JP2004104478A - Parking assist device and parking assist method - Google Patents

Abstract

A parking assist device and a parking assist method that allow a driver to easily and accurately recognize a distance to an obstacle and a direction in which the obstacle is located are provided.
A parking assist device (1A) is installed in a host vehicle and captures an image of the surroundings of the host vehicle. An image recognition unit 7 that calculates a distance and a direction in which the obstacle is located; an image creation unit 9 that creates an overhead view representing a positional relationship between the host vehicle and the obstacle based on a calculation result of the image recognition unit 7; A display device 15 installed on the vehicle to display the bird's-eye view created by the image creating unit 9.
[Selection diagram] Fig. 1

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a parking assist device and a parking assist method that assist driving during parking.
[0002]
[Prior art]
Conventionally, as a parking assist device of this type, an ultrasonic sensor is installed at a corner of a vehicle, and a driver is notified of an approaching state of an obstacle detected by the ultrasonic sensor using a buzzer or the like. A so-called corner sensor system that prevents collision with obstacles, or a camera installed at the rear of the vehicle, and an image captured by the camera behind the vehicle is displayed on a display device in the vehicle cabin so that the driver can visually recognize the obstacle. A so-called rear-view camera system for preventing collision with an object is known.
[0003]
[Problems to be solved by the invention]
The corner sensor system only informs the driver of the approaching state of the obstacle (distance to the obstacle), but cannot make the driver accurately recognize the direction in which the obstacle is located.
[0004]
On the other hand, the above rear-view camera system allows the driver to accurately recognize the direction in which the obstacle is located, but displays an image captured from a substantially horizontal viewpoint on the display device for the distance to the obstacle. Since the displayed image is displayed, accurate recognition of the distance to the obstacle from the displayed image requires the experience and feeling of the driver, and not everyone can easily and accurately recognize the distance. This is the same in the forward parking where the driver can directly see.
[0005]
Therefore, an object of the present invention is to provide a parking assist device and a parking assist method that enable a driver to easily and accurately recognize the distance to an obstacle and the direction in which the obstacle is located.
[0006]
[Means for Solving the Problems]
In order to solve the above problem, a parking assist device according to claim 1 is provided on an own vehicle, an imaging unit configured to image the periphery of the own vehicle, and a plurality of images captured by the imaging unit. An image recognition unit that calculates a distance from the own vehicle to the obstacle reflected in the captured image and a direction in which the obstacle is located, based on a calculation result of the image recognition unit, An image creating means for creating a bird's-eye view showing a positional relationship, and a display means installed in the host vehicle and displaying the bird's-eye view created by the image creating means.
[0007]
According to a second aspect of the present invention, in the parking assist apparatus, the image creating unit displays the own vehicle and the obstacle on the bird's-eye view using pictorial image images each of which is symbolized.
[0008]
The parking assist apparatus according to claim 3, wherein the predicted course calculated by the predicted course calculated by the predicted course calculated by the predicted course calculated by the predicted course calculated by the predicted course calculated by the predicted course calculated based on the steering angle of the steering wheel of the vehicle. Image synthesizing means for superimposing and displaying on the overhead view displayed on the display means.
[0009]
The parking assist device according to a fourth aspect further includes an image synthesizing unit that displays the captured image of the imaging unit on the display unit together with the overhead view.
[0010]
The parking assist device according to claim 5, wherein the imaging unit includes a plurality of cameras installed at different positions from each other and oriented in substantially the same direction, and the image recognition unit performs a relative operation between the plurality of cameras. From the information about the exact position and the information about the position of the same obstacle on each of the captured images of the plurality of cameras, based on the principle of triangulation, the distance to the obstacle and the This is to calculate the direction of the position.
[0011]
The parking assist device according to claim 6, further comprising a displacement calculating unit that calculates a displacement of the imaging unit accompanying the movement of the host vehicle at regular time intervals, wherein the imaging unit includes a single camera, The image recognizing means includes information on the position of the same obstacle in each of the captured images captured by the single camera at the fixed time intervals on each of the captured images, and the displacement calculating means sets the information on each of the fixed time intervals. From the calculated displacement of the camera, the distance to the obstacle and the direction in which the obstacle is located are calculated based on the principle of triangulation.
