JP2006044596A - Display device for vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a display device for a vehicle with high dead angle reduction effect constituted such that a driver easily, instantaneously and intuitively grasp which direction a projected image is directed. <P>SOLUTION: The display device 100 for the vehicle is provided with an outdoor camera 11 for photographing a periphery of the vehicle 41; a view recognition unit 21 for recognizing a spatial position of the view of the driver seated on a driver's seat; an image data processing device 23 for producing outdoor scene which should be projected to eyes of the driver when it is assumed that a shutting off article such as a side pillar and a door for shutting off a visual field of the driver is transparent by processing peripheral image data obtained by the outdoor camera 11 in response to recognition result of the view recognition unit 21; and displays 31-38 for displaying the image produced by the image data processing device 23. The displays 31-38 are mounted to the shutting off article such as a pillar. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a display device for a vehicle that can reduce a driver's blind spot by capturing an image of the surroundings of the vehicle and displaying the image on a display in a passenger compartment.

Conventionally, an attempt has been made to capture a driver's blind spot with a camera and display it on a display or hologram (Patent Documents 1 to 4 below).
JP-A-6-227318 Japanese Unexamined Patent Publication No. 7-17328 JP-A-7-223487 JP 2000-71877 A

  In a conventional display device, for example, an image behind the vehicle is displayed on a display arranged on a dashboard. In other words, since the display arrangement and the driver's blind spot direction do not match, it is difficult for the driver to instantly grasp what the displayed image shows.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a vehicle display device with a high effect of reducing blind spots, which is configured so that a driver can easily and intuitively know in which direction an image displayed on a display or the like is. Is to provide.

Means for Solving the Problems and Effects of the Invention

  In order to solve the above problems, a display device for a vehicle according to the present invention includes an imaging means for imaging the periphery of a vehicle, a viewpoint recognition means for recognizing a spatial position of the viewpoint of the driver seated on the driver's seat, and the driver If the obstacles such as side pillars and doors that block the field of view are transparent, the image of the scenery outside the vehicle that should be seen by the driver and the peripheral image data obtained by the imaging means are the recognition results of the viewpoint recognition means The main feature is that it includes image generation means that is generated by processing according to the above, and display means that displays an image generated by the image generation means at the position of the shielding object.

  The vehicle display device according to the present invention has a display means such as a display installed on a shielding object that generates a blind spot such as a side pillar or a door, and displays the scenery outside the vehicle as if the shielding object is transparent. It is. In addition to attaching a display to the shield, a method of displaying a hologram can also be considered. One of the technical hurdles of such a vehicle display device is to match the scenery actually reflected through the windshield or window glass with the image projected at the position of the shield. In the present invention, a method is employed in which the viewpoint position of the driver is obtained in advance, and the image obtained by the imaging means is processed according to the viewpoint position. According to such a configuration, since the viewpoint of the actual driver is the reference, what is the difference between the scenery actually reflected through the windshield and the window glass and the image projected at the position of the shielding object? It can be solved flexibly even for a physique driver. Accordingly, it is possible to effectively reduce the blind spots of all drivers without causing a sense of incongruity, and it is possible to realize a vehicle display device that can contribute to early danger detection and safe driving.

  Specifically, the viewpoint recognition unit described above determines the position of the driver's viewpoint in the three-dimensional coordinates from the in-vehicle camera provided in the vehicle interior in order to image the driver from the front and the image data acquired by the in-vehicle camera. It can comprise as a viewpoint recognizer to calculate. As described above, if the driver's viewpoint is specified on the three-dimensional coordinates, it is easy to obtain a range obstructed by a shield such as a pillar.

  The viewpoint recognizing means obtains an instruction for stabilizing the viewpoint from the viewpoint recognizer when the in-vehicle camera is operated to image the driver, and the driver recognizes the viewpoint based on identification information such as voice. It can be configured to include a dialogue controller that performs guidance for stabilization. The driver does not always maintain the driving posture at the timing when the driver is imaged by the in-vehicle camera. In this case, there is a possibility that the process of recognizing the driver's viewpoint cannot be performed smoothly. Therefore, it is preferable to encourage cooperation in the process for viewpoint recognition by voice guidance using a dialog controller.

