JP5146330B2 - Vehicle road sign recognition device - Google Patents

Description

The present invention relates to a vehicle road sign recognition equipment recognizes the vehicular road sign of lane marker and the like. The information based on the recognition can be provided to, for example, a lane departure warning device, a lane keeping support device, or the like for use in these devices.

  It is equipped with a wide-angle camera and a telephoto camera that can move in the horizontal direction. Has already been proposed (see, for example, Patent Document 1).

Japanese Patent No. 3588728

However, the technique described in Patent Document 1 does not consider the mounting height of the wide-angle camera and the telephoto camera depending on the pitch angle of the vehicle and the vertical distance of the vehicle from the traveling road surface. For this reason, an error occurs when data representing edge points in an image is converted into road system coordinates, and there remains a problem that it is difficult to ensure the estimation accuracy of the road shape.
The present invention has been made to correspond to the situation as described above, basic capable of becoming capable of acquiring a road sign recognition equipment for vehicle recognition data of the vehicle road sign for estimating the road shape with sufficient precision The purpose is to provide.

In order to solve the above-described problem, in the present invention, the first variable group calculation unit in the traveling road is based on the image signal corresponding to the first imaging target area defined by a relatively narrow field angle and a long imaging distance. Calculate the curvature of the curve.
In addition, based on the image signal corresponding to the second imaging target area defined by a relatively wide angle of view and a short imaging distance, the second variable group calculation unit compares the vehicle width direction relative to the vehicle road marking in the vehicle width direction. The position, the vertical distance of the host vehicle from the traveling road surface, the displacement amount of the host vehicle in the pitching direction, the displacement amount of the host vehicle in the yaw direction, and the like are calculated.
Further, based on the data of the variable groups acquired in the first variable group calculation unit and the second variable group calculation unit, the vehicle road marking recognition processing unit recognizes the position of the vehicle road marking relative to the host vehicle and recognizes the position. Recognition data representing is obtained.
Then, a first imaging unit that acquires an image signal corresponding to the first imaging target region and supplies the image signal to the first image signal input unit, acquires an image signal corresponding to the second imaging target region, and inputs a second image signal Second imaging means to be supplied to the vehicle, wherein the first imaging means is a camera provided at the front of the host vehicle so that the first imaging target area faces the front of the vehicle, and the second imaging means Is a rear monitoring camera provided at the rear of the host vehicle so that the second imaging target area faces the rear of the vehicle, and the first imaging means and the second imaging means perform imaging operations synchronized with each other. .

  In the present invention, the relative position of the vehicle in the vehicle width direction with respect to the vehicle road marking, the distance in the vertical direction of the vehicle from the traveling road surface, the displacement in the pitching direction of the vehicle, and the displacement in the yaw direction of the vehicle Is taken into account. For this reason, when obtaining the recognition data regarding the position of the vehicle road marking, the accuracy of the data is increased.

It is a functional block diagram showing the road marking recognition apparatus for vehicles as the 1st Embodiment of this invention. It is a conceptual diagram showing the state which equipped the vehicle road marking recognition apparatus of FIG. 1 with the vehicle (own vehicle). It is a conceptual diagram showing the imaging visual field of a 1st imaging means (telephoto camera). It is a conceptual diagram showing the imaging visual field of a 2nd imaging means (wide angle camera). It is a flowchart showing the outline | summary of operation | movement of the road marking recognition apparatus for vehicles of FIG. It is a figure for demonstrating the calculation process of the movement vector of the own vehicle in the road marking recognition apparatus for vehicles of FIG. It is a figure showing the road model applied in embodiment of this invention. It is a figure showing the positional relationship of the own vehicle and the wide-angle camera attached to the own vehicle in a vehicle side view, and a road surface. It is a figure showing the lane shape in vehicle top view. It is a figure showing the positional relationship of the own vehicle in a vehicle side view, the telephoto camera attached to the own vehicle, and a road surface. It is a functional block diagram showing the road marking recognition apparatus for vehicles as the 2nd Embodiment of this invention. It is a flowchart showing the road marking recognition method for vehicles as the 3rd Embodiment of this invention.

