JP4917961B2 - In-vehicle vehicle light detection device and vehicle illumination device - Google Patents

Description

  The present invention relates to an in-vehicle vehicle light detection device and a vehicle illumination device.

There is a technique of detecting another vehicle in front of the host vehicle and switching the headlamp of the host vehicle to a high beam (for traveling) and a low beam (for passing) (Patent Documents 1, 2, etc.).
In this type of technology, FIGS. 10, 11, 12, and 13 are known as arrangements in an image sensor with a color filter. In other words, in each pixel arrayed vertically and horizontally, as shown in FIG. 10, the “up and down arrangement” is made such that the red filter is on the top and the filter is not on the bottom, each pixel is made a Bayer type as shown in FIG. Are arranged in a checkered pattern, or filters are arranged only in an arbitrary range of the image sensor as shown in FIG.

In Patent Document 1, a light source that seems to be red based on the grayscale ratio of each pixel of red and other color filters or only the red filter, with a vertical arrangement, a Bayer type arrangement, and a checkered arrangement as shown in FIGS. Is extracted and determined as a taillight (preceding vehicle). In Patent Document 2, a red light source is extracted from a general YUV color specification and determined as a taillight (preceding vehicle).
JP 2005-534903 A JP 2005-92857 A

  In Patent Document 1, the filter setting is performed in consideration of the spectral characteristics of the vehicle light. The gradation value of each pixel of red and other color filters is compared, and if the ratio or gradation value is equal to or greater than the threshold value, the red color is set. As such, it is considered that the performance is inferior simply by determining the taillight (leading vehicle). Further, although a configuration in which the red filter is arranged twice is proposed, it is considered that there is no specific clear means and the ratio of the red filter is simply large, and the specificity is lacking.

  Although Patent Document 2 describes a countermeasure against saturation, it merely extracts red from the YUV color determination, and does not reflect the spectral characteristics of the vehicle light, and the performance is inferior as the determination. It lacks concreteness.

  The present invention has been made paying attention to the above-mentioned problems, and an object thereof is to provide an in-vehicle vehicle light detection device and a vehicle illumination device that can detect a vehicle light with high accuracy.

In order to solve the above-described problem, in the invention described in claim 1, an image sensor having an array of pixels, an image sensor that captures a traveling direction in front of the vehicle, and an image acquired by the image sensor are processed. A vehicle light detection device comprising: vehicle light detection means for determining whether the light source is a taillight of a preceding vehicle or a headlight of an oncoming vehicle and detecting a taillight of a preceding vehicle or a headlight of an oncoming vehicle. In the image sensor of the image sensor, at least three types of optical filters having different wavelength ranges to be passed are arranged on the pixel array, and the vehicle light detection means includes at least three types of optical filters in the image sensor of the image sensor. The same tendency by the gradation value change tendency determination function and the gradation value change tendency determination function that determine whether the change tendency of the gradation value of the pixels marked with the same tendency as the taillight of the preceding vehicle or the headlight of the oncoming vehicle Is determined to indicate A symmetry determination function for determining whether the light source has left-right symmetry and a light source that has been determined to have left-right symmetry by the symmetry determination function from the preceding vehicle from at least one of the moving direction and the moving amount in successive images And a tracking function for determining whether it is a headlight of an oncoming vehicle or a headlight of an oncoming vehicle .

According to the first aspect of the present invention, at least three types of optical filters having different wavelength ranges to be passed are arranged on the pixel array in the image sensor of the image sensor, and the gradation value of the vehicle lamp detection means By the change tendency determination function, it is determined whether the change tendency of the gradation value of the pixel to which the image sensor of the image sensor is attached has the same tendency as the tail light of the preceding vehicle or the headlight of the oncoming vehicle. . Accordingly, it is determined whether the light source is the taillight of the preceding vehicle or the headlight of the oncoming vehicle, and the taillight of the preceding vehicle or the headlight of the oncoming vehicle is detected. Therefore, it is possible to detect the vehicle light, that is, the tail light of the preceding vehicle or the headlight of the oncoming vehicle with high accuracy in consideration of the spectral characteristics of the vehicle light as the light source and other light sources. Further, the symmetry determination function for determining whether or not the light source determined to exhibit the same tendency by the gradation value change tendency determination function has left-right symmetry makes the vehicle lamp more excellent in detectability. Furthermore, for a light source determined to have left / right symmetry by a symmetry determination function, a tracking function for determining whether the light is a taillight of a preceding vehicle or a headlight of an oncoming vehicle from a moving direction and / or a moving amount in successive images. This makes it easier to detect vehicle lights.
It will be.

