KR102044071B1 - Apparatus and method for analyzing body shape - Google Patents

Abstract

Receiving a three-dimensional body data associated with a body analysis device and method, specifically including a plurality of markers attached to the user's body; Separating the three-dimensional body data into a marker region and a non-marker region; Clustering the markers for the marker regions using a density based clustering (DBSCAN) method; Calculating three-dimensional coordinates of the clustered markers; And determining the body shape of the user using the 3D coordinates of the clustered markers.

Description

Body analysis device and method {APPARATUS AND METHOD FOR ANALYZING BODY SHAPE}

The present invention relates to a device and a method for analyzing a user's body shape, and more particularly, to a device and a method for receiving 3D body data and analyzing a user's body shape based on a marker.

In today's society, convenience and customization of health services and related systems are being strengthened by diversifying customer demands and improving expectations according to changing consumer consciousness, and preventing lifestyle-related diseases based on accumulated personal health data. And personalized healthcare projects such as weight management are growing rapidly.

Conventionally, after simply measuring the height and weight of the user, based on the analysis of the user's body type to provide a result, but does not provide accurate and detailed data on the user's body type.

Therefore, there is a need for an apparatus that provides accurate and detailed body shape information of a user based on three-dimensional body data of the user.

Korean Patent No. 10-1602674 (announced March 11, 2016) proposes a posture analysis method using a reference model. The present invention relates to a method of aligning an input body shape in the same direction as a reference model using a transform vector.

According to one embodiment, a method performed at least temporarily by a computer, the method comprising: receiving three-dimensional body data including a plurality of markers attached to a body of a user; Separating the three-dimensional body data into a marker region and a non-marker region; Clustering the markers for the marker regions using a density based clustering (DBSCAN) method; Calculating three-dimensional coordinates of the clustered markers; And determining a body shape of the user using the 3D coordinates of the clustered markers.

According to another embodiment, the separating of the marker and the non-marker region may be a body analysis method of separating using the color space information and the clustering algorithm, and further comprising interpolating a surface of the non-marker region. Figure analysis is also possible.

According to another embodiment, the determining of the body shape of the user may further include aligning a posture of a body shape reference model that is previously specified using a rigid body transformation vector T.

According to another embodiment, a method performed at least temporarily by a computer, the method comprising: receiving three-dimensional body data including a plurality of markers attached to a user's body; Calculating three-dimensional coordinates of the marker; Determining whether the marker is a spherical marker or a planar marker to remove the spherical marker if the marker is a spherical marker; Interpolating the area from which the spherical marker is removed; And determining a body shape of the user using the calculated coordinates of the marker.

According to another embodiment of the present invention, the method comprises: calculating a number of spherical markers; And calculating the number of the planar markers.

According to one embodiment, a method performed at least temporarily by a computer, the method comprising: receiving three-dimensional body data including a plurality of markers attached to a body of a user; Separating the three-dimensional body data into a marker region and a non-marker region; Clustering the markers for the marker regions using a density based clustering (DBSCAN) method; Calculating three-dimensional coordinates of the clustered markers; Determining whether the marker is a spherical marker or a planar marker to remove the spherical marker if the marker is a spherical marker; Interpolating the area from which the spherical marker is removed; And determining a body shape of the user using the 3D coordinates of the clustered markers.

According to one side, the three-dimensional body data including a plurality of markers attached to the user's body is separated into a marker region and a non-marker region, and the marker region is clustered using a density-based clustering (DBSCAN) method, A calculator for calculating three-dimensional coordinates of the clustered markers; And a determiner configured to determine a body shape of the user by using the 3D coordinates of the clustered markers.

The calculation unit may be separated by using color space information and a clustering algorithm, and may interpolate the surface of the non-marker region.

In addition, the determination unit may be aligned with the posture of the body shape reference model that is previously specified using the rigid body transformation vector T.

According to the other side receives three-dimensional body data including a plurality of markers attached to the user's body to calculate the three-dimensional coordinates of the marker, and determine whether the marker is a spherical (spherical) or planar markers A marker for removing the spherical marker and interpolating an area from which the spherical marker is removed; And a body shape analysis device including a determination unit for determining the body shape of the user using the three-dimensional coordinates of the marker is also disclosed.