Hereinafter, the present invention will be clarified by describing embodiments of the present invention in detail with reference to the drawings.
In each of the drawings referred to below, for the sake of convenience, the main part that is the subject of the description is exaggerated as appropriate, and other than the main part is appropriately simplified or omitted.
(Vehicle road marking recognition apparatus as a first embodiment of the present invention)
FIG. 1 is a functional block diagram showing a vehicular road marking recognition apparatus according to an embodiment of the present invention. The vehicle road sign recognition apparatus 100 includes the following units.

That is, the first image signal input unit 101 and the second image signal input unit 102 that receive the image signals, the first image memory 103 and the second image memory 104 that hold the received image signals, and the first variable group calculation unit 105, respectively. And a second variable group calculation unit 106 and a vehicle road sign recognition processing unit 107.
The first image signal input unit 101 receives an image signal corresponding to a first imaging target area defined by a predetermined first imaging angle of view and a predetermined first imaging distance.

In the present embodiment, the image signal received by the first image signal input unit 101 is relatively distant from the first imaging field angle that is a relatively narrow so-called narrow field angle, which is the above-described first imaging target region. It is an image signal corresponding to a first imaging target area defined by a first imaging distance which is a so-called long imaging distance.
An image signal corresponding to such a first imaging target region is acquired by a telephoto camera as the first imaging means 201 that adjusts the optical system to the telephoto side (long focal length) or includes an optical system with a telephoto specification, This is supplied to the first image signal input unit 101 described above.

On the other hand, the second image signal input unit 102 receives an image signal corresponding to a second imaging target area defined by a predetermined second imaging angle of view and a predetermined second imaging distance.
The image signal received by the second image signal input unit 102 is a so-called short imaging distance that is relatively close to a relatively wide second imaging angle of view, which is the above-described second imaging target region. It is an image signal corresponding to the second imaging target area defined by the second imaging distance.

Such an image signal corresponding to the second imaging target area is acquired by the wide-angle camera 202 as the second imaging means 202 having the optical system adjusted to the wide-angle side (short focal length) or provided with a wide-angle optical system. , And supplied to the second image signal input unit 102 described above.
The first image signal input unit 101 and the second image signal input unit 102 are portions that function as an interface unit that receives a corresponding image signal. These can be configured as connectors with conductor contacts, but can take the form of non-contact interfaces with optical signals and short-range wireless signals.

On the other hand, the first variable group calculation unit 105 uses the information included in the image signal as described above received by the first image signal input unit 101 to calculate the curvature of the curve on the corresponding travel route. This calculation will be described in detail later.
Further, the second variable group calculation unit 106 uses a plurality of information used for recognizing the traveling state of the own vehicle using the information included in the image signal as described above received by the second image signal input unit 102. A corresponding interval between the vehicle road markings and a relative position in the vehicle width direction of the own vehicle with respect to the plurality of vehicle road markings are calculated. Furthermore, the second variable group calculation unit 106 calculates the distance in the vertical direction of the host vehicle from the traveling road surface, the amount of displacement in the pitching direction of the host vehicle, and the amount of displacement in the yaw direction of the host vehicle. This calculation will be described in detail later.

The vehicle road marking recognition processing unit 107 uses the information included in the calculation result in the first variable group calculation unit 105 and the calculation result in the second variable group calculation unit 106 to display the vehicle road marking for the vehicle. Recognition data representing the recognition is obtained by recognizing the position.
The first image memory 103 described above holds the image signal received by the first image signal input unit 101 so as to be accessible from the first variable group calculation unit 105 at an arbitrary timing.

Therefore, the first variable group calculation unit 105 uses the image data acquired by the telephoto camera as the first imaging unit 201 and stored in the first image memory 103 at an appropriate timing, and calculates the value of the corresponding variable.
Similarly, the second image memory 104 holds the image signal received by the second image signal input unit 102 so as to be accessible from the second variable group calculation unit 106 at an arbitrary timing.

Therefore, the second variable group calculation unit 106 uses the image data acquired by the wide-angle camera as the second imaging unit 202 and stored in the second image memory 104 at an appropriate timing, and calculates the value of the corresponding variable.
On the other hand, the vehicular road sign recognition apparatus according to the present embodiment has a configuration including the telephoto camera 201 as the first imaging unit 201 and the wide-angle camera 202 as the second imaging unit 202 as described above. Can do.