  Further, there may be a plurality of shielding objects such as left and right side pillars, doors of a driver seat and a passenger seat, and a dashboard. In such a case, a thin display can be adopted as display means for making the shielding objects pseudo transparent and can be individually attached to the shielding objects. The viewpoint recognizing means can be configured to perform a process of collecting the motion when the driver moves his / her line of sight to each display or the posture after the movement as image data when moving the line of sight, A storage unit for storing image data is further provided. That is, if the operation of the driver during driving is always imaged, and if it is determined whether or not the driver has performed the previously stored operation when moving the line of sight, the images outside the vehicle are always displayed on all displays. It is possible to perform a process of selectively displaying an image outside the vehicle only on a display that is in the driver's field of view.

  In another preferred embodiment, the display means is a display attached to the shield, and the image generation means specifies installation position data for specifying the installation position of the display on a three-dimensional coordinate and the shape of the display. It has a storage unit for storing shape data in advance, and is viewed from the driver from the viewpoint position of the driver obtained by the viewpoint recognition means, the installation position data read from the storage unit, and the shape data as well. In some cases, the peripheral image data obtained by the imaging means may be cut out in such a manner that only the image blocked by the shielding object is extracted. According to such a configuration, it is possible to smoothly eliminate the deviation between the scenery reflected through the windshield and the window glass and the image projected at the position of the shielding object.

  The imaging means is preferably a digital outside camera equipped with a fish-eye lens. For example, if two fisheye lenses are used, the entire periphery of the vehicle can be imaged. The image generation means performs processing for transforming an image obtained from the outside camera by coordinate transformation, calculates a range where the field of view is blocked by the shielding object, and obtains pixel data of the blocked range from the outside camera. It is possible to perform the process of extracting from the above and setting the graphic memory for display. If a fisheye lens is used, the entire periphery of the vehicle can be imaged with a small number of cameras. Since the process of extracting only the necessary pixel data from the digital image data can be performed at a relatively high speed, there is no possibility of delay in the image displayed on the display.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a block diagram showing a configuration of a vehicle display device according to the present invention. FIG. 2 is a schematic view of a vehicle showing an installation example of the vehicle display device. As shown in FIG. 1, the vehicle display device 100 includes an outside camera 11 that captures an image of the periphery of the vehicle, a viewpoint recognizer 21 that recognizes the spatial position of the viewpoint of the driver seated in the driver's seat, and the outside camera 11. The image data processing device 22 that processes the image obtained by the above and displays 31 to 38 that display the image processed by the image data processing device 22. Images are displayed on the displays 31 to 38 attached to the pillars and doors that obstruct the driver's view as if the pillars and doors are transparent. As a result, the blind spot of the driver can be reduced.

  As shown in FIG. 2, the vehicle exterior camera 11 is installed at the front end of the vehicle 41. The camera 11 outside the vehicle is a CCD camera with a fisheye lens attached. Other types of digital cameras such as a digital camera using a CMOS-IC can also be employed. By using the fisheye lens, it is possible to image a range of the solid angle 2π in the traveling direction. The fish-eye lens includes, for example, an equidistant projection method (fθ), a flat projection method (2f · tan (θ / 2)), an equal solid angle projection method (2f · sin (θ / 2)), and an orthographic projection method (f · sinθ). ) Etc. may be used. The equidistant projection method is preferable in terms of performance and cost. In addition, the equidistant projection method or the like destroys the information amount in the periphery of the screen, whereas the flat projection method is suitable for obtaining a clear image because the information amount in the peripheral portion is not lost. It is also possible to provide a plurality of such external cameras 11. For example, by providing a pair of external cameras 11 at diagonal positions between the front end portion and the rear end portion, it is possible to clearly capture not only the front but also the rear of the vehicle.

  The image picked up by the outside camera 11 is sent to the outside camera controller 15. The vehicle exterior camera controller 15 constitutes the vehicle display device 100. In addition, the camera 11 outside the vehicle is provided with a sensor for detecting the environment (temperature, humidity, etc.) outside the vehicle, and has a function of removing its own optical system, the temperature of the CCD, the fogging of the cover, etc. so that a clear image can be obtained. Have. The vehicle camera controller 15 receives an image from the vehicle camera 11, converts it into digital data, and sends it to the image data processing device 23. Alternatively, digitally converted image data may be stored in a memory (not shown) of the vehicle camera controller 15 so that the image data processing device 23 can obtain the image data for processing. In this case, since the vehicle camera controller 15 periodically rewrites the digital data, if the digital data cannot be taken from the outside, the past data is discarded (erased).