  As described in claim 2, in the in-vehicle vehicle light detection device according to claim 1, the light transmitted by at least three types of optical filters is at least one of blue light, green light, red light, and near infrared light. It is good that there are three.

  According to a third aspect of the present invention, in the in-vehicle vehicle light detection device according to the first or second aspect, the vehicle light detection means has a function of the imaging device as a processing function prior to the processing by the gradation value change tendency determination function. The other wavelength range light gradation that calculates the gradation value in the corresponding pixel for the light in the other wavelength range with respect to the light in the wavelength range detected by the corresponding pixel from the gradation value in the surrounding pixel for the corresponding pixel. For each pixel in the image sensor, the brightness at the corresponding pixel is calculated from the gradation value detected by the corresponding pixel and the gradation value calculated by the other wavelength range light gradation value calculation function. A luminance calculation function to calculate, and a light source candidate determination function to determine whether or not the luminance by the luminance calculation function in each pixel in the image sensor is greater than a threshold value, and to determine that the luminance is a light source candidate if larger More vehicle light detection And it is excellent in.

The invention according to claim 4 comprises the on-vehicle vehicle light detection device according to any one of claims 1 to 3 , and the detection of the taillight of the preceding vehicle or the headlight of the oncoming vehicle by the vehicle light detection means. The gist of the present invention is a vehicular illumination device including a headlamp control means for controlling at least one of switching and light distribution of a vehicle headlamp based on the result.

According to the invention described in claim 4 , the headlamp control means switches and distributes the headlamps of the vehicle based on the detection result of the taillight of the preceding vehicle or the headlamp of the oncoming vehicle by the vehicle lamp detection means. At least one of them is controlled.

  Note that the switching of the headlamp means switching between the headlight for passing and the headlight for traveling, and the light distribution means changing the irradiation distance and direction by changing the optical axis of the irradiation light, or irradiation. Is to adjust the amount.

DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, an embodiment of the invention will be described with reference to the drawings.
FIG. 1 shows an electrical configuration of the vehicular lighting device according to the present embodiment.
The vehicular lighting device includes a left headlamp 10 and a right headlamp 20. The left headlamp 10 installed on the left side of the front of the vehicle illuminates the front left side of the vehicle, and includes a traveling lamp 11, a passing lamp 12, and a drive control unit 13. The traveling lamp 11 is a so-called high beam lamp. The passing lamp 12 is for a so-called low beam. The drive control unit 13 controls the driving of the traveling lamp 11 and the passing lamp 12 (controls the lamp applied voltage). Similarly, the right headlamp 20 installed on the front right side of the vehicle illuminates the front right side of the vehicle, and includes a travel lamp 21, a passing lamp 22, and a drive control unit 23. The so-called high beam is used, and the passing lamp 22 is a so-called low beam. The drive control unit 23 controls the driving of the traveling lamp 21 and the passing lamp 22 (controls the lamp applied voltage).

The vehicle lighting device includes a headlamp control device 30. The headlamp control device 30 includes a light distribution determination / control unit 32 and a swivel / leveling determination unit 31.
The headlight driving mechanisms 41 and 42 are composed of a leveling mechanism that adjusts the optical axis in the vertical direction and a swivel mechanism that changes the irradiation range and direction by moving the optical axis or the headlamp itself in the horizontal direction. It is driven by. Switching between the driving lamps 11 and 21 and the passing lamps 12 and 22 by controlling the drive control units 13 and 23 of the headlamps 10 and 20 using the headlamp control device 30 (light distribution determination / control unit 32). Etc. can be performed. Further, the headlamp control mechanism 30 (swivel / leveling determination unit 31) is used to control the headlamp driving mechanisms 41 and 42 so that the optical axis of the headlamps 10 and 20 is adjusted by the leveling mechanism. The irradiation range and direction can be adjusted by the mechanism.

  The headlight control device 30 includes a display 43, a vehicle-mounted vehicle light detection device 44, a headlight operation switch 45, an engine start device 46, a wiper device 47, a vehicle speed detection device 48, a steering angle detection device 49, a vehicle state. A detection device 50 and a road information providing device (navigation) 51 are connected.

The indicator 43 informs the driver of the currently used headlamp state, and is arranged as an indicator on an instrument panel or the like.
The vehicle-mounted vehicle light detection device 44 is a device that determines a vehicle light with a forward visual field that is the traveling direction. Details will be described later.

  The headlamp operation switch 45 is normally disposed in the vicinity of the steering so that the driver can perform a manual switching operation. The operation mode is composed of a light-off mode, a passing mode, a traveling mode, and an automatic mode. When the automatic mode is set, the vehicle-mounted vehicle light detection device 44 is used.

