KR20130081037A - Apparatus for measuring body size and method thereof - Google Patents

Apparatus for measuring body size and method thereof Download PDF

Info

Publication number
KR20130081037A
KR20130081037A KR1020120001990A KR20120001990A KR20130081037A KR 20130081037 A KR20130081037 A KR 20130081037A KR 1020120001990 A KR1020120001990 A KR 1020120001990A KR 20120001990 A KR20120001990 A KR 20120001990A KR 20130081037 A KR20130081037 A KR 20130081037A
Authority
KR
South Korea
Prior art keywords
human body
measurement data
information
user
body part
Prior art date
Application number
KR1020120001990A
Other languages
Korean (ko)
Inventor
김호원
이승욱
이지형
구본기
Original Assignee
한국전자통신연구원
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 한국전자통신연구원 filed Critical 한국전자통신연구원
Priority to KR1020120001990A priority Critical patent/KR20130081037A/en
Publication of KR20130081037A publication Critical patent/KR20130081037A/en

Links

Images

Classifications

    • AHUMAN NECESSITIES
    • A41WEARING APPAREL
    • A41HAPPLIANCES OR METHODS FOR MAKING CLOTHES, e.g. FOR DRESS-MAKING OR FOR TAILORING, NOT OTHERWISE PROVIDED FOR
    • A41H1/00Measuring aids or methods
    • A41H1/02Devices for taking measurements on the human body
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/103Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/107Measuring physical dimensions, e.g. size of the entire body or parts thereof

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Engineering & Computer Science (AREA)
  • Pathology (AREA)
  • Medical Informatics (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Physics & Mathematics (AREA)
  • Textile Engineering (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Dentistry (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

PURPOSE: A human body size measuring system and a method thereof are provided to measure the size of a human body based on multi-view image data recorded through multiple cameras installed around the human body and provide the measured human body size through a terminal like a smartphone, etc. CONSTITUTION: A human body size measuring system includes the following parts: an external shape restoration part (101) restoring a voxel-based external shape of a human body based on a multi-view image of the human body recorded through multiple cameras (10) and generating three-dimensional human body shape information, and detecting skeleton information to define each region of the human body; a multi-view mask image generation part (102) generating a multi-view mask image, where human body region information is recorded, for the measurement of human body regions at a terminal (300); a transmission part transmitting measurement data containing the restored voxel-based external shape, skeleton information, and multi-view mask image; and the terminal receiving the transmitted measurement data and measuring and displaying the size of a human body region, which is selected by a user, based on the received measurement data. [Reference numerals] (10) Human body recorded through multiple cameras; (101) External shape restoration part; (102) Multi-view mask image generation part; (103) Transmission unit; (20) Multi camera control unit; (301) Measurement data input unit; (302) GUI processing unit; (303) User definition region measurement unit; (304) Output unit; (AA,BB) Measurement data; (CC) Changed measuremnet data

Description

Human body size measurement system and its method {APPARATUS FOR MEASURING BODY SIZE AND METHOD THEREOF}

The present disclosure relates to an anthropometric measurement system and method thereof.

In general, the dimensional measurement (body measurement) method of the human body measures the length or circumference of a body part using a tape measure mainly used in tailor-made clothes. Another method for measuring the size of a human body (body measurement) can automatically detect the required length and circumference of the body part from three-dimensional shape information obtained by scanning the human body through an active scanner using a laser or patterned light.

The method of measuring the body shape depending on the a priori experience of the person has a disadvantage that depends on the experience of the practitioner and the situation at the time of the measurement. The other method of measuring the size of the human body (body measurement) has the disadvantage of requiring expensive equipment, an error caused by the movement of the human body during shooting, and a considerable time in the method of processing the measured data. The human body size and clothing number measuring system and method are also disclosed in Korean Patent Application No. 10-2006-0055267.

The present specification measures a human body size based on multiview image information photographed through multiple cameras installed around a human body, and a human body size measurement system and method for providing the measured human body size through a terminal. The purpose is to provide.

Human body measurement system according to an embodiment of the present disclosure, based on a multi-view image of the human body taken through the multi-camera, generates three-dimensional human body shape information by restoring the voxel-based appearance of the human body, the human body Appearance restoration unit for detecting the skeleton information for defining each part of the; A multi-view mask image generation unit generating a multi-view mask image in which human body information is recorded for measuring a human body part in the user terminal based on the 3D human body shape information; A transmitter for transmitting measured data including the reconstructed voxel appearance, skeleton information, and a multiview mask image; And a terminal configured to receive the transmitted measurement data and measure and display dimensions of a human body part selected by a user based on the received measurement data.

