JP3468658B2 - Perimeter recognition device for vehicles - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle peripheral visual recognition device which is mounted on a front portion or a rear portion of a vehicle, images the left and right side visual fields of the vehicle, and presents left and right side visual field images to a driver. Regarding

[0002]

Related Background Art Recently, a camera mounted on a front portion or a rear portion of a vehicle captures a field of view on the left and right sides of the vehicle,
Development of a vehicle peripheral visual recognition device that displays an image using a display provided inside the vehicle is underway. According to this type of device, for example, as shown in FIG.
When the vehicle 1 enters an intersection from an alley or the like, the road (alley)
Since the field of view is blocked by the walls 2 on both sides and cannot be seen directly by the driver, the left and right field of view in front of the vehicle can be confirmed from the image obtained by the camera provided in the front part of the vehicle. It is possible to improve safety when entering an intersection.

[0003]

By the way, the visual field (viewing angle and direction) of the left and right lateral images taken by the camera in this type of vehicle peripheral visual recognition device is usually uniquely determined. . However, when entering an intersection with poor visibility, it is not always necessary to check the left and right lateral views under the same conditions depending on the width of the intersection road and the right and left visibility at the intersection. Rather, it is desirable to obtain more information on the lateral field of view in the direction in which the degree of danger is judged to be greater, and to confirm its safety.

For example, when entering a highway with two lanes on each side from a side road, since the traveling direction of the vehicle on the highway is fixed, the left side is a degree of looking through the sidewalk along the shoulder of the highway. It is sufficient to secure the field of view in the narrow field of view, but it is desirable to secure the field of view in the wide field including the main road as well as the sidewalk on the right side.

When the left and right visual field regions are changed in this manner, it is conceivable to change the imaging visual field angles of the cameras which respectively capture the left and right visual fields.
Specifically, it is conceivable to attach a zoom lens to a camera and change the imaging viewing angle by changing the focal length. However, if the imaging viewing angle (viewing area) is changed in this manner, the scale of the image input by imaging changes accordingly. In the image, the narrower the viewing angle of the same subject, the larger the subject image and the impaired perspective, and thus the display image is likely to be erroneously recognized. Moreover, using a zoom lens not only complicates the configuration of the camera itself,
There arises a problem that the control system becomes complicated.

The present invention has been made in consideration of such a situation, and an object thereof is to adjust the left and right sides according to the situation of intersections, the situation of line-of-sight, etc. without changing the scales of the left and right visual field images. Another object of the present invention is to provide a vehicle peripheral visual recognition device having a simple and highly practical configuration capable of appropriately changing the size of the other visual field region.

[0007]

[Summary of the peripheral visual confirmation apparatus for a vehicle according to the present invention in order to achieve the above object, a predetermined field of view is attached to the front or rear of the vehicle at left and right sides of the vehicle, for example, the field of view An image pickup unit for picking up a visual field corresponding to a natural visual angle of less than 90 °, and an image of a visual field range of interest in each of the left and right visual field images picked up by the image pickup unit is a left visual field image and a right visual field image. And an image synthesizing unit that forms a single display image by arranging the left-view image and the right-view image, respectively, and a display unit that is provided inside the vehicle to display the display image. Further environment detection to detect the driving environment of the vehicle
Means and a field-of-view setting for determining the screen ratio of the left-view image and the right-view image extracted from each view image by setting the range of the view field of interest in each of the left and right views captured by the image pickup means. Means and the set screen ratio to the running ring.
With learning means to store according to the boundary and the environment detection means
Detected driving environment and information stored by the learning means
And display control means for changing the screen ratio according to the above .

Particularly, as described in claim 2, the visual field setting means is provided with means for variably setting the range of the visual field of interest in each of the left and right visual field images in response to the operation of the manually operated selection switch. According to a third aspect of the present invention, the visual field setting means includes information on the traveling environment of the vehicle detected by the environment detection means, for example, information on the approach angle of the vehicle to the intersection indicated by the navigation information, It is characterized in that means is provided for variably setting the range of the visual field of interest in the left and right side view images according to the steering wheel angle information of the vehicle and the like. At this time, it is preferable to store the screen ratio of the left visual field image and the right visual field image set as described above in association with the intersection .

[0009] That is, the peripheral visual confirmation apparatus for a vehicle according to the present invention, the in each field image in the right and left sides of the vehicle to be imaged by the imaging means (camera), and a left view image and the right view image partially When extracting and arranging these visual field images to form one display image for display on a display means (display), the range of the visual field regions to be noticed on the left and right depending on the situation of the intersection is selected, for example, by a selection switch. upon a manual operation, or set according to information of an intersection, such as indicated by the navigation information, also before the set screen ratio
Memorized according to the driving environment,
According to the above, the screen ratio between the left-view image and the right-view image is variably set.