As illustrated in FIG. 2, the telephoto camera 201 and the wide-angle camera 202 acquire the image signal corresponding to the first imaging target area and the image signal corresponding to the second imaging target area as described above, as illustrated in FIG. Attach to the corresponding vehicle in the installation form that matches each different application.
FIG. 2 is a conceptual diagram showing a state in which the vehicle road marking recognition apparatus 100 of FIG. 1 having the telephoto camera 201 and the wide-angle camera 202 is mounted on the own vehicle 200 (own vehicle).

The telephoto camera as the first imaging unit 201 is an image corresponding to a first imaging target area defined by a relatively narrow first imaging angle of view and a relatively far first imaging distance as described above. Install so that it matches the purpose of acquiring signals.
That is, the telephoto camera 201 is installed so that the mounting level position is as high as possible in the center of the host vehicle 200 so as to form a shallow depression angle such that the imaging optical axis is slightly downward from the horizontal.

On the other hand, the wide-angle camera as the second imaging unit 202 corresponds to the second imaging target area defined by the relatively wide second imaging angle of view and the relatively second imaging distance as described above. Install so that it matches the purpose of acquiring image signals.
That is, for example, the wide-angle camera 202 has a relatively deep depression angle whose imaging optical axis is relatively deep so that a wide range of lane markings as close as possible to the vehicle 200 can be accommodated in the imaging field of view at the central portion of the front surface of the vehicle 200. Install to make

The installation form of the first imaging means (telephoto camera) 201 and the second imaging means (wide-angle camera) 202 as described above results in the following.
That is, the telephoto camera 201 has an attachment level position higher than the attachment level position of the wide-angle camera 202, and the depression angle of the imaging optical axis is shallower than the depression angle of the imaging optical axis of the wide-angle camera 202. Install to make

If the relative relationship is viewed with the wide-angle camera 202 side as a reference, the wide-angle camera 202 is mounted at a position where the mounting level position is lower than the mounting level position of the telephoto camera 201 and the depression angle of the imaging optical axis is the telephoto camera. It is installed so as to form an angle deeper than the depression angle of the imaging optical axis 201.
On the other hand, in the embodiment of the present invention, the first image pickup means (telephoto camera) 201 and the second image pickup means (wide-angle camera) 202 use an appropriate common synchronization signal (its supply system is not shown). Thus, imaging operations synchronized with each other with respect to the time and cycle of imaging are performed.

Therefore, for the first image pickup means (telephoto camera) 201 and the second image pickup means (wide-angle camera) 202, the image acquisition simultaneity is ensured.
For this reason, using the calculation result by the first variable group calculation unit 105 and the calculation result by the second variable group calculation unit 106, the accuracy when the vehicle road sign recognition processing unit 107 executes integrated processing is improved. Can be maintained at a level.

FIG. 3 is a conceptual diagram showing the imaging field of view of the first imaging means (telephoto camera) 201, and FIG. 4 is a conceptual diagram showing the imaging field of view of the second imaging means (wide-angle camera) 202.
As apparent from the comparison between FIG. 3 and FIG. 4, when the preceding vehicle 300 of the own vehicle is approaching, the telephoto camera 201 is obstructed as shown in FIG. , 302 cannot be seen and the image pickup signal cannot be acquired.

On the other hand, in the wide-angle camera 202, even when the preceding vehicle 300 is approaching as in the case of FIG. 3, most of the imaging field of view is not disturbed as in FIG. The extension direction can be seen and the imaging signal can be acquired.
That is, according to the wide-angle camera 202, acquisition of the imaging signal can be maintained along the extending direction of the lane markings 301 and 302 unless the preceding vehicle 300 approaches abnormally.

In reality, in a lane departure warning device (LDP: Lane Departure Prevention) or a lane keeping support device (LKS) that is assumed to use the recognition data of the vehicle road sign recognition device 100, The vehicle speed is equal to or higher than a predetermined value, and the inter-vehicle distance from the preceding vehicle is naturally separated to some extent.
For this reason, even with the wide-angle camera 202, there is almost no possibility that it is difficult to acquire the imaging signal along the extending direction of the lane markings 301 and 302.

FIG. 5 is a flowchart showing an outline of the operation of the vehicular road marking recognition apparatus 100 of FIG.
First, from the first image signal input unit 101 and the second image signal input unit 102, the first image memory 105 and the second image memory 106 acquires the image signal which is an image pickup output of telecamera 201 and the wide-angle camera 20 2 Each is held separately (S501).

