JP2006036048A - Vehicle light controlling device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle light controlling device capable of lighting up vehicle lights properly according to the surrounding situation of the vehicle depending upon its running condition without providing a mechanism to move the optical axis of a camera. <P>SOLUTION: The vehicle light controlling device 1 is equipped with a controller 2 to perform automatic lighting of the vehicle lights 7, wherein the controller 2 controls the lighting-up of the lights 7 on the basis of the image in the vehicle advancing direction photographed by the camera 3. The controller 2 determines one point on the image from the characteristic point thereon, shifts the monitoring region with that point determined assumed as the vehicle advancing direction, and thereby controls the lighting-up of the lights 7 on the basis of the brightness in the monitoring region. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a vehicle light control device that automatically turns on a vehicle light (for example, a headlamp, a small light, etc.) when the surroundings become dark while the vehicle is running. The present invention relates to a vehicle light control device that controls lighting of a vehicle light based on the captured video.

  Conventionally, a vehicle light control device is provided with an auto light switch, and if the auto light switch is turned on, the vehicle light automatically turns on when the vehicle enters a dark part such as a tunnel while driving. However, a system in which vehicle lights are automatically turned off when exiting a tunnel is known as a conlight system in the vehicle field.

Such a device is equipped with a illuminometer for detecting illuminance on the vehicle and a camera for imaging the traveling direction of the vehicle, and calculates the proportion of the dark part in the image data of the front view image, and this proportion is determined in advance. When the first dark part threshold is exceeded, the vehicle light is automatically turned on in preference to the lighting control based on the measurement value of the illuminometer (for example, Patent Document 1).
JP 2001-39210 A (first page)

  In the device disclosed in Patent Document 1 described above, when there is a dark place such as a tunnel in the traveling direction, the vehicle light can be automatically turned on sufficiently before entering the dark place.

  However, in the apparatus shown in Patent Document 1, the dark portion detection range is such that the size of the frame centered on the image changes depending on the speed of the vehicle, and when traveling on a curve, the light of the camera is changed according to the steering angle of the steering. The shaft moves left and right. For this reason, the above apparatus requires a mechanism for changing the optical axis of the camera.

  Therefore, the present invention has been made in view of the above problems, and the lighting of the vehicle light is appropriately performed according to the situation of the vehicle according to the traveling state of the vehicle without providing a mechanism for moving the optical axis of the camera. It is an object of the present invention to provide a device that can perform the above.

Means taken in order to solve the above-mentioned problem includes a control means for automatically turning on the vehicle light, and the control means is based on an image of the traveling direction of the vehicle photographed by the imaging means. In the vehicle light control device that controls lighting,
Feature point extracting means for obtaining a feature point imaged on the image sensor of the imaging means from the video, and a traveling direction extracting means for obtaining a first point indicating the traveling direction of the vehicle on the video or image from the feature point. The control means is configured to change the monitoring area based on the information based on the first point, and to turn on the vehicle light based on the luminance in the monitoring area.

  In this case, vehicle detection means for detecting the presence of the vehicle in the traveling direction from the feature points may be provided, and the monitoring area may be changed depending on the presence or absence of the vehicle in the traveling direction.

  Further, when a vehicle is present in the traveling direction, the monitoring area is moved on the road surface located behind the vehicle, and the number of pixels whose luminance value in the monitoring area on the image is equal to or less than a predetermined threshold is equal to or more than the predetermined pixel. The control means may turn on the vehicle light based on the traveling speed.

  When there is no vehicle in the traveling direction, the control means moves the monitoring area around the first point, and when the number of pixels whose luminance value in the monitoring area is equal to or less than a predetermined threshold is equal to or greater than a predetermined pixel, The vehicle light should be turned on based on.

  Further, it is preferable to provide a steering detection means for detecting the steering angle of the vehicle, and to change the monitoring area from the first point based on the steering angle.

  Furthermore, it comprises a speed detection means for detecting the traveling speed of the vehicle and a tunnel / dark place detection means for detecting a tunnel or dark place existing in the traveling direction of the vehicle, and the control means detects the tunnel or dark place, The vehicle light should be turned on based on the running speed.

