Vehicle position detection device, vehicle lamp system, and vehicle position detection method

By analyzing the image of the front area of ​​the vehicle and adjusting the dynamic light distribution, the glare problem of the adaptive high beam system on the driver of the vehicle ahead during night or tunnel driving is solved, and the driver's visual recognition and accuracy of vehicle position detection are improved. enhances driving safety.

CN118906976APending Publication Date: 2024-11-08KOITO MFG CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202411212807.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2019-10-29
Filing Date
2020-10-05
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Existing adaptive high beam systems are difficult to effectively avoid glare on the driver of the vehicle ahead at night or during tunnel driving, while improving the driver's visual recognition still has challenges, especially when detecting the vehicle ahead. question.

Method used

By taking images in the front area of ​​the vehicle, the presence area of ​​the vehicle in front is detected, and the margin processing target area is added to the vehicle width direction, the light shielding part is determined based on the pixel value of the light spot in the image, and the light distribution pattern is dynamically adjusted to Avoid glare and improve visual recognition.

Benefits of technology

It realizes the reduction of glare on the drivers of the vehicle in front, improves the visual recognition and safety of the drivers of the vehicle, and provides more accurate vehicle position detection and enhances driving safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118906976A_ABST
    Figure CN118906976A_ABST
Patent Text Reader

Abstract

A light distribution control device (6) is provided with: a vehicle detection unit (14) that detects a preceding vehicle by analyzing an image obtained from an imaging device (4) that captures an image of a region ahead of the vehicle; a region determination unit (16) that determines a processing target region in which a predetermined margin is added in the vehicle width direction to the presence region of the preceding vehicle; and a pattern determination unit (18) that, in parallel with the detection of the preceding vehicle by the vehicle detection unit (14), determines a light-blocking section in an image obtained from the imaging device (4) on the basis of pixel values of pairs of light spots arranged in the vehicle width direction included in the processing target region, and determines a light distribution pattern (PTN) including the light-blocking section.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a light distribution control device, a vehicle lamp system and a light distribution control method, and more particularly to a light distribution control device, a vehicle lamp system and a light distribution control method used in automobiles, etc. In addition, the present invention relates to a vehicle position detection device, a vehicle lamp system and a vehicle position detection method. Background Art

[0002] Vehicle lighting plays an important role in safe driving at night or in tunnels. If the front of the vehicle is illuminated brightly over a wide area to give priority to the driver's visibility, there is a problem of causing glare to the driver of the preceding vehicle or the oncoming vehicle in front of the vehicle.

[0003] In recent years, ADB (Adaptive Driving Beam) control has been proposed, which dynamically and adaptively controls the light distribution pattern of the high beam based on the state of the vehicle's surroundings (for example, see Patent Document 1). ADB control uses a camera to detect the presence of a light-shielding object in front of the vehicle that should be protected from high-brightness light, and dims or turns off the light in the area corresponding to the light-shielding object.

[0004] [Prior art literature] [Patent Literature] Patent Document 1: Japanese Patent Application Publication No. 2015-064964 Summary of the invention

[0005] [Problems to be solved by the invention] According to the above-mentioned ADB control, glare to the front vehicle such as the preceding vehicle or the oncoming vehicle can be avoided, and the visual recognition of the driver of the vehicle can be improved. By improving the visual recognition, the driver can more reliably identify the obstacles in front, so the driving safety is improved. On the other hand, in order to further improve the safety, there is always a demand to further improve the visual recognition of the driver. In addition, on the basis of realizing the ADB control, it is important to grasp the position of the front vehicle.

[0006] The present disclosure has been made in view of the above situation, and one of its purposes is to provide a technology for improving the visual recognition of the driver. In addition, another purpose of the present invention is to provide a new technology for detecting the position of a vehicle.

[0007] [Technical solutions for solving technical problems] In order to solve the above-mentioned problems, one solution of the present invention is a light distribution control device. The device includes: a vehicle detection unit that detects a front vehicle by analyzing an image obtained from a camera that captures a front area of ​​the vehicle; an area determination unit that determines a processing target area that is a region where the front vehicle exists plus a predetermined margin in the vehicle width direction; and a pattern determination unit that determines a light shielding portion based on the pixel values ​​of a pair of light points arranged in the vehicle width direction in the processing target area in the image obtained from the camera in parallel with the detection of the front vehicle by the vehicle detection unit, and determines a light distribution pattern including the light shielding portion.

