KR100982171B1 - The System for Capturing 2D Facial Image - Google Patents

Abstract

The present invention relates to a face image photographing apparatus, comprising: a camera for acquiring a front image of a target person in real time, a memory in which the image photographed by the camera is stored, a display unit for outputting the image photographed by the camera, and a front image At the time of shooting, the front image obtained by the camera is calibrated by using feature point information including the position information of the center point of both eyes, the area information of both ears, the position information of the eyelid on the center point of the eye, and the position information of the apex of the ear. And a control unit configured to store the front image in a memory when the calibration reference information satisfies the set reference range, and obtain a face image of an accurate posture using a feature point of a face. Feature point information can be extracted through the image information of the correct posture, Short-term accurate diagnosis is possible effect.

Facial image, frontal, correction, axis rotation, camera, drive, posture, diagnosis

Description

System for Capturing 2D Facial Image

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a facial imaging apparatus, and more particularly, to a facial imaging apparatus for photographing and correcting a front image to facilitate feature point extraction.

In general, feature points in images of a specific person are used in various technical fields. In particular, the feature points extracted from the face image of the target person may be usefully used in facial recognition technology to determine the identity of the target person through the image.

Feature points in these facial images can also be used in the medical field. With the development of current science and technology, technologies are being developed that can measure the degree of health or abnormality by analyzing images of patients at a distance without facing doctors and patients directly. It is also possible to apply facial diagnosis in home-care systems or ubiquitous-healthcare systems.

As described above, it is important to accurately extract the feature points in order to perform diagnosis through the face image of the patient. If the face is tilted when taking a picture, the accuracy of feature point extraction is degraded. In this case, there is a problem that it is difficult to properly diagnose through facial images. Accordingly, there is a need for a face image photographing apparatus capable of correcting a posture of a target person to obtain a face image having an accurate posture.

In addition, it can be applied to the generation and animation of two-dimensional image-based three-dimensional personal face model using a human face image [1], and can be used for games and entertainment using three-dimensional faces in places such as mobile phones.

[references]

[1] Jin-Woo Kim; Han Seok Ko; Hyung-Kon Kim; Ahn, Sang-Chul, “Creation and Animation of 3D Personal Face Model Based on 2D Image,” Korean Society of Broadcast Engineers 99, 15-20 (6), 1999.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems of the prior art, and an object thereof is to provide a facial imaging apparatus for acquiring a face image of a posture suitable for a set standard by correcting a posture of a target person manually or automatically. .

The face image photographing apparatus according to the first aspect of the present invention for solving the above problems is a camera for acquiring a front image of a target person in real time, a memory in which the image captured by the camera is stored, the image captured by the camera Display unit for outputting an image, when photographing the front image includes the position information of the center point of both eyes, the area information of both ears, the position information of the eyelid above the center point of the pupil and the position information of the apex of the ear And an image analyzer configured to generate calibration reference information using the feature point information, and a controller configured to store the front image in a memory when the calibration reference information satisfies a set reference range.

Here, the image analyzer calculates first calibration reference information that is an angle formed by a straight line connecting the center points of both eyes and an imaginary horizontal line by finding the coordinates of the center points of both eyes in the front image.

The image analyzer calculates second calibration reference information which is an area ratio of both ears in the front image.

In addition, the image analyzer calculates third calibration reference information which is a vertical distance between a straight line connecting eyelids on a center point of both eyes and a straight line connecting vertices of both ears in the front image.

The image analyzer sequentially calculates the first to third calibration reference information, and calculates calibration reference information in the following order when the first to third calibration reference information satisfy each set reference range.

The image analyzer outputs a green indicator to the display unit when the first to third calibration reference information satisfy each set reference range, and outputs a red indicator to the display unit when the first to third calibration reference information do not satisfy each set reference range. Then, the calibration reference information is recalculated for the front image newly acquired by the camera.

In addition, the face image photographing apparatus according to the second aspect of the present invention for solving the above problems is a camera for acquiring a front image of a target person in real time, a drive module for changing the position or angle of the camera, the camera The display unit outputs the image captured by the camera, and when the front image is photographed, the front image obtained by the camera includes the position information of the center point of both eyes, the area information of both ears, the position information of the eyelid on the center point of the pupil, and the position of the apex of the ear. The image analyzer for generating calibration reference information using the feature point information including the information and if the calibration reference information does not satisfy the set reference range, correct the position or the photographing angle of the camera by an error from the set reference range. Control the drive module or camera to retake the image, and the calibration reference information. When satisfied with the set reference range, and a controller for storing the face image in the memory.

Here, the controller controls the driving module to move the camera along a virtual sphere having a radius of the distance between the camera and the target person.