  According to the first invention, a feature point imaged on the image sensor of the imaging means is obtained from the video, and a first point (for example, the feature point on the road surface) indicating the traveling direction of the vehicle on the video or the image from the feature point In this case, a point (disappearance point) where the extension line of the road surface intersects at one point (a vanishing point), a point due to an optical flow, etc.) can be obtained. Then, the control means knows the traveling direction of the vehicle from the information based on the first point obtained from the feature points, changes the monitoring area where the lighting is determined, and lights the vehicle light based on the luminance in the monitoring area Can do. This is because the direction in which the vehicle should travel is known from the feature points of the image captured by the imaging means, so there is no need for a mechanism to change the optical axis of the camera, the monitoring area is changed according to the traveling direction, and the vehicle running state Accordingly, the vehicle light can be turned on at an appropriate timing. The light can be turned on by capturing an image of the traveling direction of the vehicle by the imaging means and comparing the luminance values of the pixels in the monitoring area in the traveling direction.

  According to the second aspect of the invention, it is possible to obtain one point indicating the traveling direction of the vehicle from the road surface or the installation position of the camera with respect to the vehicle and the video information during traveling. This enables light control to a dark place that exists far away in the traveling direction of the vehicle.

  According to the third invention, the vehicle detection means for detecting the presence of the vehicle in the traveling direction from the feature point is provided, and when the vehicle is present in the traveling direction by changing the monitoring area depending on the presence or absence of the vehicle in the traveling direction. The light control can be performed by setting a region on the road surface located behind the preceding vehicle and on the road surface ahead of the host vehicle as the monitoring region. On the other hand, when there is no preceding vehicle, it is possible to perform appropriate light control according to information around the vehicle by setting the distant region as the monitoring region.

  According to the fourth invention, when a vehicle exists in the traveling direction, the monitoring area is moved behind the preceding vehicle, and the number of pixels whose luminance value in the monitoring area is equal to or less than a predetermined threshold is equal to or more than the predetermined pixel. In this case, when the control means turns on the vehicle light based on the traveling speed, the light control based on the luminance value in the monitoring area can be performed even when there is a preceding vehicle.

  According to the fifth aspect, when there is no vehicle in the traveling direction, the monitoring area is moved around the first point. When the number of pixels whose luminance value in the monitoring area of the first point is equal to or smaller than a predetermined threshold is equal to or larger than the predetermined pixel, the control means turns on the vehicle light based on the traveling speed, and the luminance when there is no preceding vehicle Write control based on the value.

  According to the sixth aspect of the present invention, when the monitoring area is changed by changing the position of the first point based on the steering angle, the monitoring area is detected according to the steering angle. Therefore, the light control according to the traveling state of the vehicle can be performed.

  According to the seventh aspect of the invention, a tunnel or dark place existing in the traveling direction of the vehicle is detected, and the lighting timing of the vehicle light is determined according to the distance to the tunnel in the traveling direction of the vehicle based on the speed of the vehicle. It becomes possible to control appropriately.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a system configuration of the vehicle light control device 1. The vehicle light control device 1 automatically turns on / off the vehicle light 7 when an auto light switch (auto light SW) 6 provided in the driver's seat is on. In the present embodiment, the vehicle light indicates a headlamp that illuminates the front of the vehicle and light generated by the tail lamp, the small light, and the license plate lamp provided at the rear of the vehicle. For example, when the surroundings of the host vehicle are bright, the tail lamp and the head lamp are turned off. However, when it becomes dark for a moment or when it is dim, only the tail lamp is turned on. When the surroundings are dark or when it becomes dark for a moment, both the tail lamp and the head lamp are turned on. When the auto light SW 6 is in an off state, the light is turned on / off by an on / off operation of a manual light switch (for example, a headlight switch, a small light switch).

  The vehicle light control device 1 includes a ROM 2 for storing a light control program, a RAM for temporarily storing information necessary for executing the program, a controller 2 having a CPU for executing the program, a camera 3 and various sensors. And vehicle lights 7 are provided. A video signal from the camera 3 is input to the controller 2, and signals from a vehicle speed sensor 4 and a steering angle sensor 5 that detect the traveling state of the vehicle are input. With this controller 2, when the auto light SW6 is in an on state, light control can be performed according to the running state.