[0008] Another embodiment of the present invention is a vehicle lighting system, which includes: a camera for photographing the area in front of the vehicle; a variable light distribution lamp capable of irradiating the area in front with a visible light beam having a variable intensity distribution; a light distribution control device of the above embodiment; and a light control device for controlling the variable light distribution lamp in a manner of forming a light distribution pattern.

[0009] In addition, another embodiment of the present invention is a light distribution control method. The control method includes the following steps: detecting a vehicle ahead by analyzing an image obtained from a camera that captures an area ahead of the vehicle, determining a processing target area that is a region where the vehicle ahead exists with a predetermined margin added to the vehicle width direction, and in parallel with the detection of the vehicle ahead, determining a light shielding portion based on pixel values ​​of pairs of light points arranged in the vehicle width direction in the processing target area in the image obtained from the camera, and determining a light distribution pattern including the light shielding portion.

[0010] In addition, another embodiment of the present invention is a vehicle position detection device that performs a first expansion process and a first contraction process on an image based on an image capturing device capturing an area in front of a vehicle using a first structural element of a predetermined shape that is long in the vehicle width direction, generates a lateral expansion area in which pairs of light spots arranged in the vehicle width direction included in the image are connected to each other, and detects the position of a vehicle in front based on the lateral expansion area.

[0011] In addition, another embodiment of the present invention is a vehicle lighting system. The system includes: a camera for photographing a front area of ​​a vehicle; a variable light distribution lamp capable of irradiating a visible light beam with a variable intensity distribution to the front area; a light distribution control device having a vehicle position detection device of the above embodiment and a pattern determination unit for determining a light distribution pattern including a light shielding unit based on a detection result of the vehicle position detection device; and a light control device for controlling the variable light distribution lamp in a manner to form a light distribution pattern.

[0012] In addition, another embodiment of the present invention is a vehicle position detection method. The control method includes the following steps: performing a first expansion process and a first contraction process on an image based on an image capturing device capturing an area in front of a vehicle using a first structural element of a predetermined shape that is long in the vehicle width direction, generating a lateral expansion area in which pairs of light spots arranged in the vehicle width direction included in the image are connected to each other, and detecting the position of a vehicle in front based on the lateral expansion area.

[0013] Furthermore, any combination of the above-described constituting elements and any form of expression of the present invention in the form of methods, apparatuses, systems, etc. may also be effective as additional modes of the present invention.

[0014] [Effects of the Invention] According to the present invention, the visibility of the driver can be improved. In addition, according to the present invention, a new technology for detecting the position of a vehicle can be provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a block diagram of the vehicle lighting system according to the first embodiment.

[0016] Figure 2 It is a diagram for explaining the operation of the light distribution control device.

[0017] Figure 3 (A)~ Figure 3 (E) is a diagram for explaining the operation of the pattern determination unit.

[0018] Figure 4 (A) and Figure 4 (B) is a flowchart showing an example of ADB control executed by the light distribution control device according to the first embodiment.

[0019] Figure 5 This is a block diagram of a vehicle lighting system according to a second embodiment.

[0020] Figure 6 It is a diagram for explaining the flow of control executed by the light distribution control device.

[0021] Figure 7 (A)~ Figure 7 (C) is a diagram showing an image generated by the control performed by the light distribution control device.

[0022] Figure 8 (A)~ Figure 8 (D) is a diagram showing structural elements used in the control performed by the light distribution control device or an image generated.

[0023] Fig. 9 (A)~ Fig. 9 (C) is a diagram showing an image generated by the control performed by the light distribution control device.

[0024] Fig.10 (A)~ Fig.10 (F) is a diagram showing structural elements used in the control performed by the light distribution control device or an image generated.

[0025] Fig.11 (A)~ Fig.11 (F) is a diagram showing structural elements used in the control performed by the light distribution control device or an image generated.

[0026] Fig.12 (A)~ Fig.12 (D) is a diagram showing structural elements used in the control performed by the light distribution control device or an image generated. DETAILED DESCRIPTION

[0027] Hereinafter, the present invention will be described based on a preferred embodiment and with reference to the accompanying drawings. The embodiment does not limit the invention but is merely an illustration, and all the features and combinations thereof described in the embodiment are not necessarily the essential contents of the invention. The same or equivalent constituent elements, components, and processes shown in the drawings are marked with the same figure numerals, and repeated descriptions are appropriately omitted. In addition, for ease of description, the scales and shapes of the various parts shown in the figures are conveniently set, and are not to be interpreted restrictively unless otherwise mentioned. In addition, when the terms "first", "second", etc. are used in this specification or claims, unless otherwise mentioned, the terms do not indicate any order or importance, but are to distinguish a certain structure from other structures. In addition, in each of the drawings, a part of the components that are not important when describing the embodiment is omitted and displayed.