[0012]
In the parking assist method according to the seventh aspect, the periphery of the host vehicle is imaged by an imaging unit provided in the host vehicle, and the captured image is reflected from the host vehicle based on a plurality of images captured by the imaging unit. The distance to the obstacle and the direction in which the obstacle is located are calculated, and a bird's-eye view showing the positional relationship between the host vehicle and the obstacle is created based on the calculation result. Is displayed on the display means installed in the device.
[0013]
The parking assist method according to claim 8, wherein the own vehicle and the obstacle are respectively displayed on the bird's-eye view using pictorial image images that are symbolized.
[0014]
The parking assist method according to claim 9, wherein an expected course of the own vehicle is calculated based on a turning angle of a steering wheel of the own vehicle, and the estimated course is displayed on the overhead view displayed on the display means. Is what you do.
[0015]
According to a tenth aspect of the present invention, in the parking assist method, the captured image of the imaging unit is displayed on the display unit together with the bird's-eye view.
[0016]
The parking assist method according to claim 11, wherein the imaging unit is configured by a plurality of cameras installed at different positions from each other and directed substantially in the same direction, and information on a relative position between the plurality of cameras is provided. Based on triangulation principles, calculate the distance to the obstacle and the direction in which the obstacle is located, based on information about the position of the same obstacle in each of the captured images of the plurality of cameras on the captured image. Is what you do.
[0017]
The parking assist method according to claim 12, wherein the displacement of the imaging unit accompanying the movement of the host vehicle is calculated at regular time intervals, the imaging unit is configured by a single camera, and the camera is used at every constant time period. From the information on the position of the same obstacle on each of the captured images captured in each of the captured images and the displacement of the camera calculated at each fixed time, based on the principle of triangulation, the And the direction in which the obstacle is located.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
<First embodiment>
FIG. 1 is a schematic diagram of a configuration of a parking assist device according to a first embodiment of the present invention, FIG. 2 is an example of an installation position of each camera in FIG. 1, and FIGS. FIG. 5 is an example of a captured image, and FIG. 5 is an example of a display image of the display device of FIG.
[0019]
As shown in FIGS. 1 and 2, a parking assist device 1A according to this embodiment includes first and second cameras (plural cameras) 5L and 5R that are installed in the own vehicle 3 and image the surroundings of the own vehicle. An image recognition unit 7 that calculates the distance from the vehicle 3 to the obstacle reflected in the captured image and the direction in which the obstacle is located by image recognition based on the captured images of the first and second cameras 5L and 5R; An image creating unit 9 for creating an overhead view representing a relative positional relationship between the own vehicle 3 and the obstacle based on a calculation result of the image recognizing unit 7; and an own vehicle based on a steering angle of the steering wheel of the own vehicle 3. 3, an expected route calculating unit 11 for calculating an expected route, an image synthesizing unit 13 for synthesizing (editing) the processing results of the constituent elements 9 and 11, and an image synthesizing unit 13 installed in the cabin of the host vehicle 3. And a display device 15 for displaying the displayed image. .
[0020]
As shown in FIG. 2, the first and second cameras 5L and 5R are arranged at different positions, for example, at the rear of the host vehicle 3, and both have an imaging direction (imaging optical axis) of, for example, substantially behind the vehicle. They are installed facing the same direction. More specifically, the cameras 5L and 5R are, for example, both positioned at the same height and at the same front-rear position and are arranged side by side at a certain interval from each other. 3 is set directly behind. Here, the camera 5L (5R is set to the left side (right side) facing the rear of the vehicle. In addition, 5La and 5Ra in FIG. 2 indicate, for example, the horizontal angle of view of the cameras 5L and 5R, respectively.
[0021]
The image recognition unit 7 uses, for example, a so-called stereo method (here, a so-called binocular stereo method using two cameras) to determine the distance to the obstacle reflected in the captured image and the direction in which the obstacle is located. calculate. The stereo method uses a plurality of captured images (here, a pair of left and right images captured by a pair of left and right cameras 5L and 5R) obtained by capturing the same object from different viewpoints, and The positions on the captured images are detected (that is, the same target portions are associated (so-called matching) between the captured images), and the parallax (binocular parallax) obtained from the information on the positions is used for triangulation. Based on the principle, the position in the three-dimensional space of the surface of the object (the surface shown in the camera) (that is, the distance to the object point, the direction in which the object is located, the height of the object, etc.) Information).
