CN113545773B - Human body size measuring method and device, electronic equipment and readable storage medium - Google Patents
Human body size measuring method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN113545773B CN113545773B CN202110827397.1A CN202110827397A CN113545773B CN 113545773 B CN113545773 B CN 113545773B CN 202110827397 A CN202110827397 A CN 202110827397A CN 113545773 B CN113545773 B CN 113545773B
- Authority
- CN
- China
- Prior art keywords
- human body
- model
- arm area
- region
- registration
- 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
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1072—Measuring physical dimensions, e.g. size of the entire body or parts thereof measuring distances on the body, e.g. measuring length, height or thickness
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1079—Measuring physical dimensions, e.g. size of the entire body or parts thereof using optical or photographic means
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Dentistry (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention relates to the field of computer vision and computer graphics, and discloses a human body dimension measuring method, a human body dimension measuring device, electronic equipment and a readable storage medium. In the invention, a human body scanning model and a preset template model are obtained, wherein the template model is a parameterized human body simulation model and comprises regions of all parts of a human body which are defined in advance; registering the template model and the human body scanning model to obtain a registered template model as a registration model; obtaining the regions of all parts of the human body in the registration model, and carrying out region division on the human body scanning model to obtain at least two human body regions; and carrying out slicing processing and skeleton point positioning on each human body region, and obtaining the dimension and the length of each human body region as the result of human body size measurement. The region division is more accurate, and the human body size measurement is more accurate.
Description
Technical Field
The embodiment of the invention relates to the field of computer vision and computer graphics, in particular to a human body dimension measuring method, a human body dimension measuring device, electronic equipment and a readable storage medium.
Background
Anthropometry provides information on the shape of the human body, such as circumference, shoulder distance, etc., and has important applications in garment design, medical science, and ergonomics. With the rise of the demands of online shopping, private customization and the like, the manual measurement method by using physical tools such as a tape measure and the like has a plurality of limitations, so that a full-automatic human body measurement technology is brought forward.
The current common full-automatic human body measuring method comprises three steps: 1) generating a three-dimensional human body model according to the human body picture; 2) dividing different regions of the three-dimensional human body model; 3) and measuring the required length or circumference value of the human body part in a specific area. For the division of the region, the conventional region division method generally includes: and detecting key points from the human body picture, restoring the detected key points into the three-dimensional human body scanning model to be used as skeleton points of the three-dimensional human body model, and using the skeleton points of the three-dimensional human body model as the basis for defining the region. However, the key points detected from the human body picture cannot accurately define the regions in the three-dimensional human body model, and the accuracy of the detected key point positions can greatly influence the accuracy of the defined regions of the three-dimensional human body model, so that the accuracy of the calculated size of each region is influenced.
Disclosure of Invention
The embodiment of the invention aims to provide a human body size measuring method, a human body size measuring device, electronic equipment and a readable storage medium, so that the region division is more accurate, and the human body size measurement is more accurate.
In order to solve the above technical problem, an embodiment of the present invention provides a human body dimension measuring method, including the steps of: acquiring a human body scanning model and a preset template model, wherein the template model is a parameterized human body simulation model and comprises regions of all parts of a human body which are planned in advance; registering the template model and the human body scanning model to obtain a registered template model as a registration model; obtaining the regions of all parts of the human body in the registration model, and carrying out region division on the human body scanning model to obtain at least two human body regions; and carrying out slicing processing and skeleton point positioning on each human body region, and acquiring the dimension and the length of each human body region as the result of human body size measurement.
An embodiment of the present invention also provides a human body dimension measuring device, including: the model acquisition module is used for acquiring a human body scanning model and a preset template model, wherein the template model is a parameterized human body simulation model and comprises regions of all parts of a human body which are defined in advance; the model acquisition module is used for registering the template model and the human body scanning model to obtain a registered template model as a registration model; the region division module is used for acquiring regions of all parts of the human body in the registration model and carrying out region division on the human body scanning model to obtain at least two human body regions; and the size calculation module is used for carrying out slicing processing and skeleton point positioning on each human body region, acquiring the dimension and the length of each human body region as the result of human body size measurement.
An embodiment of the present invention also provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the body dimension measuring methods.
Embodiments of the present invention also provide a computer-readable storage medium, which when executed by a processor implements any one of the human body dimension measuring methods.
The human body size measuring method, the device, the electronic equipment and the readable storage medium provided by the embodiment of the invention register the template model and the human body scanning model to obtain the registered template model as a registered model, the essence of the registration is to adjust the template model to ensure that the template model is superposed with the human body scanning model as much as possible, because the template model comprises the pre-defined accurate regions of each human body part, the registered model also has the accurate regions of each human body part, the human body scanning model is subjected to region division based on the regions of each human body part in the registered model, because the human body region division is not carried out depending on key points in images, the problem of inaccurate human body region division caused by the inaccuracy of key point measurement and calculation in the images can be avoided, the region division precision of the human body scanning model is improved, the region division is closer to the region division of a real human body, and further, the size of the human body measured according to the divided human body regions is more accurate.