According to another aspect, three-dimensional body data including a plurality of markers attached to a user's body are separated into a marker region and a non-marker region, and the marker regions are clustered by using a density-based clustering method. Calculating a three-dimensional coordinate of the marker and determining whether the marker is a spherical marker or a planar marker, and if the marker is a spherical marker, a calculator configured to remove the spherical marker and interpolate an area from which the spherical marker is removed; And a body shape analysis device including a determination unit for determining the body shape of the user using the three-dimensional coordinates of the clustered marker is disclosed.

According to an embodiment, a computer-readable recording medium including a program for performing the body type analyzing method is also disclosed.

1 illustrates a configuration of a body type analyzer according to an exemplary embodiment.
2 is a flowchart of a body type analyzing method according to an exemplary embodiment.
3 illustrates a marker detection flow according to one embodiment.
4 is a flowchart illustrating a body analysis method including automatic marker division according to an embodiment.
5 illustrates a flow for separating markers and non-marker regions using clustering of individual markers, according to one embodiment.
6 illustrates input body shape data and reference body shape data, according to an exemplary embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of rights is not limited or limited by these embodiments. Like reference numerals in the drawings denote like elements.

The terminology used in the description below has been selected to be general and universal in the art to which it relates, although other terms may vary depending on the development and / or change in technology, conventions, and preferences of those skilled in the art. Therefore, the terms used in the following description should not be understood as limiting the technical spirit, and should be understood as exemplary terms for describing the embodiments.

In addition, in certain cases, there is a term arbitrarily selected by the applicant, and in this case, the meaning thereof will be described in detail in the corresponding description. Therefore, the terms used in the following description should be understood based on the meanings of the terms and the contents throughout the specification, rather than simply the names of the terms.

1 illustrates a configuration of a body type analyzer according to an exemplary embodiment. An apparatus for analyzing a body type according to an exemplary embodiment may include a calculator 110 and a determiner 120.

The calculation unit 110 receives the 3D body shape data and finally calculates the 3D coordinates of the marker, and the determination unit 120 determines the body shape of the user using the calculated 3D coordinates of the marker.

More specifically, the calculation unit 110 receives the 3D body data of the user. The 3D body data may be a torso or whole body image of a user including a marker. Depending on the range you want to analyze, you can use only the torso information or telegraph information. In some cases, it is possible to utilize only the lower body information.

The calculation unit 110 may be separated into a marker region and a non-marker region by using the input 3D body data. For example, by comparing the number of points belonging to each area, an area including a smaller number of points may be divided into a marker area and an opposite area into a non-marker area. Even if a plurality of markers are attached to the body of the user to be measured, it is assumed that the area where no marker is present is larger than the marker area in the entire area to be photographed. In some cases, if the marker is attached very much and the marker area becomes larger than the area where the marker does not exist, it is also possible to determine the opposite. However, in general, since the marker region is smaller than the non-marker region, an area including fewer points may be divided into marker regions.

The calculation unit 110 may cluster individual markers with respect to the separated marker region. In clustering individual markers, a density-based clustering of application with noise (DBSCAN) algorithm may be applied. Limits such as the distance between each grouped group and the number of points that can be included in each group may be excluded from the marker candidate if they do not meet the above criteria.

Finally, the calculator 110 may calculate coordinates for each group of the markers. The group coordinates can be used to determine a representative value, and obtain a clustered three-dimensional coordinates of the marker attached to the main position of the user's body surface.

The determination unit 120 determines the body shape of the user using the 3D coordinates of the clustered markers calculated by the calculation unit 110. The determination unit 120 may determine a body shape that is closest to the user's body shape measured based on various body information previously stored. For example, the body shape may be determined as upper body development type, lower body development type, muscle type, skinny type, slim type, and barrel type.

2 is a flowchart of a body type analyzing method according to an exemplary embodiment. According to an embodiment, the body analysis method includes three-dimensional body data input step 210, marker / non-marker area separation step 220, density-based clustering step 230, three-dimensional coordinate calculation step 240, and body shape of the marker. Decision step 250 can be configured.

Specifically, the three-dimensional body data input step 210 is a step in which the calculator receives the three-dimensional body data measured by the user. The three-dimensional body data may be, for example, a 3D image of the user's body shape.

In the marker / non-marker region separation step 220, the calculator separates the marker region and the non-marker region. For example, the calculation unit may use color space information and clustering to separate an area. The color space information may be any one of, but not limited to, an RBG (Red, Blue, Green) color space, an HSV (Hue, Saturation, Value) color space, or HSL (Hue, Saturation, Lightness). Clustering may also be any of, but not limited to, K-means clustering based on mean values, or medoid based K-medoids clustering with high frequency. That is, the marker and the non-marker region may be separated by using the color space information and the clustering algorithm. In particular, the non-marker region is a concept opposite to the marker region, and may be a garment region that is in close contact with the skin or the body.

