JP2014150304A - Display device and display method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a display device and a display method therefor capable of improving visibility by correcting a distortion in the depth direction of a display image.SOLUTION: A head up display (HUD device) 100 includes: display means 30 for displaying a display image including a left-eye image and a right-eye image; display control means for allowing the display means 30 to display the display image; image separation means 40 for separating the display image displayed by the display means 30 into the left-eye image and the right-eye image; and position detection means capable of detecting the viewpoint position of a driver 102, and projects the display image onto a wind shield 101. The display control means corrects a deviation amount between the left-eye image and the right-eye image in order to correct a distortion in the depth direction of the display image when projected onto the wind shield 101 on the basis of the viewpoint position detected by the position detection means.

Description

  The present invention relates to a display device and a display method thereof.

  A so-called head-up display device (hereinafter referred to as a HUD device) is known as a display device that displays information to a driver of a vehicle such as a conventional automobile. The HUD device projects a display image displayed on a display unit such as a liquid crystal display panel onto a projection target unit such as a windshield so that a driver can visually recognize a virtual image of the display image.

  In the HUD device, a display image is distorted by enlarging the display image by an optical system such as a projection target or a concave mirror. Since the visibility decreases when the display image is distorted, a technique for correcting the distortion by deforming the display image in advance reversely is known (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 10-149085

  However, the technique disclosed in Patent Document 1 corrects the distortion in the horizontal direction of the display image, but cannot correct the distortion generated in the depth direction, and has a problem of reducing visibility. It was.

  Accordingly, an object of the present invention is to provide a display device and a display method thereof that can solve the above-described problems, correct distortion in the depth direction of a display image, and improve visibility.

In order to solve the above problems, the present invention provides a display unit that displays a display image including the left-eye image and the right-eye image, a display control unit that displays the display image on the display unit, Image display means for separating the display image displayed on the display means into the image for the left eye and the image for the right eye, and a position detection means capable of detecting the viewpoint position of the observer, the display image A projection device for projecting onto a projection member,
The display control unit is configured to correct the distortion in the depth direction of the display image when projected onto the projection member based on the viewpoint position detected by the position detection unit, and the left eye image and the right eye. It is characterized in that the amount of deviation from the image for use is corrected.

In order to solve the above problem, the present invention provides a display unit that displays a display image including a left-eye image and a right-eye image, and the display image displayed on the display unit is the left-eye image. A display device display method comprising: an image separation unit that separates the right-eye image; and a position detection unit that is capable of detecting an observer's viewpoint position, and projects the display image onto a projection member.
Deviation between the left-eye image and the right-eye image to correct distortion in the depth direction of the display image when projected onto the projection member based on the viewpoint position detected by the position detection means It is characterized by correcting the amount.

  The present invention can achieve the intended purpose, correct distortion in the depth direction of the display image, and improve visibility.

It is the schematic of the head-up display apparatus of embodiment of this invention. It is a figure explaining the example of the three-dimensional display in the same as the above. It is a block diagram which shows the electric constitution of a control part same as the above. It is a flowchart figure which shows the control method in the same as the above. It is a figure explaining the viewpoint detection in the same as the above. It is a flowchart figure which shows the image correction method in the same as the above. It is a figure which shows the relationship between the viewpoint angle in the same as above, and a parallax correction parameter. It is a figure which shows an example of the parallax correction parameter in the same as the above. It is a figure explaining the relationship between the shift | offset | difference of the corresponding point of the image for left eyes and the corresponding point of the image for right eyes, and a three-dimensional image in the same as the above. It is a figure which shows the distortion of the depth direction of the virtual image in the same as the above.

Hereinafter, an embodiment in which the present invention is applied to a head-up display (HUD) device 100 will be described with reference to the accompanying drawings.
The HUD device 100 is installed in a dashboard in a vehicle interior of a vehicle, and as shown in FIG. 1, an imaging unit (position detection unit) 10, a control unit 20, a display unit 30, and image separation are performed. Means 40 and a concave mirror 50. The display light L indicating a predetermined image output from the HUD device 100 is reflected by the windshield 101 of the vehicle and reaches the eyes of the driver (observer) 102. The driver 102 visually recognizes the virtual image V displayed in front of the windshield 101.