In addition, according to the human body size measuring method provided by the embodiment of the invention, the arm area is divided into an upper arm area and a lower arm area according to the obtained skeleton point of the arm area; acquiring directions of the upper arm area and the lower arm area respectively; determining a first slicing direction corresponding to the upper arm area and a second slicing direction corresponding to the lower arm area according to the directions of the upper arm area and the lower arm area; and respectively slicing the upper arm area and the lower arm area through the first slicing direction and the second slicing direction to obtain the dimensions of the upper arm area and the lower arm area. The arm area is further divided, slicing is not simply carried out in the horizontal direction, the slicing direction is attached to the direction of the real area, and slicing accuracy is guaranteed.
Drawings
One or more embodiments are illustrated by way of example in the accompanying drawings, which correspond to the figures in which like reference numerals refer to similar elements and which are not to scale unless otherwise specified.
FIG. 1 is a flow chart of a body dimension measuring method provided by an embodiment of the present invention;
2(a) -2 (c) are schematic diagrams illustrating initial registration provided by an embodiment of the present invention;
3(a) -3 (c) are schematic diagrams illustrating the fine registration provided by the embodiment of the present invention;
FIG. 4 is a schematic diagram of region partitioning of a registration model provided by an embodiment of the present invention;
FIG. 5 is a schematic diagram of region division of a human body scanning model provided by the embodiment of the invention;
FIG. 6 is a schematic diagram of a peripheral point set corresponding to a section provided by an embodiment of the present invention;
FIGS. 7(a) -7 (b) are schematic diagrams of arm region slices provided by embodiments of the present invention;
FIGS. 8(a) -8 (b) are schematic views of an embodiment of the present invention providing a breast region for slice processing;
FIGS. 9(a) -9 (b) are schematic views of a skeleton point of a chest region provided in an embodiment of the present invention;
FIG. 10 is a diagram of an embodiment of the present invention providing anthropometric results;
FIG. 11 is a schematic structural view of a human body dimension measuring apparatus according to a third embodiment of the present invention;
fig. 12 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, it will be appreciated by those of ordinary skill in the art that numerous technical details are set forth in order to provide a better understanding of the present application in various embodiments of the present invention. However, the technical solution claimed in the present application can be implemented without these technical details and various changes and modifications based on the following embodiments.
The following embodiments are divided for convenience of description, and should not constitute any limitation to the specific implementation manner of the present invention, and the embodiments may be mutually incorporated and referred to without contradiction.
Embodiments of the present invention relate to a human body size measuring method. The specific flow is shown in figure 1.
Step 101: the method comprises the steps of obtaining a human body scanning model and a preset template model, wherein the template model is a parameterized human body simulation model and comprises regions of all parts of a human body which are defined in advance.
Step 102: and registering the template model and the human body scanning model to obtain a registered template model as a registration model.
The human body scanning model is a three-dimensional model and comprises a point set of a certain number of points, and the points in the point set are distributed in the three-dimensional model to form the human body scanning model of the human body to be detected. The template model is a preset parameterized human body simulation model divided into regions of each part of the human body. And carrying out registration processing on the acquired template model and the acquired human body scanning. The posture of the human body scanning model is not fixed, and can be an A-position (namely the normal standing posture of the human body), a T-position (namely the two arms are unfolded and form a T-shaped posture with the trunk), an arm lifting posture and the like. That is, the human body scanning model can be suitable for any postures in which the human body regions do not overlap and do not overlap. The posture of the template model is not fixed, and like the human body scanning model, the template model can be an A-position (namely the normal standing posture of the human body), a T-position (namely the two arms are unfolded and form a T-shaped posture with the trunk), an arm lifting posture and the like. But in practice a-dose or T-dose will generally be used as the pose for the phantom. When the postures of the human body scanning model and the template model are more similar, the adjustment amount and the calculation amount are smaller when the model registration is subsequently carried out, and the accuracy of the model registration can be further influenced to a certain extent.
It should be noted that, in this embodiment, registering the template model and the human body scanning model refers to an operation of adjusting the template model so that the template model may coincide with the human body scanning model.
In the present embodiment, the points in the human scan model and the registration model are feature points that constitute the peripheral shape of the model. That is, the human body scan model and the template model are constituted by feature points representing the peripheral shape of the model.
And adjusting the positions and postures of the human body scanning model and the template model so as to minimize the distance between corresponding points of the human body scanning model and the template model.