  On the other hand, displays 31 to 38 are installed in the dashboard, the left and right side pillars, the left and right doors, and the rear seat, respectively, in the vehicle. As the displays 31 to 38, an organic EL (Electro Luminescence) display using a flexible substrate can be suitably employed. Each of these displays 31 to 38 can display a separate image. An image obtained by the outside camera 11 is displayed on these displays 31 to 38. For example, on the right pillar display 32, a scene blocked by the side pillar on the driver's seat side is displayed. If the rear of the vehicle cannot be seen completely behind the backrest of the rear seat, a screen-like display may be hung from the ceiling and displayed on the display.

  An in-vehicle camera 13 is installed in the vehicle at a position where the driver can be imaged, for example, at a position adjacent to the rearview mirror. An in-vehicle camera controller 17 that sharpens an image obtained by the in-vehicle camera 13 or causes the in-vehicle camera 17 to image the inside of the vehicle in response to an external command is provided as a part of the vehicle display device 100. The in-vehicle camera controller 17 is input with detection data of a sensor 19 such as an illuminance sensor for detecting the illuminance in the vehicle and a temperature sensor for detecting the temperature in the vehicle, and controls for creating an in-vehicle environment in which an image in the vehicle can be clearly captured. Also do it.

  The seat 43 of the driver seat is provided with a seating sensor 25 that detects whether the driver is seated. Such a seating sensor 25 is composed of a known load sensor or limit switch. When the driver is seated in the driver's seat 43, the driver's seat is detected by the seating sensor 25 installed in the driver's seat 43. The detection result is sent to the viewpoint recognizer 21, and the driver's viewpoint position is measured in response to this detection result. The viewpoint position of the driver is specified on the three-dimensional coordinates in which the origin is determined at an appropriate position. As shown in FIG. 2, the three-dimensional coordinates are defined with the vehicle width direction as the X axis, the vehicle length direction as the Y axis, and the vehicle height direction as the Z axis. The in-vehicle camera 13, the in-vehicle camera controller 17, and the viewpoint recognizer 21 measure the position of the driver's viewpoint in the three-dimensional coordinates.

  A method for measuring the viewpoint position of the driver will be described in detail. FIG. 3 is a flowchart of control performed by the viewpoint recognizer 21. The viewpoint recognizer 21 checks whether the driver is seated in the driver seat by detecting the output of the seating sensor 25 (S1). When the driver is seated in the driver's seat, a command for operating the in-vehicle camera 13 is output to the in-vehicle camera controller 17 (S11). Taking this command output as an opportunity, the viewpoint recognizer 21 and the in-vehicle camera controller 17 shift to a mode for imaging the driver (S12).

  When the driver imaging mode is entered, it is checked whether the driver's operation is stable (S13). For example, if the seating sensor 25 is composed of a load sensor and the driver sees the pressure fluctuation exerted on the seat 34, it can be determined to some extent whether or not the driver is stationary. Further, it may be determined that the driver is stationary by wearing the seat belt. As described above, it is possible to determine whether or not the driver's operation is stable by setting a reference in advance.

  When the driver's motion is not stable for the set time or longer, the viewpoint recognizer 21 requests the dialog controller 27 (agent) to output an instruction for stabilizing the motion to the driver (S17). In response to this request, the dialogue controller 27 asks the driver by voice guidance or the like and prompts the driver to remain stationary for a while. For example, by performing voice guidance such as “Please stop for a while because the viewpoint position is measured”, the viewpoint position of the driver can be measured with high accuracy.

  In addition, the display that the driver wants to see (the display in a position where the view is not obstructed) is the dashboard display 31, left and right pillar displays 32 and 33, front door displays 34 and 36, rear door display 35, 37 and an input for selecting the rear seat display 38 may be prompted by voice guidance from the dialogue controller 27. By receiving an input from the driver, the image data processing device 23 recognizes and distinguishes between a display to be activated (displaying an image of the blind spot of the driver) and a display to be invalidated (not displaying an image). The image obtained by the outside camera 11 is processed and displayed only on the converted display.

  In this way, when the driver's operation is stabilized, the driver is imaged from the front by the in-vehicle camera 13, and a driver image as shown in FIGS. 4 (a) and 4 (b) is obtained. The viewpoint position E of the driver to be obtained is, for example, the center of both eyes. The coordinates of the viewpoint position E can be specified as follows. First, the driver image shown in FIG. 4 is analyzed. In the analysis of the driver image, a plurality of feature points (both eyes, joints, etc.) of the driver are obtained. On the other hand, the viewpoint recognizer 21 stores a plurality of measurement reference points as representing the shape data of the seat 43 of the driver's seat shown in FIG. Then, the coordinates (x, y, z) of the viewpoint position E are calculated and recorded from the measurement reference points and the driver feature points obtained by analyzing the driver image (S15, S16).