The facial imaging apparatus according to the present invention configured as described above can acquire the facial image of the correct posture by using the feature points of the face, show the facial image acquisition posture in real time, and can immediately correct the posture when shooting. In addition to the dimensional image, it is possible to acquire the image of the correct posture in three dimensions, and feature point information can be extracted through the image information of the correct posture, thereby making it possible to accurately diagnose the diagnosis using the image. In addition, it has a wide range of applications as it can be used for games and entertainment by using two-dimensional image-based three-dimensional personal face model generation and animation.

Hereinafter, with reference to the accompanying drawings will be described specific details and embodiments of the present invention.

1 is a block diagram showing the configuration of a facial imaging apparatus according to a first embodiment of the present invention.

Referring to FIG. 1, a facial image capturing apparatus according to a first exemplary embodiment of the present invention may include a camera 10 for obtaining a front image of a target person in real time, and a display unit 20 for outputting an image captured by the camera. And an image analyzer 30 for outputting a posture correction message to a target person by analyzing the image acquired by the camera, and a controller 40 storing the posture correction image in a memory when the posture correction image is finally determined. It is composed.

The camera 10 obtains an image of a target person, converts the image into an electrical signal, and transmits the image to the image analyzer 30. The type of camera is not limited and generally a digital camera can be used.

The display unit 20 outputs the image acquired by the camera in real time. In general, an LCD display device may be used.

In addition, the display unit 20 outputs whether or not posture correction is necessary for the target person according to the image analyzed by the image analyzing unit, that is, whether it satisfies a set standard, and displays a portion or position that needs to be corrected in the image of the user. It also displays and guides the user how to correct his posture.

Hereinafter, a process of obtaining a front image of a corrected posture using FIGS. 2A to 2C will be described.

When the front image is taken, the image analyzer 30 captures the front image obtained by the camera, the position information of the center point of both eyes, the area information of both ears, the position information of the eyelid on the center point of the pupil, and the position information of the apex of the ear. The calibration reference information is generated using the feature point information.

The image analyzer 30 analyzes the front image of the target person and displays information necessary for posture correction on the image of the target person through the display unit such that Z-axis correction, Y-axis correction, and X-axis correction are sequentially performed. In addition, a message indicating whether the calibration is suitable for the subject is displayed.

Here, the Z-axis refers to the axis indicating the shooting direction of the camera, the Y-axis refers to the axis perpendicular to the ground or the image perpendicular to the image of the camera, the X-axis refers to the axis horizontal to the ground or the axis of the image of the camera. That is, the Y-axis and the X-axis are based on the ground when the camera is installed horizontally with the ground, but when the camera is not horizontal with the ground, the Y-axis and the X-axis are the vertical and horizontal axes respectively in the image acquired by the camera.

The image analyzer 30 calculates first to third calibration reference information for calibrating the front image.

2A is a diagram illustrating a method of calculating first calibration reference information according to the present invention.

The image analyzer 30 first coordinate reference information θ1 which is an angle formed by a straight line L1 connecting the center points of both eyes in the front image and connecting the center points of the two eyes with a virtual horizontal line L2. Calculate

In addition, the image analyzer 30 outputs a straight line L1 and a virtual horizontal line L2 or first calibration reference information connecting the center points of the two eyes through the display unit 20 so that the user can see and correct it. This can help to make it easier.

The image analyzing unit 30 determines whether the first correction reference information θ1 is 0, that is, whether both eyes are horizontal, and satisfies the set reference range T1 when both eyes are horizontal. A green indicator is output through the display unit 20. In this case, the reference range T1 may be set to a specific range close to zero.

If the first calibration reference information θ1 does not satisfy the set reference range, that is, the angle is not 0, the image analyzer 30 outputs a red indicator light through the display unit 20. do. In this case, the target person who is the user corrects his posture through the display unit so that the first calibration reference information θ1 becomes zero. That is, the user tilts the head from side to side with respect to the Z axis to calibrate.

The image analyzer 30 generates second calibration reference information for Y-axis calibration when the first calibration reference information θ1 satisfies 0. However, when the first calibration reference information θ1 satisfies 0, the image analyzer 30 generates a second calibration reference information. The image is newly acquired, and the first calibration reference information θ1 is recalculated to determine whether the set reference range is satisfied.

2B is a diagram illustrating a method of calculating second calibration reference information according to the present invention.

Referring to FIG. 2B, first, the image analyzer 30 calculates second calibration reference information which is an area ratio of both ears in the front image.

The image analyzer 30 detects an ear from the front image and calculates areas A and B of both ears. The area ratio (A / B) of both ears, which is the second calibration reference information, is calculated and compared with the set reference range T2. When the second calibration reference information is not 1, that is, when the areas of both ears are different, Since the posture of the person is skewed, the second calibration reference information, that is, the area ratio of both ears, is output to the display unit so that the target person corrects the posture, and the second calibration reference information is calculated again through the newly obtained image. do. In this case, the area of both ears may be displayed on the display unit for visual viewing.