As an example related to the present specification, the apparatus may further include a database storing the measurement data transmitted from the transmitter.

As an example related to the present specification, the appearance restoring unit may compare the cross-sectional color distribution of the human body photographed by the first camera among the multi-cameras with the cross-sectional color distribution of the human body photographed by the second camera. By detecting a boundary and removing voxels that are out of the improved corresponding point for each search line of the cross-section of the human body, the voxel restoration accuracy may be improved.

As an example related to the present specification, the multi-view mask image generator may generate a multi-view mask image in which the human body part is indexed by adding an index to a region corresponding to the human body part based on the body part information. .

As an example related to the present specification, an area corresponding to the human body part may be an area within a binned foreground area of the multi-view image.

As an example related to the present specification, the transmission unit may transmit geometric correction information of the human body front view image and the multi-view image from which the user can select each human body part to the terminal.

As an example related to the present specification, the terminal may include: a measurement data input unit configured to receive the transmitted measurement data and parse the received measurement data to perform a user-defined measurement service; A user GUI processor configured to display a human body front view image in the parsed measurement data on a screen, and detect a human body part selected by the user from the displayed human body front view image; A user defined part measuring unit measuring a dimension of the human body part selected by the user; It may include an output unit for displaying the measured human body dimensions on the screen or stored in a database.

As an example related to the present specification, the user defined part measuring unit may measure the circumference of the human body part based on the length information of the convex hull corresponding to each cross section of the human body part.

The human body dimensional measurement system according to an embodiment of the present disclosure includes a voxel appearance, skeleton information, a multiview mask image, a human body front view image, and geometric correction information of the multiview image reconstructed based on a multiview image of the human body. A measurement data generator for generating and transmitting measurement data; And a terminal configured to receive the transmitted measurement data and measure and display dimensions of a human body part selected by a user based on the received measurement data.

Human body measurement system according to an embodiment of the present disclosure, based on a multi-view image of the human body taken through the multi-camera, generates three-dimensional human body shape information by restoring the voxel-based appearance of the human body, the human body Detecting skeleton information for defining each part of the apparatus; Generating a multi-view mask image in which human body information is recorded for measuring a human body part in a user terminal based on the 3D human body shape information; Transmitting measurement data including the reconstructed voxel appearance and skeleton information and a multiview mask image; Receiving the transmitted measurement data through the terminal, and measuring and displaying the dimensions of the human body selected by the user based on the received measurement data.

Human body size measurement system and method according to an embodiment of the present disclosure, the dimensions of the human body based on the multi-view image information taken through the multiple cameras installed around the human body (for example, the circumference of various parts of the human body Information), and the measured human body dimensions can be provided through a terminal such as a smartphone.

1 is a block diagram showing the configuration of a human body size measurement system according to an embodiment of the present invention.
2 is an exemplary view showing a multi-camera applied to an apparatus for measuring anthropometry according to an embodiment of the present invention.
3 and 4 is an exemplary view showing a human body part measured according to an embodiment of the present invention.
5 is an exemplary diagram illustrating a visual hull reconstruction method through silhouette intersection of a multiview image.
6 is an exemplary diagram illustrating an error between a body part boundary and a body part boundary in an actual image according to a visual hull restoration result.
FIG. 7 is an exemplary diagram illustrating a search line for improving restoration accuracy when a restoration error occurs due to self occlusion.
8 is an exemplary diagram of boundary extraction of the region 1 using two camera images sharing a specific region of the region of the region.
9 is a diagram illustrating each of the search selection colors of FIG. 8 in a color space.
10 is an exemplary view showing measurement data according to an embodiment of the present invention.
11 is an exemplary view illustrating an image of a human body front view displayed on a screen of a terminal (for example, a smartphone).
12 is an exemplary view illustrating a process of measuring a site selected by a user.

It is to be noted that the technical terms used herein are merely used to describe particular embodiments, and are not intended to limit the present invention. It is also to be understood that the technical terms used herein are to be interpreted in a sense generally understood by a person skilled in the art to which the present invention belongs, Should not be construed to mean, or be interpreted in an excessively reduced sense. In addition, when the technical terms used herein are incorrect technical terms that do not accurately express the spirit of the present invention, they should be replaced with technical terms that can be understood correctly by those skilled in the art. In addition, the general terms used in the present invention should be interpreted according to a predefined or prior context, and should not be construed as being excessively reduced.

Also, the singular forms "as used herein include plural referents unless the context clearly dictates otherwise. In the present application, the term "comprising" or "comprising" or the like should not be construed as necessarily including the various elements or steps described in the specification, Or may be further comprised of additional components or steps.