Since the template model is a parameterized model, relevant details of the template model, such as shape, posture, position, etc., can be changed to some extent by adjusting parameters.
Optionally, performing initial registration on the template model according to the height and the position of the human body scanning model; and performing fine registration on the template model subjected to initial registration according to a preset loss function so as to minimize the distance between corresponding points of the human body scanning model and the registration model, and taking the adjusted template model as the registration model.
The process of registering the template model and the human body scanning model comprises the following steps: initial registration and fine registration.
The initial registration optionally includes: carrying out scaling processing on the template model according to the human body scanning model; calculating to obtain the centroid distance and the lowest point distance between the human body scanning model and the template model after scaling; and carrying out translation processing on the template model after the scaling processing according to the centroid distance and the lowest point distance.
Specifically, the initial registration includes: and scaling and translating the template model according to the human body scanning model. And adjusting the height of the template model according to the height of the human body scanning model, namely the distance from the lowest point to the highest point of the human body scanning model, so that the height of the human body scanning model is the same as that of the template model. Respectively calculating the mass centers of the human body scanning model and the scaled template model, wherein the calculation method of the mass centers takes the mass center calculation of the human body scanning model as an example: and acquiring the position coordinates of all points in the human body scanning model, and calculating to obtain the average value of the position coordinates of all points as the mass center of the human body scanning model. And translating the template model to ensure that the mass center of the human body scanning model is superposed with the mass center of the template model. And acquiring the lowest points of the human body scanning model and the template model, and translating the template model to enable the lowest point of the human body scanning model and the lowest point of the template model to be on the same horizontal plane. The human body scanning model and the template model are overlapped through the translation of the centroid and the lowest point. The initial registration is shown in fig. 2, in which fig. 2(a) is a template model, fig. 2(b) is a human body scanning model, and fig. 2(c) is an initial registration result of the template model and the human body scanning model, that is, the template model after the initial registration.
The fine registration optionally includes: calculating the loss value of the distance between each point of the human body scanning model and the corresponding point in the template model after initial registration according to the loss function; and respectively adjusting the human body scanning model and the template model subjected to initial registration according to the loss value so as to minimize the distance between the template model subjected to initial registration and corresponding points of the human body scanning model, and taking the adjusted template model as a registration model.
And carrying out fine registration on the human body scanning model and the template model subjected to initial registration through a first loss function. In this example, the template model subjected to the initial registration is referred to as an intermediate model for ease of understanding. The closest points of the human scan model and the intermediate model are obtained through a data structure constructed by a k-dimensional tree (k-dTree). The k-dTree segments the data structure of the k-dimensional data space for searching the key data of the multi-dimensional space. In this embodiment, the closest point between the human scan model and the intermediate model is searched for in three-dimensional space by k-dTere. And calculating loss values between the nearest points through the first loss function, and adjusting the human body scanning model and the intermediate model simultaneously by adjusting parameters involved in the first loss function to minimize the loss values so as to realize fine registration of the intermediate model. And taking the intermediate model subjected to fine registration as a registration model. The fine registration is shown in fig. 3, wherein fig. 3(a) the intermediate model, fig. 3(b) the phantom, fig. 3(c) the result of the fine registration of the intermediate model and the phantom.
In addition, besides adjusting parameters of the human body scanning model and the intermediate model simultaneously, parameters of the intermediate model can be adjusted independently, and the intermediate model is adjusted by calculating a loss value of the second loss function and adjusting parameters related in the second loss function, so that the loss value is minimized.
Specifically, the first loss function is shown as formula (1), the intermediate model is a parameterized model, the result of formula (1) is minimized by adjusting the parameters of the intermediate model and the coefficients of the human body scanning model, and the intermediate model and the human body scanning model are adjusted by the values of the parameters and the coefficients corresponding to the result of formula (1) when the result of formula (1) is minimized, so as to complete fine registration.
The parameters comprise shape parameters and posture parameters which are respectively used for adjusting the shape and the posture of the middle model; the coefficients include a rotation coefficient, a translation coefficient, and a scale coefficient, wherein the rotation coefficient is used to rotate a point in the scan model, the translation coefficient is used to translate a point on the scan model, the scale coefficient is used to adjust the scale of the registration model, and when the units of the intermediate model and the scan model are not consistent, the intermediate model is adjusted to the same unit as the scan model, for example, the scale of the intermediate model is in centimeters, the scale of the scan model is in meters, and the registration model is adjusted to be in meters by the scale coefficient.
In the formula (1), R is a rotation coefficient, t is a translation coefficient, s is a scale coefficient, beta is a shape parameter, theta is an attitude parameter, and p isiObtaining a closest point, q, of a scan model and a registration model of a human body for a data structure constructed by a k-dimensional tree (k-dTere)iAnd (β, θ) is a point on the intermediate model.