[0022]
That is, the image recognizing unit 7 outputs, for example, each image block (for example, 16 pixels) in a right captured image (this is used as a reference image) for a pair of left and right captured images captured simultaneously by the pair of left and right cameras 5L and 5R. Identify an image block (also a 16 × 16 pixel partial image area) in the left captured image (this is a comparison image) in which the same object as the target object reflected in the × 16 pixel partial image area is projected, and The same target portion is associated (matched) between the captured images.
[0023]
More specifically, for example, as shown in FIG. 4, the image recognition unit 7 takes each image block 17Ra in a grid pattern (that is, does not overlap each other) on the right captured image 17R side as shown in FIG. On the left captured image 17L side, the respective image blocks 17La are superimposed on each other in the vertical direction or the horizontal direction, for example, by shifting one pixel at a time, and the left image capturing is performed on each image block 17Ra of the right captured image 17R. The sum of the absolute values of the luminance differences between the pixels arranged at the same position in the image block is calculated with respect to all the image blocks 17La in the image 17L, and the sum is minimized (that is, the best match is obtained). ) The image block 17La on the left captured image 17L side is replaced with the image block 17 on the left captured image 17L side showing the same target object as the target object shown in the image block 17Ra of the right captured image 17R. a and determined (specified), and the left and right captured images 17L, the matching between 17R. In FIGS. 3 and 4, each of the image blocks 17La and 17Ra is shown larger for convenience.
[0024]
In this embodiment, since the cameras 5L and 5R are arranged side by side, the position of the same object shown in each of the left and right captured images 17L and 17R is different between the left and right captured images 17L and 17R. It shifts only in the direction. In consideration of this, in the image recognition unit 7, the difference absolute value of the luminance difference of each pixel between each image block 17Ra of the right captured image 17R and all the image blocks 17La of the left captured image 17L is determined. Without calculating the sum, the sum of the absolute values of the brightness differences is calculated only between the image block 17Ra and each of the image blocks 17La in the left captured image 17L located in the vertical direction on the same image. Even if the image block 17La on the left captured image 17L having the smallest sum is determined to be the image block 17La on the left captured image 17L on which the same object as the object shown in the image block 17Ra is shown. Good. By doing so, the processing time for matching between the left and right captured images 17L and 17R can be reduced.
[0025]
Then, the image recognition unit 7 determines, for each image block 17Ra of the right captured image 17R, the position of the image block 17Ra on the captured image 17R and the image block on the left captured image 17L side corresponding to the image block 17Ra. From the position of the 17La on the captured image 17L and the distance between the cameras 5L and 5R, the position of the surface of the object shown in the image block 17Ra in the three-dimensional space is calculated based on the principle of triangulation. By performing this calculation for all the image blocks 17Ra of the right captured image 17R, the distances to the surfaces of all the objects shown in the right captured image 17R, the directions in which the objects are located, and the height of the object Is found. That is, the distance to the surface of all obstacles (here, the stopped vehicle 19 and the wall 21), the direction in which the obstacles 19 and 21 are located, and the height of the obstacles 19 and 21 are obtained. The positions of the image blocks 17Ra and 17La on the captured images 17R and 17L are defined by, for example, the positions of the center positions of the image blocks 17Ra and 17La on the captured images 17R and 17L.
[0026]
In addition, the image recognition unit 7 regards only an object having a height equal to or higher than a predetermined height (a height that hinders traffic of the own vehicle) as an obstacle, and regards an object having a height lower than the predetermined height as an obstacle. Not considered. Thus, only the target object such as the stopped vehicle 19, the pillar (not shown), the wall 21, and the like are appropriately regarded as obstacles.
