JP2016138755A - Vehicle detection device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle detection device with which it is possible to discriminate other vehicles from road signs, etc., and accurately determine a preceding vehicle and an oncoming vehicle.SOLUTION: A vehicle detection device 1 comprises an image-capturing device 4 for capturing the image of an external area of the vehicle itself, and a vehicle detection unit 5 for detecting other vehicles present in the external area on the basis of light spots in the captured image. The image-capturing device 4 has an image-capturing element 43 provided with a green light sensor 432G for receiving green-based light and a red light sensor 432R for receiving red-based light. The vehicle detection unit 5 discriminates between other vehicles and objects on the basis of the received light values of the green light sensor 432G and the red light sensor 432R. Furthermore, a preceding vehicle and an oncoming vehicle are discriminated from the arithmetic result of these received light values.SELECTED DRAWING: Figure 5

Description

  The present invention relates to an apparatus for detecting other vehicles existing around the host vehicle when the vehicle is in a state where a lamp is lit, such as during night driving.

  In order to appropriately control the light distribution of the headlamp of the host vehicle during night driving, a vehicle detection technique for detecting a preceding vehicle or an oncoming vehicle in front of the host vehicle has been proposed. Particularly, in recent years, there has been a demand for light distribution control such as ADB (Adaptive Driving Beam) that illuminates the front area of the host vehicle as brightly as possible while preventing the dazzling of preceding and oncoming vehicles. In order to realize the control, it is necessary to accurately detect the preceding vehicle and the oncoming vehicle at night.

  As such a vehicle detection device, there is a technology in which an imaging device that images a front region of the host vehicle is arranged in an automobile, and an image of the front region captured by the imaging device is analyzed to detect a preceding vehicle or an oncoming vehicle. Proposed. For example, in Patent Document 1, an optical filter that transmits the wavelength region A of the red light of the tail lamp of the preceding vehicle and an optical filter that transmits the wavelength region B of the blue-white light of the headlamp of the oncoming vehicle are arranged in the imaging device. ing. Then, by detecting the light spots of the wavelength regions A and B among the light spots of the lamp imaged by the imaging device, the light spot is regarded as the light spot by the light of the tail lamp or the headlamp, and the preceding vehicle and the oncoming vehicle Is detected.

JP 2008-68700 A

  The technique of Patent Document 1 measures the size of the light spot regions of the wavelength regions A and B imaged by the imaging device, and detects the preceding vehicle and the oncoming vehicle based on the measured region size. However, the light spot from the preceding vehicle or oncoming vehicle that is far ahead of the host vehicle is captured as a small image, and the illumination light or road sign light of a white or red building that is in the vicinity of the host vehicle is captured as a large image. Therefore, it is difficult to detect all preceding vehicles and oncoming vehicles only with the size of such a region.

  Moreover, in patent document 1, it is discriminate | determining a preceding vehicle and an oncoming vehicle by detecting whether the imaged light spot is the light spot of the wavelength range A or B. This technique is effective when the spectral characteristics of the headlamp light and taillight light are clearly distinguished, but the white light of normal headlamps often contains the red wavelength region, so the wavelength region The light spot detected as A light is not necessarily the light spot of the tail lamp, and it is difficult to reliably distinguish the preceding vehicle and the oncoming vehicle at this point.

  As described above, in the technique of Patent Document 1, it is difficult to accurately discriminate between a preceding vehicle and an oncoming vehicle as a road sign light, and to accurately discriminate between the preceding vehicle and the oncoming vehicle. Thus, there is a problem that it is difficult to realize the light distribution control of brightly illuminating the front area of the host vehicle.

  An object of the present invention is to select other vehicles from road sign lights and the like, and to accurately discriminate between a preceding vehicle and an oncoming vehicle, and thereby it is possible to suitably illuminate the front area of the host vehicle. A vehicle detection device is provided.