  The calculation procedure of the coordinates (x, y, z) of the viewpoint position E will be described in more detail. As shown in FIG. 4A, since the coordinates of the measurement reference point A (headrest corner) and the measurement reference point B (sheet corner) are known in advance, the difference between these measurement reference points and feature points can be obtained. The coordinates (x, y, z) of the viewpoint position E can be obtained. First, the actual coordinates of the measurement reference point A are (x1, y1), and the actual coordinates of the measurement reference point B are (x2, y2). If the number of pixels in the height direction of the rectangle formed between the measurement reference point A and the measurement reference point B is Ny, the number of pixels in the horizontal direction is Nx, and the pixel at the measurement reference point A is (1, 1). The pixel at the measurement reference point B is (Nx, Ny).

On the other hand, the position El of the left eye is measured from the captured image of the in-vehicle camera 13 at the iris position of the eye, and the coordinates of the El are counted to the right and down from the pixel A point (Elx, Ely) (where Elx and Ely are (Integer). The same applies to the right eye (Erx, Ery). Then, the viewpoint position E becomes ((Elx + Erx) / 2, (Ely + Ery) / 2). When this is normalized by the number of pixels Nx and Ny, ((Elx + Erx) / (2Nx), (Ely + Ery) / (2Ny)) is obtained. Furthermore, the coordinates (Ex, Ey) of the viewpoint position E are obtained by converting this into real coordinates.
Ex = x1 + (x2-x1) (Elx + Erx) / (2Nx)
Ey = y1 + (y2-y1) (Ely + Ery) / (2Ny)

  Alternatively, the following procedure can also be adopted. First, the driver image in FIG. 4 is analyzed (pattern recognition or the like), and the position of the driver's viewpoint position E in the XZ plane is specified. On the other hand, a displacement sensor (not shown) for measuring the amount of movement in the front-rear direction and the angle of the backrest is attached to the seat 43 of the driver's seat. A seat posture such as a seat movement amount and a backrest angle is detected by the displacement sensor and input to the viewpoint recognizer 21. From the detection result of the displacement sensor, the position / posture of the seat 43 can be known, and the position in the front-rear direction (Y-axis direction) where the driver's head is located can be identified almost accurately. That is, the Y coordinate of the viewpoint position E can be specified. In this way, the three-dimensional coordinates of the viewpoint position E are known. The obtained coordinates of the viewpoint position E are used as data for converting an image obtained by the vehicle exterior camera 11 into an image viewed from the viewpoint of the driver (for viewpoint conversion).

  In addition, when measuring the driver's viewpoint as described above, the dialog controller 27 (agent) is used to ask the driver to "Look at the left pillar" or "Look at the right pillar" It is also possible to measure the plurality of viewpoint positions E according to the driver's line-of-sight direction by directing the directions of the displays 31 to 38 individually. When the driver looks at the right pillar direction, when looking at the left pillar direction, when looking at the rear door orientation, the viewpoint position E is considered to change depending on the direction in which the eyes are facing. Therefore, when coordinate data of the viewpoint position E is stored in a form corresponding to each of the displays 31 to 38 and an image obtained from the outside camera 11 is cut out, the viewpoint corresponding to each of the displays 31 to 38 is stored. The coordinate data of the position E may be used. In this way, a blind spot image with little deviation can be displayed on all the displays 31 to 38.

  For example, the line of sight can be measured in more detail by measuring the iris positions of the left and right eyes. When the driver swings his / her head to the left or right, the eye position deviates from the left / right symmetry as compared to when the driver faces the front. The rotation angle of the neck (face) is calculated from the amount of shift of the eye position when both eyes are facing straight forward and when the neck is bent. Next, the positions of the left and right black eyes are measured with a camera installed almost in front of the face, and the direction in which the eyes are looking with the neck bent is measured. Normally, if the black eye is in the center, the line-of-sight direction is in front, and if the black eye is shifted to the right, the line of sight is in the right direction, and vice versa. Since the line-of-sight position varies from person to person, characters such as markers and agents are displayed on a plurality of displays (pillar displays 32 and 33 and door displays 34 and 36) that are viewed with the neck bent. May calibrate the line of sight when bending the neck and gazing at those characters, and reflect the calibration results in the processing and generation of images displayed on each display.