  The camera 3 includes a CCD image sensor (for example, a pixel composed of 640 × 480) that forms an image inside. The camera 3 is integrally attached to a front grille in front of the vehicle or a rear view mirror that can confirm the rear. The optical axis of the camera 3 faces the front of the vehicle, and the surrounding situation of the host vehicle can be photographed up to several meters ahead (about 100 m) in the traveling direction of the vehicle. The vehicle speed sensor 4 outputs a pulse signal (vehicle speed signal) to the controller 2 in accordance with the vehicle speed, and can detect the vehicle speed of the host vehicle based on the time interval of the pulse signal input to the controller 2. .

  Further, the steering angle sensor 5 is provided with two hall elements having different phases on the steering column shaft, and a pulse signal (steering angle signal) is generated when the steering shaft rotates in response to the steering wheel turning. From the pulse signal from the steering angle sensor 5, the controller 2 can detect the steering angle θ based on the rising edge / falling edge pattern of the two pulse signals and the time interval between the edges. That is, the steering wheel sensor operation by the driver can be detected by the steering angle sensor 5, and as shown in FIG. 2, it can be detected in which direction the tire is steered with respect to the vehicle body.

  Next, the light control based on the video imaged by the camera 3 will be described below.

(Embodiment 1): Light control based on vanishing point on road (1) When no other vehicle exists in front of the host vehicle, vanishing point Pv on the road surface is obtained on the camera image. The vanishing point Pv is defined as follows. That is, a plurality of parallel projected images of a straight line intersect at one point on the camera image. In this case, a projected image of a point at an infinite distance on the straight line (infinity point) becomes a vanishing point Pv. The vanishing point Pv can be determined by a known method from the internal parameters and external parameters of the camera 3 (camera attachment position and camera angle with respect to the road surface), and detailed description of the determination method is omitted here.
When the vanishing point Pv is obtained, the traveling direction is determined in order for the host vehicle to travel. That is, the steering angle θ (the counterclockwise direction shown in FIG. 2 is positive) is detected by the steering angle sensor 5. When the tire is steered while the vehicle is running, the position of the vanishing point Pv on the road surface moves in the horizontal direction by ua on the camera image. In this case, ua = k · x, x = f · tan θ, k is the number of pixels per unit length, and f is the focal length (see FIG. 3). Next, a direct point P (Xc, Yc, Zc) on the road surface at the center of the lens 12 is obtained for the lens 12 provided inside the camera 3, and the projected point of the direct point P on the plane including the image sensor 11 is obtained. Find Pi. The (X, Y, Z) coordinates here are camera coordinates with the camera lens center as the origin, and are a right-handed system with the optical axis as the Z axis and the horizontal direction as the X axis. A straight line on the image connecting the vanished point Pv and the projected point Pi already determined is determined as shown in FIG. 5, and this straight line is defined as the traveling direction (forward direction) of the vehicle on the image.

  When the vanishing point Pv is obtained by a known method, a monitoring area (size is arbitrary, for example, an area of 160 × 120 pixels or an area of 80 × 60 pixels) 10 is provided around the vanishing point Pv. Based on the luminance value of the monitoring area 10, the light control for the vehicle light 7 is performed. That is, the vehicle light 7 is turned on (on) if a pixel having a small luminance value in the monitoring area 10 provided around the vanishing point Pv is equal to or more than a predetermined pixel. In this case, the threshold value (number of pixels) of the predetermined pixel for turning on the vehicle light 7 is not limited to this. For example, assuming that the vehicle is traveling straight, the vehicle traveling direction (for example, 30 m) It can be set based on the number of pixels corresponding to the size of the tunnel 8 when the tunnel 8 exists first. In this light control, when the tunnel 8 exists in the traveling direction, the lighting timing of the vehicle light 7 is predicted by the controller 2 based on the traveling speed of the own vehicle, based on the traveling speed of the own vehicle. If control is performed to turn on the vehicle light 7 when the estimated time comes after detecting 8, the delay in lighting timing can be prevented.

  The condition for switching the vehicle light 7 from the on state to the off state by turning off the light is that the on condition does not satisfy the on condition continues for a predetermined time and the luminance value in the pixel of the image portion excluding the road surface in the monitoring area 10 is The vehicle light 7 is turned off when a predetermined luminance or more (a situation where the surrounding situation is bright) continues for a certain period of time.