[0028] (Implementation method 1) Figure 1 FIG. 1 is a block diagram of a vehicle lighting system according to Embodiment 1. Figure 1 In the figure, some of the components of the vehicle lighting system 1 are depicted as functional blocks. These functional blocks are implemented as hardware components by components and circuits such as a CPU and a memory of a computer, and as software components by computer programs, etc. It is obvious to those skilled in the art that these functional blocks can be implemented in various forms by combining hardware and software.

[0029] The vehicle lighting system 1 includes a variable light distribution lamp 2, a camera 4, a light distribution control device 6, and a light control device 8. All of them may be built into the same housing, or some of them may be provided outside the housing, in other words, on the vehicle side.

[0030] The variable light distribution lamp 2 is a white light source capable of irradiating a visible light beam L1 with a variable intensity distribution to the area in front of the vehicle. The variable light distribution lamp 2 receives data indicating a light distribution pattern PTN from the lamp control device 8, emits a visible light beam L1 with an intensity distribution corresponding to the light distribution pattern PTN, and forms the light distribution pattern PTN in front of the vehicle. The configuration of the variable light distribution lamp 2 is not particularly limited, and for example, it may include a semiconductor light source such as an LD (laser diode) or an LED (light emitting diode), and a lighting circuit that drives the semiconductor light source to light it.

[0031] In order to form an illumination distribution corresponding to the light distribution pattern PTN, the variable light distribution lamp 2 may also include a matrix-type pattern forming device such as a DMD (Digital Mirror Device) or a liquid crystal device, or a scanning optical pattern forming device that scans the front of the vehicle with light from a light source. The resolution of the variable light distribution lamp 2 is, for example, 1000 to 300,000 pixels. In addition, the time required for the variable light distribution lamp 2 to form one light distribution pattern PTN is, for example, 0.1 to 5 ms.

[0032] The camera 4 has sensitivity in the visible light region and captures the area in front of the vehicle. The camera 4 of this embodiment includes a high-speed camera 10 and a low-speed camera 12. The frame rate of the high-speed camera 10 is relatively high, for example, 200fps~10000fps (each frame is 0.1~5ms). On the other hand, the frame rate of the low-speed camera 12 is lower than that of the high-speed camera 10, for example, 30fps~120fps (each frame is about 8~33ms).

[0033] In addition, the resolution of the high-speed camera 10 is relatively small, for example, 300,000 pixels to less than 5 million pixels. On the other hand, the resolution of the low-speed camera 12 is relatively large, for example, more than 5 million pixels. Therefore, the image IMG1 generated by the high-speed camera 10 is relatively low-precision, and the image IMG2 generated by the low-speed camera 12 is relatively high-precision. That is, the image IMG1 is lower in precision than the image IMG2, and the image IMG2 is higher in precision than the image IMG1. In addition, the resolutions of the high-speed camera 10 and the low-speed camera 12 are not limited to the above-mentioned values, and can be set to any value within the scope of technical integration.

[0034] The high-speed camera 10 and the low-speed camera 12 capture the reflected light L2 of the visible light beam L1 from the object in front of the vehicle. The high-speed camera 10 and the low-speed camera 12 only need to have sensitivity at least in the wavelength range of the visible light beam L1. Preferably, the high-speed camera 10 and the low-speed camera 12 are arranged in a manner that their field of view angles are consistent with each other.

[0035] The light distribution control device 6 performs ADB control for dynamically and adaptively controlling the light distribution pattern PTN supplied to the light distribution variable lamp 2 based on the image obtained from the camera 4. The light distribution pattern PTN is understood as a two-dimensional illuminance distribution of an illumination pattern 902 formed by the light distribution variable lamp 2 on a virtual vertical screen 900 in front of the vehicle. The light distribution control device 6 may be composed of a digital processor, for example, a combination of a microcomputer including a CPU and a software program, or may be composed of an FPGA (Field Programmable Gate Array), an ASIC (Application Specified IC), or the like.