The image signal (image data) acquired and held in step S501 may be image data of a luminance (monochrome) image with a luminance component not including a color signal component.
Next, a differential image corresponding to the image based on the image signal obtained in step S501 and separately stored in the first image memory 105 and the second image memory 106 is generated (S502).
In the differential image generation processing in step S502, the product sum of the weight matrix and the luminance value is obtained by applying the kernel of the Sobel operator as shown in FIG.

That is, the horizontal differential kernel of FIG. 6A is applied from the left to the right along the main scanning direction to obtain a differential image output in the horizontal direction, and the vertical differential kernel of FIG. Are applied from top to bottom along the sub-scanning direction to obtain a differential image output in the vertical direction.
Here, in the differential image in the horizontal direction, an edge extending mainly in the vertical direction (up and down direction) in the screen is detected, and in the differential image in the vertical direction, it extends mainly in the horizontal direction (left and right direction) in the screen. Edges are detected.

In any of the above cases, the differential value of the edge portion changing from low luminance to high luminance is positive, the differential value of the edge portion changing from high luminance to low brightness is negative, and the luminance change is small The derivative value of becomes substantially zero.
In the present embodiment, a binary image is obtained using the data related to the differential image in the horizontal direction among the above (S503).
In the processing in step S503, a binary image (hereinafter referred to as a positive edge image) in which the pixel value having a differential value equal to or greater than a predetermined threshold TH_BIN_H is 1 and the pixel value less than 0 is set to 0 for the differential image generated in step S502. Generate.
In addition, a binary image (hereinafter referred to as a negative edge image) is generated in which a pixel value having a differential value less than a predetermined threshold TH_BIN_L is 1 and a pixel value greater than that is 0.

Here, the threshold value TH_BIN_H is set through experiments so that the left edge point of the lane marking is 1 and the road surface is 0 in the positive edge image.
Similarly, the threshold TH_BIN_L is set so that the edge point on the right side of the lane marking is 1 and the road surface is 0 in the negative edge image.
Next, the degree of matching between the image data acquired by the wide-angle camera 202 and the point sequence on the image represented by the road model formula described later and the positive edge image and the negative edge image generated in step S503 is calculated, and the matching is performed. The road model parameter that maximizes the degree is determined (S504).

Hereinafter, the process in step S504 will be described in detail with reference to the drawings.
FIG. 7 is a diagram illustrating a road model applied in the present embodiment.
In FIG. 7, a coordinate system (XYZ) centered on the camera is defined. The Z-axis is defined as the vehicle traveling direction, which also coincides with the optical axis direction of the camera projected on the horizontal plane.
The Y axis is defined vertically upward, and the X axis is defined leftward with respect to the vehicle traveling direction. FIG. 7 shows the host vehicle 700 and the left and right lane markings 701 and 702 in a top view of the vehicle.
If the own vehicle 700 is positioned with an angle of φ [rad] counterclockwise with respect to the lane, and the lane can be approximated by a straight line in the vicinity of the own vehicle imaged by the wide-angle camera 202, the left and right lane divisions The equations of the lines 701 and 702 can be expressed by the following equation 1.

Here, y _s is the lateral displacement of the vehicle from the right side of the lane marker 702, W is distance between the left and right lane markings 701 and 702, i.e., the width of the lane, the average as the W lane width Set in advance.
On the other hand, FIG. 8 is a diagram showing the positional relationship between the own vehicle 700, the wide-angle camera 202 attached to the own vehicle, and the road surface 800 in a side view of the vehicle. Assuming that the optical axis of the camera 202 is attached to the road surface 800 at an angle of η2 [rad], the road surface equation can be expressed by Equation 2.

Here, h2 represents the height of the camera 202 with respect to the road surface (the height from the traveling road surface of the imaging viewpoint regarding acquisition of the second image signal).
Consider a two-dimensional coordinate system xy on an image acquired by the camera 202. The relationship between the three-dimensional coordinate system XYZ and the two-dimensional coordinate system xy depends on the intrinsic value of the camera related to the characteristics of the optical system of the camera 202.
Here, assuming that the camera-specific function is f, the left and right lane markings are expressed as in Expression 3 on the image.