Further, the screen ratio between the left-view image and the right-view image is stored in association with the intersection so that the setting of the view range at the same intersection can be automated.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION A vehicle peripheral visual recognition apparatus according to an embodiment of the present invention will be described below with reference to the drawings, taking a nose view apparatus as an example. Although a nose view device that captures the field of view of the left and right sides of the vehicle with a camera provided at the front of the vehicle will be described as an example, the left and right sides of the vehicle with a camera provided at the rear of the vehicle will be described. The same can be applied to a vehicle rear left and right view device that images the field of view of the vehicle.

FIG. 1 is a diagram showing a functional schematic structure of a nose view device. Reference numerals 11 and 12 are left visual fields as image pickup means which are attached to the front part of a vehicle and respectively image the visual fields on the left and right sides of the vehicle. A camera and a right-view camera. Each of these cameras 11 and 12 is incorporated, for example, on the front side of the front fender of the vehicle, in particular, at a notch on the extension line of the front bumper, and the like. The fields of view on the left and right sides of the part are respectively imaged.

The image pickup viewing angles FL and FR of the cameras 11 and 12 are set within a range of, for example, a natural visual field of less than 90 ° in the horizontal direction (a visual field which can be visually recognized by a human in a natural state). Is defined as a viewing angle corresponding to a standard lens. In other words, the viewing angles FL and FR of the cameras 11 and 12 collectively capture the left and right fields of view in the front part of the vehicle 10 with almost no geometric distortion in the field of view (horizontal direction). It is defined as the viewing angle to obtain.

On the other hand, the display 13 as an image display means is incorporated in the interior of the vehicle 10, particularly in the instrument panel or console panel in front of the driver's seat, and is composed of, for example, a liquid crystal display. The display 13 is used for displaying the left and right side view images described above and also for displaying navigation information (map information indicating the road around the vehicle and the vehicle position etc.) by the navigation system 14 described later. Is also used.

The feature of this apparatus is that in the nose view system 15 constructed by a microprocessor or the like, in particular, the nose view selection control function 1
6, a left / right visual field synthesis ratio control function 17, a left / right visual field changing function 18, and a learning unit (memory) 19 as an auxiliary function. Whether or not the nose view selection control function 16 uses the display 13 to perform image display based on the view images VL and VR of the front left and right sides of the vehicle 10 imaged and input by the cameras 11 and 12 described above. Is a function of controlling the selection, and executes the selection control based on the information of the intersection given from the navigation system 14 and the instruction information from the nose view selection switch as will be described later.

The left and right visual field synthesis ratio control function 17 also has left and right visual fields cut out from the visual images VL and VR on the front left and right sides of the vehicle 10 imaged and input by the cameras 11 and 12, respectively, as will be described later. When the images DL and DR are arranged side by side to form one display image D, it plays a role of setting the screen ratio (combining ratio) of the left visual field image DL and the right visual field image DR according to an instruction from the compositing ratio switch. The left / right visual field synthesis ratio control function 17 operates in accordance with the screen ratio set as described above to obtain the front left / right side view images VL,
Left and right visual field images DL, DR are extracted from VR with a predetermined visual field width, and these visual field images DL, DR are arranged side by side to 1
A single display image D is formed. This display image D is displayed on the display 13 under the control of the nose view selection control function 16 and presented to the driver.

Further, the left and right visual field changing function 18 is used to selectively extract the left and right visual field images DL, DR from the front left and right lateral visual field images VL, VR according to the screen ratio set as described above. It plays a role of changing and setting which region of the images VL and VR is extracted, that is, which direction of the visual field region is extracted. The change of the orientation of the visual field region is performed in accordance with, for example, a steering wheel angle signal or instruction information from the visual field selection switch.

The learning unit 19 includes the functions 16,
The operation mode (operation state) of 17, 18 is learned in accordance with, for example, the traveling environment obtained from the navigation system 14. The learning unit 19 collects, for example, the use of the nose view function at a specific intersection and the usage state (form) of the use as history information, and learns at which intersection the nose view function is used and in what form. To be done. Specifically, the screen ratio between the left visual field image DL and the right visual field image DR and the directions of the visual fields of the left visual field image DL and the right visual field image DR are stored in association with the intersection. This learning result is the same as the nose view system 15 described above.
It is used for automatic control of each of the functions 16, 17, and 18 in the above.