  The present invention is a vehicle detection apparatus that includes an imaging device that captures an external region of a host vehicle, and a vehicle detection unit that detects other vehicles existing in the external region based on a light spot in the captured image. The apparatus includes a first optical sensor that receives light in the first spectral region and a second optical sensor that receives light in a second spectral region different from the first spectral region. In addition, the vehicle detection unit selects other vehicles and other objects based on the light reception values of the first light sensor and the second light sensor, and calculates the light reception values of the first light sensor and the second light sensor. The preceding vehicle and the oncoming vehicle are distinguished from each other.

  In the present invention, the imaging device includes an imaging device in which a large number of photosensors are arranged, a first optical filter that transmits light in the first spectral region on the light receiving surface side of the numerous photosensors, It is preferable that one of the second optical filters that transmit light in the two spectral regions is disposed.

  As a preferred embodiment of the present invention, the imaging apparatus includes a green light sensor that receives light in a green spectral region and a red light sensor that receives light in a red spectral region, and the vehicle detection unit includes a green light sensor and The other vehicle and the other object are selected based on the light reception value of the red light sensor, and the preceding vehicle and the oncoming vehicle are discriminated from the calculation result based on the light reception values of the green light sensor and the red light sensor.

  For example, the vehicle detection unit selects that the captured light spot is the lamp light of the other vehicle when the light reception level of the green light sensor or the light reception level of the red light sensor is equal to or higher than a reference value. In this case, the vehicle detection unit divides the captured area into a plurality of areas, and performs light spot selection based on the light reception level of the captured light spot and the area including the light spot.

  Further, the vehicle detection unit determines that the light spot imaged when both the received light amount of the green light sensor and the received light level of the red light sensor are equal to or higher than the reference level is the lamp light (head lamp light) of the oncoming vehicle, Assume that the light spot captured when only the light receiving level of the optical sensor is equal to or higher than the reference level is the lamp light (tail lamp light) of the preceding vehicle.

  According to the present invention, other vehicles and other objects can be selected based on the light reception values of the first optical sensor and the second optical sensor, for example, the light reception level. Further, the preceding vehicle and the oncoming vehicle can be determined from the calculation result of the light reception values of the first light sensor and the second light sensor, for example, the difference in the amount of light received.

The schematic block diagram of embodiment which applied the vehicle detection apparatus of this invention to the light distribution control system of the headlamp of a motor vehicle. FIG. 3 is a cross-sectional view of an imaging device and an enlarged cross-sectional view of an image sensor. The figure of the spectral characteristics of a headlamp and a tail lamp, and the spectral transmittance of the optical filter of an imaging device. The figure which shows the example of an arrangement | sequence of an optical filter typically. The block block diagram of the light distribution control system of embodiment. FIG. 6 is a block diagram of a first area to a third area of an imaging area. The circuit diagram of a calculating part. The figure which shows typically the modified example of a different arrangement | sequence of an optical filter. The expanded sectional view of the modification of an image sensor. The figure which shows the imaged image typically.

  Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of an embodiment in which the vehicle detection device of the present invention is configured as one of components of a light distribution control system for a headlamp of an automobile. The vehicle CAR is provided with a vehicle detection device 1 that detects a preceding vehicle and an oncoming vehicle existing in front of the host vehicle. Further, the headlamps HL disposed on the left and right of the front part of the car body of the automobile CAR are provided with a light distribution control unit 2, which is connected to the preceding vehicle detected by the vehicle detection device 1. The light distribution of the headlamp HL is controlled corresponding to the oncoming vehicle.

  In this embodiment, as can be seen with reference to FIG. 5, the headlamp HL includes an LED unit 3 in which a plurality of light emitting diodes (LEDs) 32 are mounted on a required circuit board 31, and the light distribution control unit 2. The light distribution of the headlamp HL is switched by controlling the light emission of the LED unit 3. For example, when all the LEDs 32 are caused to emit light, control is performed with a high beam distribution, and from this state, a part of the LEDs 32 is extinguished to control with a low beam distribution, or an area where a preceding vehicle or an oncoming vehicle is present is controlled. It is possible to control the ADB light distribution by stopping the light irradiation to the illuminating LED 32.