As described above, when the road model defined by Equation 1, Equation 2, and Equation 3 is applied, there are four parameters of the road model: yaw angle φ, lateral displacement amount ys, pitch displacement amount η2, and camera height h2.
This corresponds to a case where the road curvature ρ is set to 0 in the road model shown in Japanese Patent No. 3733875 of the present applicant.

In the processing in step S504, the parameters of the road model formula described above are optimized by an optimization method, for example, simulated annealing (SA).
At that time, in the predetermined area on the screen, the number of points where the point sequence of the lane marking line on the left side of the vehicle according to Equation 3 overlaps the point of pixel value 1 of the negative edge image, and the lane segmentation on the right side of the vehicle according to Equation 3 A value obtained by summing the number of points where the point sequence of the line overlaps the pixel value 1 of the positive edge image is set as an evaluation value of the degree of matching of the parameters.

Next, whether or not there is a preceding vehicle is applied to the image acquired by the telephoto camera 201 by applying a known vehicle recognition algorithm, for example, the technology of Patent No. 3146809 owned by the present applicant. Is determined (S505).
When there is a preceding vehicle (S505: Yes), as described with reference to FIG. 3, in the telephoto camera 201, the preceding vehicle interferes with the field of view and the lane line is hidden, so that the correct lane recognition is performed. I can't expect it.

Therefore, in this case, the four road model parameters calculated in the process in step S504 are output as recognition results, and the process ends.
When there is no preceding vehicle (S505: No), the road curvature ρ is calculated from the image acquired by the telephoto camera 201 based on the four road model parameters calculated in the process in step S504 (S506). Five parameters including ρ are output as recognition results, and the process ends.

Next, a method for calculating the road curvature ρ will be described with reference to FIGS. 9 and 10.
FIG. 9 is a diagram illustrating a lane shape in a vehicle top view.
FIG. 10 is a diagram illustrating the positional relationship between the own vehicle 700 and the telephoto camera 201 attached to the own vehicle 700 and the road surface 800 in a side view of the vehicle.
The coordinate system centered on the camera, the yaw angle of the vehicle, the pitch displacement amount, the camera height, and the lateral displacement amount have the same definitions as in step S504.
However, the wide-angle camera 202 and the telephoto camera 201 have different mounting positions and postures as shown in FIGS.

Therefore, the pitch displacement amount η1 of the telephoto camera 201 and the camera height h1 (the height from the traveling road surface of the imaging viewpoint regarding acquisition of the first image signal) are the pitch displacement amount η2 and the camera height h2 of the wide-angle camera 202. Different.
However, since the positional relationship between the wide-angle camera 202 and the telephoto camera 201 is unchanged, the pitch displacement amount η1 and the camera height h1 of the telephoto camera 201 are relative to the pitch displacement amount η2 and the camera height h2 of the wide-angle camera 202, respectively. Is calculated automatically.
Assuming that the lane shape is an arc as shown in FIG. 9 and its curvature is ρ, the equation of the left and right lane markings in the vehicle top view and the equation of the road surface in the vehicle side view can be expressed by Equation 4.

ここで、車線幅Ｗは既述の例と同様に定数とし、横変位量ｙ_s、ヨー角φ、ピッチ変位量ηはステップＳ５０４で算出された値を適用する。
カメラ高さｈ１はステップＳ５０４で算出された広角カメラ２０２の高さｈ２から相対的に算出される望遠カメラ２０１の高さとする。
即ち、ここで用いる道路モデルのパラメータのうち、未知のパラメータは道路曲率ρだけである。
従って、道路曲率ρの最適化にはステップＳ５０４で用いた上述の最適化手法ＳＡよりも簡素な手法、例えば黄金分割法などを用いることが可能であり、必要な計算量はステップＳ５０４における最適化に比べて大幅に削減される。

Here, the lane width W is a constant as in the above-described example, and the values calculated in step S504 are applied to the lateral displacement amount y _s , the yaw angle φ, and the pitch displacement amount η.
The camera height h1 is the height of the telephoto camera 201 calculated relatively from the height h2 of the wide-angle camera 202 calculated in step S504.
That is, of the road model parameters used here, the only unknown parameter is the road curvature ρ.
Therefore, a method simpler than the above-described optimization method SA used in step S504, such as the golden section method, can be used for optimization of the road curvature ρ, and the necessary calculation amount is optimized in step S504. It is greatly reduced compared to.