  The imaging unit 10 includes a known imaging device such as a CCD and peripheral circuits and lenses that control the imaging device. The imaging unit 10 is installed in front of the driver 102 and images the face of the driver 102 and outputs a captured image. .

  The control unit 20 is a so-called microcomputer. The control unit 20 is connected to the imaging unit 10 and the display unit 30, performs a viewpoint position detection process for analyzing the face image of the driver 102 output from the imaging unit 10 and calculating the viewpoint position of the driver 102. Based on the calculation result, a control signal is output to the display means 30 to perform image display control. That is, the control unit 20 functions as the position detection unit of the present invention together with the imaging unit 10 and also functions as the display control unit of the present invention. The control unit 20 cuts out a striped divided image from the image for the left eye when the object is viewed from the left eye and the image for the right eye when the object is viewed from the right eye, and alternately converts the divided images in the horizontal direction. A composite image (display image) including the left-eye image and the right-eye image is generated side by side, and this composite image is displayed on the display means 30.

  The display unit 30 includes, for example, a backlight light source 31 and a dot matrix type liquid crystal display panel 32, and emits the composite image as display light L toward the concave mirror 50. The display unit 30 may be any unit that emits the composite image as display light L, and may be a self-luminous display panel such as an organic EL panel.

  The image separating means 40 is composed of, for example, a parallax barrier (parallax barrier) provided with a plurality of stripe-shaped slit portions that transmit light, and is provided on the emission surface side of the display means 30 with a predetermined interval. The display light L passes through the slit portion of the image separation means 40, so that the composite image is divided into a left-eye image and a right-eye image, and the left-eye image light and the right-eye image light are divided by the driver 102. Can be incident separately on the left and right eyes. A lenticular lens may be used as the image separation means 40.

  The concave mirror 50 irradiates the windshield 101 with the display light L so that the display light L is focused on the left and right eyes of the driver 102. The concave mirror 50 includes a rotating shaft (not shown). The rotation axis is perpendicular to the drawing of FIG. 1, and the angle of the concave mirror 50 relative to the windshield 101 can be changed around the rotation axis. In addition, the concave mirror 50 is provided with a motor (not shown) as a driving means in order to change the angle of the concave mirror 50. By this motor, the angle of the concave mirror 50 with respect to the windshield 101 is changed, and the angle is changed. Can be maintained. As described above, the display light L can be adjusted to the driver's 102 line of sight.

When the left-eye image light enters the left eye and the right-eye image enters the right eye of the display light L, the driver 102 recognizes the virtual image V as a three-dimensional image due to a shift (parallax) between the two images. can do. FIG. 2 shows a state in which the composite image including two images for the left eye and two for the right eye (four in total) is displayed on the display unit 30 to perform stereoscopic display. E1 to E4 in FIG. 2 indicate each region (lobe) in the observation region E where the image can be observed. When a composite image in which two divided images of the left-eye image 60L and two divided images of the right-eye image 60R are alternately arranged is displayed on the display means 30 (display of the divided image of the left-eye image 60L in FIG. 2). The position is “L” and the display position of the divided image of the right-eye image 60R is “R”), and the driver 102 observes the composite image through the slit portion 41 of the image separation means 40. Thus, different images are observed for the left eye and the right eye. In FIG. 2, the right-eye image 60R can be observed in the areas E1 and E2 of the observation area E, and the left-eye image 60L can be observed in the areas E3 and E4. Therefore, when the right eye of the driver 102 is positioned between the areas E1 and E2 and the left eye is positioned between the areas E3 and E4, stereoscopic display is possible. Further, in the parallax barrier method or the lenticular method in which an image is divided and stereoscopic display is performed, the areas E1 to E4 exist repeatedly in the observation area E due to the spread of light from the display means 30. In the present embodiment, among the observation areas E, the areas E1 to E4 located in the center are main lobes that are the observation positions of the designed image (virtual image V), and the areas E1 to E4 located outside the main lobes are the side lobes. Become a robe.
In FIG. 2, the concave mirror 50 and the windshield 101 are omitted to simplify the drawing.