  The vehicle detection device 1 includes an image pickup device 4 that is disposed at a front part of the car CAR, in this case, at a position inside the windshield, and picks up a front area of the host vehicle, and an image of an image taken by the image pickup device 4. A vehicle detection unit 5 that detects a preceding vehicle and an oncoming vehicle based on the signal is provided. The image pickup device 4 is configured as an image pickup camera as shown in a schematic configuration in FIG. 2A, and includes a camera case 41 and a preceding vehicle that is built in the camera case 41 and exists in a front area of the automobile. An imaging lens 42 for optically imaging an object including an oncoming vehicle, and a large number of light receiving cells comprising a CCD, a CMOS, or the like that photoelectrically converts the object image formed by the imaging lens 42 and outputs an electrical signal An image sensor 43 in which (optical sensors) are arranged in a matrix is provided.

  FIG. 2B is an enlarged cross-sectional view of a portion B in FIG. 2A, and the image sensor 43 has an optical sensor 432 that is intrinsically a large number of pixels formed in a matrix on the surface of the semiconductor substrate 431. A light receiving surface is formed, and an inner lens 433, an optical filter 434, and a micro lens 435 are laminated on the light receiving surface. The object image light imaged by the imaging lens 42 is condensed by the micro lens 435 and the inner lens 433 in units of pixels, and the light selected by the optical filter 434 is received by the optical sensor 432. . A light reception signal obtained by each optical sensor 432 is output to the vehicle detection unit 5 as an imaging signal of an object, that is, an image signal of a front area of the automobile.

  As shown in FIG. 3, the optical filter 434 includes a first optical filter (hereinafter referred to as a green filter) 434G that selectively transmits light in a green wavelength region of 530 to 570 nm, and a red color of 590 to 650 nm. It is configured by any one of a second optical filter (hereinafter, red filter) 434R that selectively transmits light in the wavelength region. FIG. 3 shows a headlamp (LED-HL) using an LED as a light source, a headlamp (halogen HL) using a halogen bulb as a light source, a tail lamp (LED-TL) using an LED as a light source, and an incandescent bulb. 2 also shows the spectral characteristics of a tail lamp (incandescent TL) having a light source as a light source.

  Therefore, the image sensor 43 receives only light that passes through the green filter 434G, that is, light of the LED-HL and the halogen HL, and the light sensor (hereinafter, green light sensor) 432G in which the green filter 434G is disposed. Then, a light reception signal is output. On the other hand, the light sensor (hereinafter, red light sensor) 432R provided with the red filter 434R is light that passes through the red filter 434R, that is, light of LED-HL and halogen HL, light of LED-TL and incandescent TL. Is received and outputs a light reception signal.

  In other words, the green light sensor 432G detects only the headlamp light of the oncoming vehicle, and the red light sensor 432R detects the headlamp light of the oncoming vehicle and the tail lamp light of the preceding vehicle. On the other hand, the green light sensor 432G and the red light sensor 432R are respectively installed in light having a wavelength region of 530 nm or less, 570 nm to 590 nm, or 650 nm or more, for example, on the road, by the spectral transmittance of the green filter 434G and the red filter 434R. It does not detect the light of street signs or street sign lamps that emit yellow or orange light for road lighting.

