KR20150101836A - Full body high speed three dimesional scanning apparatus - Google Patents

Abstract

The present invention relates to a high speed three dimensional scanning apparatus for a body to scan the motion and face of a person by reducing scan time to 0.5~2 seconds. It includes two pairs of supporters which are concentrically symmetrical around a body, which is a scanning body, and are separated from each other by an angle of azimuth, at least one 3D scanner sensor which is installed in each of supporters and obtains the partial 3D information of the body as seen from each direction, and a computer which successively controls the scanning of the 3D scanner sensor to remove interference, and obtains the total 3D information of the body by using partial 3D information of the body obtained from each of the 3D scanner sensors.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a high-speed three-

The present invention relates to a human body stereoscopic scanning apparatus, and more particularly, to a human body high-speed stereoscopic scanning apparatus capable of precisely scanning a human's facial expression and motion by shortening a scan time to 1 to 2 seconds or less.

Stereoscopic scanning technology is a technology to digitize 3D shapes and surface colors of objects in the world. Recently, 3D printer technology has been attracting attention due to the development of 3D printer technology, and remarkable technological progress has been made.

In the field of computer graphics such as virtual reality, game, and animation, skilled designers use 3D CAD programs to manually create 3D models, which take a lot of time to produce and have a disadvantage in that the quality varies according to the skill of the designer . In recent years, even when creating real 3D models of people including heads, it is possible to make 3D models of people by modifying eyes, nose, mouth, ears, do. However, in such a case, there is a disadvantage in that it is necessary to repeatedly repeat the process of measuring the photographs of the human body and measuring the images one by one, and it is also difficult to perform detailed modeling. Stereoscopic scanning technology is different from the manual CAD modeling work of the designer, and various kinds of technologies are already utilized in many application fields.

The 3D scanning technique of the object can be classified into contact type and non-contact type. The contact type 3D scanning technique is a method in which the sensor directly acquires the three-dimensional coordinates while the measurement object parts are in contact with each other. However, there is a disadvantage that it can not be applied when the shape changes when the pressure is applied, such as the surface of the human body, clothes, or the like. Therefore, non-contact scanning technology is used when the shape changes when pressure is applied, such as the surface of a human body, clothes, and the like. Non-contact type is a method of making a 3D model by measuring the amount of energy reflected or transmitted from an object. To measure the energy reflected from an object in order to measure the shape of the object, an optical method using computer vision It is mainly used.

The optical 3D scanning technique can be divided into a passive mode and an active mode depending on the sensing method. The manual method is a technology applied to objects made of lines and surfaces such as buildings and automobiles. It uses intensity, parallax, etc., of images taken without artificially projecting energy to objects. To calculate the three-dimensional position. The manual method is somewhat less accurate than the active method, but has the advantage of being easy to use and acquiring the texture directly from the input image. However, it is difficult to distinguish the boundary line between the face and the face in the human face, especially the expression, so the active method is used in the 3D scanning field of the human body rather than the manual method.

The active method is a method of projecting a predefined pattern or sound wave onto an object, and restoring the 3D shape of the object by measuring the reflected energy of the projected pattern image or sound wave. Typical methods include structured light or laser light Is projected onto an object to measure the phase change according to the distance.

In the case of using laser light, a laser having a strong energy is projected on an object, and a 3D image of an object can be scanned by rotating the object 360 degrees while photographing an image of the projected object using the camera. When scanning a person, 3D scanned results should be able to express even detailed information such as facial expressions. In this case, it is the pupil that plays an important role about 90% in the facial expression. Using a strong laser beam makes it possible to scan an object very precisely in 3D, but since it is harmful to irradiate a person with a strong laser beam, the laser beam is not used for the human body.

Instead of one-dimensional linear light such as a laser beam, a two-dimensional structured light composed of infrared light is projected onto an object, and an image of the object projected by the structural light is photographed using a camera. An active 3D scanning method using structural light that scans 3D information of an object while rotating is also used to scan the human body. At this time, a method of calculating the 3D distance information by shooting the result image projected on the human body with an infrared camera is mainly applied.