Next is a density based clustering step 230. Density-based clustering (DBSCAN) is performed on the marker regions to be separated. Individual markers are clustered by density-based clustering, and the markers that do not meet the criteria are limited by limiting the distance between each grouped group and the number of points that can be included in each group.

Next, three-dimensional coordinates are calculated for the clustered markers by the density-based clustering method (240). For example, the calculator may calculate a 3D coordinate of the marker attached to the main position of the body surface by defining a representative value of the group coordinates of the grouped markers.

Finally, the body shape determination step 250 is performed. The determination unit may determine the body shape of the user using the three-dimensional coordinates of the clustered markers obtained according to a series of processes performed by the calculation unit. As described above with reference to FIG. 1, the body shape determination may determine a body shape closest to the body shape information previously designated as the body shape of the measurement user.

3 illustrates a marker detection flow according to one embodiment. 3 illustrates a marker detection flow based on a state in which a plurality of markers are attached to the back of a measurement user.

Reference numeral 310 denotes three-dimensional data input to the calculator, and 320 illustrates a marker region in the three-dimensional data in solid line. 330 illustrates the non-marker region of the 3D data in solid line. 340 illustrates the density-based clustering of the marker region, and 350 illustrates the merge of the density-based clustered marker region and the non-marker region.

The input 3D data 310 may be a body image of a user including a marker. The marker can be attached at a suitable position to measure the body shape and can be determined clinically.

Color space information and a clustering algorithm may be used to separate the marker region 320 and the non-marker region 330 from the 3D data. Various methods may be used for the color space information and the clustering algorithm. For example, HSV color space and K-means clustering may be used, but the present invention is not limited thereto.

Density-based clustering (DBSCAN) may be performed on the obtained marker region 320. It can be seen that the density-based clustering is performed at 340.

Finally, data obtained by combining the density-based clustered marker region 340 and the non-marker region 330 may be obtained. The calculator may calculate 3D coordinate information on the body with respect to the clustered marker region, and the determiner may determine a body shape using the 3D coordinate information.

4 is a flowchart illustrating a body analysis method including automatic marker division according to an embodiment. Commonly used markers include spherical markers and plane markers, but the body shape analysis method according to an embodiment may determine a body shape regardless of the type of the marker.

According to an embodiment, a body shape analysis method may include inputting a three-dimensional body shape data 410, calculating a three-dimensional coordinates marker 420, determining whether a marker is a spherical marker 430, removing a marker, and removing a region 440. Body shape determination step 450.

In the same manner as the body shape analysis method described with reference to FIG. 2, 3D body data may be input (410). Next, a step 420 of calculating three-dimensional coordinates with respect to the marker may be performed. Through the marker detection algorithm described above, the coordinates of the marker may be calculated.

In the determining of whether the marker is a spherical marker (430), it is determined whether each marker is a spherical marker by using the 3D body data inputted by the calculator. If the marker is a spherical marker, the spherical marker may be removed and interpolated with the removed spherical marker region (440). After the removal of the spherical marker region and the interpolation 440 or when the marker is not the spherical marker, the body shape determination step 450 is performed immediately.

Let's look at how to determine whether or not it is a spherical marker in detail. Adjacent surrounding areas based on the position of each marker area may be extracted as a 3D point cloud. By way of example but not limitation, a cube space of 5 cm on one side from the center of the marker region can be extracted into the point cloud.

The peripheral area may be composed of a marker point cloud area and a non-marker point cloud area, and assumes a non-marker point cloud area as a plane and performs fitting. Since the peripheral region is extracted based on the position of the marker region obtained by using the color space information of the marker, a point cloud clustered at the center of the peripheral region among the point clouds not fitted to the plane may be set as the marker point cloud region. . If the center of the peripheral area is not set as an outlier or the distance from the fitted plane is smaller than a specific threshold, it may be determined as a planar marker rather than a spherical marker.

5 illustrates a flow for separating markers and non-marker regions using clustering of individual markers, according to one embodiment.

5A shows an enlarged view of individual markers. In detail, it shows a state in which the adjacent peripheral region around the position of the individual markers extracted with a three-dimensional point cloud.