Furthermore, terms including ordinals such as first, second, etc. used in this specification can be used to describe various elements, but the elements should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, wherein like reference numerals refer to like or similar elements throughout the several views, and redundant description thereof will be omitted.

In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. It is to be noted that the accompanying drawings are only for the purpose of facilitating understanding of the present invention, and should not be construed as limiting the scope of the present invention with reference to the accompanying drawings.

The human body dimensional measurement system and method according to an embodiment of the present invention obtains 3D human body shape information through a camera capable of instant shooting instead of an active scan method, and the human body (or body) selected by a user through a terminal. It can be measured. For example, the human body measurement system and method according to an embodiment of the present invention, based on the multi-view image information obtained by photographing the human body through a multi-camera installed to surround the human body, height, leg length, arm length, Measure the length and circumference of the main part of the body (body) such as thigh circumference, waist circumference, then measure the body part corresponding to the user's request or change the body part measurement point according to the user's request, In order to re-measure the changed measuring points, the information required to measure the specific area of the human body can be minimized and stored and transmitted.

In the following description, a human body dimension (eg, length and circumference information of a main body of a human body) is measured from a multi-view image of a human body photographed through a multi-camera surrounding the human body, and the measured human body dimension is stored and stored at low capacity. And a measurement method for allowing a user to directly manipulate measurement information of a specific human body part selected through a low specification terminal, will be described with reference to FIGS. 1 to 12.

1 is a block diagram showing the configuration of a human body size measurement system according to an embodiment of the present invention.

As shown in FIG. 1, a human body dimensional measurement system according to an exemplary embodiment of the present invention includes a measurement data generator 100 generating measurement data for measuring human body dimensions; The terminal 300 receives the transmitted measurement data and measures and displays the dimensions of the human body part selected by the user based on the received measurement data.

The measurement data generating apparatus 100 generates (generates) three-dimensional human body shape information by restoring a voxel based appearance of the human body based on a multi-view image of the human body, and generates each part of the human body. An appearance restoration unit 101 for detecting skeleton information for definition; A multi-view mask image generation unit 102 for generating (generating) a multi-view mask image in which human body information for recording a human body part is recorded in the user terminal based on the 3D human body shape information; And a transmitter 103 for transmitting the measured data including the reconstructed voxel shape, skeleton information, and a multi-view mask image to an external device.

The transmitter 103 may transmit the measurement data to a database (DB) 200 and the user terminal 300 through a wireless / wired communication network.

Hereinafter, the configuration and operation of the human body measurement system according to an embodiment of the present invention will be described with reference to FIGS.

2 is an exemplary view showing a multi-camera applied to an apparatus for measuring anthropometry according to an embodiment of the present invention.

As shown in FIG. 2, the multi-camera 10 applied to the apparatus for measuring a human body size according to an exemplary embodiment of the present invention captures a human body by being composed of a plurality of cameras to surround the human body. The cameras may be arbitrarily distributed to maximize the dimensional information of the human body with respect to the human body, and center the global coordinate system located at the center of gravity of the multiple cameras for individual images obtained from the distributed cameras. Camera geometric calibration can also be performed. The multi-camera controller 20 obtains a multi-view image of the human body photographed by the multi-camera 10.

3 and 4 is an exemplary view showing a human body part measured according to an embodiment of the present invention.

As shown in FIG. 3, length information such as height, leg / arm length of the human body, and circumference information of main parts such as thighs and waist circumference are measured from the multi-view image captured by the multi-camera. In this case, the circumference information may be length information of a convex hull corresponding to each cross section of the human body, as shown in FIG. 4, for utilization of obesity management or tailored clothing.

Since the method for acquiring the multi-view image and the geometric correction information itself is a known technique, a detailed description thereof will be omitted. In addition, the contour restoration unit 203 separates the foreground region of the human body from the multi-view image using chroma key technology, restores the voxel appearance using the separated foreground silhouette information, and restores the human body from the restored voxel appearance. Three-dimensional skeleton information of the body can obtain information about the five parts of the body, arms, legs of the body.

On the other hand, since the voxel shape restored by the shape restoring unit 101 is restored to a visual hull level obtained through silhouette intersection of a multiview image as in the example of FIG. It is restored to the form of the convex hull (convex hull), which can be seen in the results of Figure 6 applied to the actual image.

5 is an exemplary diagram illustrating a visual hull reconstruction method through silhouette intersection of a multiview image.

6 is an exemplary diagram illustrating an error between a body part boundary and a body part boundary in an actual image according to a visual hull restoration result.