  It is also preferable to record as an image how the driver views the left and right pillar displays 32 and 33. If it is predicted from this image data that the driver will turn to the left, right, or rear from the front, the image obtained by the outside camera 11 is selectively displayed only on the display in the direction that the driver faces according to the timing. Can be controlled.

  When the viewpoint position E of the driver is measured as described above, among the images obtained from the vehicle exterior camera 11 from the coordinate data of the viewpoint position E and the installation position data and shape data of the displays 31 to 38, the displays 31 to 31 are displayed. The cutout range of the portion to be displayed at 38 is determined. Hereinafter, detailed control of the image data processing device 23 will be described with reference to the flowchart of FIG.

  The image data processing device 23 determines whether or not the image data of the outside camera 11 is sent from the outside camera controller 15 (S1). When the image data is sent, the image data is stored (S2). Next, it is checked whether data indicating the driver's viewpoint position (viewpoint position data) is stored in the viewpoint recognizer 21 (S3). As described above, the viewpoint recognizer 21 stores the coordinate data of the viewpoint position E. The image data processing device 23 acquires the coordinate data of the viewpoint position E from the viewpoint recognizer 21 and stores it (S4). Alternatively, the coordinate data of the viewpoint position E may be passed from the viewpoint recognizer 21 to the image data processing device 23 when the viewpoint position E is measured.

  Note that, as described above, the viewpoint position E is not limited to a single point, and is individually associated with the displays 31 to 38, such as the right pillar direction, the left pillar direction, the rear left direction, the rear right direction, and so on. It can be measured. Therefore, in this case, the image data processing device 23 stores the correspondence between the displays 31 to 38 and the viewpoint position E.

  Next, the attribute data of the displays 31 to 38 is read out. The attribute data here is coordinate data for specifying the installation positions (installation ranges) of the displays 31 to 38 and is set in advance in the image data processing device 23. For example, if the display is a quadrangle matched to the side pillar, the three-dimensional coordinates of the four corners are set as attribute data. The displays 31 to 38 also have attribute data such as ID, display size, and display shape. The ID includes an ID unique to each display and a local ID linked to the vehicle. By having a unique ID as attribute data, different images can be displayed on the plurality of displays 31 to 38. When the display size can be selected, a plurality of display sizes may be provided as attribute data. The position and shape of the display can be represented by a plurality of points on the same three-dimensional coordinates as the driver's viewpoint position E.

  In S5 of the flowchart of FIG. 5, after the display attribute data is read, the displays 31 to 38 attached to the shielding such as a side pillar are displayed from the display attribute data and the coordinate data of the driver's viewpoint position E. When the driver sees, it is calculated which areas outside the vehicle are hidden by the shielding. A polygonal area (for example, a quadrangular area) representing the hidden portion is recorded. In this way, the area that becomes the blind spot of the driver is obtained.

  On the other hand, the conversion necessary for displaying the image obtained by the outside camera 11 in accordance with the shape of the displays 31 to 38 is performed (S6). As shown in FIG. 7, the image formed by the fisheye lens is reduced in the left and right direction. The actual three-dimensional coordinates are converted into the X and Y coordinates of the fisheye lens by the formula in FIG. Here, X and Y take a range from -1 to 1. If the distance z to the object is known, x and y can be calculated from X and Y.

  After obtaining the image without the distortion by performing the coordinate conversion of the image data from the vehicle camera 11 as described above, the image data is cut out in such a manner that only the area where the blind spot is obtained previously is left. The transmitted image (pixel data) is sent to the display control device 29 (S6, S7). As a result, the image data received by the display control device 29 is set in the graphic memory, and an image to be displayed on the display 31 is generated.

  The coordinate-converted image data is sent to the display control device 29 together with the coordinate data indicating the region that is the driver's blind spot, and the display control device 29 performs a cropping operation for regions other than the blind spot, and the cropped image (pixel data). ) May be set in the graphic memory of the displays 31-38. The display control device 29 adjusts the data set timing according to the drawing performance of the displays 31 to 38. When a plurality of displays 31 to 38 for reducing the driver's blind spot are provided as in this embodiment, such a process is performed for each display.

  In addition, as a method of cutting out an image directly from a fisheye lens image, a pixel corresponding to the driver's viewpoint position E may be sent directly to the display. According to this method, display speed can be increased. The corresponding pixel according to the driver's viewpoint changes according to the height of the driver's eyes.