  When the driver steers the steering wheel while driving, the following occurs.

When the steering angle based on the steering angle sensor 5 is steered by θ as shown in FIG. 2 as compared with the straight traveling state, as described above, as shown in FIG. 4, the lens 12 provided in the camera is provided. The point P directly below the center of the lens forms an image at the projection point Pi on the image sensor. When steering from the projection point Pi by θ, the position of the vanishing point Pv moves by ua as shown in FIG. 5, so that the monitoring area 10 of the camera 3 is moved according to the amount of movement, Control of turning on / off the vehicle light 7 is performed.
(2) When another vehicle is present in front of the host vehicle When the presence of another vehicle ahead (indicated by an elliptical broken line in FIG. 5) in the traveling direction on the image is detected by image processing, The monitoring area is moved behind the other vehicle (on the vehicle side) as shown by the arrow in FIG. 5, the road surface behind the other vehicle in the monitoring area is extracted, and the luminance value of the rear road surface is obtained. And if the brightness value of the road surface is below a predetermined brightness value and there are a predetermined number or more of pixels (dark portions), the vehicle light 7 is turned on. In this case, the position of the monitoring area 10 for detecting the surrounding brightness moves the monitoring area 10 from the vicinity of the vanishing point Pv to the rear of the preceding other vehicle. The region 10 is moved again to the position of the vanishing point Pv.

  Next, the write control based on the vanishing point Pv will be specifically described with reference to the flowchart of FIG. In the flowchart shown below, the details of each step of the program will be described according to the numbers at the beginning of the drawing (indicating the step number “S”).

  In the process shown in FIG. 6, first, the steering angle θ is detected based on the signal output from the steering angle sensor 5 in S1, and the forward direction on the image is determined by the movement of the vanishing point Pv based on the steering angle θ. Determine the direction in which the vehicle will travel. In the subsequent process of S2, it is determined whether or not there is another vehicle ahead of the host vehicle. Here, the determination as to whether or not the other vehicle is ahead in the traveling direction is made according to the flowchart shown in FIG.

  In FIG. 7, three camera images at time t, (t + Δt), and (t + 2Δt) are obtained by the camera 3, and the obtained camera images are stored in a predetermined memory (for example, RAM). Then, a point Pv is obtained from the stored image at time t by a known method, and a feature point m (u, v) around the vanishing point Pv including the vanishing point Pv is obtained. Here, u: a row pixel in the camera image and v: a column pixel in the camera image are two-dimensionally shown. In this case, a plurality of feature points m (u, v) can be obtained.

Thereafter, an epipolar line on the image at time t + Δt is obtained. This epipolar line can be obtained as follows. That is, m (u, v): the position of the feature point on the camera image at time t, m ′ (u ′, v ′): the feature point on the camera image at time (t + Δt) m (u, v) If the position of the corresponding point is, the following equation is established.

Then, the epipolar equation is expressed by the following equation.

A: camera internal parameters, f: focal length, ku: number of pixels per unit length in the u-axis direction, kv: number of pixels per unit length in the v-axis direction, (u ₀ , v ₀ ): Assuming that the center of the image, the translation amount of the camera between T: Δt, [T] × R: the outer product of T and R, θx: the angle between the horizontal direction and the camera optical axis direction, and V: the vehicle speed, the following equation is established. .

The matrix component of F is known, and when [Equation 4] is defined and substituted into the epipolar equation of [Equation 2], the epipolar line [Equation 5] can be calculated on the image at time (t + Δt).

  When the epipolar line is calculated by the above method, returning to FIG. 7, the corresponding point of the feature point m (u, v) shown in FIG. 8 on the epipolar line is calculated and obtained by known pattern matching. The corresponding point obtained by pattern matching is m ′ (u ′, v ′). When the corresponding points are obtained, three-dimensional data in real space is obtained from the feature points m (u, v) and the corresponding points m ′ (u ′, v ′) by using a known method (x, y, z). Is calculated. The (x, y, z) coordinates here are camera coordinates with the lens center of the camera as the origin, and are a right-handed system with the optical axis as the Z axis and the horizontal direction as the X axis.