1 is a schematic diagram of a 3D scanner sensor 1 (hereinafter simply referred to as "scanner sensor" or "sensor") in which an infrared projector, an infrared camera, and a color image camera are integrated. In the infrared projector 3 of the 3D scanner sensor 1 shown in FIG. 1, the infrared light used in the active system is projected onto a human body to be scanned. The result projected on the human body is infrared, so that it can not be seen by human eyes. It is not captured by a general color image camera 5 but is captured by an infrared camera 7 and used for calculating 3D distance information. In the color image camera 5, a texture such as the color of the human body is stored. A detailed 3D scan model of the human body can be obtained by combining the 3D distance information calculated by the infrared projector and the infrared camera and the texture information (hereinafter, referred to as "three-dimensional information of the human body" In order to obtain the stereoscopic information of the entire body of a human body with such a 3D scanner sensor 1, as shown in FIG. 2, while another person holds a single scanner sensor around the human body 9 to be scanned, It is necessary to carefully and slowly scan all the directions and altitudes thoroughly so that the three-dimensional information of the human body necessary for 3D scanning modeling of the entire human body can be obtained.

However, the stereoscopic scanning method of the human body takes too much time to obtain a 3D scanning model for the following reasons, and accurate 3D scanning information can not be obtained, and sharp and accurate 3D scanning information can not be obtained.

In order to obtain full stereoscopic information of the human body, the scan operator not only scans the scanner sensor while moving the 360 degrees of the scan target body and the altitude from the foot to the head, Moving, as with taking a picture with a camera, can lead to inaccurate results as a result of a 3D scan, so be very careful. Therefore, in the case of the conventional stereoscopic scanning method, there is a disadvantage that the scan time is excessively long in order to obtain the entire stereoscopic information of the human body.

When the scan time for the entire human body takes a long time, it is difficult for the human body to maintain the same posture or expression during the scan period, and it is difficult to obtain accurate stereoscopic information because the expression is changed or moved.

The 3D scanner sensor, which integrates an infrared projector, infrared camera and color video camera, uses an optical method that uses a 3D scan, such as when the subject is shaken when taking pictures in everyday life, or when a camera is shaken, The fine movement of the target person and the fine tilting of the hands of the 3D scan performer degrade the sharpness and accuracy of the 3D scan result. Therefore, in the conventional stereoscopic scanning method, it is also difficult to obtain clear and accurate stereoscopic information of the human body.

A first object of the present invention is to provide a stereoscopic image capturing apparatus and a stereoscopic image capturing method capable of capturing a stereoscopic image of a human body in a short time (0.5 seconds Speed stereoscopic scanning apparatus capable of acquiring accurate stereoscopic information about the entire human body within a short time (for example, 2 seconds to 2 seconds).

The second problem to be solved by the present invention is to provide a human body high-speed stereoscopic scanning apparatus which can obtain stereoscopic information with high clarity and accuracy for the entire human body.

The above-described problems to be solved by the present invention can be attained by a medical treatment apparatus comprising two or more pairs of supporting rods paired in a symmetrical direction concentric with a human body to be scanned and spaced apart from each other by a predetermined azimuth angle and at least one pair of supporting rods A 3D scanner sensor for acquiring partial stereoscopic information, a control unit for sequentially controlling the scanning of the 3D scanner sensor so as to avoid interference with each other, and acquiring all the stereoscopic information of the human body using the partial stereoscopic information of the human body acquired by each of the 3D scanner sensors A high-speed stereoscopic scanning apparatus including a computer for carrying out the present invention.

Preferably, the 3D scanner sensor is a 3D scanner sensor integrated with an infrared projector, an infrared camera and a color image camera, and the infrared scanner, the infrared camera, and the color image camera are installed on the support table Do.

At least two or more 3D scanner sensors are installed at regular intervals with different altitudes, and a 3D scanner sensor installed at each support is installed at the same height as a 3D scanner sensor installed on a support in a symmetrical direction, It is preferable to control the 3D scanner sensors so as to acquire the partial stereoscopic information of the human body simultaneously from the 3D scanner sensors installed at different heights among the 3D scanner sensors installed on the support of the point.