FIG. 5B illustrates a peripheral region divided into a marker point cloud region and a non-marker point cloud region, and is a plane fit to the non-marker point cloud region. The space between two parallel straight lines is a planar fitted non-marker region and corresponds to an inlier.

FIG. 5C shows a non-marker point cloud present at the fit point, and FIG. 5D shows a marker region that is an outlier.

Through the above method, the marker region may be further divided into a marker point cloud region and a non-marker point cloud region, and further, a spherical marker and a planar marker may be distinguished.

Color space information and clustering are divided into marker regions and non-marker regions, and density-based clustering is performed on the marker regions to cluster them. Next, the clustered marker regions may be divided into a marker point cloud region and a non-marker point cloud region. That is, by setting a peripheral region around the clustered marker regions and subdividing the peripheral region into a marker point cloud region and a non-marker point cloud region, the marker region and the non-marker region can be distinguished more clearly.

The calculation unit according to an embodiment may be applied even when the color and the total number of markers and the type of the markers are not known. Therefore, the color, number and type of markers can be freely selected. For example, a flat marker may be attached to the user's back and a spherical solid marker may be attached to the shoulder.

6 illustrates input body shape data and reference body shape data, according to an exemplary embodiment.

Alignment is necessary between the finally obtained input body shape and the reference body body that is the standard for body shape determination. FIG. 6A shows the input body shape and FIG. 6B shows the reference body shape.

Specifically, referring to FIG. 6A, the input body is looking at the right oblique direction, but the reference body of FIG. 6B is looking at the left oblique direction. Therefore, the rigid body transformation vector T should be used to align the input body shape with the same reference point and direction as the reference body shape.

The rigid transformation vector T may be a vector representing a rotation and translation transformation about the X, Y, and Z axes. In addition, the rigid transformation vector T may be rotated and translated about a X, Y, and Z axes.

For example, the reference model may be positioned at the origin of the X, Y, and Z axes at the position of the navel, and only the upper body is displayed for convenience, but may also be applied to the whole body.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable arrays (FPAs), It may be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For the convenience of understanding, a processing device may be described as one being used, but a person skilled in the art will appreciate that the processing device includes a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as parallel processors.

The software may include a computer program, code, instructions, or a combination of one or more of the above, and configure the processing device to operate as desired, or process independently or collectively. You can command the device. Software and / or data may be any type of machine, component, physical device, virtual equipment, computer storage medium or device in order to be interpreted by or to provide instructions or data to the processing device. Or may be permanently or temporarily embodied in a signal wave to be transmitted. The software may be distributed over networked computer systems so that they are stored or executed in a distributed manner. Software and data may be stored on one or more computer readable recording media.

Method according to the embodiment is implemented in the form of program instructions that can be executed by various computer means may be recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the media may be those specially designed and constructed for the purposes of the embodiments, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer readable recording media include magnetic media such as hard disks, floppy disks and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine code, such as produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described with reference to the accompanying drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques may be performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. may be combined or combined in a different form than the described method, or other components. Or even if replaced or substituted by equivalents, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are within the scope of the claims that follow.

Claims (15)