  FIG. 3 shows the electrical configuration of the control unit 20. As illustrated in FIG. 3, the control unit 20 includes a CPU 21, a ROM 22, a general-purpose memory 23, a video memory 24, and an external device I / F 25.

  The CPU 21 reads out and executes a predetermined program stored in advance in the ROM 22, temporarily stores the captured image from the imaging unit 10, a processed image obtained by performing image processing on the image, a calculation result such as a viewpoint position, and the like in the general-purpose memory 23. The synthesized image to be displayed on the display means 30 is stored in the video memory 24 using a circuit that performs image processing for creating the synthesized image.

  The ROM 22 stores a program for operating the CPU 21, parallax correction parameters described later, and the like.

  The general-purpose memory 23 temporarily stores a captured image from the imaging unit 10, a processed image obtained by performing image processing on the captured image, a calculation result such as a viewpoint position, and the like.

  The video memory 24 stores the composite image to be displayed on the display means 30.

  The external device I / F 25 electrically connects the control unit 20 to a vehicle control system unit 103 including a computer that controls the imaging unit 10, the display unit 30, and the vehicle. The external device I / F 25 acquires a captured image from the imaging unit 10, outputs a control signal to the display unit 30, and acquires vehicle information such as a vehicle speed from the vehicle control system unit 103. CPU21 can display vehicle information on the display means 30 based on the acquired vehicle information.

  Next, viewpoint detection processing and image display control in the control unit 20 will be described with reference to FIG.

  First, in step S <b> 1, the CPU 21 detects the left eye and the right eye of the driver 102 from the captured image obtained by capturing the face of the driver 102 output from the imaging unit 10. Specifically, as shown in FIG. 5, for example, template matching which is a well-known image processing technique is performed on the captured image P including the face of the driver 102, and a predetermined template T is moved on the captured image P. Then, the location that matches the template T is detected as the left eye and the right eye of the driver 102 on the captured image P.

  Next, in step S2, the CPU 21 determines the left eye viewpoint position and the right eye viewpoint position in the space of the driver 102 based on the positions of the left eye and the right eye of the driver 102 on the captured image P detected in step S1. And calculate. The position information of the left eye viewpoint position and the right eye viewpoint position may be two-dimensional information or three-dimensional information. Note that the imaging unit 10 captures the face of the driver 102 even when the left eye or the right eye of the driver 102 is outside the main lobe of the observation area E, that is, inside the side lobe and outside the observation area E. It is assumed that the CPU 21 can calculate the left eye viewpoint position and the right eye viewpoint position of the driver 102.

  Next, in step S3, the CPU 21 generates the composite image in which the divided image of the left-eye image 60L and the divided image of the right-eye image 60R are arranged in a predetermined order. At this time, the CPU 21 corrects the shift amount between the left eye image 60L and the right eye image 60R so as to correspond to the left eye viewpoint position and the right eye viewpoint position of the driver 102 calculated in step S2, and Correct horizontal distortion. Then, the CPU 21 arranges the corrected divided image of the left eye image 60L and the divided image of the right eye image 60R in a predetermined order, and generates the composite image. The correction method for the left-eye image 60L and the right-eye image 60R will be described in detail later.

  Next, in step S4, the CPU 21 displays the composite image on the display unit 30, and displays the left-eye image 60L and the right-eye image 60R side by side.

  The CPU 21 executes steps S1 to S4 in an infinite loop until the power is turned off.

  Next, an example of a correction method for the left-eye image 60L and the right-eye image 60R will be described with reference to FIG.

  First, in step S11, the CPU 21 calculates a viewpoint angle from the left eye viewpoint position and right eye viewpoint position calculated in step S2. Specifically, with reference to the perpendicular line from the focal plane of the virtual image V, the horizontal position formed by the center of the viewpoint position (the center between the left eye viewpoint position and the right eye viewpoint position) and the perpendicular line from the focal plane of the virtual image V The angle is calculated as the viewpoint angle.