  The engine starter 46 outputs a start / stop signal when starting the engine. The wiper device 47 outputs a wiping speed signal. The vehicle speed detection device 48 outputs a vehicle speed signal. The steering angle detection device 49 outputs a steering state signal, and the vehicle state detection device 50 includes a yaw rate sensor, a tilt sensor, and the like, and outputs a vehicle state signal. The road information providing device (navigation) 51 outputs an information signal such as a road shape. The headlight control device 30 and the vehicle-mounted vehicle light detection device 44 acquire these signals (start / stop signal, wiping speed signal, vehicle speed signal, steering state signal, vehicle state signal, navigation signal, etc.).

A specific configuration of the in-vehicle vehicle light detection device 44 in FIG. 1 is shown in FIG.
In FIG. 2, the vehicle-mounted vehicle light detection device 44 includes an image sensor 59 and a microcomputer 60. The image sensor 59 includes an optical lens 61, an image sensor 62, an optical filter 63, and a circuit unit 64. An optical filter 63 is attached to the image sensor 62. The image sensor 62 with the optical filter 63 and the optical lens 61 are installed in the front part of the room mirror. The image sensor 62 is composed of a CCD or a CMOS, and images the traveling direction ahead of the vehicle through the optical lens 61. The circuit unit 64 includes an A / D conversion circuit and a timing circuit, and converts the light received by the image sensor 62 into electrical information (A / D conversion).

  Functionally, the microcomputer 60 includes an image input unit 65, an image processing unit 66, a calculation determination processing unit 67, and a drive control unit 68. The drive control unit 68 drives the image sensor 59. The image input unit 65 acquires an image after A / D conversion by the circuit unit 64 of the image sensor 59. The image processing unit 66 processes the image after A / D conversion and analyzes the image. The calculation determination processing unit 67 determines whether the light source is a vehicle light, that is, a taillight of a preceding vehicle or a headlight of an oncoming vehicle from data after image processing when the automatic mode is set. Or the headlamp of an oncoming vehicle is detected, and the headlamp of the own vehicle is controlled based on this result.

Details of the image sensor 62 and the optical filter 63 of FIG. 2 are shown in FIG.
3A is a plan view showing a part of the image sensor, FIG. 3B is a schematic longitudinal sectional view taken along line AA in FIG. 3A, and FIG. It is a schematic longitudinal cross-sectional view in the -B line.

  In FIG. 3, a large number of light receiving elements 81 are arranged vertically and horizontally on the upper surface of a substrate 80. A red filter 85 that selectively transmits red light (wavelength λ is 630 ± 50 nm) is disposed on the light receiving element 81. A green filter 84 that selectively transmits green light (wavelength λ is 530 ± 50 nm) is disposed on another light receiving element 81. A blue filter 83 that selectively transmits blue light (wavelength λ is 430 ± 50 nm) is disposed on another light receiving element 81. On the other light receiving element 81, a near infrared filter 82 that selectively transmits light in the near infrared region (light having a wavelength λ of 730 nm or more) is disposed.

  In this way, with respect to the arrangement of the image pickup device with a filter, the planar configuration of FIG. 3A is a pixel with a near infrared filter as indicated by IR, a pixel with a blue filter as indicated by B, and a pixel with a green filter as indicated by G. , Pixels with red filter are arranged as indicated by R. The arrangement is described with reference to FIG. 3A. A near-infrared filter-equipped pixel (IR) is arranged on the upper left, and a blue-filter-equipped pixel (B) is arranged on the right side in the following order. IR) and blue filter-equipped pixels (B) are alternately arranged. The arrangement in the second column from the top in FIG. 3A will be described. A pixel with a green filter (G) is arranged on the leftmost side, and a pixel with a red filter (R) is arranged on the right side with a green filter. Pixels (G) and red filter-equipped pixels (R) are alternately arranged. The arrangement in the third column from the top in FIG. 3A will be described. Similarly to the first column, the pixels with near infrared filter (IR) and the pixels with blue filter (B) are alternately arranged. The arrangement in the fourth column from the top in FIG. 3A will be described. Similarly to the second column, the pixels with green filter (G) and the pixels with red filter (R) are alternately arranged. The arrangement is the same in the following.

  For the four types of band-pass filters (optical filters) having different transmission bands, the four types of pass bands are selected in consideration of the spectral characteristics and visibility of vehicle lights and lighting. Further, the types are not limited, and may be four or more types or three types.