In this case, a red indicator is output to the display unit 20 to inform the user that the posture is incorrect. In this case, the user may rotate the head with respect to the Y axis while looking at the display device to correct the posture so that the areas of the left and right ears are the same.

If the second calibration reference information is 1 or within a specific range set very close thereto, the image analysis unit 30 outputs a green indicator through the display unit 20 to maintain a posture. The third calibration reference information is calculated for the X axis calibration.

2C is a diagram illustrating a method of calculating third calibration reference information according to the present invention.

Referring to FIG. 2C, first, the image analyzer 30 calculates second calibration reference information which is an area ratio of both ears in the front image.

In the front image, the image analyzing unit 30 is a third distance between the straight line L3 connecting the eyelids on the center point of both eyes and the straight line L4 connecting the vertices of both ears. ) Is compared with the set reference range T3. When the third calibration reference information is not 0, that is, when the height of the eyelid and the height of the apex on the pupil do not match, the posture of the target person is Since it means a skew, the two straight lines L3 and L4 or the distance D1 between the two straight lines are output to the display unit so that the target person corrects the posture, and then the third correction reference information is again obtained through the newly obtained image. Calculate

In this case, a red indicator is output to the display unit 20 to inform the user that the posture is incorrect. In this case, the user may lower his or her head on the X axis while looking at the display device and correct the posture so that the height of the eyelid and the height of the apex on the pupil coincide with each other.

If the third calibration reference information is 0 or within a specific range set very close thereto, the image analysis unit 30 outputs a green indicator through the display unit 20 to maintain a posture. do.

The controller 40 finally stores the front image of the target person in the memory when the first to third calibration reference information satisfies each of the predetermined criteria T1, T2, and T3 sequentially.

3 is a block diagram showing the configuration of a facial imaging apparatus according to a second embodiment of the present invention.

In the face image photographing apparatus according to the second embodiment of the present invention, a camera 10 for acquiring a front image of a target person in real time, a display unit 20 for outputting an image captured by the camera, a position of the camera or A driving module 60 for changing a photographing angle, an image analyzing unit 30 for analyzing calibration of the image acquired by the camera, and calculating calibration reference information, and controlling the operation of the camera or the driving module according to the calibration reference information. It is configured to include a control unit 40.

In the first exemplary embodiment of the present invention, the image analyzer 30 analyzes the acquired image and outputs a message through the display unit 20 to correct the posture by the target person. In the second exemplary embodiment of the present invention, The controller 40 is configured to adjust the photographing angle or position by driving the camera or the driving module in consideration of the calibration reference information calculated by the image analyzer 30 through the driving module 60 installed in the camera 10. .

The drive module 60 includes one or more drive motors, which are controlled by the controller 40 and adjust the photographing angle of the camera or the position of the camera.

The image analyzer 30 generates the first to sixth calibration reference information in the same manner as the first embodiment, and the controller 40 satisfies the reference range in which the first to sixth calibration reference information is set. If not, the position or photographing angle of the camera is corrected by an error from the set reference range.

The controller 40 controls the driving module to move the camera along a virtual sphere having a radius of the distance between the camera and the target person. Therefore, a support or the like for guiding the operation of the camera may be installed in the camera so that the camera can move along the virtual spherical surface as described above.

When the camera is photographed while moving along a virtual spherical surface, the same effect as that of the subject correcting the posture may be obtained in the first embodiment.

Hereinafter, an operation of the facial imaging apparatus according to the second exemplary embodiment of the present invention will be described with reference to FIGS. 4A to 4C.

4A to 4C illustrate the operation of the camera in the case of the front image according to the second embodiment of the present invention.

4A illustrates a state in which calibration is performed based on the Z axis in the front image.

Referring to FIG. 4A, when the first calibration reference information is not 0 °, the controller 40 captures the photographing direction of the camera 10 such that the first calibration reference information is 0 °. The camera is rotated with respect to the photographing direction axis (Z axis) while being fixed to face (). That is, the photographing angle of the camera is adjusted so that a straight line connecting the centers of two eyes of the target person in the front image is horizontal in the image of the camera. The movement of the camera is performed by the drive module 60.

In the case of Figure 4b is a view showing the correction is made based on the Y axis in the front image.

Referring to FIG. 4B, when the second calibration reference information is not 1, the control unit 40 fixes the photographing direction of the camera so that the second calibration reference information is 1 so as to face the target person. Move the camera left and right. That is, the camera 10 is moved to a position where the areas of both ears are the same in the front image. Similarly, the camera movement is performed by the drive module 60. At this time, the camera does not move in a straight line, but moves in a semicircular curvature around the subject.

In the case of Figure 4c is a view showing the correction is made based on the X-axis in the front image.