[0036] The light distribution control device 6 includes a vehicle detection unit 14, an area determination unit 16, and a pattern determination unit 18. Each unit operates by executing a program stored in a memory as an integrated circuit constituting the unit. Figure 2 1 and 12 are diagrams for explaining the operation of the light distribution control device 6. The upper part shows the output result of the vehicle detection unit 14, the middle part shows the output result of the area determination unit 16, and the lower part shows the output result of the pattern determination unit 18. In addition, the four diagrams arranged horizontally in each side are the output results of each unit at time t1 to t4.

[0037] The vehicle detection unit 14 detects the vehicle 100 in front by image analysis of the image obtained from the camera 4. The vehicle detection unit 14 of the present embodiment detects the vehicle 100 in front based on the high-definition image IMG2 obtained from the low-speed camera 12. The vehicle 100 in front includes a preceding vehicle and an oncoming vehicle. The vehicle 100 in front has a pair of light spots 102 corresponding to lamps. The pair of light spots 102 corresponds to headlights when the vehicle 100 in front is an oncoming vehicle, and corresponds to taillights when the vehicle 100 in front is a preceding vehicle. The taillights include parking lights and taillights. The pair of light spots 102 includes a left light spot 102a and a right light spot 102b.

[0038] The vehicle detection unit 14 performs high-precision image analysis using well-known methods such as algorithm recognition and deep learning, and outputs the analysis results at a low speed. For example, the vehicle detection unit 14 can output the detection result of the front vehicle 100 every 30 ms. Figure 2 In the illustrated example, the vehicle detection unit 14 outputs detection results at time t1 and time t4 .

[0039] As a result of detecting the front vehicle 100, the vehicle detection unit 14 generates angle information of the area 20 where the front vehicle 100 exists relative to the host vehicle. The angle information includes the angle θ of the left end of the front vehicle 100. L The angle θ between the right end of the vehicle in front and R . Left end angle θ L and the right end angle θR The positions of the left and right ends of the front vehicle 100 on the image IMG2 are equal, corresponding to the field of view of the low-speed camera 12. The vehicle detection unit 14 sends a signal indicating the detection result to the area determination unit 16.

[0040] The region determination unit 16 determines the processing target region 22 by adding a predetermined margin M in the vehicle width direction (left-right direction) to the region 20 where the leading vehicle 100 exists. The region determination unit 16 determines the left end angle θ L Add the left margin M L , angle θ to the right end R Add the right margin M R Therefore, the processing target area 22 is wider than the area 20 where the leading vehicle 100 exists by a margin M in the vehicle width direction. L and right margin M R The region determination unit 16 generates angle information of the processing target region 22 as information indicating a result of determination of the processing target region 22 , and transmits a signal indicating the result of determination to the pattern determination unit 18 .

[0041] The processing performed by the area determination unit 16 is only to give the margin M to the existing area 20 determined by the vehicle detection unit 14. Therefore, the processing speed of the area determination unit 16 is faster than the processing speed of the vehicle detection unit 14. For example, the area determination unit 16 can output the determination result of the processing target area 22 every 0.1 to 5 ms. Figure 2 In the illustrated example, the region determination unit 16 outputs determination results at time t1 to time t4.

[0042] The pattern determination unit 18 determines the light distribution pattern PTN in which the portion corresponding to the front vehicle 100 is shielded based on the image obtained from the camera 4. The pattern determination unit 18 of the present embodiment determines the light distribution pattern PTN based on the image IMG1 obtained from the high-speed camera 10. "Shielding a certain portion" means not only making the brightness (illuminance) of the portion completely zero, but also including reducing the brightness (illuminance) of the portion.

[0043] The pattern determination unit 18 determines the light shielding portion 24 based on the pixel values ​​of the pairs 102 of light spots arranged in the vehicle width direction in the processing target region 22 in the image IMG1 in parallel with the detection of the leading vehicle 100 by the vehicle detection unit 14 . Figure 3 (A)~ Figure 3 (E) is a diagram for explaining the operation of the pattern determination unit 18 .

[0044] like Figure 3As shown in (A), first, the pattern determination unit 18 associates the angle information of the processing target area 22 received from the area determination unit 16 with the image IMG1, and determines the processing target area 22 on the image IMG1. Then, the pattern determination unit 18 extracts the pair of pixels 26 corresponding to the pair of light spots 102 based on the pixel value of each pixel in the processing target area 22 of the image IMG1, specifically, based on the brightness value or color value. The pair of pixels 26 includes a left light spot pixel 26a overlapping the left light spot 102a and a right light spot pixel 26b overlapping the right light spot 102b.