The characteristic functions of the vehicle peripheral visual recognition device (nose view device) constructed in this way and the operation thereof will be described in more detail according to the basic control flow shown in FIG. Specifically, it is started by determining whether or not the nose view selection switch is selected (turned on) [step S1]. If the nose view selection switch is not selected (OFF), this process is stopped until the nose view selection switch is selected (ON), or the learning by the learning unit 19 although no control flow is shown. Nose view control is executed based on the result. The nose view control based on this learning result will be described later.

When the nose view selection switch is selected (turned on), it is then determined based on the vehicle speed signal whether or not the vehicle speed at that time is a low speed of 20 km / h or less [for example]. Step S2]. This decision is
Based on the fact that the driving environment that requires a nose view is when traveling at low speeds, such as when entering an intersection with poor visibility, and it is almost unnecessary to check the left and right views of the front of the vehicle when traveling at high speeds. There is. In other words, when it is necessary to check the left and right field of view of the front portion of the vehicle, it is naturally based on the fact that the vehicle is decelerating, traveling at a low speed, or temporarily stopped. Therefore, when the vehicle 10 is traveling at high speed, the nose view process is stopped.

The nose view selection control function 16 executes the selection control for the on / off judgment of the nose view selection switch and the operation of the nose view function based on the vehicle speed judgment. Then, if the necessary condition for operating the nose view function is determined by the nose view selection control function 16, next, according to the position information of the vehicle 10 obtained from the navigation system 14, the intersection where the vehicle 10 is located is determined. It is determined whether the intersection is a specific intersection stored in the learning unit 19 [step S3]. If the intersection is already stored, it is determined whether or not the composite ratio switch for variably setting the screen ratios of the left and right visual field images DL, DR is in the operating state (ON) [step S4].

If the intersection where the vehicle 10 is about to enter is not the intersection already registered in the learning section 19,
Alternatively, if the composition ratio switch is on even at an already registered intersection, the left visual field image DL and the right visual field image DR extracted from the visual field images VL and VR respectively according to an instruction from the composite ratio switch. The area width (horizontal image width) is variably set, and the screen ratio of the left visual field image DL and the right visual field image DR in forming the display image D is set. Then, with the screen width (region width) determined by this screen ratio, the left visual field image DL and the right visual field image DR are respectively cut out from the respective visual field images VL, VR, and these left visual field image DL and right visual field image DR are cut out. Are arranged side by side to form one display image D, which is displayed on the display 13 [step S5].

When the intersection where the vehicle 10 is about to enter is an intersection already registered in the storage unit 19 and the combination ratio switch is in the OFF state, the left visual field image DL and the right visual field image are displayed. Since it is indicated that it is not necessary to change the setting of the screen ratio with respect to DR, the left visual field image DL and the right visual field image DR are selected from the visual field images VL and VR according to the screen ratio stored in the learning unit 19. Each is cut out to form one display image D and displayed on the display 13 [step S6].

At this time, for example, according to an instruction from a manually operated field-of-view selection switch, the left and right visual fields FL, F are set.
By changing and setting the directions AL and AR of the visual field of interest in R, the images of the predetermined visual field angles PL and PR determined under the screen ratio with reference to the set visual field directions AL and AR are set to the left and right visual field images DL, Field images VL and VR of the front left and right sides as DR
You may make it selectively extract from each.

When no instruction to change the direction of the field of interest is given, the direction of the field of interest set according to the directions AL and AR of the field of view set as default values or according to the steering wheel angle. AL, A
R, or direction AL, A of the visual field of interest set according to the approach angle to the intersection indicated by the navigation information
According to R, the areas of the visual field images DL and DR cut out from the visual field images VL and VR may be specified.

When the display image D consisting of the visual field images DL and DR cut out from the visual field images VL and VR is displayed on the display 13 as described above, the screen ratio used for this display is stored next. It is judged [Step S
7]. This determination is made, for example, by checking whether or not the screen ratio of the nose view regarding the intersection that has already been stored, whether or not the screen ratio has been changed, and further whether or not a memory instruction has been given by the driver. It In other words, it is determined whether or not the intersection is a frequently-passed intersection such as commuting, and it is determined whether or not the screen ratio is stored in association with the intersection according to the determination result.

Then, according to the result of this judgment, the learning unit 1
9 is activated, and the screen ratio is stored in association with the intersection at that time [step S8]. Alternatively, if the screen ratio has already been stored for that intersection, the stored data is updated with the new screen ratio that has been changed and set.
The information stored in the learning unit 19 is, for example, information that identifies an intersection and the direction of approach to the intersection, and the visual field images DL and DR displayed on the display at that time.
Information of the screen ratio and the directions AL and AR of the field of view. Further, as the information of the screen ratio, the visual field image VL,
Cutout widths WL and WR of the visual field images DL and DR during VR
And the cut-out position may be directly stored.