As shown in FIG. 6, the left leg is restored larger than the actual size by the right leg, and the depth boundary between the two legs obtained through z-buffering by projecting it on the corresponding image and the actual size. This can be confirmed by the difference between the color boundary generated between the two legs of the image. When restoring the appearance of the human body, the reduction of visual hull restoration precision due to self occlusion as shown in FIG. 6 is an inevitable part of the method using only the silhouette intersection of the multiview foreground region.

In the present invention, in order to solve the reduction in the precision of the cross-sectional measurement caused by the self occlusion, as shown in FIG. 7, two actual objects generated inside the foreground area of the multiview image by self occlusion (for example, For example, the color boundary information generated at the boundary of the left leg and the right leg) area is used.

FIG. 7 is an exemplary diagram illustrating a search line for improving restoration accuracy when a restoration error occurs due to self occlusion.

As illustrated in FIG. 7, an example in which boundary information of the actual surface of the region 1 is utilized to improve the reduction of the cross-sectional restoration accuracy of the region 1 caused by the region 2 will be described.

First, a first camera cam1 having a projection cross section for the most pixels and a second camera cam2 having a smallest projection cross section for the corresponding search line Ray are determined. The first and second cameras select cameras having a minimum dot product of the two camera direction information using the camera direction information of the camera looking at the object included in the search line and the geometric correction information of the individual cameras. Can be determined.

As shown in FIG. 8, the second camera cam2 shares surface color information corresponding to the cross section of the actual part 1 with the first camera cam1.

8 is an exemplary diagram of boundary extraction of the region 1 using two camera images sharing a specific region of the region of the region.

As shown in FIG. 8, the first camera cam1 when the pixels on which the cross section of the first camera cam1 is projected has surface color information corresponding to the cross section of the actual portion 1 of the second camera cam2. It can be determined as a result of the projection of the cross section of the portion 1 of the pixels. On the other hand, the two search lines shown in the example of FIG. 8 outside of the region of region 1 are pixels in which the surface color of region 2 is projected, and thus the colors different from the surface colors supported by the second camera cam2 are shown. To be excluded from the search. The projection of these two parts is actually shown as a color boundary in the first camera cam1, and the color boundary pixels are found in the projection pixels of the corresponding cross-section in the foreground area of the first camera cam1, and in the pixels before and after the boundary pixels. By determining the similarity with the surface color information of the corresponding section photographed by the second camera cam2, the problem of deterioration in restoration accuracy caused by self occlusion of the region 1 and region 2 can be solved. In this case, since the second camera sees only the color value of the first part, the color of the second part projected on the first camera does not have similarity, so that the silhouette information of the correct part 1 of the first camera can be extracted.

The determination can be described by diagramming in the color space as shown in FIG.

9 is a diagram illustrating each of the search selection colors of FIG. 8 in a color space.

As shown in FIG. 9, the color of the cross section of the human body of the second camera cam2 and the color distribution of the cross section of the human body of the first camera cam1 may be expressed as shown in FIG. 9, and the human body of the first camera cam1. The cross-sectional pixels are inspected, and the color boundary of the pixel of the first camera cam1 can be easily found by comparing with the cross-sectional color of the human body of the second camera cam2. By applying this process to the cross sections sampled along the skeleton of each human body, the voxels can be improved by removing the voxels that are out of the improved correspondence points by the search lines of the cross sections for the individual cross sections, thereby improving the voxel restoration accuracy. On the basis of this, the index of the body parts of the skeletons and the restored voxels by body parts can be updated.

The multi-view mask image generating unit 102 corresponds to a human body part (for example, a body, an arm, and a leg) in a binary foreground area of an individual multi-view image using the body part information obtained from the appearance restoring unit 101. Add an index to the area to The index addition process projects five areas to a multiview image using the geometric correction information of the camera, and uses the depth information of the projected voxels to index the area of the voxel nearest the camera for each pixel of the foreground area of the image. It can be processed quickly with a z-buffering method that records. Using the added part index information of the binary foreground area and the extracted skeleton information for each part, the user terminal 300 can measure the circumference information of any part of the body desired by the user in a low computing power environment. The circumference measurement method will be described an implementation example in the human body measuring unit 303.

The measurement data generating apparatus 100 may generate a multiview mask image in which appearance information expressed as a voxel, three-dimensional skeleton information for each five parts of the body, and body part information are indexed through the above process. The transmission unit 103 efficiently stores the measured data obtained as in the example of FIG. 10 in the DB 200 and transmits the measured data to a user terminal 300 such as a PC or a smartphone through a wired / wireless environment. Transmitting the measured data generated by the user terminal 300 to proceed with the user-centered measurement service.