(Operations and effects of the first embodiment)
(1-1) The first image signal input unit 101 that receives an image signal acquired by the camera (telephoto camera) 201 whose optical system is relatively adjusted to the telephoto side, and the optical system is relatively adjusted to the wide-angle side. And a second image signal input unit 102 that receives an image signal acquired by the camera (wide-angle camera) 202.
A first image memory 103 that holds an image signal received by the first image signal input unit 101 and a second image memory 104 that holds an image signal received by the second image signal input unit 102 are provided.
Therefore, the image signal (image data) from the telephoto camera 201 held in the first image memory 103 and the image signal (image data) from the wide-angle camera 202 held in the second image memory 104 are used according to different uses. It can be used effectively and the vehicle road marking can be accurately recognized.

(1-2) First variable group calculation that calculates the curvature of the curve on the corresponding travel path using image data from the telephoto camera 201 received by the first image signal input unit 101 and held in the first image memory 103 Part 105 is provided.
For this reason, when the field of view taken by the telephoto camera 201 is secured without being obstructed by a preceding vehicle or the like, the curvature of the curve on the corresponding travel path can be calculated quickly and with high accuracy.

(1-3) Using the image data from the wide-angle camera 201 received by the second image signal input unit 102 and stored in the second image memory 104, the distance between the lane markings and the distance between the lane markings and the own vehicle A second variable group calculation unit 106 is provided that calculates the relative position, the distance in the vertical direction of the host vehicle from the traveling road surface, the amount of displacement in the pitching direction of the host vehicle, and the amount of displacement in the yaw direction of the host vehicle.
For this reason, using the image data from the wide-angle camera 201 whose imaging field of view is not easily obstructed by the preceding vehicle or the like, the relative position described above, the distance in the vertical direction of the vehicle from the traveling road surface, the amount of displacement in the pitching direction of the vehicle, And the displacement amount of the own vehicle in the yaw direction can be accurately calculated.

(1-4) Using the information included in the calculation result of the first variable group calculation unit 105 and the calculation result of the second variable group calculation unit 106, the position of the vehicle road marking relative to the host vehicle is recognized and the A vehicular road marking recognition processing unit 107 is provided for obtaining recognition data representing recognition.
For this reason, when obtaining recognition data regarding the position of the vehicle road marking, the relative position in the vehicle width direction of the vehicle relative to the vehicle road marking, the vertical distance of the vehicle from the traveling road surface, the pitching direction of the vehicle Considering the amount of displacement and the amount of displacement of the vehicle in the yaw direction, the accuracy of the recognition data is increased.

(1-5) A camera (telephoto camera) 201 that supplies the first image signal input unit 101 with an image signal obtained by adjusting the optical system relatively to the telephoto side, and the optical system is relatively adjusted to the wide-angle side. A camera (wide-angle camera) 202 that supplies the acquired image signal to the second image signal input unit 102 is provided.
Therefore, the image data of the imaging field foresee the distant by telecamera 201, and the image data of the imaging field of view with a spread of interfering with less susceptible vehicle near the preceding vehicle by the wide-angle camera 20 2 is obtained, separately It can be used effectively according to the usage of the vehicle, and the vehicle road marking can be accurately recognized.

(1-6) The first imaging unit (telephoto camera) 201 and the second imaging unit (wide-angle camera) 202 are configured to perform imaging operations synchronized with each other.
For this reason, the simultaneous acquisition of images by both cameras 201 and 202 is ensured, and the vehicle road marking recognition process is performed using the calculation result by the first variable group calculation unit 105 and the calculation result by the second variable group calculation unit 106. The accuracy when the integrated processing is executed by the unit 107 can be maintained at a high level.

(1-7) The telephoto camera 201 is mounted so that its mounting level position is higher than the mounting level position of the wide-angle camera 202, and the depression angle of its imaging optical axis is smaller than the depression angle of the imaging optical axis of the wide-angle camera 202. Install at a shallow angle.
For this reason, the telephoto camera 201 occupies a mounting position suitable for acquiring image data of an imaging field of view far away, and the wide-angle camera 202 is image data of an imaging field of view having a spread in the vicinity of the own vehicle that is not easily disturbed by the preceding vehicle. It can occupy a mounting position suitable for obtaining.