  In this way, the image sensor 59 has the imaging element 62 in which pixels are arranged, images the traveling direction in front of the vehicle, and allows the imaging element 62 to pass through a wavelength range (transmission) on the pixel arrangement. 4 types of optical filters 82 to 85 having different bands), and in a broad sense, at least three types of optical filters are arranged. The transmission band of the optical filter is set in consideration of the spectral characteristics of the vehicle light and other light sources. Specifically, the light transmitted through at least three types of optical filters (82 to 85) is blue light and green light. And at least three of red light and near infrared light.

Here, the image example of FIG. 8 is demonstrated.
FIG. 8 is an example of an image when the vehicle-mounted vehicle light detection device 44 (image sensor 59 or the like) is mounted at the front of the rear mirror (front windshield side).

  In FIG. 8, there are vehicle lights (a headlamp 94a of the oncoming vehicle 94 and a taillamp 93a of the preceding vehicle 93) as light sources. The headlamp 94a of the oncoming vehicle 94 is a white lamp, and a halogen lamp or the like is used. The tail lamp 93a of the preceding vehicle 93 is a red lamp. In FIG. 8, reference numeral 95 denotes a vanishing point of the road in the image.

Next, the operation of the vehicle lighting device will be described.
A signal for starting the image sensor 59 is output from the microcomputer 60 of FIG. 2, and a forward image in the traveling direction is captured by the image sensor 59. Image processing is performed on the image in the image processing unit 66 and an arithmetic determination processing unit In 67, detection of vehicle lights in the surrounding vehicles and control processing of the headlights of the own vehicle are performed. This process will be described according to the flow shown in FIGS.

  In FIG. 4, the microcomputer 60 captures an image in step 100. In step 101, the microcomputer 60 calculates the gradation value of each pixel based on the image data captured by the image sensor 59. Tone values other than the color (wavelength range) in charge (for example, the tone value for blue and the tone value for green in the pixel with the red filter) are determined from the tone values of the surrounding pixels, for example, adjacent pixels. It is calculated from the gradation ratio between them. In other words, the gradation value at the corresponding pixel for light in the other wavelength region relative to the light in the wavelength region detected by the corresponding pixel is calculated from the gradation value at the surrounding pixel with respect to the corresponding pixel in the image sensor 62. To do.

  Specifically, as shown in FIG. 7, in the pixel with a red filter, as shown by reference numeral R22, as an operation of other gradations in the pixel, the outputs (tone values B12, From B32), a gradation value B22 is calculated when it is assumed that a blue filter is attached with an arithmetic average (= (B12 + B32) / 2). In addition, as indicated by reference numeral R22, in the pixel with a red filter, as an operation of other gradations in the pixel, an arithmetic average (= () is obtained from outputs (tone values G21 and G23) of the left and right green filter pixels as peripheral pixels. The gradation value G22 when it is assumed that the green filter is attached in G21 + G23) / 2) is calculated.

  Further, as shown in FIG. 7, as indicated by reference numeral G <b> 43, the gradation value B <b> 32 of the blue filter pixel located adjacent to the diagonal direction as the peripheral pixel is calculated as another gradation in the pixel with the green filter. , B34, B52, and B54, a gradation value B43 is calculated when it is assumed that a blue filter is attached with an arithmetic average (= (B32 + B34 + B52 + B54) / 4). Further, as indicated by reference numeral G43, in the pixel with the green filter, as an operation of other gradations in the pixel, an arithmetic average (= (R42 + R44) is obtained from the gradation values R42 and R44 of the pixels with the red filter located on the left and right as the peripheral pixels. ) / 2), the gradation value R43 when the red filter is assumed is calculated.

  Further, in step 101 of FIG. 4, the microcomputer 60 determines, for each pixel in the image sensor 62, the gradation value detected by the corresponding pixel and the gradation value calculated as described above, at the corresponding pixel. Calculate the brightness. Specifically, the luminance (brightness) Y of each pixel is calculated by the following equation.

Y = 0.299 × R + 0.587 × G + 0.114 × B
Here, R is a gradation value for red, G is a gradation value for green, and B is a gradation value for blue.

  Next, in step 102 of FIG. 4, the microcomputer 60 determines whether or not the luminance Y obtained from the gradation value of each pixel is greater than a predetermined light source extraction threshold. As a result, when the luminance Y obtained from the gradation value of each pixel is smaller than the threshold, the microcomputer 60 proceeds to step 103 and determines that there is no light source or disturbance.