Besides rotating and translating the points on the human body scanning model, the offset of the middle model can be adjusted, wherein the offset refers to the offset of each point of the middle model in the normal direction of the middle model, and the middle model can be attached to the human body scanning model more closely by adjusting the offset. As shown in formula (2), the shape parameter of the intermediate model, the attitude parameter of the intermediate model, and the offset of the intermediate model are adjusted to minimize the value of formula (2). In formula (2), D is the offset of the intermediate model, β is the shape parameter, θ is the pose parameter, pi is the closest point of the human body scanning model and the intermediate model obtained by the data structure constructed by the k-dimensional tree (k-dTree), and qi (β, θ) is the point on the intermediate model.
The offset of the intermediate model can also be adjusted separately, as shown in equation (3). Wherein D is the offset of the intermediate model, pi is the closest point of the human body scanning model and the intermediate model obtained by a data structure constructed by a k-dimensional tree (k-dTree), and q is the offset of the intermediate modeliAnd (β, θ) is a point on the intermediate model.
The fine registration of the intermediate model and the human body scanning model is realized through formulas (1), (2) and (3).
The initial registration and the fine registration maximize the coincidence degree of the human body scanning model and the template model.
Step 103: and obtaining the regions of all parts of the human body in the registration model, and carrying out region division on the human body scanning model to obtain at least two human body regions.
Registering the template model with the scan body model is essentially by adjusting the template model so that the template model is adjusted to coincide with the scan body model. The registered template model is used as a registration model, and since the template model defines the regions in advance, the regions of each part of the human body in the registered template model, that is, the regions of each part of the human body in the registration model, as shown in fig. 4, including the chest region, the waist region, the arm region, the leg region, and the like, can be obtained. Since the human body scanning model and the registration model are overlapped to the maximum extent, the human body scanning model is subjected to region division according to the region defined by the registration model, and the region division of the human body scanning model according to the registration model is shown in fig. 5. In step 102, the initial registration and the fine registration are performed to maximize the degree of coincidence between the human body scanning model and the registration model, so that the human body region of the segmented human body scanning model is more accurate according to the regions of the parts of the human body in the registration model. After accurate human body region division is realized, the dimension and the length of each region obtained through calculation are more accurate.
Step 104: and carrying out slicing processing and skeleton point positioning on each human body region, and acquiring the dimension and the length of each human body region as the result of human body size measurement.
When the dimension of the human body area is calculated, the human body area is sliced, and the dimension of the human body area is calculated through a section obtained by slicing. However, if the human body regions are sliced in the same direction, erroneous calculation results, such as arm regions and leg regions, are obtained for some human body regions with special angles. When the human body to be measured is scanned, because the standing postures of the human body are different, the directions of the arm area, the leg area and the like of the model obtained by scanning are not completely uniform with those of other human body areas, so that if the same slicing direction is adopted, the accuracy of the measurement result is influenced. Therefore, the human body area is further divided, and the slicing direction is independently set according to different directions of different human body areas.
It should be noted that, the human body scanning model is divided into different human body regions, and the human body regions include: chest area, waist area, arm area, leg area, etc. When determining the direction of each human body region, because the human body is in a normal standing posture, the trunk (especially, the trunk is a part of the human body without head, neck and limbs, and at least includes the chest region and the waist region in the present scheme) of the human body can be determined to be in an upright state, i.e., perpendicular to the ground horizontal plane, that is, the directions of the chest region and the waist region are both perpendicular to the directions of the four horizontal planes and are normal vectors of the ground horizontal plane. For the arm area and the leg area, the directions may not be completely the same as the direction of the trunk due to different standing postures of the human body, so the directions of the arm area and the leg area need to be determined separately according to the skeleton point. The template model further includes pre-labeled human skeleton points, and the preliminary registration and the precise registration in step 102 enable the registration model to be approximately equal to the human scanning model, so that the registration model also has accurate skeleton points, and the skeleton points in the registration model are equivalent to the skeleton points of the human scanning model, so that the positions of the skeleton points of the registration model can be used as the skeleton points of the human scanning model. Taking the leg region as an example, skeleton points of the leg are determined in the human body scanning model, and the direction of a vector formed by coordinates of the skeleton points is taken as the direction of the leg region.
When the leg area and the arm area are sliced, the direction of the slice is determined according to the direction of the human body area determined by the skeleton point. Specifically, a vector of the human body region is composed of skeleton points of the human body region, a plane perpendicular to the vector of the human body region is determined, and a plane obtained by intersecting the perpendicular plane and the human body region is used as a tangent plane obtained by slicing. Since the human body scan model is a model composed of a certain number of points and including only the peripheral shape, a portion where the human body scan model intersects with the perpendicular plane is actually a certain number of point sets. For each human body region, slicing is performed at intervals to obtain a plurality of slices, and a peripheral point set corresponding to each slice is shown in fig. 6.