According to the present invention having the above-described configuration, a plurality of scanner sensors are fixed by a double-sided projection method at an azimuth and at an elevation so as to obtain human stereoscopic three-dimensional information, The stereoscopic information of the entire human body can be acquired in a short period of time in which the scan target human body does not change its expression or posture by sequentially ordering the infrared rays of the scanner sensor so as not to interfere with each other. In addition, according to the present invention, not only a scanner sensor and a human body can be scanned at a constant scan rate while maintaining a constant interval in a stopped state, but also a projector projecting structured light from a unit scanner sensor and a projected image The operation of the scanner sensors located in the normal direction and the opposite direction is naturally blocked by the human body to be scanned in 3D so that no interference occurs between the scanning of each scanner sensor, Can be obtained.

1 is a schematic diagram of a 3D scanner sensor to which the present invention is applied.
FIG. 2 is a state diagram illustrating a scanning method using the 3D scanner sensor shown in FIG. 1. FIG.
3 is a layout diagram of a support and a 3D scanner sensor of a human body high-speed stereoscopic scanning apparatus according to the present invention.
4 is a configuration diagram of a human body high-speed stereoscopic scanning apparatus according to the present invention.
5 to 12 are scanning procedures using a human body high-speed stereoscopic scanning apparatus according to the present invention.

Hereinafter, a detailed embodiment of a human body high-speed stereoscopic scanning apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

5 to 12, in the present invention, a plurality of scanner sensors h1 to h4 are arranged in advance in a 360-degree orientation of a human body to be scanned in 3D, The time consumed by mechanical movements such as a movement by an expert or a rotating device is eliminated and the structure light of each of the scanner sensors h1 to h4 is projected so as not to interfere with the structure light of the other scanner sensors h1 to h4 Speed stereoscopic scanning apparatus that acquires 3D scan data at a high speed by controlling only an operation for photographing while taking a photograph.

To this end, as shown in Figs. 3 and 4, two or more pairs of supporting rods 1a to 1h are provided, which are paired in a concentric direction on the center of the human body 9 to be scanned and spaced apart by a predetermined azimuth angle 3D scanner sensors h1 to h4 for acquiring partial stereoscopic information of the human body 9 viewed from each direction and installed on at least one of the support rods 1a to 1h and 3D scanner sensors h1 to h4, Dimensional information of the human body 9 acquired by the respective 3D scanner sensors h1 to h4 so as to prevent the interference of the structured light (structured light formed by the infrared rays) 9 to acquire the entire stereoscopic information.

FIGS. 3 to 12 show a configuration in which 360 degrees of orientation are arranged in advance at eight fixed angles by applying eight supports 1a to 1h. In practical use, if the accuracy is to be reduced, the support (1a to 1h) may be used at 0 degrees, 180 degrees, 270 degrees, or 0 degrees and 60 degrees, 120 degrees, 180 degrees, 240 degrees, It is also possible to use six bearings. Also, the number of scanner sensors h1 to h4 used for the respective supports 1a to 1h at different altitudes is not limited to four as shown in the drawings of the embodiment, but a scanner sensor for a face and a scanner sensor for an upper half It is also possible to use only three of the scanner sensors for the lower half.

1 and 3, the 3D scanner sensors h1 to h4 include 3D scanner sensors h1 to h4 in which an infrared projector 3, an infrared camera 7 and a color video camera 5 are integrated, The infrared scanner 3, the infrared camera 7, and the color image camera 5 are installed in the vertical direction on the supports 1a to 1h, .

As shown in FIG. 3, two or more 3D scanner sensors h1 to h4 are installed at predetermined intervals at different altitudes on the respective support rods 1a to 1h and installed on the support rods 1a to 1h It is preferable that the 3D scanner sensors h1 to h4 are installed at the same height as the 3D scanner sensors h1 to h4 installed in the symmetrical supports 1a to 1h.