Now, the above-mentioned left and right visual field images VL, V
The selective extraction of the left-view image DL and the right-view image DR from within R and the combination of the extracted view images DL and DR will now be described in a little more detail. This process is performed by the cameras 11 and 12 as shown in FIG. A partial image of the area specified by the directions AL and AR of the visual field of interest and the viewing angles PL and PR determined by the screen ratio from the left and right visual field images VL and VR indicated by the input visual fields FL and FR. Is cut out as the visual field images DL and DR, respectively.

That is, the directions AL and AR of the visual field of interest are, as described above, the manual operation of the visual field selection switch, the steering wheel angle,
Alternatively, it is set according to the approach angle to the intersection indicated by the navigation information. Also, the viewing angles PL and PR are
Widths of view images DL, DR (horizontal image area widths) WL, WR
Is specified and is variably set by the manual operation of the above-described composition ratio switch. Field of view images DL, DR
Since the widths of one display image D are defined, the widths WL and WR of the whole are reduced, for example, by increasing the width WL of the left visual field image DL and narrowing the width WR of the right visual field image DR. Are determined in association with each other so that the width W of each is constant. In this case, the minimum width of the visual field images DL and DR may be defined in advance. Therefore, the widths WL and WR of the visual field images DL and DR
, Respectively, the screen ratios of the visual field images DL and DR are determined.

However, the cutout of the visual field images DL, DR from the respective visual field images VL, VR on the left and right sides based on the screen ratio is partially cut out as the visual field images DL, DR as shown in FIG. This corresponds to determining the widths WL and WR of the. In addition, the viewing angles PL, PR according to the directions AL, AR of the target visual field
Cutting out the visual field images DL and DR of just the width W
Set the position of the image area indicated by L and WR to A
This is equivalent to sliding to the left or right according to L and AR.

Therefore, the nose-view device according to this embodiment uses the directions AL, A of the visual field of interest set as described above, from the left and right visual field images VL, VR fixedly imaged and input by the cameras 11, 12. Field of view image DL according to AR,
One of the display images D is formed by selectively extracting DR according to the screen ratio and arranging the visual field images DL and DR, and the display image D is displayed on the display 13.

Therefore, according to the present device, depending on the condition of the intersection to be approached, for example, the right-side view image DR can be obtained at a wide angle, more right-side information can be obtained, and the safety confirmation can be surely performed. It will be possible. In this case, in particular, by narrowing the width of the left-side visual field image DL that does not require a large amount of information, the right-side visual field image DR is secured while securing the necessary information indicated by the left-side visual field image DL. As much information as you can get. Moreover, as described above, since the widths WL and WR of the left and right visual field images DL and DR that are partially cut out from the visual field images VL and VR are simply variably set according to the screen ratio, the left and right visual field images DL and D are set.
Even if the widths W L and W R of R are changed, the image scale hardly changes. Therefore, the widths WL and WR of the visual field images DL and DR
There is no visual discomfort even if is changed. Therefore, it is possible to accurately check the right and left side safety according to the intersection.

Even if the approach angle of the vehicle 10 to the intersection is inclined, the visual field images DL and DR are as described above.
Since the cutout position of is changed, it is possible to obtain an image of the side of the vehicle in the direction corresponding to the approach angle. Moreover, it is possible to obtain the images DL and DR in the desired visual field direction on the side of the vehicle without changing the directions of the visual fields of the cameras 11 and 12 themselves. Therefore, there is no need for a mechanism or the like for changing the directions of the fields of view of the cameras 11 and 12, and as described above, the cameras 11 and 12 can be fixedly attached to the front fenders, etc. It can be simplified.

Further, as described above, since the left and right visual field images VL and VR themselves have almost no distortion in the visual field direction (horizontal direction), distortion correction is performed according to the cutout positions of the visual field images DL and DR. It is not necessary to perform the above process, and by simply cutting out the visual field images DL and DR having a predetermined width from the visual field images VL and VR, it is possible to obtain a vehicle side image with high naturalness, that is, without a sense of discomfort.

Further, in the present apparatus, when it is determined that the intersection is a specific intersection based on the navigation information, the screen ratio and the like are automatically set based on the information. Therefore, it is possible to eliminate the inconvenience of performing the setting operation for the nose view each time when entering an intersection that is routinely passed by commuting or the like.