(Second Embodiment)
FIG. 11 is a conceptual diagram showing a second embodiment of the present invention. In the example of FIG. 11, the vehicle road sign recognition apparatus 100 of FIG. 1 having a telephoto camera 201 and a wide-angle camera 202a is installed in the own vehicle 200 (own vehicle).
The telephoto camera as the first imaging unit 201 is an image corresponding to a first imaging target area defined by a relatively narrow first imaging angle of view and a relatively far first imaging distance as described above. Install so that it matches the purpose of acquiring signals.
That is, for the telephoto camera 201, as in the example described with reference to FIG. 2, the attachment level position is as high as possible in the center of the host vehicle 200, and the imaging optical axis is slightly downward from the horizontal. Install so as to form a shallow depression.

On the other hand, the wide-angle camera as the second imaging unit 202a corresponds to the second imaging target area defined by the relatively wide second imaging angle of view and the relatively second imaging distance as described above. It is the same as the example described with reference to FIG. 2 in that it is installed so as to match the application for acquiring the image signal.
However, in this second embodiment, the wide-angle camera 202a is installed, for example, rearwardly at the central portion of the rear surface of the host vehicle 200, and a wide range of lane markings as close as possible to the host vehicle 200 is located behind the host vehicle. It is installed so that its imaging optical axis forms a relatively deep depression angle so that it can be accommodated in the imaging field of view.

The installation form of the first imaging means (telephoto camera) 201 and the second imaging means (wide-angle camera) 202a as described above results in the following.
That is, the telephoto camera 201 is positioned so that its mounting level position is higher than the mounting level position of the wide-angle camera 202 a, and the depression angle of the imaging optical axis is shallower than the depression angle of the imaging optical axis of the wide-angle camera 202. Install to make

If the relative relationship is viewed with the wide-angle camera 202a as a reference, the wide-angle camera 202a is mounted at a position where the mounting level position is lower than the mounting level position of the telephoto camera 201, and the depression angle of the imaging optical axis is the telephoto camera. It is installed so as to form an angle deeper than the depression angle of the imaging optical axis 201.
Also in the second embodiment, the first imaging means (telephoto camera) 201 and the second imaging means (wide-angle camera) 202a use an appropriate common synchronization signal (its supply system is not shown). Thus, imaging operations synchronized with each other with respect to the time and cycle of imaging are performed.

Therefore, as for the first image pickup means (telephoto camera) 201 and the second image pickup means (wide-angle camera) 202a, the simultaneity of image acquisition by them is ensured.
Therefore, the accuracy in executing integrated processing in the vehicular road marking recognition processing unit 107 using the calculation result by the first variable group calculation unit 105 and the calculation result by the second variable group calculation unit 106 is high. Can be maintained.

It should be noted that the preferred mounting position as described above with respect to the second imaging means (wide-angle camera) 202a substantially coincides with the mounting position of the rear monitoring camera for vehicles that are already widely used.
Therefore, if the image data acquired by such an existing rear monitoring camera is supplied to the second image signal input unit 102 of FIG. 1 and further stored in the second image memory 104, the subsequent vehicle The road marking recognition process is the same as in the first embodiment described above.

(Operations and effects of the second embodiment)
(2-1) The angle of attachment of the wide-angle camera 202a is lower than the attachment level position of the telephoto camera 201, and the depression angle of the imaging optical axis is deeper than the depression angle of the imaging optical axis of the telephoto camera 201 It is installed in the center part on the rear surface of the host vehicle 200 so as to face backward.
In the case of adopting this configuration, a vehicle rear monitoring camera that has already been widely used can be applied as the second imaging means (wide-angle camera) 202a.

(Vehicle road marking recognition method as a third embodiment of the present invention)
A vehicle road marking recognition method according to a third embodiment of the present invention will be described with reference to FIG.
The vehicle road marking recognition method as an embodiment of the present invention includes a first image signal receiving step (S1201), a second image signal receiving step (S1202), a first variable group calculating step (S1203), A bivariate group calculation step (S1204) and a vehicle road marking recognition step (S1205) are included.
In the first image signal receiving step (S1201), an image signal corresponding to a first imaging target area defined by a predetermined first imaging angle of view and a predetermined first imaging distance is received.