3, the supports 1a to 1h are arranged at intervals of 45 degrees along eight orientations of 360 degrees. Four scanner sensors h1 to h4 are provided on each of the supports 1a to 1h for detecting the structured light projected on the human body 9 to be subjected to the 3D scan and the projector 3 for projecting infrared light, The infrared camera 7 is fixed so as to be arranged in the vertical direction or the vertical direction. At this time, the angle at which the supports 1a to 1h are disposed is determined by a computer vision algorithm for 3D scanning, softly connecting the scanned three-dimensional information of the human body scanned by the individual azimuth and altitude using a software algorithm such as contour and mesh detection, The entire stereoscopic information is generated. Therefore, it is not absolutely necessary to arrange them with a high degree of accuracy at intervals of 45 degrees, and when the units are arranged in a unit of about 45 degrees, the operation of the present invention is accurately performed. Similarly, as shown in FIG. 3, the process of combining all the stereoscopic information of the human body in the partial stereoscopic information of the human body in the unit of 45 degrees even when the first support (1a to 1h) 11) (hereinafter simply referred to as "vision processing computer" or "computer") using an algorithm such as contour and mesh detection, it is possible to perform position and angle correction using separate hardware, It is not necessary to perform the preliminary 3D scanning operation of the < RTI ID = 0.0 >

4, the communication lines and the power supply lines are displayed so as to be exposed to the outside for convenience of explanation. However, the individual scanner sensors h1 to h4 are connected to the vision processing computer through communication lines and power lines provided inside the supports 1a to 1h Dimensional information of the human body photographed by the entire scanner sensors h1 to h4 in real time is stored in the internal storage space of the computer 11 in real time and then the vision processing algorithm To complete the stereoscopic information of the human body. The operation speed of this vision processing algorithm depends on the computation ability of the computer and the performance of the graphic module. In the case of complicated 3D scan data, it takes more than 30 minutes. However, the time required for the vision processing of this portion does not affect the advantage that the human body can capture the vivid 3D scan data of the present invention, which is the time taken after the human face or the detailed posture of the 3D scan data is captured with a unique attitude.

In summary, in the present invention, in order to precisely and precisely scan the facial expression or posture of the human body 9, the scanner sensors h1 to h4 provided at the 360 degrees fixed angle are operated within 0.5 to 2 seconds After storing and storing the partial 3D scan data (partial stereoscopic information of the human body) of the scanner sensors h1 to h4 on the computer, the computer vision processing algorithm is repeatedly performed several times over about 30 minutes, And calculate accurate 3D scan data of the human body (total stereoscopic information of the human body). In the future, if the computer technology develops more or the dedicated vision processing hardware is used, this part can be done in 1 to 2 seconds in real time.

A high-speed human body scanning method using a human body high-speed stereoscopic scanning apparatus according to the present invention will now be described with reference to FIGS. 5 to 12. FIG.

5 to 12, the computer 11 includes 3D scanner sensors h1 to h4 installed at different heights among the 3D scanner sensors h1 to h4 installed on the supports 1a to 1h symmetrical to each other, h4 to obtain the stereoscopic information of the human body 9 at the same time from the 3D scanner sensors h1 to h4. When the structured light of the scanner sensors h1 to h4 is projected, if the structured light of the adjacent scanner sensors h1 to h4 is simultaneously projected, the area of the overlapping projected images in the adjacent scanner sensors h1 to h4 In order to prevent high speed 3D scanning while preventing it from being interfered with due to interference of structured light, a pair of support rods 1a and 1b located at opposite directions such as 0 degrees and 180 degrees are controlled by being paired, As shown in FIG. 5, the scanner sensors h1 and h3, which are skipped so that the structured light does not interfere with the top end of the support 1a located at the 0-degree position, and the 0 6, at the same time, the structured light is projected and photographed simultaneously with the scanner sensors h2 and h4 located at positions shifted between the scanner sensors h1 and h3 used at the point of time, 0 degrees and 180 degrees The remaining scanner sensors h1 to h4 of the supporting members 1a and 1b on the supporting points 1a and 1b also project the structured light and photograph them so that the supporting members 1c to 1h, To 1h), so that the 3D scan of the human body can be performed at high speed without interference by the structured light in a short time, so that the sharpness and precision of the 3D scan data due to the movement of the human body are not affected.