Optionally, the human body region at least includes an arm region, and the dimension and the length of each human body region are obtained by calculating each human body region respectively, including: obtaining skeleton points of an arm area; dividing an arm area into an upper arm area and a lower arm area according to a skeleton point; respectively acquiring the directions of an upper arm area and a lower arm area; determining a first slicing direction corresponding to the upper arm area according to the direction of the upper arm area, and determining a second slicing direction corresponding to the lower arm area according to the direction of the lower arm area; slicing the upper arm area through a first slicing direction to obtain a plurality of first tangent planes of the upper arm area, and calculating the dimension of the upper arm area through the first tangent planes; and slicing the lower arm area through the second slicing direction to obtain a plurality of second sections of the lower arm area, and calculating the dimension of the lower arm area through the second sections.
Taking the arm area as an example, because the direction of the arm area of the human body is not consistent with the directions of the human body areas such as the trunk area of the human body, that is, the vector corresponding to the arm direction has an included angle with the vector corresponding to the trunk direction, and when the human body arm naturally droops, the direction of the upper arm area of the human body is also inconsistent with the direction of the lower arm area of the human body, that is, the upper arm area and the lower arm area are not on the same straight line, and the vector corresponding to the direction of the upper arm area and the vector corresponding to the lower arm area have an included angle, if the slicing processing is performed on the whole arm area in the same slicing direction, the accuracy of the arm dimension obtained by calculating the slicing processing on the slice obtained by the slicing processing will be affected. The template model further includes pre-labeled human skeleton points, and the registration model can be approximately equal to the human scanning model by performing registration in step 102, so that the registration model also includes the human skeleton points, and the skeleton points on the registration model are equivalent to the skeleton points of the human scanning model, and therefore, the arm region of the human scanning model is divided into an upper arm region and a lower arm region by using the skeleton points on the registration model as a segmentation basis. The corresponding slice directions are determined for the directions of the upper arm region and the lower arm region, respectively. And respectively carrying out multiple slicing treatments on the upper arm area and the lower arm area at intervals of a certain distance in a first slicing direction corresponding to the upper arm area and a second slicing direction corresponding to the lower arm area to obtain a plurality of first sections and a plurality of second sections.
Optionally, slicing the upper arm region through the first direction to obtain a plurality of first facets of the upper arm region, and calculating a dimension of the upper arm region through the first facets, includes: acquiring points on the periphery corresponding to each first tangent plane to obtain a peripheral point set of each first tangent plane; for each first tangent plane, determining the minimum convex hull length of the first tangent plane according to the peripheral point set corresponding to the first tangent plane; selecting one minimum convex hull length meeting preset conditions from the minimum convex hull lengths corresponding to the first tangent planes as the dimension of the upper arm area; slicing the lower arm area through the second slicing direction to obtain a plurality of second sections of the lower arm area, and calculating the dimension of the lower arm area through the second sections, wherein the step comprises the following steps: acquiring points on the periphery corresponding to each second tangent plane to obtain a peripheral point set of each second tangent plane; for each second tangent plane, determining the minimum convex hull length of the second tangent plane according to the peripheral point set corresponding to the second tangent plane; and selecting one minimum convex hull length meeting a preset condition from the minimum convex hull lengths corresponding to the second tangent planes as the dimension of the lower arm area. Fig. 7 is a schematic view of slicing an arm region, fig. 7 a is a schematic view of slicing without division of upper and lower arm regions, and fig. b is a schematic view of slicing with division of the upper and lower arm regions.
After the tangent plane is determined, a peripheral point set of each tangent plane is obtained according to the points on the periphery corresponding to the slices, and the peripheral point set is calculated to obtain the minimum convex hull length of the point set. The convex hull is a minimum external polygonal convex edge containing all the points in the point set, in a two-dimensional Euclidean space, the convex hull can be imagined as a rubber band just covering all the points, and the fitting degree of the tape measure when the human body is measured can be simulated. In this embodiment, the convex hull is calculated using the GrahamScan algorithm to obtain the minimum convex hull length. However, the method for calculating the convex hull is not limited to this, and the minimum convex hull length can be calculated by using other algorithms besides the GrahamScan algorithm.
In a human body region, a plurality of sections exist, each section has a minimum convex hull length, and the minimum convex hull length meeting a preset condition is selected from the minimum convex hull lengths of each section as the dimension of the human body region. The preset conditions may be: the minimum convex hull length with the largest value among all the minimum convex hull lengths in the human body region, or the minimum convex hull length with the smallest value among all the minimum convex hull lengths in the human body region. In addition, the average of all minimum convex hull lengths can be set as the dimension of the human body region.