10 is an exemplary view showing measurement data according to an embodiment of the present invention.

As shown in FIG. 10, the measurement data according to the embodiment of the present invention includes the restored voxel appearance, skeleton information for each body part, and a multi-view mask image, and includes a graphical user interface (GUI). It may further include a front view image for. For example, the transmission unit 103 may measure the measurement data including the restored voxel appearance, skeleton information for each body part, a multi-view mask image, and a front view image for a graphical user interface (GUI). It transmits to the DB 200 and the user terminal 300.

The voxel appearance information is generally expressed as NxNxN voxels in a three-dimensional space, and as the number of N increases, a detailed shape can be expressed, but there is a disadvantage in that the amount of data increases rapidly. In addition, since the measurement is performed based on the image, the higher the resolution of the image is, the more accurate measurement is possible, but also the increase in resolution in the use of a multiview image has the disadvantage of a sudden increase in the amount of data.

Therefore, in the present invention, using the voxel appearance information using the low resolution voxel precision and the high resolution multiview image to prevent the explosion of the measurement data capacity while maintaining the measurement accuracy at the measurement using the high resolution image and the detailed voxel precision Provide a method for To this end, in the present invention, the voxel appearance restoration information, the three-dimensional skeleton information for each part, the multi-view mask image indexed to the human body part, the front picture for inputting the user part selection through the user GUI, the geometric correction information of the multi-view image 5 Branch information may be stored and transmitted.

The extraction of the skeleton of the five body parts does not require a high-precision voxel appearance. Therefore, simple appearance information of the human body is restored using the voxel resolution of 128x128x128 level, and the 3D skeleton information is extracted (detected) by the appearance restoration unit 203 from the restored appearance. In this case, the presence or absence of a voxel in the voxel space is defined by (binary binarization), and thus the voxel appearance information can be recorded with data of approximately 260 KBytes. It can be defined as a continuous line consisting of three-dimensional points inside and outside the dog, so that skeleton information can be recorded with data of approximately 300 bytes.

In the case of a multi-view image, it is important information to determine the measurement accuracy of the present invention. Therefore, it is necessary to store index information of the foreground area with the highest resolution. The mask image in which the index of each region is recorded for each multiview image may be compressed and recorded. Assuming a high resolution image of 12M pixels or more, such as a digital single-lens reflex camera (DSLR) image, assuming that an average of about 2 line segments occurs per horizontal line for a 4000x3000 image, approximately 60 KByte for each multiview image Index information of five body parts can be stored as data. For example, when 15 multiview images are used, the entire multiview mask image information can be recorded with 1 MByte data.

In addition, a low resolution RGB image corresponding to the front view is generated for the user to intuitively select a region, and the generated region information of the body is stored through an alpha channel or a separate data channel in each generated pixel. When considering 0.4M pixel, which is a common smartphone resolution, an additional 1.5 MByte is required for BMP video.In case of saving the video through JPEG data compression, 150KByte data capacity is required. Is added.

Finally, geometric correction information of a multiview image should be recorded, and about 100 bytes of data are required for each image. In the case of 15 multiview images, a data capacity of 1.5 KBytes is required. That is, based on the above example, about 1.5MByte of measurement data need to be stored and transmitted for measuring the arbitrary part desired by the user terminal for one human body. In an environment, it can be transmitted without load.

The measurement data generated in this way is stored in the DB (database) 200, and can be accessed from a terminal such as a PC or a smart phone through a wired / wireless network through an authentication procedure such as login at the user's request. The terminal 300 provides the user with dimension measurement information of a desired human body part using the measured data transmitted in this way.

The terminal 300 receives the measurement data transmitted from the DB 200 or the transmission unit 103 and measures the data input unit for parsing the received measurement data to perform a user-defined measurement service. 302; A user GUI processor (302) for displaying an image of the human body front view in the parsed measurement data on a screen for the user to see, and detecting a portion (touched) selected by the user from the displayed human body front view image; A user defined part measurer 303 for measuring circumference information corresponding to the human body part selected by the user; It is composed of an output unit 304 for displaying the measurement information of the parts defined by the user on the screen for viewing by the user or stored in the DB (200).