[0045] Then, if Figure 3 As shown in (B), the pattern determination unit 18 determines the left light spot pixel 26a and the predetermined number of left upper pixels 28a arranged above the left light spot pixel 26a in the image IMG1 as the left upper expansion group 30a. In addition, the pattern determination unit 18 determines the right light spot pixel 26b and the predetermined number of right upper pixels 28b arranged above the right light spot pixel 26b in the image IMG1 as the right upper expansion group 30b.

[0046] The pattern determination unit 18 of this embodiment uses the first structural element 32 of a predetermined shape that is long in the vertical direction, and makes the pixel at the upper end of the first structural element 32 correspond to the focus pixel 32a, and performs the dilation process on the processing target area 22 of the image IMG1. As a result, in the image IMG1, the pixel value of the upper left pixel 28a is converted into the pixel value of the left light spot pixel 26a, and the pixel value of the upper right pixel 28b is converted into the pixel value of the right light spot pixel 26b. As a result, the upper dilation pattern image IMG1a including the upper left dilation group 30a and the upper right dilation group 30b is generated.

[0047] Then, if Figure 3 As shown in (C), the pattern determination unit 18 determines the left upper expansion group 30a and the predetermined number of right-column pixels 34 arranged to the right from the left upper expansion group 30a in the upper expansion pattern image IMG1a as the right expansion group 36. The pattern determination unit 18 of this embodiment uses the second structural element 38 of a predetermined shape that is long in the vehicle width direction, associates the pixel at the right end of the second structural element 38 with the pixel of interest 38a, and performs the expansion process on the processing target area 22 of the upper expansion pattern image IMG1a. The second structural element 38 has at least a length that connects the left upper expansion group 30a and the right upper expansion group 30b by the expansion process. As a result, the pixel values ​​of the right-column pixels 34 in the upper expansion pattern image IMG1a are converted into the pixel values ​​of the left upper expansion group 30a. As a result, the right expansion pattern image IMG1b including the right expansion group 36 is generated.

[0048] In addition, if Figure 3As shown in (D), the pattern determination unit 18 determines the right upper expansion group 30b and the predetermined number of left-column pixels 40 arranged to the left from the right upper expansion group 30b in the upper expansion pattern image IMG1a as the left expansion group 42. The pattern determination unit 18 of this embodiment uses the third structural element 44 of a predetermined shape that is long in the vehicle width direction, associates the pixel at the left end of the third structural element 44 with the pixel of interest 44a, and performs the expansion process on the processing target area 22 of the upper expansion pattern image IMG1a. The third structural element 44 has at least a length that connects the left upper expansion group 30a and the right upper expansion group 30b by the expansion process. As a result, the pixel values ​​of the left-column pixels 40 in the upper expansion pattern image IMG1a are converted into the pixel values ​​of the right upper expansion group 30b. As a result, the left expansion pattern image IMG1c including the left expansion group 42 is generated.

[0049] Then, if Figure 3 As shown in (E), the pattern determination unit 18 includes the pixel area 46 where the right expansion group 36 and the left expansion group 42 overlap in the light-shielding portion. The pattern determination unit 18 of this embodiment synthesizes the right expansion pattern image IMG1b and the left expansion pattern image IMG1c, that is, performs an AND operation on the right expansion pattern image IMG1b and the left expansion pattern image IMG1c, determines the pixel area 46 where the right expansion group 36 and the left expansion group 42 overlap, and determines the light-shielding portion 24 (see FIG. 24 ) so as to include the pixel area 46. Figure 2 ).

[0050] Then, the pattern determination unit 18 determines the light distribution pattern PTN including the light shielding portion 24. For example, the pattern determination unit 18 determines a predetermined first illuminance for the area excluding the light shielding portion 24, and determines a second illuminance lower than the first illuminance for the light shielding portion 24. As described above, the second illuminance may be zero, or an illuminance higher than zero and lower than the first illuminance. The pattern determination unit 18 transmits data indicating the light distribution pattern PTN to the lamp control device 8.