When the human body region is sliced, the human body region itself is sliced only in the slice direction corresponding to the human body region, and the other human body regions are not sliced in the slice direction. For example, as shown in fig. 8, fig. 8 is a schematic diagram of slicing a chest region, fig. 8(a) shows a schematic diagram of slicing without accurately dividing a slice region, and fig. 8(b) shows a schematic diagram of slicing with accurately dividing a slice region, that is, only a divided chest region is sliced.
When the length of the human body region is calculated, the Euclidean distance between skeleton points of the same human body region corresponding to the registration model after primary registration and fine registration is used as the length of the region. Fig. 9 is a skeleton point diagram of the registration model after the preliminary registration and the fine registration, in which fig. 9(a) is a front view of the skeleton point diagram, and fig. 9(b) is a side view of the skeleton point diagram.
The measurement result of the human body obtained by the human body dimension measurement method according to the present embodiment is exemplified by the measurement result of the human body shown in fig. 10. In fig. 10, 26 types of size measurement results are included, for example, height, shoulder width, and the like are length information, and a measurement result is obtained by a length calculation method, and waist length, hip circumference, and the like are dimension information, and a measurement result is obtained by a dimension calculation method.
Compared with the related art, the embodiment of the invention has the advantages that the positions and postures of the sample model and the human body scanning model are closer by adjusting the sample model, the similarity between the sample model and the human body scanning model is higher, the region division of the human body scanning model is closer to the region division of the sample model when the region of the human body scanning model is divided according to the sample model, the region division accuracy is higher, the region division is closer to the region division of a real human body, and the human body size measured according to the divided region is more accurate.
The steps of the above methods are divided for clarity, and the implementation may be combined into one step or split some steps, and the steps are divided into multiple steps, so long as the same logical relationship is included, which are all within the protection scope of the present patent; it is within the scope of the patent to add insignificant modifications to the algorithms or processes or to introduce insignificant design changes to the core design without changing the algorithms or processes.
An embodiment of the present invention relates to a human body size measuring apparatus, as shown in fig. 11, including:
the model acquisition module 1101 is used for acquiring a human body scanning model and a preset template model, wherein the template model is a parameterized human body simulation model and comprises regions of each part of a human body which are defined in advance;
a model obtaining module 1102, configured to register the template model with the human body scanning model, and obtain a registered template model as a registration model;
the region division module 1103 is configured to obtain regions of all parts of the human body in the registration model, and perform region division on the human body scanning model to obtain at least two human body regions;
and the size calculation module 1104 is configured to perform slicing processing and skeleton point positioning on each human body region, and obtain dimensions and lengths of each human body region as a result of human body size measurement.
It should be understood that this embodiment is an example of an apparatus corresponding to the other embodiments, and that this embodiment can be implemented in cooperation with the other embodiments. Related technical details mentioned in other embodiments are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to other embodiments.
It should be noted that each module referred to in this embodiment is a logical module, and in practical applications, one logical unit may be one physical unit, may be a part of one physical unit, and may be implemented by a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, elements that are not so closely related to solving the technical problems proposed by the present invention are not introduced in the present embodiment, but this does not indicate that other elements are not present in the present embodiment.
An embodiment of the present invention relates to an electronic device, as shown in fig. 12, including:
at least one processor 1201; and a memory 1202 communicatively coupled to the at least one processor 1201; wherein the memory 1202 stores instructions executable by the at least one processor 1201, the instructions being executable by the at least one processor 1201 to enable the at least one processor 1201 to perform the body dimension measuring method of any of the embodiments.
The memory and the processor are connected by a bus, which may include any number of interconnected buses and bridges, linking together one or more of the various circuits of the processor and the memory. The bus may also link various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor is transmitted over a wireless medium via an antenna, which further receives the data and transmits the data to the processor.
The processor is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And the memory may be used to store data used by the processor in performing operations.
Embodiments of the present invention relate to a computer-readable storage medium, and a computer program is executed by a processor to implement the above-described human body dimension measuring method.
Those skilled in the art can understand that all or part of the steps in the method of the foregoing embodiments may be implemented by a program to instruct related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, etc.) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the invention, and that various changes in form and details may be made therein without departing from the spirit and scope of the invention in practice.