[0027]
The image creating unit 9 creates, for example, a plan view as the overhead view. More specifically, as shown in FIG. 5, the image creation unit 9 represents the own vehicle 3 and the obstacles 19 and 21 as pictorial images 3a, 19a and 21a, respectively, which are symbolized, and calculates the calculation results of the image recognition unit 7. Based on the relative positional relationship between the host vehicle 3 and the obstacles 19 and 21 (ie, the distance relationship between the host vehicle 3 and the surfaces of the obstacles 19 and 21 and the directional relationship between the host vehicle 3 and the obstacles 19 and 21). In consideration of the above, a plan view 23 including the host vehicle 3 and the obstacles 19 and 21 is created. Here, the own vehicle 3 is displayed by a pictorial image 3a in the shape of an automobile, and the obstacles 19 and 21 are displayed by pictorial images 19a and 21a in which their occupied areas are modeled.
[0028]
The image synthesizing unit 13 superimposes and synthesizes the trajectory 25 of the predicted course of the vehicle 3 calculated by the predicted course calculation unit 11 on the plan view 23 created by the image creation unit 9 as shown in FIG. The composite image is displayed on the display device 15. In addition, in response to an instruction from a predetermined operation switch, the image synthesizing unit 13 displays, for example, a captured image (actual image: reference image) 17R of the right camera 5R together with the plan view 23 as shown in FIG. 15 are displayed simultaneously. As a method of simultaneous display, for example, the plan view 23 may be displayed so as to overlap a partial area on the captured image 17R of the right camera 5R, or the display screen may be divided into two parts, for example, left and right. The plan view 23 and the captured image 17R of the right camera 5R may be displayed side by side.
[0029]
The plan view 23 is sequentially updated and displayed on the display device 15 as time passes or as the vehicle 3 moves. Also, the trajectory 25 of the predicted course of the host vehicle 3 superimposed and synthesized on the plan view 23 is updated according to the change in the steering angle of the steering wheel of the host vehicle 3.
[0030]
According to the parking assist device 1A configured as described above, an image is captured from the host vehicle 3 based on the plurality of captured images 17L and 17R around the host vehicle captured by the cameras 5L and 5R installed in the host vehicle 3. The distances to the obstacles (here, 19 and 21) reflected in the images 17L and 17R and the direction in which the obstacles are located are calculated, and the positions of the own vehicle 3 and the obstacles 19 and 21 are calculated based on the calculation results. A bird's-eye view 23 representing the relationship is created and the bird's-eye view 23 is displayed on the display device 15 installed in the host vehicle 3, so that the positional relationship between the host vehicle 3 and the obstacles 19, 21 is given to the driver. Can be recognized easily and accurately.
[0031]
Further, since the own vehicle 3 and the obstacles 19 and 21 are displayed on the bird's-eye view 23 by pictorial image images 3a, 19a and 21a symbolized respectively, the own vehicle 3 and the obstacles 19 and 21 are displayed to the driver. Can be visually recognized easily and accurately.
[0032]
In addition, the expected course of the own vehicle 3 is calculated based on the turning angle of the steering wheel of the own vehicle 3, and the trajectory 25 of the expected course is displayed on the bird's-eye view 23 displayed on the display device 15 so as to be accurate. The own vehicle 3 can be guided.
[0033]
Further, for example, the captured image (real image) 17R of the camera 5R is displayed on the display device 15 together with the bird's-eye view 23, so that the driver can compare the real image 17R with the bird's-eye view 23 and The positional relationship between the vehicle 19 and the obstacles 19 and 21 can be recognized more easily and accurately.
[0034]
Further, the image pickup means is composed of two cameras 5L and 5R installed at different positions from each other and directed substantially in the same direction, and information on the relative position between the two cameras 5L and 5R and the two cameras 5L and 5R. The distance to the obstacles 19 and 21 based on the principle of triangulation is obtained from information on the positions of the same obstacles 19 and 21 in the captured images 17L and 17R of the 5R on the captured images 17L and 17R. Since the direction in which the obstacles 19 and 21 are located is calculated, the distance to the obstacles 19 and 21 and the direction in which the obstacles 19 and 21 are located can be calculated quickly and accurately.