  The green filter 434G and the red filter 434R are arranged in a required pattern on the imaging surface side of the imaging element 43. FIG. 4 is a diagram schematically showing the arrangement of the green filter 434G and the red filter 434R, and corresponds to the arrangement of the optical sensors 432 of the image sensor 43. Here, the green filter is indicated by the symbol G and the red filter is indicated by the symbol R. In the configuration of FIG. 4A, the green filter G and the red filter R are alternately arranged in the row direction and the column direction of the photosensor. Alternatively, as shown in FIG. 4B, the green filters G and the red filters R may be alternately arranged in units of photosensors, and further, green in units of photosensors as shown in FIG. 4C. Filter G and red filter R may be arranged alternately. In any case, the green filter G and the red filter R are preferably arranged adjacent to each other.

  As shown in FIG. 5, the vehicle detection unit 5 includes a green light detection unit 51 that detects a light reception level of the green light sensor 432G on which the green filter G is disposed, and a red color on which the red filter R is disposed. A red light detection unit 52 that detects the light reception level of the optical sensor 432R is provided. The green light detection unit 51 and the red light detection unit 52 can also detect the coordinate position (X, Y) on the imaging screen of the received green light sensor 432G and red light sensor 432R.

  Further, the vehicle detection unit 5 selects whether the received light is the light of the head lamp or tail lamp of the vehicle based on the light reception values detected by the green light detection unit 51 and the red light detection unit 52. In the processing unit 53, a calculation unit 54 that performs calculation on the received light values of the green light detection unit 51 and the red light detection unit 52, and the vehicle selection processing unit 53 based on the calculation result of the calculation unit 54 A vehicle discrimination processing unit 55 is provided for discriminating whether the selected light is the headlamp light of the oncoming vehicle or the taillamp light of the preceding vehicle.

  Furthermore, the vehicle detection unit 5 detects road information ahead of the host vehicle based on road information from a navigation device (not shown) mounted on the host vehicle, for example, and captures an image area captured by the imaging device 4. Is provided with a region setting unit 56 that partitions the region into a plurality of regions.

  FIG. 6A shows an example of areas set by the area setting unit 56. The first area A1 is a preceding vehicle existence area, the second area A2 is an oncoming vehicle existence area, and the third area A3 is an exclusion area. . For example, when the road ahead of the host vehicle is a straight path, the left and right end points of the horizontal line H passing through the reference point P1 that is far forward in the direction of the host vehicle and the horizontal predetermined width region including the reference point P1 A lower region surrounded by a left inclined line Ll and a right inclined line Lr extending obliquely downward to the left and right is defined as a first region A1. This first area A1 corresponds to the traveling area of the host vehicle imaged in the front area of the host vehicle. In the right region of the first region A1, a region between the right inclined line Lr and the upper right inclined line Lrr extending along the upper side is defined as a second region A2. This second area A2 corresponds to the traveling area of the oncoming vehicle. A region not included in the first region A1 and the second region A2 is a third region A3.

  In the vehicle detection device 1 according to the embodiment having the above-described configuration, the imaging element 43 of the imaging device 4 includes the green light sensor 432G and the red light sensor 432R having the spectral light receiving characteristics illustrated in FIG. It does not detect light spots from road signs such as street lights and delineators that emit yellow or orange light for road lighting installed on roads that do not include light in these green and red wavelength regions. . Therefore, the light spot imaged by the imaging device 4 is limited to the light spot by the head lamp and tail lamp of the automobile, or the light spot by the white or red street lamp or building illumination light.

  The vehicle detection unit 5 detects the light reception level as the light reception value for all the green light sensors 432G and the red light sensors 432R of the image sensor 43 in the green light detection unit 51 and the red light detection unit 52. Then, the light reception level is compared with a predetermined reference value, and an optical sensor having a light reception level equal to or higher than the reference value is detected. At the same time, the coordinate position (X, Y) of the detected optical sensor is detected.

  The vehicle sorting processing unit 53 contrasts the detected optical sensor, in other words, the coordinate positions of the light spots BP1 and BP2 detected by the optical sensor with the first to third regions A1 to A3 set by the region setting unit 56. The light spots existing in the first area A1 and the second area A2 are selected. As a result, light spots detected in the third region A3, that is, light spots caused by white or red street lamps or building illumination light are excluded. Here, as shown in FIG. 6B, only the light spot BP1 due to the tail lamp of the preceding vehicle existing in the first area A1 and the light spot BP2 due to the headlamp of the oncoming car existing in the second area A2 are selected. .