Claims (9)
1. A method of measuring a dimension of a human body, comprising:
acquiring a human body scanning model and a preset template model, wherein the template model is a parameterized human body simulation model and comprises regions of all parts of a human body which are planned in advance;
registering the template model and the human body scanning model to obtain a registered template model as a registration model;
obtaining the regions of all parts of the human body in the registration model, and carrying out region division on the human body scanning model to obtain at least two human body regions, wherein the human body regions at least comprise arm regions;
slicing each human body region and positioning skeleton points to obtain the dimension and length of each human body region as the result of human body size measurement;
wherein, the slicing processing and length positioning are performed on each human body region, the dimension and length of each human body region are obtained, and the obtained dimension and length are used as the result of human body size measurement, and the method comprises the following steps: dividing the arm area into an upper arm area and a lower arm area according to the acquired skeleton point of the arm area; acquiring directions of the upper arm area and the lower arm area respectively; determining a first slicing direction corresponding to the upper arm area and a second slicing direction corresponding to the lower arm area according to the directions of the upper arm area and the lower arm area; and respectively slicing the upper arm area and the lower arm area through the first slicing direction and the second slicing direction to obtain the dimensions of the upper arm area and the lower arm area.
2. The human body dimension measuring method according to claim 1, wherein the human body scanning model and the template model are constituted by a set of points;
registering the template model and the human body scanning model to obtain a registered template model as a registration model, wherein the registering comprises:
carrying out initial registration on the template model according to the height and the position of the human body scanning model and the parameter data of the template model;
and carrying out fine registration on the template model and the human body scanning model after the initial registration according to a preset loss function to obtain the registration model, wherein the distance between corresponding points of the registration model and the human body scanning model is minimum.
3. The method of claim 2, wherein the initially registering the template model according to the height and position of the scan body model and the parameter data of the template model comprises:
carrying out scaling processing on the template model according to the human body scanning model;
acquiring a centroid distance and a lowest point distance between the human body scanning model and the template model after scaling;
and carrying out translation processing on the template model after the scaling processing according to the centroid distance and the lowest point distance.
4. The method for measuring the human body size according to claim 2, wherein the fine registration of the template model and the human body scanning model after the initial registration according to a preset loss function to obtain the registration model comprises:
calculating the loss value of the distance between each point of the human body scanning model and the corresponding point in the template model after the initial registration according to the loss function;
and respectively adjusting the human body scanning model and the template model subjected to initial registration according to the loss value so as to minimize the distance between the template model subjected to initial registration and the corresponding point of the human body scanning model, and taking the adjusted template model as the registration model.
5. The method according to claim 1, wherein the obtaining dimensions of the upper arm region and the lower arm region by slicing the upper arm region and the lower arm region in the first slicing direction and the second slicing direction, respectively, comprises:
obtaining at least two first sections corresponding to the upper arm area and at least two second sections corresponding to the lower arm area according to the first slicing direction and the second slicing direction;
acquiring points on the periphery corresponding to each first tangent plane and each second tangent plane to obtain a peripheral point set of each first tangent plane and each second tangent plane;
for each first tangent plane or second tangent plane, determining the minimum convex hull length of the first tangent plane or second tangent plane according to the peripheral point set corresponding to the first tangent plane or the second tangent plane;
and selecting one minimum convex hull length meeting preset conditions from the minimum convex hull lengths corresponding to the first tangent plane and the second tangent plane respectively as the dimension of the upper arm area and the dimension of the lower arm area.
6. The human body dimension measuring method according to claim 1 or 5, wherein the slicing processing is performed on the upper arm region and the lower arm region respectively through the first slicing direction and the second slicing direction, and after obtaining dimensions of the upper arm region and the lower arm region, the method further comprises:
and positioning the positions of skeleton points of the upper arm area and the lower arm area to obtain the distance between the skeleton points, wherein the distances are respectively used as the length of the upper arm area and the length of the lower arm area.
7. A human body dimension measuring device, comprising:
the model acquisition module is used for acquiring a human body scanning model and a preset template model, wherein the template model is a parameterized human body simulation model and comprises regions of all parts of a human body which are defined in advance;
the model acquisition module is used for registering the template model and the human body scanning model to obtain a registered template model as a registration model;
the region division module is used for acquiring regions of all parts of the human body in the registration model and carrying out region division on the human body scanning model to obtain at least two human body regions, wherein the human body regions at least comprise arm regions;
the size calculation module is used for carrying out slicing processing and skeleton point positioning on each human body region, and obtaining the dimension and the length of each human body region as the result of human body size measurement;
the slicing processing and length positioning are performed on each human body region, the dimension and the length of each human body region are obtained, and the obtained dimension and length are used as the result of human body size measurement, and the method comprises the following steps: dividing the arm area into an upper arm area and a lower arm area according to the acquired skeleton point of the arm area; acquiring directions of the upper arm area and the lower arm area respectively; determining a first slicing direction corresponding to the upper arm area and a second slicing direction corresponding to the lower arm area according to the directions of the upper arm area and the lower arm area; and respectively slicing the upper arm area and the lower arm area through the first slicing direction and the second slicing direction to obtain the dimensions of the upper arm area and the lower arm area.
8. An electronic device, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the body dimension measuring method of any of claims 1-6.