[0035]
In this embodiment, the case where two cameras are used has been described, but three or more cameras may be used. In that case, one camera is used as the reference image camera, and the remaining cameras are used as comparative image cameras. Then, matching is performed between each image block in the reference image captured by the reference image camera and all image blocks in each comparative image captured by all remaining comparative image cameras, and the best matching is performed. Is determined as an image block on the comparison image side in which the same object as the object shown in the image block on the reference image side is shown. Then, the position of the image block on the reference image on the reference image, the position of the image block on the comparison image corresponding to the image block on the reference image on the comparison image, the reference image camera and the reference Using the relative positional relationship with a comparative image camera that has captured a comparative image including an image block on the comparative image corresponding to the image block on the image side, triangulation as in the case of using two cameras The position in the three-dimensional space of the object shown in the image block on the reference image side is calculated based on the principle of the above.
[0036]
As described above, when three or more cameras are used (when two or more comparative image cameras are used), the comparative image captured by one comparative image camera does not include the target object shown in the reference image. In some cases, the comparison image captured by another comparison image camera is reflected in the comparison image. By using the other comparison image, matching between the reference image and the comparison image can be appropriately performed, and the accuracy can be improved. Good matching can be performed.
[0037]
In this embodiment, the case has been described where the imaging directions of the cameras 5L and 5R are oriented substantially in the horizontal direction. However, when the cameras 5L and 5R are installed with the imaging directions of the cameras 5L and 5R inclined, for example, downward, The triangulation may be calculated in consideration of the depression angle.
[0038]
It goes without saying that the parking assist method is also included in the category of this embodiment.
[0039]
<Second embodiment>
FIG. 6 is a schematic diagram of a configuration of a parking assist device according to a second embodiment of the present invention, and FIG. 7 is an example of an installation position of the camera in FIG. 6 and a change in an imaging direction of the camera in FIG. It is a figure showing an example.
[0040]
The parking assist device 1B according to this embodiment uses a so-called monocular stereo method (a stereo method using a captured image captured by a single camera) to locate a distance to an obstacle and the position of the obstacle. As shown in FIG. 6, only one camera (imaging means) 5 is used as a camera for imaging the surroundings of the host vehicle as compared with the parking assist device 1A according to the first embodiment, as shown in FIG. It has the same configuration except that it is provided and that it further comprises a camera displacement calculator 27. Hereinafter, the same components as those of the parking assist device 1A are denoted by the same reference numerals, and description thereof will be omitted. Only different components will be described.
[0041]
As shown in FIG. 7, the camera 5 is located, for example, at the rear of the vehicle 3, and is installed with its imaging direction facing, for example, the rear of the vehicle (for example, just behind a substantially horizontal direction). Note that 5a in FIG. 7 indicates a horizontal angle of view of the camera 5, for example.
[0042]
The camera displacement calculation unit 27 calculates the displacement of the camera 5 (the moving distance and moving direction of the camera 5, in other words, the camera) based on the vehicle speed and the turning angle (steering angle of the steering wheel) of the own vehicle 3, for example. 5) is calculated at predetermined time intervals with respect to the object (obstacle).
[0043]
The image recognizing unit 7B according to the present embodiment, in comparison with the image recognizing unit 7 according to the first embodiment, sequentially picks up each of the images captured at regular intervals among the images captured by the camera 5. The camera displacement calculator 27 calculates a reference image, and a distance between a camera used for triangulation calculation and a point used as a comparison image with a captured image captured a fixed time after the reference image is captured. Similarly, the same matching method and triangle are used except that the displacement of the camera 5 (that is, the displacement of the camera 5 during the above-mentioned fixed time between the time of capturing the reference image and the time of capturing the comparative image) is used. By the calculation based on the principle of the survey, the distance to all obstacles shown in the reference image and the direction in which the obstacles are located are calculated.
[0044]
In this embodiment, the camera position at which the reference image is taken and the camera position at which the comparison image is taken may be shifted back and forth with respect to the same object with the movement of the vehicle 3. As a result, the size of the same object appearing in the reference image and the comparison image may be different. Therefore, when matching between the reference image and the comparison image is performed, for example, in the image recognition unit 7B, based on the above-mentioned longitudinal displacement amount of the camera 5 calculated by the camera displacement calculation unit 27, for example, the image of the comparison image Matching may be performed after the size is enlarged or reduced so that the size of the same object appearing in the reference image and the comparison image is approximately the same. In this case, the accuracy of matching can be increased, and erroneous matching can be prevented.