  On the other hand, if the luminance Y obtained from the gradation value of each pixel in step 102 is larger than the threshold value, the microcomputer 60 determines that a light source candidate has been extracted (determines that it is a light source candidate). In this way, as shown in FIG. 9, light source candidates (the headlamp 94a of the oncoming vehicle 94 and the taillamp 93a of the preceding vehicle 93 in FIG. 8) are extracted from the image.

  When the light source candidates are extracted, the microcomputer 60 executes steps 104 and 105 in FIG. 4 to calculate the average luminance of the extracted light source candidates, and uses the average luminance of the light source candidates as the first threshold value for vehicle light detection and Compare with the second threshold. If the average luminance of the light source candidate is smaller than the first threshold value in step 104, the microcomputer 60 compares the average luminance of the light source candidate with the second threshold value in step 105.

  If the average luminance of the light source candidates is greater than the first threshold value in step 104, the microcomputer 60 proceeds to step 106 assuming that there is a high possibility that the light source candidate light source is a vehicle light candidate light source, particularly a headlight or a short-distance taillight. If the average luminance of the light source candidates is smaller than the second threshold value in step 105, the microcomputer 60 proceeds to step 103 and determines that there is no light source or there is a disturbance.

  In step 106, the microcomputer 60 compares the luminance (gradation value) of each light source extracted as a candidate with the headlamp determination value (whiteness value), and the luminance of the light source is larger than the headlamp determination value. Then, the microcomputer 60 determines that the headlamp has been extracted as a light source candidate, proceeds to step 107, and determines the wavelength range of the extracted candidate. That is, in step 107, the microcomputer 60 determines whether the change tendency of the gradation value of the pixel attached with the optical filters 82 to 85 in the image sensor 62 of the image sensor 59 shows the same tendency as the headlight of the oncoming vehicle (four pieces). The change tendency of each wavelength region is analyzed from the gradation value of the, and compared with the change tendency of the headlight, it is determined whether the same tendency is exhibited). Specifically, for example, when B <G <R <IR is satisfied, a headlamp (halogen lamp) is estimated. However, B is a gradation value in a pixel with a blue filter, G is a gradation value in a pixel with a green filter, R is a gradation value in a pixel with a red filter, and IR is a gradation value in a pixel with a near-infrared filter. Value. That is, the gradation value (B) of the pixel with blue filter, the gradation value (G) of the pixel with green filter, the gradation value (R) of the pixel with red filter, and the gradation value (IR) of the pixel with near infrared filter ) In order, it is estimated to be a headlamp (halogen lamp). As a result, the headlamp is extracted as a light source candidate from the color (wavelength range) of the light source. The change tendency is selected in consideration of the spectral (light emission) characteristics of the vehicle light, the spectral sensitivity characteristics of the image sensor, the spectral transmission characteristics of the filter, and the like.

  Then, in step 107 of FIG. 4, when the change tendency of the gradation value of the pixel with the optical filters 82 to 85 in the image sensor 62 shows the same tendency as the headlight of the oncoming vehicle (for example, B <G <R When <IR is satisfied), the microcomputer 60 executes the process of FIG.

  On the other hand, if the luminance of the light source is greater than the second threshold value in step 105 of FIG. 4, the microcomputer 60 proceeds to step 112 and compares the luminance (gradation value) of the light source with the taillight determination value (redness value). . As a result, if the luminance of the light source is greater than the taillight determination value, the microcomputer 60 determines that the taillight has been extracted as a light source candidate, proceeds to step 113, and determines the wavelength range of the extracted candidate. That is, in step 113, the microcomputer 60 determines whether the change tendency of the gradation value of the pixel attached with the optical filters 82 to 85 in the image sensor 62 of the image sensor 59 shows the same tendency as the taillight of the preceding vehicle (four floors). Analyzing the change tendency of each wavelength range from the tone value and comparing it with the tail lamp change tendency, it is determined whether the same tendency is shown). Specifically, for example, when B≈0 and G <R are satisfied, a taillight (red light) is estimated. However, B is a gradation value in a pixel with a blue filter, G is a gradation value in a pixel with a green filter, and R is a gradation value in a pixel with a red filter. That is, when the gradation value (B) of the pixel with the blue filter is substantially “0” and the gradation value (R) of the pixel with the red filter is larger than the gradation value (G) of the pixel with the green filter. Estimated as a taillight (red light). As a result, the taillight is extracted as a light source candidate from the color (wavelength range) of the light source.

  Then, in step 113 of FIG. 4, when the change tendency of the gradation value of the pixel to which the optical filters 82 to 85 are attached in the image sensor 62 shows the same tendency as the taillight of the preceding vehicle (for example, B≈0 and G <R If the condition is satisfied, the microcomputer 60 executes the process of FIG.