The present invention is not limited to the above embodiment. For example, a camera is built into the outer quarter panel of the vehicle. The left and right side visual fields in the rear part of the vehicle may be similarly imaged and displayed as an image. Further, the direction AR of the visual field of interest with respect to the right visual field is set in association with the setting of the direction AL of the visual field of interest with respect to the left visual field, or the directions AL, AR of the visual field of interest are set in association with each other in the opposite relationship. May be. Further, the directions AL and AR of the visual fields of interest with respect to the left and right visual fields may be set independently of each other.

Regarding the directions AL and AR of the visual fields of interest, it is preferable to automatically set the reference directions in association with the setting of the screen ratio of the left and right visual images DL and DR. That is, when the screen ratio is changed and set, the visual field images DL, DR are referenced with respect to a certain angle on the side of the vehicle.
It suffices to determine the cutout area of. In addition, the present invention can be variously modified and implemented without departing from the scope of the invention.

[0038]

As described above, according to the present invention, from the respective visual field images in which the predetermined visual fields on the left and right sides of the vehicle are respectively imaged by the cameras attached to the front or rear of the vehicle, When partial images of the visual field of interest are extracted as the left and right visual field images respectively, and the images of these visual field regions are arranged side by side to form one display image to be displayed on the display, according to the situation of the intersection, etc. The screen ratio of the left and right field images can be variably set and
The screen ratio set for is stored according to the running environment,
According to the learning result and the traveling environment, the cutout of the left and right visual field images from the respective visual field images is controlled under this screen ratio.

Therefore, according to the present invention, a side image having a desired visual field width is obtained without changing the direction of the visual field of the camera itself or changing the visual field angle of the camera.
It can be combined into a single image and displayed together. Therefore, it is possible to simply and accurately check the required situation on the side of the vehicle according to the situation such as an intersection, and a great effect can be obtained in terms of safety confirmation.

In particular, in the visual field setting means, the width of the visual field of interest in each visual field image is variably set according to the manual operation of the selection switch, and further, the width of the visual field of interest in accordance with the traveling environment of the vehicle detected by the environment detection means. Is set, it is possible to obtain an effect such that a vehicle side view image having a view width corresponding to a situation such as an intersection can be reliably obtained.

[Brief description of drawings]

FIG. 1 is a functional schematic configuration diagram of a nose view device that is an example of a vehicle surrounding visual recognition device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a relationship between a left and right lateral field of view at the front of the vehicle by a left field of view camera and a right field of view camera incorporated in the nose view device shown in FIG.

FIG. 3 is a diagram showing a flow of a basic control procedure in the embodiment of the present invention.

FIG. 4 is a diagram showing a relationship between left and right visual field images VL and VR, visual field images DL and DR cut out from these visual field images VL and VR, and a composite image D thereof.

FIG. 5 is a diagram showing a function of a nose view device at an intersection.

[Explanation of symbols]

10 vehicles 11 Left-view camera 12 Right-view camera 13 Display 14 Navigation system 15 Nose View System 16 Nose view selection control function 17 Left and right visual field combination ratio control function 18 Left and right view change function 19 Learning section (memory)

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Satoshi Omuro 5-3-8 Shiba, Minato-ku, Tokyo Mitsubishi Motors Corporation (56) Reference JP-A-8-18949 (JP, A) JP Flat 6-227318 (JP, A) (58) Fields surveyed (Int.Cl. ⁷ , DB name) B60R 1/00

Claims (3)

(57) [Claims] 1. An image pickup means attached to a front portion or a rear portion of a vehicle for picking up images of the left and right side fields of view of the vehicle, respectively, and attention in each of the left and right side view images picked up by the image pickup means. Image combining means for extracting images in the visual field range as a left visual field image and a right visual field image, respectively, and arranging the left visual field image and the right visual field image to form one display image; Display means for displaying the display image and a ring for detecting the traveling environment of the vehicle.
The boundary detection means and the range of the visual field of interest in the left and right visual fields imaged by the imaging means are set to determine the screen ratio of the left visual field image and the right visual field image extracted from the visual field images. The field of view setting means and the set screen ratio
Learning means for storing according to the driving environment, and the environment detection hand
It is stored in the running environment detected by the step and the learning means.
Display control means for changing the screen ratio according to the information
A peripheral visibility device for a vehicle, comprising: Wherein said field setting means comprises a manual setting unit that includes a selection switch which is manually operated, variably setting the width of the target field in the field image of the left and right sides receiving operation of the selection switch The vehicle peripheral visual recognition device according to claim 1, wherein: 3. The visual field setting means comprises automatic setting means for variably setting the width of the visual field of interest in the left and right side view images in accordance with the information obtained by the environment detecting means. The vehicle peripheral visual recognition device according to claim 1.