  Next, in step S12, the CPU 21 acquires a parallax correction parameter corresponding to the viewpoint angle calculated in step S11. For example, as shown in FIG. 7, a plurality of parallax correction parameters are set according to a plurality of viewpoint angles and stored in the ROM 22. In FIG. 7, five parallax correction parameters RP1 to RP5 are set for five viewpoint angles of viewpoint angles 60 °, 30 °, 0 °, −30 °, and −60 °. In this embodiment, five viewpoint angles are provided at equal intervals, but the number of viewpoint angles set and the interval are arbitrary. As shown in FIG. 8, each of the plurality of parallax correction parameters RP1 to RP5 set is displayed on the display surface of the display means 30 for a plurality of corresponding points respectively set for the left-eye image 60L and the right-eye image 60R. Have a value (the number of dots from the origin) indicating the position in the x-axis (horizontal axis) direction and the position in the y-axis (vertical axis) direction. The corresponding point indicates a dot that is determined to be the same or approximate in the left-eye image 60L and the right-eye image 60R. FIG. 8 shows an example in which the values in the x-axis direction and the value in the y-axis direction of the nine corresponding points 1 to 9 are set in the parallax correction parameters RP1 to RP5, but the number of corresponding points is arbitrary. It is.

The parallax correction parameters RP1 to RP5 are respectively the corresponding points of the left-eye image 60L and the corresponding points of the right-eye image 60R to enable stereoscopic viewing by the left-eye image 60L and the right-eye image 60R. A predetermined deviation amount (parallax amount) is set between the two. The pop-out amount (or depth amount) of the stereoscopic image is determined by the amount of deviation between each corresponding point of the left-eye image 60L and each corresponding point of the right-eye image 60R. The amount of projection of the virtual image V of the stereoscopic image from the focal plane is obtained by the following equation 1.
(Formula 1)
3D image projection amount = (deviation amount of left and right corresponding points) / (deviation amount of left and right corresponding points) + pupil interval) × distance to the focal plane of the virtual image V Also, from the focal plane of the virtual image V of the stereoscopic image The depth amount is obtained by the following equation 2.
(Formula 2)
3D depth amount = (deviation amount between left and right corresponding points) / (deviation amount between left and right corresponding points−pupil interval) × distance to the focal plane of virtual image V FIG. 9 shows corresponding points 70L of left-eye image 60L. The relationship between the shift between the corresponding point 70 </ b> R and the corresponding point 70 </ b> R of the right-eye image 60 </ b> R and the projection or depth of the stereoscopic image F is shown. 9 shows a state in which the virtual image V is viewed from above. The lower side of FIG. 9 shows the near side (driver 102 side), and the upper side shows the depth side (front side). When obtaining a three-dimensional image F that protrudes toward the front with respect to the focal plane G of the virtual image V, as shown in FIG. 9A, the left-eye image 60L with respect to the corresponding point 70R of the right-eye image 60R. The corresponding point 70L is shifted to the right in the x-axis direction to provide a right parallax. On the other hand, when obtaining a stereoscopic image F having a depth with respect to the focal plane G of the virtual image V, as shown in FIG. 9B, the left eye with respect to the corresponding point 70R of the right-eye image 60R. The corresponding point 70L of the work image 60L is shifted leftward in the x-axis direction to provide a leftward parallax. As shown in FIG. 9C, when the corresponding point 70R of the right-eye image 60R and the corresponding point 70L of the left-eye image 60L match, a stereoscopic image F is formed on the focal plane G of the virtual image V. Is done.