[0051] The processing object of the pattern determination unit 18 is limited to the processing object area 22 in the image IMG1. In addition, the pattern determination unit 18 determines the shading portion 24 based on the pixel value (brightness value or color value) of the image IMG1. Specifically, the pattern determination unit 18 performs the above-mentioned dilation processing on the image IMG1 to convert the pixel value of each pixel, thereby determining the shading portion 24. Therefore, the processing performed by the pattern determination unit 18 is completed in a shorter time compared to the high-level image analysis performed by the vehicle detection unit 14. Therefore, the processing speed of the pattern determination unit 18 is faster than the processing speed of the vehicle detection unit 14. For example, the pattern determination unit 18 can output the determination result of the light distribution pattern PTN every 0.1~5ms. Figure 2In the illustrated example, the pattern determination unit 18 outputs determination results at time t1 to time t4.

[0052] also, Figure 3 The production process of the right expansion pattern image IMG1b shown in (C) and Figure 3 The order of creating the left expansion pattern image IMG1c shown in (D) is different, and it can also be processed in parallel. Figure 3 (B) ~ Figure 3 In (D), the first structural element 32 to the third structural element 44 are schematically shown, and the number of pixels constituting the first structural element 32 to the third structural element 44 is not limited to the number shown in the figure. In addition, the existing area 20 and the processing target area 22 may be defined not only in the vehicle width direction but also in the vertical direction.

[0053] The lamp control device 8 controls the variable light distribution lamp 2 in such a manner that the visible light beam L1 having an intensity distribution corresponding to the light distribution pattern PTN determined by the pattern determination unit 18 is emitted. For example, when the variable light distribution lamp 2 includes a DMD, the lamp control device 8 controls the lighting and extinguishing of the light source and the on / off switching of each reflector element constituting the DMD. The lamp control device 8 can send a driving signal to the variable light distribution lamp 2 every 0.1 to 5 ms, for example.

[0054] Thus, a light distribution pattern PTN having a light shielding portion 24 overlapping with the front vehicle 100 is formed, and the visual recognition of the driver of the vehicle 100 can be improved without causing glare to the driver of the front vehicle 100. As described above, the pattern determination unit 18 has a faster processing speed than the vehicle detection unit 14, and can determine the light shielding portion 24 and the light distribution pattern PTN at a higher frequency than the frequency of outputting the detection result of the vehicle detection unit 14. Therefore, compared with the case where the light shielding portion 24 is directly determined based on the detection result of the vehicle detection unit 14, the light shielding portion 24 can follow the movement of the front vehicle 100 with high accuracy.

[0055] In addition, as described above, the processing speed of the vehicle detection unit 14 is relatively slow (that is, slower than the area determination unit 16). Figure 2In the example shown, the detection results are output at time t1 and time t4. Therefore, at time t2 and t3, the front vehicle 100 deviates from the existence area 20 determined by the vehicle detection unit 14. On the other hand, the processing speed of the area determination unit 16 is relatively faster (that is, faster than the vehicle detection unit 14), and the processing target area 22 is determined at a frequency shorter than the frequency of the vehicle detection unit 14 outputting the detection results. Therefore, at time t2 and t3, the processing target area 22 is determined for the existence area 20 output from the vehicle detection unit 14 at time t1. That is, the area determination unit 16 repeatedly determines the processing target area 22 for the same detection result (existence area 20) obtained from the vehicle detection unit 14 at a predetermined timing (time t2, t3).

[0056] As the number of times the processing target area 22 is determined for the same existing area 20 increases, the area determination unit 16 gradually increases the size of the margin M. Therefore, the processing target area 22 gradually expands. As a result, even if the front vehicle 100 deviates from the existing area 20, the possibility that the front vehicle 100 deviates from the processing target area 22 can be reduced. Therefore, the front vehicle 100 can be shielded more reliably. In addition, by reducing the margin M at the initial stage of setting the processing target area 22, the processing speed of the pattern determination unit 18 can be increased. In addition, when receiving a new detection result from the vehicle detection unit 14, the area determination unit 16 returns the margin M to the initial value.

[0057] The pattern determination unit 18 may also determine whether the pair 102 of light points included in the processing target area 22 is the headlight of the oncoming vehicle or the taillight of the preceding vehicle. For example, the pattern determination unit 18 performs grayscale conversion processing on the processing target area 22. Then, the brightness value of each pixel is binarized, thereby extracting the headlight. In addition, the pattern determination unit 18 performs HSV conversion processing on the processing target area 22. Then, the color value of each pixel is binarized, thereby extracting the taillight.

[0058] Then, an OR operation is performed on the extraction result image of the headlights and the extraction result image of the taillights to generate an image including the paired pixels 26 corresponding to the headlights and the paired pixels 26 corresponding to the taillights. Based on this image, an upper expansion pattern image IMG1a including the left upper expansion group 30a and the right upper expansion group 30b is obtained.

[0059] As an example, when the pattern determination unit 18 sets the shading portion 24 for the pixel 26 corresponding to the headlight of the oncoming vehicle, the pixel area 46 where the right expansion group 36 and the left expansion group 42 overlap is determined as the shading portion 24. On the other hand, when the shading portion 24 is set for the pixel 26 corresponding to the rear light of the preceding vehicle, the area of ​​the predetermined margin on the left and right sides of the pixel area 46 is set as the shading portion 24. The predetermined margin corresponds to the area overlapping with the side mirror of the preceding vehicle. Thus, the glare caused to the driver of the preceding vehicle can be suppressed.

[0060] Figure 4 (A) and Figure 4 (B) is a flowchart showing an example of ADB control performed by the light distribution control device 6 of embodiment 1. This process is performed by, for example, a light switch (not shown) to instruct the execution of ADB control, and is repeatedly performed at a predetermined timing when the ignition is turned on, and ends when the execution instruction of ADB control is released (or a stop instruction is given) or the ignition is turned off. In addition, Figure 4 The process shown in (A) and Figure 4 The processes shown in (B) are executed in parallel.

[0061] like Figure 4 As shown in (A), the light distribution control device 6 determines whether a new image IMG2 is received from the low-speed camera 12 (S101). When a new image IMG2 is received (Yes in S101), the light distribution control device 6 detects the vehicle 100 ahead based on the image IMG2 and updates the existing area 20 (S102). Next, the light distribution control device 6 updates the processing target area 22 based on the obtained existing area 20 (S103), and ends this program. When a new image IMG2 is not received from the low-speed camera 12 (No in S101), the light distribution control device 6 updates the processing target area 22 based on the already obtained existing area 20 (S103).

[0062] In addition, if Figure 4As shown in (B), the light distribution control device 6 determines whether a new image IMG1 is received from the high-speed camera 10 (S201). If the new image IMG1 is not received (No in S201), this program is terminated. If the new image IMG1 is received (Yes in S201), the light distribution control device 6 sets the processing object area 22 updated in step S103 to the image IMG1 (S202). Next, the light distribution control device 6 extracts the light points in the processing object area 22, that is, the pixels 26 (S203). Next, the light distribution control device 6 determines the light shielding portion 24 through expansion processing (S204). Then, the light distribution control device 6 sends data indicating the new light distribution pattern PTN to the light control device 8 to update the light distribution pattern PTN (S205), and this program is terminated.

[0063] As described above, the light distribution control device 6 of the present embodiment includes: a vehicle detection unit 14 that detects a leading vehicle by image analysis of an image obtained from a camera 4 that captures an area in front of the vehicle; an area determination unit 16 that determines a processing target area 22 that is a region 20 where the leading vehicle 100 exists plus a specified margin M in the vehicle width direction; and a pattern determination unit 18 that, in parallel with the detection of the leading vehicle 100 by the vehicle detection unit 14, determines a shading portion 24 based on pixel values ​​of a pair 102 of light points arranged in the vehicle width direction contained in the processing target area 22 in the image obtained from the camera 4, and determines a light distribution pattern PTN including the shading portion 24.

[0064] The pattern determination unit 18 determines the light shielding unit 24 based on the pixel value of the pixel. Therefore, the pattern determination unit 18 can perform processing at a higher speed than the vehicle detection unit 14 that performs image analysis. In addition, the image area where the pattern determination unit 18 performs processing is limited to the processing target area 22. In addition, the pattern determination unit 18 performs processing in parallel with the vehicle detection unit 14. Therefore, according to the light distribution control device 6 of this embodiment, the light distribution pattern PTN can be updated at a high frequency. As a result, the light shielding unit 24 can follow the movement of the front vehicle 100 with high accuracy.

[0065] In the conventional ADB control in which the front vehicle is detected by image analysis and the shading portion is determined directly based on the result, a margin is provided in the shading portion in consideration of the deviation between the front vehicle and the shading portion caused by the delay in the decision processing of the shading portion. In contrast, according to the present embodiment, since the shading portion 24 can be made to coincide with the front vehicle 100 with high precision, the margin provided in the shading portion can be reduced. Therefore, the visual recognition of the driver can be improved.