  On the other hand, the calculation unit 54 performs a predetermined calculation on the light reception levels of the green light sensor 432G and the red light sensor 432R. In this embodiment, one green light sensor 432G and one red light sensor 432B adjacent to each of the light sensors 432 constituting the image sensor 43 are formed as one group, and each group is formed. A logical operation of each light reception level of the green light sensor 432G and the red light sensor 432B is executed. The subtraction value obtained by this logical operation processing is output to the vehicle discrimination processing unit 55.

  For example, FIG. 7 shows an example of the arithmetic unit 54, which includes two comparison circuits CMP1 and CMP2, two logic register circuits AND1 and AND2, and one negation circuit NOT. Here, the output level of the green light detection unit 51 based on the green light sensor 432G is compared with the reference level ref1 by CMP1, and “1” is output when the level is equal to or higher than the reference level ref1. Further, the output level of the red light detection unit 52 based on the red light sensor 432R is compared with the reference level ref2 by CMP2, and “1” is output when the output level is equal to or higher than the reference level ref1. Note that these reference levels ref1 and ref2 do not necessarily match the reference values used when detecting other vehicles.

  Then, the output of CMP1 and the output of CMP2 are input to AND1, and the inverted output of CMP1 and the output of CMP2 are input to AND2. The output of AND1 is “1” when the input green light and red light are above the reference level. Further, since the input of AND2 is “0” when the green light is above the reference level, the output is “1” when the green light is less than the reference level and the red light is above the reference level. That is, the output of either AND1 or AND2 is “1” depending on the level of green light. In other words, white light is output when the output of AND1 is "1", and red light is output when the output of AND2 is "1".

  Accordingly, the vehicle discrimination processing unit 55 discriminates which of the AND1 and AND2 outputs of the calculation unit 54 is “1”, so that the light spots, here the light spots BP1 and BP2 are white light or red light. That is, it is possible to determine whether the light spot of the headlamp or the light spot of the tail lamp. In this determination, it is more reliable by referring to which one of the first area A1 and the second area A2 set by the area setting unit 56 the light spot to be determined is present. Discrimination becomes possible.

  Then, the discrimination signal discriminated by the vehicle discrimination processing unit 55 is output to the light distribution control unit 2, and the light distribution control unit 2 controls the light distribution of the headlamp HL. As described above, when the oncoming vehicle or the preceding vehicle is detected by the vehicle detection device 1, the light distribution control unit 2 controls the light distribution of the headlamp HL of the own vehicle, so that the detected oncoming vehicle or the preceding vehicle can be detected without being distracted. It goes without saying that the front area of the vehicle can be illuminated brightly. In particular, in the case of ADB light distribution control, the front area excluding oncoming vehicles and preceding vehicles can be illuminated with light distribution equivalent to a high beam, and suitable light distribution control can be realized.

  Here, in the present invention, the light spot obtained by imaging the tail lamp of the preceding vehicle and the headlamp of the oncoming vehicle existing far in front of the host vehicle with the imaging device 4 is smaller than the imaging screen of the imaging element 43. Therefore, it may be difficult to increase the accuracy of vehicle selection and vehicle discrimination based on this small light spot. Therefore, in this modified embodiment, the vehicle selection processing unit 53 performs weighting processing on the detected light reception level.

  For example, although not shown, for the light spot in the first area A1, the light receiving level detected by the red light sensor 432R is weighted more than “1” than the light receiving level detected by the green light sensor 432G. do. On the contrary, for the light spot in the second region A2, the light receiving level detected by the green light sensor 432G is weighted with a value larger than “1” than the light receiving level detected by the red light sensor 432R. For this weighting, for example, a value larger than “1” may be multiplied by the received light level or received light amount. In other words, the light reception level and light reception amount of the red light sensor or the green light sensor are selectively amplified.