  Furthermore, the parallax correction parameters RP1 to RP5 are respectively corresponding points of the left-eye image 60L and the right-eye image 60R to correct distortion in the depth direction of the virtual image V generated according to the corresponding viewpoint angle. The amount of deviation is adjusted. The virtual image V is distorted in the depth direction with the change in the viewpoint angle when the display light L is projected onto the windshield 101 due to the enlargement of the image by the concave mirror 50, the shape of the windshield 101, and the like. FIG. 10 shows an example of distortion of the virtual image V in the depth direction. 10 shows a state in which the virtual image V is viewed from above. The lower side of FIG. 10 shows the near side (driver 102 side), and the upper side shows the depth side (front side). The distortion in the depth direction is such that the virtual image V appears in front of the focal plane G as indicated by the region V1, and the virtual image V is positioned on the depth side of the focal plane G as indicated by the region V2. There is distortion. For example, as shown in FIG. 9A, when distortion in the depth direction occurs in a portion where the stereoscopic image F popping out from the virtual image V is obtained, if the distortion toward the near side occurs, the pop-out amount of the stereoscopic image F Becomes larger than the setting and the distortion to the depth side occurs, the pop-out amount of the stereoscopic image F becomes smaller than the setting, and the visibility of the stereoscopic image F is lowered in both cases. In addition, when the pop-out amount becomes excessively large, stereoscopic vision may not be obtained. On the other hand, each of the parallax correction parameters RP1 to RP5 has a parallax that gives the depth to the opposite side, that is, the stereoscopic image, at a position where distortion toward the near side occurs as distortion correction in the depth direction of the virtual image V. This is added to the amount of deviation between the corresponding point of the left-eye image 60L and the corresponding point of the right-eye image 60R. Specifically, for example, the value in the x-axis direction of the corresponding point of the left-eye image 60L is shifted to the left in the x-axis direction by an amount corresponding to the distortion on the near side with respect to the corresponding point of the right-eye image 60R. Add the amount of shift in the left direction. In addition, as a distortion correction in the depth direction of the virtual image V, in a portion where distortion to the depth side occurs, a parallax that causes the stereoscopic image to jump out to the opposite side of the depth side, i. This is added to the amount of deviation from the corresponding point of the image for use 60R. Specifically, for example, the corresponding point of the right-eye image 60R is a value obtained by shifting the value in the x-axis direction of the corresponding point of the left-eye image 60L to the right in the x-axis direction by an amount corresponding to the distortion on the depth side. Is added to the right. As the distortion correction in the depth direction of the virtual image V, the value in the x-axis direction at the corresponding point of the right-eye image 60R may be adjusted. In the case where the shift amount is corrected by the corresponding point of the right-eye image 60R, specifically, in the portion where the distortion to the near side occurs, specifically, the value in the x-axis direction of the corresponding point of the right-eye image 60R is the near side. The amount of rightward shift is added to the corresponding point of the left-eye image 60L as a value shifted to the right in the x-axis direction by an amount corresponding to the distortion of the left-eye. Further, at the location where the distortion to the depth side occurs, specifically, the value in the x-axis direction of the corresponding point of the right-eye image 60R is set to the left side in the x-axis direction by an amount corresponding to the distortion on the depth side. A shift amount in the left direction is added to the corresponding point of the left-eye image 60L as a shifted value.

The parallax correction parameters RP1 to RP5 are stored in the ROM 22 as a lookup table, and the CPU 21 acquires the parallax correction parameters RP1 to RP5 from the ROM 22 based on the calculated viewpoint angle. When the viewpoint angle is a value that does not correspond to the parallax correction parameters RP1 to RP5 (for example, when the viewpoint angle is −50 °), the parallax correction parameters RP1 to RP5 corresponding to the previous and next viewpoint angles are acquired and both The parallax correction parameter corresponding to the viewpoint angle calculated by the interpolation process may be acquired. As an example, when the midpoint between the parallax correction parameter RPn and the parallax correction parameter RPn + 1 is acquired as the interpolated parallax correction parameter RP ′, the x-axis of the m-th corresponding point of the right-eye image 60R of the parallax correction parameter RP ′ The value xR′m in the direction and the value yR′m in the y-axis direction are obtained by the following (Formula 3).
(Formula 3)
xR′m = (xRnm−xRn + 1m) / 2
yR′m = (yRnm−yRn + 1m) / (xRnm−xRn + 1m) × xR′m + yRnm
Note that xRnm and yRnm respectively indicate the value in the x-axis direction and the value in the y-axis direction of the m-th corresponding point of the right-eye image 60R in the parallax correction parameter RPn, and xRn + 1m and yRn + 1m are respectively in the parallax correction parameter RPn + 1. A value in the x-axis direction and a value in the y-axis direction of the m-th corresponding point of the right-eye image 60R are shown. Note that the value xL′m in the x-axis direction and the value yL′m in the y-axis direction of the m-th corresponding point of the left-eye image 60L of the parallax correction parameter RP ′ can be calculated in the same manner.
In addition to linear interpolation, there are three types of polynomials used as an interpolation function, such as cubic interpolation, Lagrangian interpolation, and spline interpolation. Any of them may be used for interpolation processing. In addition, a high-order polynomial is used for the interpolation function used for the interpolation process for the corresponding point located at the location where the distortion is large, and a low-order polynomial is used for the corresponding point located for the location where the distortion is small for the interpolation process. For example, the polynomial may be changed for each corresponding point. Although higher-order polynomials require more computation, distortion correction accuracy is improved. On the other hand, low-order polynomials have a low distortion correction accuracy but a small amount of calculation. By properly using the two according to the corresponding points, both the distortion correction accuracy and the calculation speed can be improved.