  5 and 6, in steps 108, 109, 111, 114, 115 and 117, the symmetry (vehicle lights are generally a pair of left and right) is extracted from the surrounding light sources showing the same characteristics, Further, the vehicle light is determined by tracking the light source from a plurality of consecutive images and determining the direction and amount of movement. That is, the positional relationship between the light sources determined as candidates (refers to whether or not there is a pair on the left and right, and the normal vehicle lights are a pair on the left and right), the moving direction and the amount are determined, If it matches the amount, it is determined as a vehicle light.

  Specifically, the microcomputer 60 determines whether or not the light source selected in step 108 in FIG. 5 has a pair of right and left as a positional relationship and the light source is symmetric. It is determined whether or not the movement amount is close to the assumed value. As a result, if the moving direction and moving amount of the light source are close to the assumed values, the microcomputer 60 determines in step 110 that the light source is the headlight of the oncoming vehicle. Thereby, the headlamp of the oncoming vehicle is detected. On the other hand, if there is no symmetry of the light source in step 108, the microcomputer 60 proceeds to step 111 and determines whether or not the moving direction and moving amount of the light source are close to the assumed values. As a result, if the moving direction and moving amount of the light source are close to the assumed values, the microcomputer 60 determines in step 110 that the light source is the headlight of the oncoming vehicle. If the moving direction and moving amount of the light source are not close to the values assumed in steps 109 and 111, the microcomputer 60 proceeds to step 103 in FIG. 4 and determines that there is no light source or a disturbance.

  Further, in step 114 of FIG. 6, the microcomputer 60 determines whether or not there is a pair of left and right symmetry as a positional relationship with respect to the candidate light source, and if there is symmetry, the moving direction and amount of movement of the light source in step 115. Is determined to be close to the assumed value. As a result, if the moving direction and moving amount of the light source are close to the assumed values, the microcomputer 60 determines in step 116 that the light source is the taillight of the preceding vehicle. As a result, the taillight of the preceding vehicle is detected. On the other hand, if there is no symmetry of the light source in step 114, the microcomputer 60 proceeds to step 117 and determines whether or not the moving direction and moving amount of the light source are close to the assumed values. As a result, if the moving direction and moving amount of the light source are close to the assumed values, the microcomputer 60 determines in step 116 that the light source is the taillight of the preceding vehicle. If the moving direction and amount of the light source are not close to the values assumed in steps 115 and 117, the microcomputer 60 proceeds to step 103 in FIG. 4 and determines that there is no light source or a disturbance.

  When the microcomputer 60 determines the positional relationship (symmetry) or the movement amount in FIGS. 5 and 6, other vehicle signals (the vehicle speed by the vehicle speed detection device 48 in FIG. 1, the steering angle by the steering angle detection device 49, The inclination may also be taken in by the vehicle state detection device 50).

  If the brightness of the light source is smaller than the headlamp determination value in step 106 of FIG. 4 or if the change in the gradation value of the pixel does not indicate the tendency of the headlamp in step 107, the microcomputer 60 proceeds to step 112. Then, the brightness (gradation value) of each light source is compared with the taillight value (redness) to determine whether the light source is a taillight. When the luminance of the light source is smaller than the taillight determination value at step 112 or when the change tendency of the gradation value of the pixel does not indicate the tendency of the taillight at step 113, the microcomputer 60 proceeds to step 103 and the light source is It is determined that there is no disturbance.

  Note that the threshold values in steps 104 and 105 in FIG. 4 may reflect the light source size, coordinate position, and the like (for example, threshold = constant × size). Alternatively, each pixel may be compared with another threshold value to calculate the number of pixels that are equal to or greater than the threshold value, and may be extracted as a light source candidate if the number of pixels is equal to or greater than the determination value. In this case, the size and position of the light source may be reflected in the determination value.

  Further, in steps 109, 111, 115, and 117 of FIGS. 5 and 6, the movement direction and movement amount of the light source having left and right symmetry are determined from the continuous images. However, the movement direction and movement are determined from the continuous images. What is necessary is just to determine at least one of quantity.

  If the microcomputer 60 determines that there is another vehicle (headlight, taillight) ahead of the host vehicle, the microcomputer 60 outputs a passing mode signal as shown in FIG. With this signal, irradiation with a passing headlamp is performed via the headlamp control device 30 in FIG. If the microcomputer 60 determines that there is no other vehicle (headlight, taillight), it outputs a travel mode signal as shown in FIG. With this signal, irradiation with the traveling headlamp is performed via the headlamp control device 30 in FIG. 1 (the beam is turned into a high beam).