[0045]
According to the parking assist device 1B configured as described above, in addition to the same effect as the parking assist device 1A according to the first embodiment, the imaging means includes the single camera 5, and Triangulation is performed based on information on the position of the same obstacle on each of the captured images captured by the camera 5 at regular intervals on the captured images and the displacement of the camera 5 calculated at regular intervals. Since the distance to the obstacle and the direction in which the obstacle is located are calculated based on the principle of the above, the distance to the obstacle and the direction in which the obstacle is located are calculated relatively accurately at a relatively high speed with a low-cost configuration. it can.
[0046]
In this embodiment, a case has been described where two captured images captured at fixed time intervals are used. However, three or more captured images captured at fixed time intervals may be used. In that case, one is set as a reference image, and all the rest are set as comparative images. Then, the position in the three-dimensional space of the object shown in each image block in the reference image is calculated by a method similar to the case where three or more cameras are used in the first embodiment. As described above, when three or more captured images are used (when two or more comparative images are used), even if the target object shown in the reference image does not appear in one comparative image, it appears in other comparative images. In some cases, by using the other comparison image, matching between the reference image and the comparison image can be appropriately performed, and accurate matching can be performed.
[0047]
It goes without saying that the parking assist method is also included in the category of this embodiment.
[0048]
【The invention's effect】
According to the first and seventh aspects of the present invention, based on a plurality of captured images around the own vehicle captured by the imaging means installed in the own vehicle, obstacles reflected in the captured images from the own vehicle. The distance to the object and the direction in which the obstacle is located are calculated, and an overhead view showing the positional relationship between the own vehicle and the obstacle is created based on the calculation result, and the overhead view is installed in the own vehicle. Since the information is displayed on the display means, the driver can easily and accurately recognize the positional relationship between the host vehicle and the obstacle.
[0049]
According to the second and eighth aspects of the present invention, the own vehicle and the obstacle are displayed on the bird's-eye view by the pictorial image that is symbolized, respectively, so that the driver can easily understand the own vehicle visually. And the positional relationship between the obstacle and the obstacle can be recognized.
[0050]
According to the third and ninth aspects of the present invention, the expected course of the own vehicle is calculated based on the turning angle of the steering wheel of the own vehicle, and the estimated course is displayed on the bird's-eye view displayed on the display means. Therefore, accurate guidance of the own vehicle can be performed.
[0051]
According to the fourth and tenth aspects of the present invention, the captured image (real image) of the imaging means is displayed on the display means together with the bird's-eye view, so that the driver can compare the real image with the bird's-eye view. However, the positional relationship between the host vehicle and the obstacle can be recognized, and the positional relationship between the host vehicle and the obstacle can be recognized more easily and accurately.
[0052]
According to the fifth and eleventh aspects of the present invention, the image pickup means is composed of a plurality of cameras installed at different positions from each other and directed together in substantially the same direction, and relates to the relative position between the plurality of cameras. Based on the information and information on the position of the same obstacle in each of the captured images of each of the plurality of cameras, based on the principle of triangulation, the distance to the obstacle and the direction in which the obstacle is located Is calculated, the distance to the obstacle and the direction in which the obstacle is located can be calculated quickly and accurately.
[0053]
According to the invention described in claim 6 and claim 12, the imaging means comprises a single camera, and each of the same obstacles appearing in each of the captured images captured by the single camera at regular time intervals. Based on the principle of triangulation, the distance to the obstacle and the direction in which the obstacle is located are calculated from the information on the position on the captured image and the displacement of the camera calculated at regular intervals. It is possible to calculate the distance to the obstacle and the direction in which the obstacle is located at a relatively high speed and with high accuracy with an inexpensive configuration.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a parking assist device according to a first embodiment of the present invention.
FIG. 2 is an example of an installation position of each camera in FIG. 1;
FIG. 3 is an example of an image captured by a second camera in FIG. 1;
FIG. 4 is an example of an image captured by a first camera in FIG. 1;
FIG. 5 is an example of a display image of the display device of FIG. 1;
FIG. 6 is a schematic configuration diagram of a parking assist device according to a second embodiment of the present invention.
7 is a diagram illustrating an example of an installation position of the camera in FIG. 6 and an example of a change in an imaging direction of the camera in FIG. 6 when the own vehicle moves.