  By performing this weighting, in the vehicle selection processing unit 53, the light spot detected in the first area A1 has the light reception level of the red light sensor 432R significantly higher than the light reception level of the green light sensor 432G. Therefore, even when the size of the light spot of the detected tail lamp of the preceding vehicle is small, red light can be amplified. Since there is a high probability that the preceding vehicle exists in the first region A1, the subtraction value in the vehicle determination process becomes significant by weighting the light reception signal of the red light sensor 432R in the first region A1.

  The same applies to the light spot detected in the second area A2, and in the second area A2, the light reception level of the green light sensor 432G is significantly higher than the light reception level of the red light sensor 432R. Therefore, even when the size of the light spot of the headlamp of the oncoming vehicle detected is small, the green light can be amplified. As a result, it is possible to increase the accuracy of the vehicle selection processing in the vehicle selection processing unit 53 or the vehicle determination processing in the vehicle determination processing unit 55.

  In addition, when the imaging area of the front area of the own vehicle in the imaging device 4 is fixed, for example, the range of the first area A1 to the third area A3 is fixed as illustrated in FIG. If there is, the arrangement number ratio of the green filter 434G and the red filter 434R in the image sensor 43 may be made different corresponding to each region.

  In other words, in the example shown in FIG. 8A, the optical filter 434 of the image sensor 43 corresponding to the first area A1 has a red filter R and green filter G arrangement ratio of 3: 1. On the other hand, each optical filter 434 of the image sensor 43 corresponding to the second region A2 has a ratio of the arrangement number of the red filter R and the green filter G of 1: 3 as shown in FIG. 8B.

  In addition, in the first region A1, three red filters R and one green filter B surrounded by a broken line are used as one block, and the received light value of the red light sensor 432R including the three red filters R in each block is used as the light reception value. And the amount of light received by the green light sensor 432G formed by one green filter G is detected. The second region A2 has the opposite configuration. By doing in this way, the light reception level of the green light sensor 432G and the red light sensor 432R can be increased and the calculation in the calculation unit 54 can be performed more clearly, and the vehicle selection processing in the vehicle selection processing unit 53, or It becomes possible to improve the accuracy of the vehicle discrimination processing in the vehicle discrimination processing unit 55.

  Alternatively, the imaging element 43 of the imaging device 4 may be configured as shown in FIG. FIG. 9 is an enlarged cross-sectional view of the same image pickup element as that in FIG. 2B, and the same portions are denoted by the same reference numerals. In the imaging device 4, the microlens 435 constituting a part of the imaging element 43 is configured as a microlens having the same or different focal length for each optical sensor 432. Here, the lens is composed of three types of lenses: a long focus lens 435t, a medium focus lens 435m having a shorter focal length, and a short focus lens 435w having a shorter focal length.

  These microlenses 435t, 435m, and 435w having different focal lengths are, as shown in FIG. 10A, long focal regions as a central region including a straight forward far front point in a front region of the host vehicle that is imaged by the imaging device 4. This corresponds to a (tele area) T, a middle focal area (middle area) M as a middle band area surrounding the outside in a frame shape, and a short focal area (wide area) W outside the middle focal area. That is, the micro lens of the photosensor in the tele area T is configured by the long focus lens 435t, the micro lens of the photo sensor in the middle area M is configured by the middle focus lens 435m, and the micro lens of the photo sensor in the wide area W is short focused. A lens 435w is used. Needless to say, the lens positions of these microlenses 435, that is, the distances from the respective optical sensors 432, are arranged so that the subject image is properly focused according to the focal length of each microlens 435.

