(2) Establishing a curved surface of a cover surface main body: establishing a curved surface of the mask, carrying out mixed reconstruction on a joint surface in the middle of the mask main body, and adjusting UVN coordinates of the curved surface; creating a cover surface on the solid model of the head and the face through a section line; the three-dimensional X, Y, Z coordinate axis shape angle rotation deformation is changed into a cross section, and the cross section line strings generated by the cross section and the entity are required to intersect within a set tolerance range and are mutually perpendicular; the mask is based on a facial curve, belongs to a complex curved surface, and needs to finish a main surface or a large surface firstly, then smooth the connecting curved surface, and finally edit and modify;
(3) Perfecting a mask three-dimensional model: constructing a three-dimensional model of the mask structure according to the ear positions and the cheek shapes, and providing a final structure model of the 3D protective mask;
(4) Extracting a plane paper pattern: using forward engineering design software to expand the 3D protective mask model from a three-dimensional model to a two-dimensional plane; selecting a laminar surface to be unfolded, after determining an unfolding point and a primary and secondary azimuth, unfolding by taking four corners as starting points, comparing the unfolded model size obtained by UG software calculation with the original head-face measured size, and taking a two-dimensional unfolding diagram with the minimum error point and the minimum size error as a final 3D protective mask paper sample;
repeating the steps to obtain the 3D child protection mask paper patterns with four different specifications S, M, L, XL.
2. The method for constructing a 3D child-resistant mask based on head-to-face modeling of claim 1, wherein: when the cross section angle between the nose tip and the small nose column on the side surface, the cross section included angle between the nose back and the face and the transverse vector diameter ratio of the head and the face of the human body are obtained, geomagic Studio software is used for carrying out secondary measurement calculation on three-dimensional point cloud data of the human body.
3. The method for constructing a 3D child-resistant mask based on head-to-face modeling of claim 1, wherein: in the step 3, in order to verify the accuracy of the head-face clustering analysis and the feasibility and accuracy of the classification rule, the discrimination analysis is performed on the initial 400 head-face data and the newly added 50 head-face feature data; the feasibility and the accuracy of the classification rule are verified through the accuracy of the classification result of the percentage discriminant analysis and the new sample, and the percentage discriminant indexes comprise ungrouped observation values, estimated categories and discriminant categories.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011441258.7A CN112471643B (en) | 2020-12-08 | 2020-12-08 | 3D child protection mask construction method based on head-face modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011441258.7A CN112471643B (en) | 2020-12-08 | 2020-12-08 | 3D child protection mask construction method based on head-face modeling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112471643A CN112471643A (en) | 2021-03-12 |
| CN112471643B true CN112471643B (en) | 2024-01-23 |
Family
ID=74941527
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011441258.7A Active CN112471643B (en) | 2020-12-08 | 2020-12-08 | 3D child protection mask construction method based on head-face modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112471643B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203538425U (en) * | 2013-09-29 | 2014-04-16 | 孙金良 | Mask |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1619969A4 (en) * | 2003-05-02 | 2007-04-18 | Op D Op Inc | Lightweight ventilated face shield frame |
| US20060023228A1 (en) * | 2004-06-10 | 2006-02-02 | Geng Zheng J | Custom fit facial, nasal, and nostril masks |
| US10220172B2 (en) * | 2015-11-25 | 2019-03-05 | Resmed Limited | Methods and systems for providing interface components for respiratory therapy |
-
2020
- 2020-12-08 CN CN202011441258.7A patent/CN112471643B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203538425U (en) * | 2013-09-29 | 2014-04-16 | 孙金良 | Mask |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112471643A (en) | 2021-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022099958A1 (en) | Head-face dimension classification method based on three-dimensional point cloud coordinates | |
| Ren et al. | Invariant-feature-pattern-based form characterization for the measurement of ultraprecision freeform surfaces | |
| CN106504331A (en) | Tooth modeling method based on three-dimensional model search | |
| CN101344373A (en) | Standardization processing method based on three-dimensional head and face curved surface modeling | |
| CN109871818B (en) | Face recognition method based on normal vector distribution histogram and covariance descriptor | |
| CN112471643B (en) | 3D child protection mask construction method based on head-face modeling | |
| Baek et al. | Statistical foot-shape analysis for mass-customisation of footwear | |
| Lee et al. | Classification of body shape characteristics of women's torsos using angles | |
| CN107133928A (en) | A kind of facial symmetry plane acquiring method based on iteration mirror image with registration | |
| CN102063719A (en) | Local matching method of three-dimensional model | |
| CN1971274A (en) | Method for assessing smoothness level of dress material based on point model | |
| Yu et al. | Rapid and precise reverse engineering of complex geometry through multi-sensor data fusion | |
| CN110307804B (en) | Quantitative evaluation method for curve/curved surface quality | |
| CN110569539B (en) | Virtual model construction method with geometric error based on actual measurement point cloud fractal fusion | |
| CN112471649A (en) | Point cloud-based real-time three-dimensional human body measurement method and system and storage medium | |
| CN104599325A (en) | Three-dimensional face resampling method | |
| CN109035404A (en) | A kind of human foot model positioning transformations method and system | |
| CN106485667B (en) | A kind of cranium face restored method | |
| Huang et al. | A sampling method based on improved firefly algorithm for profile measurement of aviation engine blade | |
| CN105868727A (en) | Three-dimensional face similarity measurement method | |
| JP4971811B2 (en) | Function database generation method and function database generation device | |
| Li et al. | A mannequin modeling method based on section templates and silhouette control | |
| CN117436280A (en) | Wave skirt rapid customization platemaking production method based on virtual technology | |
| Gao et al. | Quality judgment of 3D face point cloud based on feature fusion | |
| CN117338084A (en) | Inclined skirt modeling detection method |
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 |