In a method performed at least temporarily by a computer,
Receiving 3D body data including a plurality of markers attached to a user's body;
Separating the three-dimensional body data into a marker region and a non-marker region;
Clustering the markers for the marker regions using a density based clustering (DBSCAN) method;
Comparing the marker clusters with a predetermined criterion based on at least one of a distance between marker clusters in which the markers are clustered, and a number of points included in the marker cluster;
As a result of the comparison, excluding the marker populations that do not meet the criteria from a marker candidate;
Calculating three-dimensional coordinates of the clustered markers; And
Determining the body shape of the user using three-dimensional coordinates of the clustered markers
Body analysis method comprising a.
The method of claim 1,
Separating into the marker and the non-marker region,
Body analysis method using color space information and clustering algorithm to separate.
The method of claim 2,
Interpolating a surface of the non-marker region;
Body analysis method further comprising
The method of claim 3,
Determining the body shape of the user,
Aligning the pose of the predefined body reference model using a rigid transformation vector
Body analysis method further comprising.
In a method performed at least temporarily by a computer,
Receiving 3D body data including a plurality of markers attached to a user's body;
Calculating three-dimensional coordinates of the marker;
Determining whether the marker is a spherical marker or a planar marker to remove the spherical marker if the marker is a spherical marker;
Interpolating the area from which the spherical marker is removed; And
Determining the body shape of the user using the calculated 3D coordinates of the marker
Including,
Removing the spherical marker,
Distinguishing the area around the marker into a marker point cloud area and a non-marker point cloud area;
Fitting the nonmarker point cloud area into a plane;
Determining the marker as a spherical marker in response to the distance between the marker point cloud area and the planar fitted portion being greater than a reference value; And
In response to the marker being a spherical marker, removing the spherical marker
Body analysis method comprising a.
The method of claim 5,
Calculating the number of spherical markers; And
Calculating the number of the planar markers
Body analysis method further comprising.
In a method performed at least temporarily by a computer,
Receiving 3D body data including a plurality of markers attached to a user's body;
Separating the three-dimensional body data into a marker region and a non-marker region;
Clustering the markers for the marker regions using a density based clustering (DBSCAN) method;
Comparing the marker clusters with a predetermined criterion based on at least one of a distance between marker clusters in which the markers are clustered, and a number of points included in the marker cluster;
As a result of the comparison, excluding the marker populations that do not meet the criteria from a marker candidate;
Calculating three-dimensional coordinates of the clustered markers;
Determining whether the marker is a spherical marker or a planar marker to remove the spherical marker if the marker is a spherical marker;
Interpolating the area from which the spherical marker is removed; And
Determining the body shape of the user using three-dimensional coordinates of the clustered markers
Including,
Removing the spherical marker,
Distinguishing the area around the marker into a marker point cloud area and a non-marker point cloud area;
Fitting the nonmarker point cloud area into a plane;
Determining the marker as a spherical marker in response to the distance between the marker point cloud area and the planar fitted portion being greater than a reference value; And
In response to the marker being a spherical marker, removing the spherical marker
Body analysis method comprising a.
The three-dimensional body data including a plurality of markers attached to the user's body is separated into a marker region and a non-marker region, and the markers are clustered using a density-based clustering (DBSCAN) method to cluster the markers. A calculator for calculating three-dimensional coordinates of the marker; And
Determination unit for determining the body shape of the user using the three-dimensional coordinates of the clustered marker
Including,
The calculation unit,
Comparing the marker clusters with a predetermined criterion based on at least one of a distance between marker clusters in which the markers are clustered, and the number of points included in the marker cluster, and comparing the marker clusters with a preset criterion. By excluding marker clusters from the marker candidates, the three-dimensional coordinates of the clustered markers are calculated.
Body analysis device.
The method of claim 8,
The calculation unit,
Body analysis apparatus for separating using color space information and clustering algorithm.
The method of claim 9,
The calculation unit,
A body analyzer for interpolating the surface of the non-marker region.
The method of claim 10,
The determination unit,
Figure analysis device that uses the rigid transformation vector to align the posture of the predetermined body reference model.
When the 3D body data including a plurality of markers attached to the user's body is input, the 3D coordinates of the marker are calculated, and whether the marker is a spherical marker or a flat marker is determined as a spherical marker. A calculator configured to remove the spherical marker and interpolate an area from which the spherical marker is removed; And
Determination unit for determining the body shape of the user using the three-dimensional coordinates of the marker to be calculated
Including,
The calculation unit,
The marker area is divided into a marker point cloud area and a non-marker point cloud area,
Fitting the non-marker point cloud area into a plane,
In response to the distance between the marker point cloud area and the plane-fitted portion being greater than a reference value, determining the marker as a spherical marker,
In response to the marker being a spherical marker, removing the spherical marker
Body analysis device.
The method of claim 12,
The calculation unit,
And counting the number of the spherical markers and calculating the number of the planar markers.
3D body data including a plurality of markers attached to the user's body is separated into a marker region and a non-marker region, and the markers are clustered by clustering the markers using a density-based clustering method. Based on at least one of a distance between clusters and the number of points included in the marker cluster, comparing the marker clusters with a preset criterion, and as a result of the comparison, the marker clusters that do not meet the criterion in the marker candidate; Calculate three-dimensional coordinates of the clustered markers, and determine whether the marker is a spherical marker or a flat marker to remove the spherical marker and interpolate an area from which the spherical marker is removed. part; And
Determination unit for determining the body shape of the user using the three-dimensional coordinates of the clustered marker
Body analysis device comprising a.
Claim 1 to 7
A computer-readable recording medium containing a program for performing a body analysis method.

Images