  By configuring the imaging device 43 of the imaging device 4 in this way, as shown in FIG. 10B, the captured image of the front area of the host vehicle is a tele area T, a middle area M, and a wide area W. Images are taken as images enlarged in order. Therefore, since the oncoming vehicle and the preceding vehicle existing far ahead of the host vehicle are enlarged and imaged in the tele area T, the light spots of the oncoming vehicle and the preceding vehicle are also enlarged and imaged. Thereby, the light spot by the headlamp and tail lamp of the oncoming vehicle and the preceding vehicle existing far ahead of the host vehicle is enlarged, and the accuracy of sorting and discrimination in the vehicle sorting processing unit 53 and the vehicle discrimination processing unit 55 is improved. It becomes possible.

  In addition, since the wide area W is imaged with a wide angle of view by the short focus lens, it is possible to image the light spots of the headlamps and taillights of oncoming vehicles and preceding vehicles that are approaching very close to the host vehicle. Can be accurately detected.

  If the microlens is composed of a Fresnel lens, when the microlenses having different focal lengths are mixedly mounted, the lens thickness dimension of the microlens can be reduced even in the case of a short focal length lens, and the imaging element can be thinned. In addition, with this configuration, a hybrid image in which such a multifocal imaging screen is combined can be obtained without configuring the imaging lens 42 of the imaging device with a large number of lenses. Can be prevented.

DESCRIPTION OF SYMBOLS 1 Vehicle detection apparatus 2 Light distribution control part 3 LED unit 4 Imaging device 5 Vehicle detection part 43 Image pick-up element 51 Green light detection part 52 Red light detection part 53 Vehicle selection process part 54 Calculation part 55 Vehicle discrimination | determination process part 56 Area | region setting part 432 Optical sensor 434 Optical filter (green filter, red filter)
435 micro lens

Claims (6)

  A vehicle detection device comprising: an imaging device that captures an external region of the host vehicle; and a vehicle detection unit that detects another vehicle existing in the external region based on a light spot in the captured image. A first optical sensor that receives light in one spectral region; and a second optical sensor that receives light in a second spectral region different from the first spectral region, wherein the vehicle detection unit includes the first optical sensor. Based on the light reception values of the light sensor and the second light sensor, other vehicles and other objects are selected, and the preceding vehicle and the oncoming vehicle are discriminated from the calculation result of the light reception values of the first light sensor and the second light sensor. A vehicle detection device.   The image pickup apparatus includes an image pickup element in which a large number of photosensors are arranged, and a first optical filter that transmits light in the first spectral region is disposed on a light receiving surface side of the multiple photosensors, and the second The vehicle detection device according to claim 1, wherein any one of the second optical filters that transmit light in the spectral region is disposed.   The imaging apparatus includes a green light sensor that receives light in a green spectral region and a red light sensor that receives light in a red spectral region, and the vehicle detection unit includes the green light sensor and the red light sensor. 3. The vehicle according to claim 2, further comprising: selecting another vehicle and another object on the basis of the received light value, and discriminating a preceding vehicle and an oncoming vehicle from a calculation result of the received light values of the green light sensor and the red light sensor. The vehicle detection device described.   The vehicle detection unit, when the light reception level of the green light sensor or the light reception level of the red light sensor is equal to or higher than a reference value, selects the captured light spot as lamp light of another vehicle. Item 4. The vehicle detection device according to Item 3.   The vehicle detection unit divides a captured area into a plurality of areas, and performs light spot selection based on a light reception level of the captured light spot and a region including the light spot. 5. The vehicle detection device according to 4. The vehicle detection unit determines that a light spot captured when the light reception level of the green light sensor and the light reception level of the red light sensor are equal to or higher than a reference level is lamp light of an oncoming vehicle, and the light reception of the green light sensor 6. The vehicle detection according to claim 4 or 5, wherein when the level is less than a reference level and the light receiving level of the red light sensor is greater than or equal to a reference level, it is determined that the captured light spot is the lamp light of the preceding vehicle. apparatus.

Images