  Next, in step S13, the CPU 21 calculates the parallax between the left-eye image 60L and the right-eye image 60R based on the parallax correction parameters RP1 to RP5 acquired in step S12 (including the case acquired by interpolation processing). to correct. Specifically, each corresponding point is moved based on the value in the x-axis direction and the value in the y-axis direction of each corresponding point acquired as the parallax correction parameters RP1 to RP5. In addition, image interpolation is performed when pixel loss occurs with the movement of each corresponding point.

  Next, in step S14, the CPU 21 divides the left-eye image and the right-eye image subjected to the parallax correction in step S13 into a predetermined number of substantially triangular shapes, and distortion correction parameters for correcting the horizontal distortion. It is pasted to the triangular patch created based on (stored in the ROM 22) by texture mapping. The texture mapping may use a function implemented in a circuit that performs image processing of the CPU 21 or may be a function implemented as software.

  The CPU 21 corrects the left-eye image 60L and the right-eye image 60R by executing steps S11 to S14.

The HUD device 100 according to the present embodiment includes a display unit 30 that displays a composite image including a left-eye image 60L and a right-eye image 60R, and a display control unit (control unit 20) that displays the composite image on the display unit 30. ), An image separation unit 40 that separates the composite image displayed on the display unit 30 into a left-eye image 60L and a right-eye image 60R, and a position detection unit (imaging) that can detect the viewpoint position of the driver 102. Unit 10 and control unit 20), and a display device for projecting the composite image onto the windshield 101,
The display control means includes a left-eye image 60L to correct distortion in the depth direction of the composite image (virtual image V) when projected onto the windshield 101 based on the viewpoint position detected by the position detection means. The amount of deviation from the right-eye image 60R is corrected.

  Further, the display method of the HUD device 100 is such that the left eye corrects the distortion in the depth direction of the composite image (virtual image V) when projected onto the windshield 101 based on the viewpoint position detected by the position detection unit. The shift amount between the image for use 60L and the image for the right eye 60R is corrected.

  According to this, the distortion in the depth direction of the virtual image V is canceled by the amount of shift between the left-eye image 60L and the right-eye image 60R that is conventionally set only for stereoscopic vision. Distortion in the depth direction can be corrected, and visibility can be improved.

In addition, the HUD device 100 according to the present embodiment includes a ROM 22 that stores parallax correction parameters RP1 to RP5 of deviation amounts between the left-eye image 60L and the right-eye image 60R according to the viewpoint angle of the driver 102. ,
The display control unit calculates the viewpoint angle of the driver 102 based on the viewpoint position detected by the position detection unit, acquires parallax correction parameters RP1 to RP5 corresponding to the calculated viewpoint angle from the ROM 22, and acquires them. The shift amount between the left-eye image 60L and the right-eye image 60R is corrected based on the parallax correction parameters RP1 to RP5.

  The display method of the HUD device 100 calculates the viewpoint angle of the driver 102 based on the viewpoint position detected by the position detection unit, and acquires parallax correction parameters RP1 to RP5 corresponding to the calculated viewpoint angle from the ROM 22. The shift amount between the left-eye image 60L and the right-eye image 60R is corrected based on the acquired parallax correction parameters RP1 to RP5.