  FIG. 6 shows an operation state of the vehicle display device 100 and an image of the driver's seat corresponding to the state. As shown in the upper diagram, it is assumed that the object 1 crosses the front and the object 2 exists on the right side of the vehicle when the vehicle 41 is stopped or slowed upward. At this time, the image of the driver's seat is as shown in the figure below. The right pillar display 32 and the dashboard display 31 display the object 1 viewed from the driver's viewpoint and become a transmission image display image. The object 2 also appears to be transmitted through the image displayed on the door display 34.

  In this way, if the dashboard display 31 can be used to look over the blind spot at the front of the vehicle, it is effective for safety confirmation. In particular, in the case of a vehicle having no bonnet protrusion or a small vehicle such as a truck, bus, wagon car, etc., it is possible to widen the field of view without causing the driver to feel uncomfortable compared to a passenger car having a large bonnet protrusion. .

  Next, application examples of the present invention will be described. As shown in FIG. 8, a display 39 is provided so as to separate the driver's seat and the rear seat, and an image of the rear viewpoint is displayed on one surface side of the display 39 (the front seat side), and the other surface Entertainment information such as movies is displayed on the side (rear seat side). In this case, if the entertainment information is displayed on the displays 35 and 37 below the doors of the rear seats, a large screen can be displayed with a sense of presence. In addition, as described above, the driver can provide safe driving assistance because the surroundings can be seen. If the display 39 having the role of a large screen can be stored on the ceiling, it is convenient for communication between the passenger in the rear seat and the driver.

The block diagram which shows the structure of the display apparatus for vehicles of this invention. The vehicle schematic diagram which shows the example of installation of the display apparatus for vehicles of FIG. The flowchart of control of a viewpoint recognizer. The image figure of the driver imaged with the camera in a car. 5 is a control flowchart of the image data processing unit. An image of the driver's seat. The image image figure of a fisheye lens. The vehicle schematic diagram which shows another example of installation of the display apparatus for vehicles of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Out-of-vehicle camera 13 In-vehicle camera 21 Viewpoint recognition device 23 Image data processing device 27 Dialogue controller 29 Display control devices 31-39 Display 41, 51 Vehicle 43 Seat 100 Display device for vehicles

Claims (6)

Imaging means for imaging the periphery of the vehicle;
Viewpoint recognition means for recognizing the spatial position of the driver's viewpoint seated in the driver's seat;
When an obstacle such as a pillar or a door that blocks the driver's field of view is transparent, an image of a scenery outside the vehicle that should be seen by the driver's eyes, and the peripheral image data obtained by the imaging unit are the viewpoint. Image generating means for generating by processing according to the recognition result of the recognition means;
Display means for displaying an image generated by the image generation means at a position of the shield;
A vehicle display device comprising:   The viewpoint recognizing means calculates a position of the driver's viewpoint in three-dimensional coordinates from an in-vehicle camera provided in a vehicle interior for imaging the driver from the front and image data acquired by the in-vehicle camera. The vehicle display device according to claim 1, comprising a recognizer.   The viewpoint recognizing means acquires an instruction for stabilizing the viewpoint from the viewpoint recognizer to the driver when the in-vehicle camera is operated to image the driver, and the driving is performed based on identification information such as voice. The vehicle display device according to claim 2, further comprising a dialogue controller that provides guidance to the person to stabilize the viewpoint.   There are a plurality of shielding objects, and the display means is a display individually attached to the shielding objects, and the viewpoint recognition means operates or moves when the driver moves his / her line of sight to each of the displays. The vehicle display device according to any one of claims 1 to 3, wherein the vehicle display device is configured to perform a process of collecting a later posture as image data when the line of sight is moved, and is provided with a storage unit that stores the image data. The display means is a display attached to the shield,
The image generation means includes a storage unit that preliminarily stores installation position data for specifying the installation position of the display on three-dimensional coordinates and shape data for specifying the shape of the display, and the viewpoint recognition means Only the image that is obstructed by the obstruction when viewed from the driver is extracted from the viewpoint position of the driver obtained by the above, the installation position data read from the storage unit, and the shape data. The vehicle display device according to any one of claims 1 to 3, wherein the vehicle display device is configured to cut out the peripheral image data obtained by the imaging means.   The imaging means is a digital outside camera provided with a fisheye lens, and the image generating means performs a process of transforming an image obtained from the outside camera by coordinate transformation, and a range in which a field of view is blocked by the shielding object. The vehicle display device according to claim 5, wherein the pixel data of the blocked range is extracted from an image obtained from the outside camera and set in the graphic memory for display.

Images