The measurement data input unit 301 restores the transmitted binary state information of the voxel to the voxel space of 128x128x128 with or without the voxel in order to provide the cross-sectional information of the part in the voxel space of the arbitrary cross section. -Length Coding "Multi-view mask image information is restored and 15 multi-view mask images which store the index and foreground information about five body parts are produced | generated. In the RGB image of the human body front view compressed with “JPEG”, the RGB color value and pixel region information of each pixel of the foreground pixel are restored and stored. The camera geometry correction information is parsed to calculate a camera projection matrix that defines a correspondence between three-dimensional points of the global coordinate system and pixels in each camera image. The correspondence of voxel reconstructed surface voxels with respect to the foreground pixel of the front view is calculated and stored through z-buffering using the calculated projection matrix. Through this process, when the user selects the foreground pixel of the 2D front view, the program can calculate which voxels in the 3D space are selected.

11 is an exemplary view illustrating an image of a human body front view displayed on a screen of a terminal (for example, a smartphone).

As shown in FIG. 11, the GUI processor 302 displays the restored RGB image 11-1 of the human body front view on the screen, and displays the RGB image 11-1 of the displayed human body front view to the user. The selected site 11-2 is detected, and the detected site information 11-2 is transmitted to the user-defined site measurement unit 303. FIG. For example, the GUI processor 302 detects a coordinate value of a pixel clicked by a mouse of a personal computer in the displayed RGB image 11-1 of the human body front view, or touches a touch screen on a smartphone. The coordinate value of the pixel is detected, and which pixel of the RGB image is selected based on the coordinate value, and the area index of the selected pixel is detected. When the location index is detected, the location information or the coordinate value of the pixel selected by the user is transmitted to the user-defined site measurement unit 303.

The user-defined site measurement unit 303 may automatically calculate three-dimensional position and cross-sectional direction information of a site corresponding to the site information transmitted from the GUI processing unit 302 as shown in FIG. 12.

12 is an exemplary view illustrating a process of measuring a site selected by a user.

As shown in FIG. 12, the user-defined area measuring unit 303 is a three-dimensional skeleton as a continuous line segment of a site where a three-dimensional position value of the surface voxel corresponding to the pixel value transmitted from the GUI processing unit 302 is selected. By calculating the information on which section of the corresponding to the user can automatically calculate the three-dimensional position and the direction information of the section of the selected section. Through this process, from the simple user interface where the user selects one pixel of the front view, the program calculates the center point and direction vector of the three-dimensional plane that cuts the section of the body part selected by the user. Once the user-defined section information is determined, 360-degree search lines are generated at regular intervals of rotation, such as 1-degree intervals, based on the center point of the corresponding section. Projects the point mask image and examines the index of the search line pixels for each multiview image for the index of the corresponding part to find the pixel value of the boundary area, and projects it in 3D space to find the position of the surface in the actual 3D space. Find.

The position of the surface with the smallest distance from the center point for the entire multiview image becomes the three-dimensional intersection of the real part with respect to the search line. In this way, the intersection point of the entire search line can be found for each angle, and the circumference information of the corresponding area can be measured by measuring the length of the line connecting it.

The output unit 304 displays the cross-sectional information measured for the body part defined by the user and the circumference of the corresponding part in a graphic screen, and measures the information for reuse of the accumulated body measurement information. By adding to the generated changed measurement data, and stores it in the DB (200).

Through the above example, the user can easily receive anthropometric information of any part desired by the user in the multi-view image in the user's environment.

As described above, the human body size measurement system and the method according to the embodiment of the present invention, based on the multi-view (Multiview) image information captured by the multiple cameras installed around the human body (for example, Information about the circumference of various parts of the human body may be measured, and the measured human body dimension may be provided through a terminal such as a smartphone. That is, the terminal may provide the user with the human body size (eg, circumference information of the portion desired by the user) of the portion desired by the user. For example, in a typical user environment such as a PC or a smartphone, the user can directly measure and utilize the length or circumference information of a desired body part. In other words, when purchasing clothes through home shopping or internet shopping, you can directly check the dimensions of the parts defined by the corresponding products and purchase clothes suitable for yourself. It can also be applied to general body management.

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

100: measurement data generator 200: database
300: terminal

Claims (16)