Referring to FIG. 4C, when the third calibration reference information is not 0, the control unit fixes the photographing direction of the camera 10 toward the target person so that the third calibration reference information is 0. Move up and down. That is, the camera 10 is moved to a position where the height of the eyelid and the height of the apex of the ear just above the center of the pupil are the same. Similarly, the movement of the camera is performed by the driving module 60. At this time, the camera does not move in a straight line, but moves in a semicircular curvature around the target person.

The image analyzer 30 sequentially calculates the first to third calibration reference information, and when the first to the third calibration reference information satisfies each set reference range, the calibration reference information is performed in the following order. Calculate The controller 40 finally stores the front image satisfying the first to third calibration reference information in the memory.

The image analyzer 30 outputs a green indicator to the display unit when the first to third calibration reference information satisfy each set reference range, and outputs a green indicator to the display unit when the first to third calibration reference information do not satisfy each set reference range. After the red light is output, the calibration reference information is recalculated for the front image newly acquired by the camera. Hereinafter, operations of the remaining components are substantially the same as those of the first embodiment, and thus description thereof is omitted.

5 is a flowchart illustrating a method of acquiring a front image of a corrected posture by a facial imaging apparatus according to the present invention.

Since the detailed process in each step is the same as the contents included in the above-described first and second embodiments, a detailed description thereof will be omitted.

Referring to FIG. 5, in the facial image capturing method according to the present invention, a front image is first obtained using a camera (S10).

Next, the first calibration reference information is generated to examine whether it meets the set reference range, and if it satisfies the set reference range, the second calibration reference information is generated. Or, recalibrate the image after calibrating in the Z-axis direction by moving the camera, and generate new first calibration reference information based on the re-photographed image, compare with the set reference range, and repeat until satisfied (S20). .

Similarly, Y-axis rotation calibration is performed using the second calibration reference information (S30), and X-axis rotation calibration is performed using the third calibration reference information (S40).

The first to third calibration reference information stores the front image satisfying each set reference range in a memory (S50).

As described above, the apparatus for photographing a facial image according to the present invention has been described with reference to the illustrated drawings, but the present invention is not limited by the embodiments and drawings disclosed herein, and may be applied within a range in which technical thoughts are protected. .

1 is a block diagram showing the configuration of a facial imaging apparatus according to a first embodiment of the present invention;

2A to 2C are diagrams illustrating a method for calculating first to third calibration reference information according to the present invention;

3 is a block diagram showing the configuration of a facial imaging apparatus according to a second embodiment of the present invention;

4A to 4C are diagrams illustrating operations of a camera in the case of a front image according to the second embodiment of the present invention.

5 is a flowchart illustrating a method of acquiring a front image of a corrected posture by a facial imaging apparatus according to the present invention.

<Explanation of symbols on main parts of the drawings>

10: camera

20: display unit

30: image analysis unit

40: control unit

50: memory

60: drive module

Claims (16)

delete delete delete delete delete delete A camera for acquiring a front image of the target person in real time; A driving module for changing the position or the photographing angle of the camera; A display unit for outputting an image photographed by the camera; First calibration reference information, which is an angle formed by a straight line connecting the center points of two eyes in a front image and a virtual horizontal line in the front image, and a second calibration reference, which is an area ratio of both ears in the front image. An image analyzer configured to generate calibration reference information by calculating information and third calibration reference information which is a vertical distance between a straight line connecting eyelids on a center point of both eyes and a straight line connecting vertices of both ears in the front image; And When the first calibration reference information is not 0 °, the camera photographing direction is fixed so that the first calibration reference information becomes 0 ° or a value within a predetermined range close to 0 ° while being fixed to face the target person. Rotate the camera to the shooting direction axis, and if the second calibration reference information is not 1, the camera is fixed to the target person so that the shooting direction is set to 1 so that the second calibration reference information is 1. If the third calibration reference information is not 0, the camera is fixed while the photographing direction of the camera is set toward the target person such that the third calibration reference information becomes 0 or a value within a predetermined range close to zero. Control the driving module or the camera to move up and down, and when the calibration reference information satisfies the set reference range, Control section for storing a face image in a memory; Facial imaging device comprising a. delete delete delete delete delete delete The method of claim 7, The image analyzing unit sequentially calculates the first to third calibration reference information, and calculates calibration reference information in the following order when the first to third calibration reference information satisfy each set reference range. Facial imaging device. The method according to claim 14, The image analyzer outputs a green indicator to the display unit when the first to third calibration reference information satisfy each set reference range, and outputs a red indicator to the display unit when the first to third calibration reference information do not satisfy each set reference range. And recalculate corresponding calibration reference information with respect to the front image newly acquired by the camera. The method according to claim 14, And the controller controls the driving module to move the camera along a virtual sphere having a radius of the distance between the camera and the target person.

Images