Here, the calculation method of the three-dimensional data in the real space is described. P: camera matrix at time t, P ′: camera matrix at time (t + Δt), A: internal parameters of camera (known), θx: horizontal direction And the angle between the camera optical axis direction and V: vehicle speed, the following matrix can be used.

In this case, if X and M + are set as follows, three-dimensional data can be derived from the relational expression shown in [Equation 9].

  Such a process is repeated and performed for all feature points m (u, v). The corresponding point obtained on the image at time t + Δt is set as a feature point m ′ (u ′, v ′) at time t + Δt.

  By the same method as described above, as shown in FIG. 9, this time, the corresponding point of the feature point m ′ (u ′, v ′) on the epipolar line on the image at time t + 2Δt is calculated and obtained by known pattern matching. The corresponding point obtained by matching is m ″ (u ″, v ″). Once the corresponding points are obtained, the three-dimensional data in the real space is obtained from the feature points m ′ (u ′, v ′) and the corresponding points m ″ (u ″, v ″) using a known method ( x ′, y ′, z ′) is calculated. Such a process is repeated and performed for all feature points m ′ (u ′, v ′).

  Three-dimensional data (x, y, z) obtained between time t and time (t + Δt), three-dimensional data (x ′, y ′, z ′) obtained between time (t + Δt) and time (t + 2Δt) Is obtained. When these two three-dimensional data are obtained, if zz ′ = (VΔt) cos θx is established, it is determined that the feature point does not move over time, and D (u, v) = 0 (stationary state). On the other hand, if y−y ′ = (VΔt) cos θx does not hold, it is determined that the feature point has moved with time, and D (u, v) = 1 (moving state). Thereafter, it is determined whether or not there is a predetermined number of moving points D (u, v) = 1 in the forward direction in which the vehicle travels on the image (ND ≧ threshold NDth), and if this condition is satisfied, there is another vehicle ahead. Then, the CA flag indicating the other vehicle ahead is set (= 1), but if the condition is not satisfied, the CA flag is cleared (= 0), and the corresponding feature point from the epipolar line is displayed from the third screen. The other vehicle ahead is detected.

  When the other front vehicle is recognized in this manner, the process returns to FIG. 6 and when the other front vehicle exists in S2 (CA = 1), a predetermined monitoring area 10 is located behind the front vehicle in S3. And the luminance value in that area is obtained. Then, the number of pixels whose luminance value f (i, j) is equal to or less than a predetermined threshold fth (dark state level) is checked, and it is determined whether there are more than the predetermined number Nth of pixels N with f (i, j) ≦ fth. . In this process, if the above condition is satisfied, the vehicle light 7 is turned on in S4. If the above condition is not satisfied, the state of the vehicle light 7 is currently turned off (off) in S5. It is determined whether or not. If the light is off, the process returns to S1 again. If the light is not off at S5, the process proceeds to S6. In S6, a road surface and a region other than the road surface are distinguished by image recognition, and this time, an average of luminance values in an image portion other than the road surface is calculated. Then, the average luminance value is compared with a predetermined threshold value, and the average luminance value of the image portion other than the road surface is the other vehicle ahead and is equal to or less than the predetermined threshold value (fm (i, j) ≦ fmth) for determining lighting. Returns to the process of S1 again. However, if this does not hold, it is determined in S7 whether this state has continued for a predetermined time (t ≧ tc1). If this condition does not continue for a predetermined time in S7, the process returns to S1 again, but if it continues for a predetermined time, it is determined that the surroundings other than the road surface are bright, and the light is turned off and turned off in S8. After that, the processing from S1 is repeated.

  On the other hand, when there is no other vehicle ahead in S2 (CA = 0), a pixel whose luminance value is equal to or less than a predetermined threshold (f (i, j) ≦ fth) in the pixels around the vanishing point in S9. It is determined whether or not there is a predetermined number or more (N ≧ Nth). Here, if the above condition is satisfied, it is determined in S15 whether or not it has been in this state last time, and in this state, the process returns to S1. On the other hand, if it was not this state last time in S15, the vehicle light 7 is turned on after TL time in S16 (SW = 1). In this case, lighting the light after TL time works as follows. That is, when a tunnel exists in the traveling direction, the controller 2 predicts the time until the host vehicle enters the tunnel, and after detecting the tunnel, the vehicle light 7 can be turned on when the predicted time is reached. Therefore, the lighting timing can be determined based on the traveling speed of the vehicle, and the lighting timing of the vehicle light 7 can be made appropriate.