Based on a multi-view image of a human body photographed through a multi-camera, the 3D human body shape information is generated by restoring the voxel-based appearance of the human body, and the contour restoration for detecting skeleton information for defining each part of the human body. Wealth;
A multi-view mask image generation unit generating a multi-view mask image in which human body information is recorded for measuring a human body part in the user terminal based on the 3D human body shape information;
A transmitter for transmitting measured data including the reconstructed voxel appearance, skeleton information, and a multiview mask image;
And a terminal configured to receive the transmitted measurement data and to measure and display dimensions of a human body selected by a user based on the received measurement data.
The human body measurement system according to claim 1, further comprising a database storing the measurement data transmitted from the transmission unit. The method of claim 1, wherein the appearance restoration unit,
Among the multi-cameras, the cross-sectional color distribution of the human body photographed by the first camera and the cross-sectional color distribution of the human body photographed by the second camera are detected to detect color boundaries of each cross-section of the human body, and based on the detected color boundaries. The human body dimensional measurement system, characterized in that to improve the voxel restoration accuracy by removing the voxels out of the improved corresponding point for each search line of the cross section of the human body.
The image display apparatus of claim 1, wherein the multiview mask image generator comprises:
And a multi-view mask image in which the human body part is indexed by adding an index to an area corresponding to the human body part based on the human body part information.
The method of claim 4, wherein the region corresponding to the human body part,
And an area within a binned foreground area of the multi-view image.
The method of claim 1, wherein the transmission unit,
The human body dimension measurement system, characterized in that for transmitting the geometric correction information of the human body front view image and the multi-view image that the user can select each human body part with the measurement data.
According to claim 6, The terminal,
A measurement data input unit for receiving the transmitted measurement data and parsing the received measurement data to perform a user defined measurement service;
A user GUI processor configured to display a human body front view image in the parsed measurement data on a screen, and detect a human body part selected by the user from the displayed human body front view image;
A user defined part measuring unit measuring a dimension of the human body part selected by the user;
And an output unit configured to display the measured anatomical dimensions on the screen or to store them in a database.
The method of claim 7, wherein the user-defined site measuring unit,
And a circumference of the human body part based on length information of a convex hull corresponding to each cross section of the human body part.
A measurement data generator for generating and transmitting measurement data including a restored voxel appearance, skeleton information, a multiview mask image, a human body front view image, and geometric correction information of the multiview image based on a multiview image of the human body; ;
And a terminal configured to receive the transmitted measurement data and to measure and display dimensions of a human body selected by a user based on the received measurement data.
Generating 3D human body shape information by restoring a voxel-based appearance of the human body based on a multi-view image of the human body photographed through a multi-camera, and detecting skeleton information for defining each part of the human body; ;
Generating a multi-view mask image in which human body information is recorded for measuring a human body part in a user terminal based on the 3D human body shape information;
Transmitting measurement data including the reconstructed voxel appearance and skeleton information and a multiview mask image;
And receiving the transmitted measurement data through the terminal, and measuring and displaying the dimensions of the body part selected by the user based on the received measurement data.
The method of claim 10, wherein the restoring the voxel-based appearance of the human body comprises:
Among the multi-cameras, the cross-sectional color distribution of the human body photographed by the first camera and the cross-sectional color distribution of the human body photographed by the second camera are detected to detect color boundaries of each cross-section of the human body, and based on the detected color boundaries. The method of measuring the human body size of the human body according to the search line of the improved cross-sectional area by removing the improved voxel out of the corresponding point.
The method of claim 10, wherein the generating of the multiview mask image comprises:
And generating a multi-view mask image in which the human body part is indexed by adding an index to a region corresponding to the human body part based on the human body part information.
The method of claim 12, wherein the region corresponding to the human body part,
And an area within a binned foreground area of the multi-view image.
The method of claim 10, wherein the measurement data,
A human body dimension measurement method further comprising geometric correction information of the human body front view image and the multi-view image from which a user can select each human body part.
The method of claim 14, wherein measuring and displaying the dimensions of the human body part comprises:
Receiving the transmitted measurement data and parsing the received measurement data to perform a user defined measurement service;
Displaying a human body front view image in the parsed measurement data on a screen, and detecting a human body part selected by a user from the displayed human body front view image;
Measuring dimensions of the body part selected by the user;
And displaying the measured anatomical dimensions on the screen or storing them in a database.
The method of claim 15, wherein the measuring of the dimension of the human body part comprises:
Measuring the circumference of the human body part based on length information of a convex hull corresponding to each cross section of the human body part.
KR1020120001990A 2012-01-06 2012-01-06 Apparatus for measuring body size and method thereof KR20130081037A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020120001990A KR20130081037A (en) 2012-01-06 2012-01-06 Apparatus for measuring body size and method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020120001990A KR20130081037A (en) 2012-01-06 2012-01-06 Apparatus for measuring body size and method thereof

Publications (1)

Publication Number Publication Date
KR20130081037A true KR20130081037A (en) 2013-07-16

Family

ID=48992905

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020120001990A KR20130081037A (en) 2012-01-06 2012-01-06 Apparatus for measuring body size and method thereof

Country Status (1)