9. A computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the human body dimension measuring method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110827397.1A CN113545773B (en) | 2021-07-21 | 2021-07-21 | Human body size measuring method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110827397.1A CN113545773B (en) | 2021-07-21 | 2021-07-21 | Human body size measuring method and device, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113545773A CN113545773A (en) | 2021-10-26 |
| CN113545773B true CN113545773B (en) | 2022-05-31 |
Family
ID=78103960
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110827397.1A Active CN113545773B (en) | 2021-07-21 | 2021-07-21 | Human body size measuring method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113545773B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114373040A (en) * | 2021-12-13 | 2022-04-19 | 聚好看科技股份有限公司 | Three-dimensional model reconstruction method and acquisition terminal |
| CN114788923A (en) * | 2022-04-08 | 2022-07-26 | 青岛港国际股份有限公司 | Intelligent epidemic prevention house |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007160108A (en) * | 2005-12-12 | 2007-06-28 | Siemens Medical Solutions Usa Inc | System and method for image based physiological monitoring of cardiovascular function |
| CA3071399A1 (en) * | 2017-07-28 | 2019-01-31 | Memorial Sloan Kettering Cancer Center | Systems and methods for designing and manufacturing custom immobilization molds for use in medical procedures |
| CN108986159B (en) * | 2018-04-25 | 2021-10-22 | 浙江森马服饰股份有限公司 | Method and equipment for reconstructing and measuring three-dimensional human body model |
| CN110288646A (en) * | 2019-06-21 | 2019-09-27 | 北京邮电大学 | A kind of human dimension calculation method and device based on image |
| CN112401369A (en) * | 2020-11-23 | 2021-02-26 | 叠境数字科技（上海）有限公司 | Body parameter measuring method, system, equipment, chip and medium based on human body reconstruction |
-
2021
- 2021-07-21 CN CN202110827397.1A patent/CN113545773B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113545773A (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10460517B2 (en) | Mobile device human body scanning and 3D model creation and analysis | |
| CN113545773B (en) | Human body size measuring method and device, electronic equipment and readable storage medium | |
| CN104679831B (en) | Method and device for matching human body model | |
| US6804683B1 (en) | Similar image retrieving apparatus, three-dimensional image database apparatus and method for constructing three-dimensional image database | |
| US6664956B1 (en) | Method for generating a personalized 3-D face model | |
| CN102525662B (en) | Three-dimensional visual tissue organ operation navigation system | |
| CN101311967B (en) | Dummy body form establishment method and dummy body form based on body type of actual measurement for crowds | |
| CN106780619A (en) | A kind of human body dimension measurement method based on Kinect depth cameras | |
| CN110335297A (en) | A kind of point cloud registration method based on feature extraction | |
| CN1864074B (en) | Method for determining patient-related information, control apparatus and magnetic resonance tomography instrument | |
| CN102157013A (en) | System for fully automatically reconstructing foot-type three-dimensional surface from a plurality of images captured by a plurality of cameras simultaneously | |
| CN103829966B (en) | For automatically determining the method and system of the position line in detecting image | |
| US9349074B2 (en) | Method and apparatus for generating 3D knee joint image | |
| CN110021053A (en) | A kind of image position method, device, storage medium and equipment based on coordinate conversion | |
| Wuhrer et al. | Landmark-free posture invariant human shape correspondence | |
| CN111400830B (en) | Machining calibration method and device for three-dimensional blank workpiece | |
| CN107610121B (en) | A kind of initial pose setting method of liver statistical shape model | |
| CN112381862A (en) | Full-automatic registration method and device for CAD (computer-aided design) model and triangular mesh | |
| US8108187B2 (en) | Method and system for surface analysis and envelope generation | |
| CN109410257A (en) | Multi information 3D medical image high registration accuracy method | |
| CN110728685B (en) | Brain tissue segmentation method based on diagonal voxel local binary pattern texture operator | |
| CN114745985A (en) | Bra sizing optimization from 3D shape of breast | |
| CN116650115A (en) | Orthopedic surgery navigation registration method based on UWB mark points | |
| CN113345079B (en) | Face three-dimensional model visualization method, device, electronic equipment and storage medium | |
| Cheng et al. | Ground truth delineation for medical image segmentation based on Local Consistency and Distribution Map analysis |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220424 Address after: 230091 room 611-217, R & D center building, China (Hefei) international intelligent voice Industrial Park, 3333 Xiyou Road, high tech Zone, Hefei, Anhui Province Applicant after: Hefei lushenshi Technology Co.,Ltd. Address before: 100083 room 3032, North B, bungalow, building 2, A5 Xueyuan Road, Haidian District, Beijing Applicant before: BEIJING DILUSENSE TECHNOLOGY CO.,LTD. Applicant before: Hefei lushenshi Technology Co., Ltd |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |