JP4722101B2 - Control devices for automobiles such as light distribution - Google Patents

Description

  The present invention relates to an automobile control device that detects the situation of a preceding vehicle and an oncoming vehicle and the situation of a road environment such as a streetlight, and performs light distribution control of a headlight (headlight), automobile speed control, and the like.

  The headlights mounted on the vehicle can be switched in two stages, high beam and low beam. When driving in the suburbs at night, it is common to use a high beam when there is no preceding vehicle and an oncoming vehicle, and when there is a preceding vehicle or an oncoming vehicle, use a low beam so that glare does not occur for the other driver. . In order to automatically control switching between high beam and low beam at night, it is necessary to recognize the presence of a preceding vehicle or an oncoming vehicle. As a method for recognizing a preceding vehicle or an oncoming vehicle in order to perform such light distribution control of the headlight, a technique has been developed in which the light of the preceding vehicle or the oncoming vehicle is imaged and detected by a camera.

  Japanese Patent Application Laid-Open No. 2004-228561 discloses a device that images the front of a vehicle with a camera mounted on the vehicle, extracts a pair of vehicle lights from the binarized image, and calculates a distance to a preceding vehicle or an oncoming vehicle. The camera used in this apparatus is equipped with an image memory and a digital signal processor (DSP) in addition to the CPU. By using this camera, it is possible to distinguish the street light from the light of the preceding vehicle and the oncoming vehicle, and it is possible to calculate the distance to the preceding vehicle and the oncoming vehicle.

JP-A-8-166221

  However, when a general imaging device equivalent to VGA is used, the distance at which a pair of lights can be detected is about 100 m. When traveling on a highway, a field of view of 100 m or more is necessary. For example, when detecting a pair of lights of a preceding vehicle or an oncoming vehicle 500 m ahead, it is necessary to use a telephoto lens or a high-resolution imaging device. . However, with the telephoto lens, the vehicle in the vicinity of the host vehicle cannot be imaged, so that the preceding vehicle and the oncoming vehicle cannot be detected. Therefore, when a high-resolution imaging device is used, a large-capacity memory is required, which increases costs. Even when a general VGA-equivalent imaging device is used, if an image memory or DSP is installed in the camera, the cost increases. For the above reasons, there is a problem that the device that detects the situation of the preceding vehicle and the oncoming vehicle and the situation of the road environment and controls the light distribution of the headlight based on the situation is expensive.

  The present invention uses a simple camera and an image processing device that do not require a high-cost device, accurately detects the situation of a preceding vehicle or an oncoming vehicle, and the situation of a road environment, and provides inexpensive light distribution control and vehicle travel control. An object of the present invention is to provide an apparatus that enables the above.

  In order to achieve the above object, the present invention basically proposes the following control apparatus for an automobile. That is, an in-vehicle imaging device that images the front of the vehicle, an image processing unit that divides the captured image into an upper image region and a lower image region, and brightness of light spots detected from the upper image region and the lower image region And a control unit for controlling at least one of the light distribution of the headlamp and the running state of the vehicle.

  Moreover, the following specific aspects are proposed as an application example of the above invention. One is the control apparatus for an automobile, wherein the lower image area is set to a position for capturing a vehicle in front of the own vehicle, and the upper image area is set to a position for capturing a streetlight in front of the own vehicle. The threshold value of the object detection reference detected in the region is different.

  One is characterized in that, in the control apparatus for an automobile, a detection determination period in the upper image area is longer than a detection determination period in the lower image area.

  One is the control apparatus for an automobile, wherein the control unit is configured to detect adjacent pixels when a plurality of pixels having a luminance of a predetermined value or more are adjacent to at least one of the upper image area and the lower image area. When the number is detected at a predetermined value or more, the headlamp is set to a low beam or the amount of light of the headlamp is reduced, and the upper image area and the lower image area are both equal to or larger than the predetermined value. When the number of adjacent pixels having luminance is not detected more than a predetermined value, light distribution control is performed such that the headlamp is set to a high beam or the light quantity of the headlamp is increased.