  In this way, the visibility of the driver can be improved by optimally controlling the headlamp of the host vehicle. For example, the driving headlamp is used in an environment where there are few street lights, etc. To prevent dazzling, use a headlight for passing when there is a preceding or oncoming vehicle.

  In addition, although switching of the headlamps 10 and 20 was controlled, the light distribution of the headlamps 10 and 20 may be controlled. Alternatively, both switching control of the headlamps 10 and 20 and light distribution control may be performed. That is, at least one of switching of the headlamps 10 and 20 and light distribution of the vehicle is controlled based on the determination result of the presence or absence of another vehicle. Here, “switching of the headlamps” when controlling at least one of switching and light distribution of the vehicle headlamps 10 and 20 is switching between the headlamp for passing and the headlamp for traveling. “Light distribution” is to change the irradiation distance and direction by changing the optical axis of the irradiated light and to adjust (dimming) the irradiation amount. These can be performed by using a variable light distribution mechanism (AFS; Adaptive Front-Lighting System) or an optical axis adjustment mechanism (auto-leveling) introduced in recent years.

  When the microcomputer 60 controls at least one of switching and light distribution of the headlamps 10 and 20 of the vehicle, the microcomputer 60 sets the control speed (switching speed and light distribution speed) of the vehicle traveling information (vehicle speed, steering angle, You may make it change based on a yaw rate, the inclination of a vehicle, etc.). Specifically, using the information indicating the vehicle state from the vehicle speed detection device 48, the steering angle detection device 49, the vehicle state detection device 50, etc., the time until the headlamp is switched may be made variable. When the vehicle is traveling at high speed, the switching is performed quickly, and when the vehicle is traveling along a curve, the switching is performed slowly. This is preferable in optimizing the control speed of the vehicle headlamp.

  Further, as shown in FIG. 9, the microcomputer 60 of the in-vehicle vehicle light detection device 44 determines the brightness of the road surface from the luminance value of the arbitrary range Z1 (lower position as the road surface range) of the captured image data, The light distribution of the headlamp is controlled via the headlamp control device 30 according to the determined brightness. In determining the brightness of the road surface from this luminance value, determination is made in consideration of the visibility correction (G> B≈R gradation value) and the characteristics (preset) of the vehicle headlamp.

According to the above embodiment, the following effects can be obtained.
(1) A microcomputer 60 is provided as an in-vehicle vehicle light detection device, and the microcomputer 60 as vehicle light detection means processes an image acquired by the image sensor 59, and a light source is a taillight of a preceding vehicle or a headlight of an oncoming vehicle. In step 107 and 113 in FIG. 4, at least three types of optical filters 82 to 85 in the image sensor 62 of the image sensor 59 are detected. It has a gradation value change tendency determination function for determining whether the change tendency of the gradation value of the attached pixel shows the same tendency as the taillight of the preceding vehicle or the headlight of the oncoming vehicle. With this function, it is determined whether the light source is the taillight of the preceding vehicle or the headlight of the oncoming vehicle, and the taillight of the preceding vehicle or the headlight of the oncoming vehicle is detected. Therefore, it is possible to detect the vehicle light, that is, the tail light of the preceding vehicle or the headlight of the oncoming vehicle with high accuracy in consideration of the spectral characteristics of the vehicle light as the light source and other light sources.

  (2) The microcomputer 60 as the vehicle light detection means performs imaging as described in FIG. 7 in step 101 of FIG. 4 as a processing function prior to the processing by the gradation value change tendency determination function in (1) described above. The other wavelength for calculating the gradation value in the corresponding pixel for the light in the other wavelength region relative to the light in the wavelength region detected by the corresponding pixel from the gradation value in the surrounding pixel with respect to the corresponding pixel in the element 62 In the same way as in step 101 of FIG. 4, the pixel value in the image sensor 62 is calculated by the tone value detected by the corresponding pixel and the other wavelength range light tone value calculation function. It is determined whether the luminance by the luminance calculation function for calculating the luminance at the corresponding pixel from the gradation value and the luminance calculation function at each pixel in the image sensor 62 in step 102 in FIG. And further has a large and a light source candidate determining function determines that a candidate of the light source, the. Thereby, it becomes more excellent in the detection property of a vehicle light.

  (3) The microcomputer 60 as the vehicle light detection means determines whether the light source determined to show the same tendency in steps 107 and 113 in FIG. 4 has left-right symmetry in steps 108 and 114 in FIGS. It further has a symmetry determination function. Thereby, it becomes more excellent in the detection property of a vehicle light.