  On the other hand, in S9, if there are no more than a predetermined number of pixels whose luminance value is equal to or less than the predetermined threshold (f (i, j) ≦ fth) in the pixels around the vanishing point, the current light state is determined in S10. It is determined whether the light is off or not. Here, if the light is off, the process returns to S1. If the light is on, it is determined in S11 whether the light has continued for a predetermined time (t ≧ tc2). Here, when the lighting of the vehicle light 7 has not continued for a predetermined time, the process returns to S1, but when the vehicle light 7 has continued for a predetermined time in S11, the luminance value in the image portion other than the road surface is now changed. It is determined whether the average value is equal to or less than a predetermined threshold value (fm (i, j) ≦ fmth). If the above condition is satisfied in S12, the process returns to S1. If the above condition is not satisfied, it is determined in S13 whether this state has continued for a predetermined time (t ≧ tc3). When the condition shown in S12 does not continue for a predetermined time, the process returns to S1. However, if the condition shown in S12 continues for a predetermined time in S13, it is determined that the lighting is not necessary and the surrounding situation is sufficiently bright, and the vehicle light 7 is turned off in S14. The process is repeated at a predetermined cycle.

(Second Embodiment)
In the first embodiment, the light control is performed by obtaining the vanishing point Pv on the camera image and comparing the luminance around the vanishing point Pv. However, in the second embodiment, even if an epipole is used instead of the vanishing point Pv, the write control can be performed as shown in FIG. The process shown in FIG. 10 is basically the same as that in FIG. 6, but only S9 is different. In S9, not the vanishing point Pv but the brightness value is compared for pixels around the epipole, which is different from FIG. ing.

(Third embodiment)
FIG. 11 shows a flowchart of the third embodiment. In the third embodiment, write control is performed by optical flow. The difference in this process is that the optical flow is first calculated at the feature points on the image, the point Of having the smallest distance is obtained from each optical flow, and the same comparison as S9 shown in FIG. 6 is performed around the point Of. The point to do is different.

  The point Of is obtained as follows. In other words, at each feature point on the camera image shown in FIG. 12, a point Of at which the distance from a straight line drawn by each optical flow (for example, an arrow) is minimized is obtained as shown in FIG. Hereinafter, a method for deriving the optical flow will be described.

(X, y): position on the camera image, f (x, y): luminance value, fx, fy, ft: partial differentiation in the horizontal direction, vertical direction, and time axis direction, the following expression is obtained.

Is set as a predetermined value and is substituted into the following equation, and is repeated a predetermined number of times from k = 0 to n, and finally obtained [Expression 13] is defined as an optical flow.

  The point Of is derived by the method described above. A direct point P (Xc, Yc, Zc) on the road surface at the center of the lens is obtained, and a projection point Pi of the direct point P is obtained on a plane including the image sensor. Then, a straight line on the image connecting the already-obtained point Of and the projected point Pi is determined as shown in FIG. 14, and this straight line becomes the forward direction of the host vehicle on the camera image.

(Fourth embodiment)
FIG. 15 shows a flowchart of the fourth embodiment. In the process shown in FIG. 15, a plurality of tunnel templates shown in FIG. 16 are stored in the memory in the tunnel controller, and the tunnel in the traveling direction is detected by pattern matching using the stored template, and the light control is performed. Is to do. This process is different in S9 shown in FIG. That is, in S9, a plurality of tunnel images are detected by pattern matching around the vanishing point (or epipole). In this case, the write control is performed depending on whether there is a similarity higher than a predetermined threshold.

  As described above, according to the first to fourth embodiments, there is no need for a mechanism for moving the optical axis of the camera according to the traveling state of the vehicle, and the situation around the vehicle according to the traveling state of the vehicle. Thus, a device capable of turning on the vehicle light at an appropriate timing can be obtained. In addition, it is possible to provide a device that can be lit without causing the driver to feel uncomfortable.