  One is the control apparatus for an automobile, wherein the control unit is adjacent when a plurality of pixels having a luminance of a predetermined value or more are adjacent to only the lower image area of the upper image area and the lower image area. When the number of pixels is detected to be equal to or greater than a predetermined value, the vehicle speed is controlled or the acceleration is controlled.

  Furthermore, in the control apparatus for an automobile, the predetermined value used for the brightness detection is different in the upper image area and the lower image area.

  Further, the predetermined value used for detecting the number of adjacent pixels is different in the upper image area and the lower image area.

  According to the present invention, a simple and inexpensive camera or image processing device is used to accurately detect the situation of a preceding vehicle or an oncoming vehicle or the situation of a road environment, and light distribution control or other vehicle control based on the situation can be performed. It becomes feasible.

  FIG. 1 is a block diagram showing a light distribution control device as an example of an automobile control device according to an embodiment of the present invention, and FIG. 2 is a vehicle configuration diagram of the light distribution control device according to the present embodiment. In the vehicle 200 of FIG. 2, the camera 100 is mounted at a position where the front of the host vehicle can be imaged.

  As shown in FIG. 1, the internal configuration of the camera 100 is roughly divided into an imaging unit 109 and a CPU 110. The imaging unit 109 includes a lens 101, an imaging element 102, and an imaging element output signal conversion LSI 103. The horizontal angle of view of the lens 101 of the imaging unit 109 corresponds to the horizontal irradiation range of the headlight 112 of the vehicle 200. The headlight 112 is subjected to light distribution control by the CPU 110 via the headlight driving circuit 108, such as switching between high beam and low beam, and control of beam intensity and optical axis. The image sensor 102 has a number of pixels equivalent to CIF or VGA, and images the front of the host vehicle. Imaging may be performed in color or monochrome. When imaging in color, as described later, it is possible to discriminate between the preceding vehicle and the oncoming vehicle. The image sensor output signal conversion LSI 103 converts the analog data sent from the image sensor 102 into digital data (hereinafter referred to as image data) and transfers it to the CPU 110. As will be described later, the CPU 110 transfers the image data to the RAM 104 through the I / O (input / output unit) 105. The CPU 110 is connected to a CAN (Controller Area Network) 106 in addition to the headlight driving circuit 108 described above, and controls the vehicle to a vehicle unit 111 that controls the vehicle via a communication I / F (communication interface) 107. Send a signal.

  The CPU 110 will be described with reference to FIG. In addition to the CPU core, the CPU 110 includes a RAM 104, a ROM, an I / O 105, and the like, and is connected to the CAN 106 to output a headlight control signal and other vehicle control signals. The RAM 104 is assigned four areas A, B, C, and D for handling the above-described image data. The I / O 105 includes an image clock for grasping the pixel position of the image sensor, a horizontal synchronization signal for grasping the position of the horizontal line of the image sensor, a vertical synchronization signal for grasping the start point of image capture, and Image data (8 bits per pixel) is input.

  Here, a method in which the CPU 110 captures the image data transferred from the image sensor output signal conversion LSI 103 shown in FIG. 1 into the areas A, B, C, and D of the RAM 104 will be described with reference to FIG. FIG. 4 is a diagram showing image data for one screen, where (Ii, Jj) indicates a pixel position (1 ≦ i ≦ n, 1 ≦ j ≦ m, where n and m are horizontal directions, respectively. And the number of pixels in the vertical direction).

  The CPU 110 captures the image data transferred from the image sensor output signal conversion LSI 103 into the area A. At that time, the input of the vertical synchronization signal is detected to grasp the start point of the image data. Also, a horizontal line counter, which is a counter for grasping the horizontal line position (pixel vertical position Jj) of the image, is initialized. Then, it is detected by detecting the input of the horizontal synchronization signal that the data of (I1, J1) to (In, J1), which are the data for one line in the horizontal direction, have been captured. When the fetching of data for one line is completed, the fetched data is transferred to area B. After transferring the data to the area B, the horizontal line counter is incremented when fetching the data for the next line (I1, J2) to (In, J2). Thus, the pixel data for one horizontal line sequentially transferred to the area B is stored at different addresses for each pixel. Then, among the data for one line transferred to the area B, only pixel data having a luminance of a certain value or higher is transferred to the area C or the area D and stored. A method for determining whether to transfer the pixel data to the region C or the region D will be described later. The pixel data to be transferred has information on the luminance value of the pixel, the color of the pixel, and the position (Ii, Jj). The horizontal pixel position Ii is grasped by the image clock, and the vertical pixel position Jj is grasped by the horizontal line counter. Then, when (I1, Jm) to (In, Jm), which are the data for the lowest line, are stored in the area C or D, the image data for one screen is captured. When the input of the next vertical synchronizing signal is detected, the next captured image data starts to be taken, the horizontal line counter is cleared, and the above-described processing is repeated. As described above, the captured image data is sequentially stored in the areas C and D of the RAM 104.

  In this example, the image data is transferred from the area A to the area B for each line, but may be performed for every plurality of lines. For example, when 3 lines are transferred at a time, a Sobel filter or the like can be applied to 9 pixels of 3 × 3. It is also possible to detect oncoming vehicles or street lights.

  Here, data stored in the area C and the area D will be described. Data stored in the area C is pixel data in the first area (lower image area) 501 shown in FIG. Data stored in the region D is pixel data having a luminance of L2 or more in the second region (upper image region) 502 shown in FIG.

  The first area 501 is an area for mainly detecting preceding vehicles and oncoming vehicles. The range of the first region 501 includes a central portion of the screen (image data) and is large enough to include images of a preceding vehicle and an oncoming vehicle. In order to ensure that oncoming vehicles are included, this range may extend from the center to the lane on which the oncoming vehicle travels (the right side if the vehicle is on the left side).

  The second area 502 is located in the upper part of the screen and is an area for mainly detecting lighting such as street lamps. The range of the second area 502 is from one end of the screen to the other end in the horizontal direction, and the vertical direction includes the upper end of the screen, and is a streetlight that is a certain distance away from the host vehicle, for example, a streetlamp that is 20 meters ahead Is included.

  FIG. 6 shows a general flowchart of a light distribution control process performed after image data is stored in the area C or the area D of the RAM 104. In step 601, the image data of the first area 501 stored in the area C is used to detect the preceding vehicle and the oncoming vehicle. In the case of image data captured in color, the preceding vehicle and the oncoming vehicle can be determined and detected from the color of the light spot. In step 602, streetlight detection processing is performed using the image data of the second region 502 stored in the region D. In step 603, a headlight light distribution control process is performed in which the state of the headlight is determined based on the detection result of the preceding vehicle and the oncoming vehicle and the detection result of the streetlight. The above steps 601 to 603 will be described in detail below with reference to FIGS.

  FIGS. 7A and 7B are general flowcharts of the detection process for the preceding vehicle and the oncoming vehicle, which is performed in step 601 of FIG. FIG. 7A shows processing when an image is captured in monochrome, and FIG. 7B shows processing when an image is captured in color. These processes are performed on the data stored in the area C. In the area C, information on the luminance value, color, and position (Ii, Jj) of pixel data having a luminance equal to or higher than a certain value L1 in the first area 501 is stored. In step 601, the stored pixel data is sequentially extracted one by one, and the following processing is performed.

  First, the processing when the image shown in FIG. 7A is captured in monochrome will be described. In step 701, it is confirmed whether the luminance value of the extracted pixel data is greater than or equal to a certain value L1. If the luminance value is greater than or equal to L1, the process proceeds to step 702. In step 702, it is checked whether a pixel having a luminance value equal to or greater than L1 is adjacent to the adjacent position of the pixel data. If a pixel having a luminance value of L1 or more is adjacent to P1 or more, the process proceeds to step 703. In step 703, it is determined that the adjacent group of pixels is an image of the light of the preceding vehicle or the oncoming vehicle, and it is determined that there is a preceding vehicle or an oncoming vehicle, and this processing ends. If the process does not proceed to step 702 in step 701, or if the process does not proceed to step 703 in step 702, it is determined that there is no preceding vehicle or an oncoming vehicle, and this process is terminated.

  Next, with reference to FIG. 7B, processing when an image is captured in color will be described. In this case, a process for discriminating between the preceding vehicle and the oncoming vehicle based on the color of the light is performed. For the preceding vehicle, the taillight (taillight) is detected, so the pixel color is red. For the oncoming vehicle, the headlight is detected, so the pixel color is white or pale yellow. Determine.

  In step 705, it is confirmed whether the luminance value of the extracted pixel data is L1 or more. If the luminance value is greater than or equal to L1, the process proceeds to step 706. If the luminance value is less than L1, the process proceeds to step 712, where it is determined that there is no preceding vehicle or oncoming vehicle, and this process ends.

  In step 706, the color information of the pixel data is white or light yellow, and the number of pixels whose color information is white or light yellow and has a luminance value of L1 or more is 3 or more pixels (for example, 3 pixels). Above) Check if it is adjacent. If white or light yellow pixels having luminance values of L1 or more are adjacent to P1 or more, the process proceeds to step 707. In step 707, it is determined that the adjacent group of pixels is an image of the headlight of the oncoming vehicle, and it is determined that an oncoming vehicle exists. Then, the process proceeds to Step 709. If it is determined in step 706 that no white or light yellow pixel having a luminance value of L1 or more is adjacent to P1 or more, the process proceeds to step 708, and it is determined that there is no oncoming vehicle. Then, the process proceeds to Step 709.

  In step 709, whether the color information of the pixel data is red, and pixels having color values of red and having a luminance value of L1 or more are adjacent to the adjacent position by the number of pixels of P1 or more (for example, 3 pixels or more). Please check. If red pixels having a luminance value of L1 or more are adjacent to P1 or more, the process proceeds to step 710. In step 710, it is determined that the adjacent group of pixels is an image of the tail lamp of the preceding vehicle, and it is determined that there is a preceding vehicle, and this processing is terminated. If it is determined in step 709 that a red pixel having a luminance value equal to or greater than L1 is not adjacent to P1 or greater, the process proceeds to step 711, where it is determined that there is no preceding vehicle, and this process ends.

  As described above, in step 601, the stored pixel data is sequentially extracted one by one, and the processing of FIG. 7A or 7B is performed, so that both the preceding vehicle and the oncoming vehicle or It is possible to detect whether either one exists.

  FIG. 8 shows a general flowchart of the streetlight detection process performed in step 602 of FIG. This process is performed on the data stored in the area D. In the area D, information on the luminance value, color, and position (Ii, Jj) of pixel data having a luminance of a certain value L2 or more in the second area 502 is stored. In step 602, the stored pixel data is sequentially extracted one by one, and the following processing is performed.

  In step 801, it is confirmed whether the luminance value of the extracted pixel data is greater than or equal to a certain value L2. The purpose of streetlight detection is to determine whether or not the vehicle is traveling in an urban area, and there is no need to instantaneously determine that there is a streetlight. For this reason, it is not necessary to detect a streetlight from a distance from the host vehicle, and it is only necessary to detect the vicinity of the host vehicle. Since the street light in the vicinity of the host vehicle has a high luminance, the constant luminance value L2 in this process can be made larger than the constant luminance value L1 in step 701. If the luminance value is greater than or equal to L2, the process proceeds to step 802. If it is less than L2, the process proceeds to step 807, where it is determined that the vehicle is traveling in the suburbs, and this process is terminated.

  In step 802, it is checked whether or not a pixel having a luminance value of L2 or more is adjacent to the adjacent position of the pixel data by a number of pixels of a certain reference value P2 or more. In this process, in order to detect a streetlight in the vicinity of the host vehicle, the streetlight is greatly imaged. Therefore, it is checked whether or not a larger number of pixels are adjacent than in the case of step 702. That is, P2 can be set to a larger number of pixels (for example, 5 pixels) than P1. If a pixel having a luminance value of L2 or more is adjacent to P2 or more, the process proceeds to step 803. If it is not adjacent to P2 or more, the process proceeds to step 807, where it is determined that there is no street lamp and the vehicle is traveling in the suburbs, and this process is terminated.

  In step 803, it is determined that the adjacent group of pixels is an image of a streetlight, and it is determined that a streetlight exists.

In step 804, it is determined whether the vehicle currently traveling is an urban area or a suburb. Judgment conditions follow Table 1. If a streetlight detection time is detected at a certain rate, for example, 30% or more within a certain time, for example, 6 seconds, it is determined that the city is traveling (step 805). If so, it is determined that the vehicle is traveling in the suburbs (step 806). As described above, since it is not necessary to instantly determine that there is a streetlight, the certain time for detecting the streetlight can be longer than the period for detecting the preceding vehicle or the oncoming vehicle.
With the above processing, it can be determined whether the host vehicle is traveling in an urban area or a suburb.
Table 1

  In this example, whether or not the vehicle is traveling in an urban area is determined by the ratio of the streetlight detection time to a certain time, but may be the ratio of the travel distance while the streetlight is being detected for a certain distance.

  FIG. 9 is a general flowchart of the headlight light distribution control process performed in step 603 of FIG. In step 901, it is determined whether there is a preceding vehicle or an oncoming vehicle based on the processing result of step 601 in FIG. If there is a preceding vehicle or an oncoming vehicle, the process proceeds to step 904, where the headlight is set to a low beam. If there is no preceding vehicle or oncoming vehicle, the process proceeds to step 902. In step 902, it is determined whether the vehicle is traveling in an urban area based on the processing result of step 602 in FIG. If it is determined that the vehicle is traveling in an urban area, the process proceeds to step 904 where the headlight is set to a low beam. If it is not determined in step 902 that the vehicle is traveling in the city (when it is determined that the vehicle is traveling in the suburbs), the process proceeds to step 903, and the headlight state is changed to a high beam.

  Further, for example, when the headlight state is switched from a high beam to a low beam, control for smoothly changing the irradiation range and intensity of the light, such as switching to the low beam while gradually decreasing the intensity of the high beam, This can be done in step 904.

  Here, the execution timing of the detection process of the preceding vehicle and the oncoming vehicle or the streetlight will be described with reference to the time chart shown in FIG. In FIG. 10, time progresses in the horizontal direction, and the waveform in the upper part of the figure indicates that the vertical synchronization signal is input at a period of 16.6 ms. Imaging is performed in synchronization with the vertical synchronization signal, and the captured image data is transferred and detected in a field of one cycle of the vertical synchronization signal. In FIG. 10, (1) imaging with the image sensor 102, (2) transferring image data to area A of the RAM 104, (3) transferring image data from area A to area B, and from area B to area C or area 4 shows a state where four types of processing, that is, transfer to D and (4) detection processing of image data transferred to region C or region D, are performed in each field. The arrows in the figure indicate the flow of processing of each image data. Hereinafter, processing will be described for each field as time elapses.

  In field 1, image data a is imaged by exposure a. In field 2, image data a is transferred to area A of RAM 104 of CPU 110 shown in FIG. Similarly, in field 2, image data a is transferred from region A to region B for each horizontal line. Further, among the image data a transferred to the area B, the data belonging to the first area 501 having a luminance value of L1 or more is transferred to the area C, and the data belonging to the second area 502 having a luminance value of L2 or more. Transfer to region D. In this field 2, the next image data b is picked up by exposure b. In the field 3, the preceding vehicle and oncoming vehicle detection processing and streetlight detection processing shown in FIG. 6 are performed on the image data a transferred to the region C or D. Similarly, in the field 3, the image data b is transferred to the area C or the area D, and the next image data c is captured by the exposure c. In the field 4 as well, the same processing as described above is performed on the image data b and c, and the next image data d is captured by exposure d. Although only field 4 is shown in FIG. 10, the above processing is continued in the same manner after field 5 and imaging to detection processing are performed.

  As described above, the detection processing of the preceding vehicle and the oncoming vehicle and the detection processing of the streetlight are performed in synchronization with the image data capturing, and therefore the timing of the processing start is measured using the vertical synchronization signal as a trigger.