Country Link
KR (1) KR20130081037A (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101580506B1 (en) 2014-10-20 2015-12-28 최영진 Apparatus and method of measuring body size of customer on the internet
US10004429B2 (en) 2014-01-21 2018-06-26 My Size Israel 2014 Ltd. Method and system for measuring a path length using a handheld electronic device
KR101878095B1 (en) * 2017-02-07 2018-07-12 동의대학교 산학협력단 Method and device of Estimating the Human Height in Using Depth Image OF FRONT VIEW
JP6424309B1 (en) * 2018-08-15 2018-11-14 裕樹 有光 Program and apparatus for generating a three-dimensional model based on measurement values
KR20190002071A (en) 2017-06-29 2019-01-08 주식회사 케이피앤피인터내셔널 Flexure measuring device of body
JP2020080020A (en) * 2018-11-12 2020-05-28 裕樹 有光 Program and device for mutually generating size measurement value and depth image
KR102118723B1 (en) * 2019-10-17 2020-06-04 메디컬아이피 주식회사 Method for analyzing abdominal image and apparatus therefor
WO2020141657A1 (en) * 2019-01-03 2020-07-09 엘지전자 주식회사 Body measurement device, and control method therefor

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10004429B2 (en) 2014-01-21 2018-06-26 My Size Israel 2014 Ltd. Method and system for measuring a path length using a handheld electronic device
KR101580506B1 (en) 2014-10-20 2015-12-28 최영진 Apparatus and method of measuring body size of customer on the internet
KR20160046286A (en) 2014-10-20 2016-04-28 최영진 Apparatus and method for measuring size of part of body using smart phone on the internet
KR101878095B1 (en) * 2017-02-07 2018-07-12 동의대학교 산학협력단 Method and device of Estimating the Human Height in Using Depth Image OF FRONT VIEW
KR20190002071A (en) 2017-06-29 2019-01-08 주식회사 케이피앤피인터내셔널 Flexure measuring device of body
JP6424309B1 (en) * 2018-08-15 2018-11-14 裕樹 有光 Program and apparatus for generating a three-dimensional model based on measurement values
JP2020027526A (en) * 2018-08-15 2020-02-20 裕樹 有光 Program and device for generating three-dimensional model based on size measurement value
JP2020080020A (en) * 2018-11-12 2020-05-28 裕樹 有光 Program and device for mutually generating size measurement value and depth image
WO2020141657A1 (en) * 2019-01-03 2020-07-09 엘지전자 주식회사 Body measurement device, and control method therefor
US11375922B2 (en) 2019-01-03 2022-07-05 Lg Electronics Inc. Body measurement device and method for controlling the same
KR102118723B1 (en) * 2019-10-17 2020-06-04 메디컬아이피 주식회사 Method for analyzing abdominal image and apparatus therefor

Similar Documents

Publication Publication Date Title
KR20130081037A (en) Apparatus for measuring body size and method thereof
CN110889890B (en) Image processing method and device, processor, electronic equipment and storage medium
US10154246B2 (en) Systems and methods for 3D capturing of objects and motion sequences using multiple range and RGB cameras
CN107223269B (en) Three-dimensional scene positioning method and device
JP5093053B2 (en) Electronic camera
TWI486551B (en) Method and system for three-dimensional data acquisition
Maimone et al. Real-time volumetric 3D capture of room-sized scenes for telepresence
KR20190034092A (en) Image processing apparatus, image processing method, image processing system, and storage medium
JPWO2018047687A1 (en) Three-dimensional model generation device and three-dimensional model generation method
US20210219869A1 (en) Body measurement device and method for controlling the same
WO2013056188A1 (en) Generating free viewpoint video using stereo imaging
EP2870428A1 (en) 3-d scanning and positioning system
JP7197451B2 (en) Image processing device, method and program
KR101495299B1 (en) Device for acquiring 3d shape, and method for acquiring 3d shape
US10909709B2 (en) Body measurement device and method for controlling the same
CN112802208B (en) Three-dimensional visualization method and device in terminal building
US20210312647A1 (en) Detecting device, information processing device, detecting method, and information processing program
US11527026B2 (en) Body measurement device and method for controlling the same
JP2007295375A (en) Projection image correction device, and its program
JP5361758B2 (en) Image generation method, image generation apparatus, and program
CN111742352A (en) 3D object modeling method and related device and computer program product
JP7341736B2 (en) Information processing device, information processing method and program
CN114913287B (en) Three-dimensional human body model reconstruction method and system
CN113706692B (en) Three-dimensional image reconstruction method, three-dimensional image reconstruction device, electronic equipment and storage medium
Lai et al. Exploring manipulation behavior on video see-through head-mounted display with view interpolation

Legal Events

Date Code Title Description
WITN Withdrawal due to no request for examination