In the second image signal receiving step (S1202), a first imaging target area defined by a second imaging angle wider than the first imaging angle of view and a second imaging distance shorter than the first imaging distance. The image signal corresponding to is received.
In the first variable group calculation step (S1203), the curvature of the curve on the corresponding travel path is calculated using the information included in the image signal received in the first image signal reception step (S1201).

  In the second variable group calculating step (S1204), the information included in the image signal received in the second image signal receiving step (S1202) is used to recognize a traveling state of the own vehicle. The corresponding interval between road markings, the relative position of the vehicle in the vehicle width direction with respect to the plurality of vehicle road markings, the distance in the vertical direction of the vehicle from the traveling road surface, the amount of displacement in the pitching direction of the vehicle, and The amount of displacement in the yaw direction of the car is calculated.

In the vehicle road marking recognition step (S1205), the vehicle is used by using information included in the calculation result in the first variable group calculation step (S1203) and the calculation result in the second variable group calculation step (S1204). Recognize the position of the vehicle road marking relative to and obtain recognition data representing the recognition.
In the above, the first image signal receiving step (S1201) and the second image signal receiving step (S1202) do not ask the execution order thereof, and therefore, the order may be changed from the above, Alternatively, it may be executed in parallel.

100 vehicle road marking recognition apparatus 101 first image signal input unit 102 second image signal input unit 103 first image memory 104 second image memory 105 first variable group calculation unit 106 second variable group calculation unit 107 road marking for vehicle Recognition processing unit 200 Vehicle (own vehicle)
201 First imaging means (telephoto camera)
202 Second imaging means (wide angle camera)
202a Second imaging means (wide angle camera)
300 preceding vehicle 301,302 lane marking 700 own vehicle 701,702 lane marking 800 road surface

Claims (4)

A first image signal input unit that receives an image signal corresponding to a first imaging target area defined by a predetermined first imaging angle of view and a predetermined first imaging distance;
A second image signal that receives an image signal corresponding to a second imaging target area defined by a second imaging field angle wider than the first imaging field angle and a second imaging distance shorter than the first imaging distance. An input section;
A first variable group calculation unit that calculates the curvature of the curve on the travel path using information included in the image signal received by the first image signal input unit;
Using information included in the image signal received by the second image signal input unit, the interval between the plurality of vehicle road markings, the relative position in the vehicle width direction of the own vehicle with respect to the plurality of vehicle road markings, traveling A second variable group calculation unit for calculating the distance in the vertical direction of the vehicle from the road surface, the amount of displacement in the pitching direction of the vehicle, and the amount of displacement in the yaw direction of the vehicle;
Using the information included in the calculation result of the first variable group calculation unit and the calculation result of the second variable group calculation unit, the recognition data representing the recognition by recognizing the position of the vehicle road marking relative to the own vehicle A vehicle road marking recognition processing unit,
First imaging means for acquiring an image signal corresponding to the first imaging target area and supplying the image signal to the first image signal input unit;
Second imaging means for acquiring an image signal corresponding to the second imaging target area and supplying the image signal to the second image signal input unit;
Equipped with a,
The first imaging unit is a camera provided at a front portion of the host vehicle so that the first imaging target region faces the front of the vehicle, and the second imaging unit is configured such that the second imaging target region faces the rear of the vehicle. A vehicle road sign recognition device, characterized in that the rear imaging camera is provided at the rear of the host vehicle, and the first imaging means and the second imaging means perform imaging operations synchronized with each other. . The first imaging means is arranged such that its attachment level position is higher than the attachment level position of the second imaging means, and the depression angle of the imaging optical axis is the depression angle of the imaging optical axis of the second imaging means. The vehicular road marking recognition apparatus according to claim 1 , wherein the vehicular road sign recognition apparatus is installed so as to form a shallower angle. The second imaging means is arranged such that its attachment level position is lower than the attachment level position of the first imaging means, and the depression angle of the imaging optical axis is the depression angle of the imaging optical axis of the first imaging means. 3. The vehicular road marking recognition apparatus according to claim 1 , wherein the vehicular road marking recognition apparatus is installed so as to form a deeper angle.   A first image memory that holds the image signal received by the first image signal input unit so as to be accessible from the first variable group calculation unit, and the image signal received by the second image signal input unit is the second variable group. The vehicular road sign recognition apparatus according to claim 1, further comprising a second image memory that is accessible from the calculation unit.

Images