[Explanation of symbols]
1A, 1B parking assist device
3 own vehicle
5L, 5R, 5 camera
7,7B image recognition unit
9 Image creation unit
11 Expected course calculation section
13 Image synthesis section
15 Display device
17R, 17L captured image
19, 21 Obstacle
23 Plan view
25 Trajectory of expected route

Claims (12)

Imaging means installed in the own vehicle to image the surroundings of the own vehicle,
Image recognition means for calculating a distance from the host vehicle to an obstacle reflected in the captured image and a direction in which the obstacle is located, based on the plurality of captured images captured by the imaging means,
Based on the calculation result of the image recognition means, an image creating means for creating an overhead view representing the positional relationship between the vehicle and the obstacle,
Display means for displaying the bird's-eye view created by the image creation means installed in the vehicle,
A parking assist device comprising: 2. The parking assist device according to claim 1, wherein the image creating unit displays the own vehicle and the obstacle on the bird's-eye view by using pictorial images each of which is symbolized. 3. Anticipated course calculating means for calculating an expected course of the own vehicle based on the turning angle of the steering wheel of the own vehicle,
Image synthesis means for displaying the predicted course calculated by the predicted course calculation section on the overhead view displayed on the display section,
The parking assist device according to claim 1 or 2, further comprising: The parking assist device according to any one of claims 1 to 3, further comprising an image synthesizing unit configured to display the captured image of the imaging unit on the display unit together with the overhead view. The imaging means comprises a plurality of cameras installed at different positions from each other and directed substantially in the same direction,
The image recognition unit is configured to calculate the principle of triangulation from information on a relative position between the plurality of cameras and information on a position of the same obstacle on each of the captured images of each of the plurality of cameras. The parking assist device according to any one of claims 1 to 4, wherein a distance to the obstacle and a direction in which the obstacle is located are calculated based on the distance. Further comprising a displacement calculating means for calculating the displacement of the imaging means accompanying the movement of the vehicle at regular intervals,
The imaging means comprises a single camera,
The image recognizing means includes information on the position of the same obstacle in each of the captured images captured by the single camera at the fixed time intervals on each of the captured images, and the displacement calculating means sets the information on each of the fixed time intervals. The distance to the obstacle and the direction in which the obstacle is located are calculated from the calculated displacement of the camera based on the principle of triangulation. The parking assist device as described. The surroundings of the own vehicle are imaged by the imaging means installed in the own vehicle, and the distance from the own vehicle to the obstacle reflected in the captured image based on the plurality of captured images captured by the imaging means and the distance of the obstacle Calculating a position direction, creating an overhead view representing the positional relationship between the own vehicle and the obstacle based on the calculation result, and displaying the overhead view on display means installed in the own vehicle. Characteristic parking assist method. The parking assist method according to claim 7, wherein the own vehicle and the obstacle are displayed on the bird's eye view using pictorial image images each of which is symbolized. 8. The vehicle according to claim 7, wherein an expected course of the own vehicle is calculated based on a steering angle of the steering wheel of the own vehicle, and the estimated course is displayed on the overhead view displayed on the display unit. Item 9. The parking assist method according to item 8. The parking assist method according to any one of claims 7 to 9, wherein a captured image of the image capturing unit is displayed on the display unit together with the overhead view. The imaging unit is configured by a plurality of cameras installed at different positions from each other and directed substantially in the same direction,
From the information on the relative position between the plurality of cameras and the information on the position of the same obstacle in each of the captured images of the plurality of cameras on the respective captured images, the obstacle based on the principle of triangulation. The parking assist method according to any one of claims 7 to 10, wherein a distance to the obstacle and a direction in which the obstacle is located are calculated. Calculate the displacement of the imaging means accompanying the movement of the vehicle at regular intervals,
The imaging means is constituted by a single camera,
From the information regarding the position of the same obstacle on each of the captured images captured in each of the captured images captured by the camera at the fixed time and the displacement of the camera calculated at the fixed time, triangulation is performed. The parking assist method according to any one of claims 7 to 10, wherein a distance to the obstacle and a direction in which the obstacle is located are calculated based on a principle.

Images