  As described above, according to the present invention, it is possible to accurately detect the state of a preceding vehicle or an oncoming vehicle or the state of a road environment using a simple and inexpensive camera or image processing device, and based on the detected state. Light distribution control can be realized.

  In addition, in the said Example, as shown to the flowchart of FIG.6 and FIG.9, according to the determination of the presence or absence of a preceding vehicle or an oncoming vehicle based on FIGS. Although the light distribution control of the own vehicle headlamp is performed, the speed control and the acceleration control of the own vehicle can be performed based on this determination.

  For example, in FIG. 9, when there is a preceding vehicle or an oncoming vehicle, in consideration of safety in relation to the preceding vehicle or the oncoming vehicle, in step 904, the vehicle of the own vehicle is used instead of or in combination with the low beam control. It is also possible to limit the speed or limit the acceleration. If neither the preceding vehicle nor the oncoming vehicle exists and the vehicle is traveling in the suburbs, in step 903, it is possible to limit the vehicle speed limit and the acceleration limit in place of or in combination with the high beam control. Such vehicle speed and acceleration limitation can be realized by giving a command to the ATC (Automatic Vehicle Control Device) for throttle control of the throttle valve opening and control of the fuel injection amount.

  In other words, the vehicle control unit used in the present invention counts the number of adjacent pixels when a plurality of pixels having a luminance equal to or higher than the predetermined value L1 are adjacent to only the lower image region of the upper image region 502 and the lower image region 501. Can be controlled by controlling the speed or acceleration of the automobile.

1 is a block diagram of a light distribution control device according to an embodiment of the present invention. Vehicle configuration diagram of light distribution control device according to this embodiment Internal configuration diagram of CPU 110 Illustration of image data for one screen Explanatory drawing of the 1st field 501 and the 2nd field 502 General flow chart of light distribution control processing General flow chart for detection of preceding and oncoming vehicles for monochrome images General flow chart of the preceding and oncoming vehicle detection process for color images General flowchart of streetlight detection processing General flowchart of headlight state determination processing Timing chart from exposure to detection processing

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... Camera, 101 ... Lens, 102 ... Imaging device, 103 ... Imaging device output signal conversion LSI, 104 ... RAM, 105 ... I / O (input / output part), 106 ... CAN (Controller Area Network), 107 ... Communication I / F (communication interface), 108: headlight driving circuit, 109: camera imaging unit, 110: camera CPU, 111 ... vehicle unit, 112 ... headlight, 200 ... vehicle, 501 ... first area, 502 ... Second area.

Claims (3)

An in-vehicle imaging device that images the front of the vehicle, an image processing unit that divides the captured image into an upper image region and a lower image region, and a luminance of a light spot detected from the upper image region and the lower image region the control apparatus for vehicles comprising a control unit for controlling at least one of the traveling state of the light distribution and a vehicle headlight Te,
The lower image area is set to a position for capturing a vehicle ahead of the host vehicle, and the upper image area is set to a position for capturing a streetlight in front of the host vehicle. The values are different,
The controller is
A headlamp is used when at least one of the upper image area and the lower image area detects the number of adjacent pixels when a plurality of adjacent pixels having a luminance equal to or higher than a predetermined value are detected. Set to low beam or reduce the amount of headlight,
If the number of adjacent pixels having a luminance equal to or higher than the predetermined value is not detected in any of the upper image area and the lower image area, the headlamp is set to a high beam or the headlight. Perform light distribution control to increase the light quantity of the lamp,
The predetermined value used for the luminance detection is a value different between the upper image region and the lower image region,
The vehicle control apparatus according to claim 1, wherein the predetermined value used for detecting the number of adjacent pixels is a value different between the upper image area and the lower image area . The vehicle control apparatus according to claim 1,
A control apparatus for an automobile , wherein a detection determination period in the upper image area is longer than a detection determination period in the lower image area . The vehicle control apparatus according to claim 1,
The control unit detects a number of pixels adjacent to each other when a plurality of pixels having a luminance equal to or higher than a predetermined value are adjacent to only the lower image region in the upper image region and the lower image region. In some cases, a vehicle control device controls the speed or acceleration of the vehicle.

Images