  When this embodiment is associated with the claims, the controller 2 is control means, feature point extraction means, traveling direction extraction means, vehicle detection means, tunnel / dark place detection means, and the camera 3 is imaging means, vehicle speed. The sensor 4 is a speed detection unit, the steering angle sensor 5 is a steering detection unit, and the template 9 is a tunnel detection unit. Further, the vanishing point Pv and the point Of due to the optical flow are the first points.

It is a block diagram which shows the structure of the system of the light control apparatus in one Embodiment of this invention. It is explanatory drawing which shows the relationship between the steering angle of a vehicle, and a vehicle body. It is explanatory drawing which shows the relationship of the moving amount | distance on the image pick-up element which changes according to the steering angle shown in FIG. 2, and a steering angle. It is explanatory drawing which shows the relationship between the direct point P on a road surface, and the projection point Pi of an image pick-up element. It is explanatory drawing of the camera image which shows that the vanishing point ahead of a vehicle moves, when steering operation of steering angle (theta) is performed. It is a 1st flowchart (write control by vanishing point) in the write control of this invention. It is a flowchart which recognizes presence of other vehicles ahead. It is explanatory drawing in the case of calculating three-dimensional data from the feature point of the camera image in the time t and the time (t + (DELTA) t). It is explanatory drawing in the case of calculating three-dimensional data from the feature point of the camera image at the time (t + Δt) and the time (t + 2Δt). It is a 2nd flowchart (write control by epipole) in the write control of this invention. It is a 3rd flowchart (Light control by an optical flow) in the light control apparatus of this invention. It is explanatory drawing of the optical flow shown in FIG. It is explanatory drawing which shows the straight line group and point which an optical flow draws. It is explanatory drawing which shows the relationship between the projection point Pi of the direct point P on a road surface, and the point shown in FIG. It is a 4th flowchart (Light control by pattern matching of a tunnel) in the light control apparatus of this invention. It is explanatory drawing which shows the pattern matching by the template of a tunnel.

Explanation of symbols

1 vehicle light control device 2 controller (control means, feature point extraction means, traveling direction extraction means, vehicle detection means, tunnel / dark place detection means)
3 Camera (imaging means)
4 Vehicle speed sensor (speed detection means)
5 Steering angle sensor (steering detection means)
7 Vehicle light 8 Tunnel 9 Template (tunnel detection means)
10 Monitoring Area 11 Image Sensor 12 Lens Pv Vanishing Point (First Point)
Of optical flow point (first point)

Claims (6)

In the vehicle light control device for controlling lighting of the vehicle light based on an image of the traveling direction of the vehicle photographed by the imaging means, the control means for automatically lighting the vehicle light,
Feature point extracting means for obtaining a feature point imaged on the image sensor of the imaging means from the video, and a traveling direction extracting means for obtaining a first point indicating the traveling direction of the vehicle on the video or image from the feature point. The vehicle light control device is characterized in that the control means changes the monitoring area based on the information based on the first point, and turns on the vehicle light based on the luminance in the monitoring area. The vehicle light control device according to claim 1, further comprising vehicle detection means for detecting the presence of a vehicle in a traveling direction from the feature point, and changing the monitoring area depending on the presence or absence of a vehicle in the traveling direction. When there is a vehicle in the traveling direction, the monitoring area is moved on the road surface located behind the vehicle, and the number of pixels whose luminance value in the monitoring area on the image is equal to or less than a predetermined threshold is equal to or more than a predetermined pixel The vehicle light control device according to claim 2, wherein the control means turns on the vehicle light based on a traveling speed. When there is no vehicle in the traveling direction, the monitoring area is moved around the first point, and when the number of pixels whose luminance value in the monitoring area is equal to or less than a predetermined threshold is equal to or greater than a predetermined pixel, based on the traveling speed The vehicle light control device according to claim 2, wherein the control means turns on the vehicle light. The vehicle light control device according to any one of claims 2 to 4, further comprising: a steering detection unit that detects a steering angle of the vehicle, and changes a monitoring region from the first point based on the steering angle. A speed detecting means for detecting the traveling speed of the vehicle, and a tunnel / dark place detecting means for detecting a tunnel or dark place existing in the traveling direction of the vehicle, wherein the control means travels when the tunnel / dark place is detected. The vehicle light control device according to claim 2, wherein the vehicle light is turned on based on a speed.

Images