  (4) The microcomputer 60 determines the moving direction and moving amount of the light source determined to have left / right symmetry in steps 109, 111, 115, and 117 in FIGS. It further has a tracking function for determining whether the tail light of the preceding vehicle or the headlight of the oncoming vehicle is at least one. Thereby, it becomes more excellent in the detection property of a vehicle light.

  (5) The vehicle lighting device described above is provided as the vehicle lighting device, and the microcomputer 60 serving as the headlight control means includes a vehicle light (a taillight of a preceding vehicle or an oncoming vehicle). Based on the detection result of the headlight), at least one of switching and light distribution of the headlights 10 and 20 of the vehicle is controlled. Thereby, the headlamps 10 and 20 of the vehicle can be automatically controlled.

The electrical block diagram of the illuminating device for vehicles in this embodiment. The electrical block diagram of the vehicle-mounted vehicle light detection apparatus. (A) is a top view of an image pick-up element, (b) is a schematic longitudinal cross-sectional view in the AA line of (a), (c) is a schematic longitudinal cross-sectional view in the BB line of (a). The flowchart for demonstrating an effect | action. The flowchart for demonstrating an effect | action. The flowchart for demonstrating an effect | action. The top view of an image sensor. FIG. The image figure which extracted the light source candidate. The top view for demonstrating pixel arrangement | positioning. The top view for demonstrating pixel arrangement | positioning. The top view for demonstrating pixel arrangement | positioning. The top view for demonstrating pixel arrangement | positioning.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Headlamp, 20 ... Headlamp, 44 ... Vehicle-mounted vehicle light detection apparatus, 59 ... Image sensor, 60 ... Microcomputer, 62 ... Image sensor, 82 ... Near-infrared filter, 83 ... Blue filter, 84 ... Green Filter, 85 ... red filter.

Claims (4)

An image sensor (59) having an image pickup element (62) in which pixels are arranged and picking up a traveling direction in front of the vehicle;
Vehicle light detection that processes an image acquired by the image sensor (59) and determines whether the light source is a taillight of a preceding vehicle or a headlight of an oncoming vehicle and detects a taillight of a preceding vehicle or a headlight of an oncoming vehicle Means (60);
An in-vehicle vehicle light detection device comprising:
In the image sensor (62) of the image sensor (59), at least three types of optical filters (82 to 85) having different wavelength ranges to be passed are arranged on the pixel array,
The vehicle light detection means (60)
In the image sensor (59) of the image sensor (59), the change tendency of the gradation value of the pixel to which the at least three types of optical filters (82 to 85) are attached is the same as the taillight of the preceding vehicle or the headlight of the oncoming vehicle. A gradation value change tendency judgment function for judging whether to show a tendency ;
A symmetry determination function for determining whether the light source determined to show the same tendency by the gradation value change tendency determination function has left-right symmetry;
With respect to the light source determined to have left and right symmetry by the symmetry determination function, a tracking function for determining whether the headlight of the preceding vehicle or the headlight of the oncoming vehicle from at least one of the moving direction and the moving amount in successive images,
An on-vehicle vehicle light detection device comprising: The in-vehicle vehicle light detection device according to claim 1,
The vehicle-mounted vehicle light detection device, wherein the light transmitted by the at least three types of optical filters (82 to 85) is at least three of blue light, green light, red light, and near infrared light. The in-vehicle vehicle light detection device according to claim 1 or 2,
The vehicle light detection means (60)
As a processing function prior to the processing by the gradation value change tendency determination function,
From the gradation value at the surrounding pixel with respect to the corresponding pixel in the image sensor (62), the gradation value at the corresponding pixel for the light in the wavelength range detected by the corresponding pixel is calculated. Other wavelength range light gradation value calculation function to calculate,
For each pixel in the image sensor (62), the luminance at the corresponding pixel is calculated from the gradation value detected by the corresponding pixel and the gradation value calculated by the other wavelength band light gradation value calculation function. Brightness calculation function to calculate,
A light source candidate determination function that determines whether or not the luminance by the luminance calculation function in each pixel in the image sensor (62) is greater than a threshold value, and determines that the luminance is a light source candidate if the luminance is large;
An in-vehicle vehicle light detection device further comprising: The vehicle-mounted vehicle light detection device according to any one of claims 1 to 3 ,
A headlamp that controls at least one of switching and light distribution of the vehicle headlamps (10, 20) based on the detection result of the taillight of the preceding vehicle or the headlamp of the oncoming vehicle by the vehicle lamp detection means (60). A vehicle lighting device comprising a control means (60).

Images