First, FIG. 5 shows the first operation step, in which the scanner sensor h1 located at the top of the support la installed at 0 degrees and the scanner sensor h3 installed next to the scanner sensor h3, Is projected to the human body 9 and is photographed by the camera. The reason why the first scanner sensor h1 and the one skipped scanner sensor h3 are used is that when the first and second scanner sensors h1 and h2 simultaneously project the structured light for 3D scanning, h1, and h2) overlap each other and overlap each other. When the camera is photographed by each camera, even when a good algorithm is applied to configure the entire 3D scan data in the computer after the partial 3D scan, an individual scanner sensor h1, h2) can not remove the error from the combined partial 3D data. Therefore, in order to completely eliminate the interference effect generated when the structured light is projected from the adjacent scanner sensors h1 to h4 at the time of acquiring the partial 3D data from the individual scanner sensors h1 to h4, in the present invention, And the third scanner sensor (h1, h3). Then, the structured light projecting unit and the camera are vertically arranged in the individual scanner sensors h1 to h4, and the second scanner sensor h2 and the fourth scanner sensor h1 h4 are simultaneously operated, the human body 9, which is the target of the 3D scan, is naturally blocked, so that the time required for the scan can be made very efficient without causing an error when calculating the entire 3D scan data in the computer vision processing software Can be reduced.

5 and 6, if the individual scanner sensors h1 to h4 of the supporting members 1a to 1h provided at 0 and 180 degrees are sequentially operated one at a time, if 0.25 seconds are required per one time, 1a to 1h), it is necessary to project and irradiate the total of eight times of structured light four times from the 180 degree support rods 1a to 1h. Therefore, the necessary time is required to be 8 seconds to 0.25 seconds in 0.25 seconds, 5, partial 3D scan data of 0.25 seconds in FIG. 5 and 0.25 seconds in 0.5 seconds and 0 degrees and 180 degrees in 0.5 seconds can be stored in the computer.

FIGS. 5 to 12 sequentially show operations from 0 degrees / 180 degrees to 45 degrees / 225 degrees, 90 degrees / 270 degrees, and 135 degrees / 225 degrees. As shown in Figs. 6 and 14, in order to store the partial 3D scan data from the supporters 1a to 1h provided at the fixed angles of 8 orientations into the computer, four pairs of 0.5 seconds each takes 2 seconds. 6 and 7 can be performed within 1 second if the operation of the portion for projecting the structured light for 3D scanning and the shooting operation is more precisely controlled. For reference, the 3D scanner sensor (h1 ~ h4) technology using the structured light can scan about 30 times per second. Therefore, the apparatus of the present invention can finish the operation including storing the 3D scan data in the computer within 0.5 second as long as the operation control of the scanner sensors h1 to h4 provided at the entire fixed angle can be precisely controlled.

1a ~ 1h: Support 3: Infrared projector
5: Color video camera 7: Infrared camera
9: human body 11: computer
h, h1 to h4: 3D scanner sensor

Claims (3)

Two or more pairs of supports installed in pairs in a symmetrical direction concentric with the human body to be scanned and spaced apart by a predetermined azimuth angle;
A 3D scanner sensor installed at least on each of the support rods to acquire partial stereoscopic information of a human body viewed from each direction;
And a computer for sequentially controlling the scanning of the 3D scanner sensor so as to avoid interference with each other and acquiring all the stereoscopic information of the human body using the human stereoscopic information acquired by each of the 3D scanner sensors, .
The method according to claim 1,
Wherein the 3D scanner sensor is a 3D scanner sensor integrated with an infrared projector, an infrared camera, and a color image camera, and the 3D scanner sensor is installed in the support base such that the infrared projector, the infrared camera, and the color video camera are arranged in the vertical direction A human body high-speed stereoscopic scanning device.
3. The method according to claim 1 or 2,
At least two or more 3D scanner sensors are installed at regular intervals with different altitudes, and a 3D scanner sensor installed at each support is installed at the same height as a 3D scanner sensor installed on a support in a symmetrical direction, Wherein the controller controls the 3D scanner sensors so as to acquire partial stereoscopic information of a human body simultaneously from 3D scanner sensors installed at different heights among 3D scanner sensors installed on a support of a point.

Images