  According to this, even when the viewpoint of the driver 102 moves in the horizontal direction, the distortion in the depth direction of the display image can be corrected according to the viewpoint angle, and the visibility can be improved. .

  In the HUD device 100 according to the present embodiment, the display control unit performs an interpolation process between a plurality of the correction parameters according to the calculated viewpoint angle.

  Further, the display method of the HUD device 100 is characterized in that interpolation processing between a plurality of the correction parameters is performed according to the calculated viewpoint angle.

  According to this, it is possible to perform distortion correction in the depth direction with high accuracy without reducing the amount of data stored in the ROM 22 and increasing the size of the storage unit.

  Note that the present invention is not limited to the above-described embodiment, and various modifications (including deletion of constituent elements) can be made without departing from the scope of the invention. In the present embodiment, two divided images of the left eye image 60L and two divided images of the right eye image 60R (a total of four) are displayed side by side on the display unit 30. The eye image 60L and the right eye image 60R may be displayed one by one, or three or more of each may be displayed. Further, the display position of the left eye image 60L and the display position of the right eye image 60R on the display unit 30 are changed so as to correspond to the left eye viewpoint position and right eye viewpoint position of the driver 102 calculated in step S2. The stereoscopic view may be maintained even if the viewpoint position of the driver 102 moves. In the present embodiment, the HUD device 100 that displays the virtual image V is taken as an example of the stereoscopic display device. However, the present invention is not limited to the display image from the display means as a reflection screen (projection member) or a transmission screen (projection member). ), And may be applied to a display device for visually recognizing a real image. Moreover, although the HUD apparatus 100 of this embodiment was for vehicles, you may apply this invention to display apparatuses other than the object for vehicles.

  The present invention is applicable to a display device.

100 Head-up display device 101 Wind shield (projection member)
102 Driver (observer)
DESCRIPTION OF SYMBOLS 10 Imaging part 20 Control part 30 Display means 40 Image separation means 50 Concave mirror 60L Image for left eyes 60R Image for right eyes

Claims (6)

Display means for displaying a display image including a left-eye image and a right-eye image;
Display control means for displaying the display image on the display means;
Image separation means for separating the display image displayed on the display means into the left-eye image and the right-eye image;
A position detecting means capable of detecting an observer's viewpoint position, and projecting the display image onto a projection target member,
The display control unit is configured to correct the distortion in the depth direction of the display image when projected onto the projection member based on the viewpoint position detected by the position detection unit, and the left eye image and the right eye. A display device that corrects a deviation amount from an image for use. A storage unit that stores a correction parameter for a shift amount between the left-eye image and the right-eye image according to the viewpoint angle of the observer;
The display control unit calculates a viewpoint angle of the observer based on the viewpoint position detected by the position detection unit, acquires the correction parameter corresponding to the calculated viewpoint angle from the storage unit, and acquires the acquired The display device according to claim 1, wherein a shift amount between the left-eye image and the right-eye image is corrected based on a correction parameter.   The display device according to claim 2, wherein the display control unit performs an interpolation process between the plurality of correction parameters according to the calculated viewpoint angle. Display means for displaying a display image including a left-eye image and a right-eye image;
Image separation means for separating the display image displayed on the display means into the left-eye image and the right-eye image;
A position detection means capable of detecting a viewer's viewpoint position, and a display device display method for projecting the display image onto a projection target member,
Deviation between the left-eye image and the right-eye image to correct distortion in the depth direction of the display image when projected onto the projection member based on the viewpoint position detected by the position detection means A display method for a display device, wherein the amount is corrected. A storage unit that stores a correction parameter for a shift amount between the left-eye image and the right-eye image according to the viewpoint angle of the observer;
The viewpoint angle of the observer is calculated based on the viewpoint position detected by the position detection unit, the correction parameter corresponding to the calculated viewpoint angle is acquired from the storage unit, and the correction parameter is acquired based on the acquired correction parameter. The display device display method according to claim 4, wherein a shift amount between the left-eye image and the right-eye image is corrected.   The display method of the display device according to claim 5, wherein an interpolation process between the plurality of correction parameters is performed according to the calculated viewpoint angle.

Images