[0066] In addition, the pattern determination unit 18 of the present embodiment determines, in the image obtained from the imaging device 4, the left light spot pixel 26a overlapping the left light spot 102a of the pair of light spots 102 and the predetermined number of left upper pixels 28a arranged above the left light spot pixel 26a as the left upper expansion group 30a, and determines the right light spot pixel 26b overlapping the right light spot 102b and the predetermined number of right upper pixels 28b arranged above the right light spot pixel 26b as the right upper expansion group 30b. In addition, the pattern determination unit 18 determines the left upper expansion group 30a and the predetermined number of right column pixels 34 arranged to the right from the left upper expansion group 30a as the right expansion group 36, and determines the right upper expansion group 30b and the predetermined number of left column pixels 40 arranged to the left from the right upper expansion group 30b as the left expansion group 42. In addition, the pattern determination unit 18 includes the pixel region 46 where the right expansion group 36 and the left expansion group 42 overlap in the light shielding portion. This can further increase the processing speed of the light distribution control device 6 .

[0067] In addition, the image decision unit 18 converts the pixel value of the upper left pixel 28a into the pixel value of the left light spot pixel 26a, and converts the pixel value of the upper right pixel 28b into the pixel value of the right light spot pixel 26b in the image obtained from the imaging device 4, thereby creating an upper expansion pattern image IMG1a including the upper left expansion group 30a and the upper right expansion group 30b. In addition, the pattern decision unit 18 converts the pixel value of the right column pixel 34 in the upper expansion pattern image IMG1a into the pixel value of the upper left expansion group 30a, thereby creating a right expansion pattern image IMG1b including the right expansion group 36. In addition, the pattern decision unit 18 converts the pixel value of the left column pixel 40 in the upper expansion pattern image IMG1a into the pixel value of the upper right expansion group 30b, thereby creating a left expansion pattern image IMG1c including the left expansion group 42. In addition, the pattern decision unit 18 synthesizes the right expansion pattern image IMG1b and the left expansion pattern image IMG1c to determine the light shielding portion 24. This can further increase the processing speed of the light distribution control device 6 .

[0068] The region determination unit 16 repeatedly determines the processing target region 22 for the same detection result obtained from the vehicle detection unit 14 , and gradually increases the size of the margin M as the number of times the processing target region 22 is determined increases. This makes it possible to more reliably shield the existing region 20 of the leading vehicle 100 .

[0069] In addition, the camera 4 includes a high-speed camera 10 and a low-speed camera 12 having a lower frame rate than the high-speed camera 10, the vehicle detection unit 14 detects the front vehicle 100 based on the image IMG2 obtained from the low-speed camera 12, and the pattern determination unit 18 determines the light distribution pattern PTN based on the image IMG1 obtained from the high-speed camera 10. Thus, the ADB control can be performed with higher accuracy. In addition, by assigning cameras to the vehicle detection unit 14 and the pattern determination unit 18, respectively, it is possible to use cameras that are specialized for the performance required for each process. Therefore, according to this embodiment, the camera 4 can be made low-cost, and the vehicle lighting system 1 can be made low-cost.

[0070] Above, the first embodiment of the present invention has been described in detail. The above-mentioned first embodiment only represents a specific example when the present invention is implemented. The content of the first embodiment does not limit the technical scope of the present invention. Various design changes such as changes, additions, and deletions of constituent elements can be made without departing from the scope of the idea of ​​the invention specified in the claims. The new embodiment with the design changes has the respective effects of the combined embodiment and the deformation. In the above-mentioned first embodiment, the contents that can be subjected to such design changes are marked with expressions such as "this embodiment" and "in this embodiment" to emphasize, but the contents without such expressions also allow design changes. Any combination of the above constituent elements is also valid as a scheme of the present invention. The hatching marked on the cross section of the accompanying drawings does not limit the material of the object marked with the hatching.

[0071] The functions of both the high-speed camera 10 and the low-speed camera 12 may be realized by one camera. For example, if the high-speed camera 10 has a resolution equivalent to that of the low-speed camera 12, and if the vehicle detection unit 14 has an algorithm that can perform sufficient vehicle detection even if the resolution of the high-speed camera 10 is low, the low-speed camera 12 may be omitted. Thus, the vehicle lighting system 1 can be miniaturized.

[0072] The invention of the above-mentioned first embodiment can also be specified by the items described below.

[0073] (Item 1) A light distribution control method includes: detecting a front vehicle 100 by image processing of an image obtained from a camera 4 that captures an area in front of the vehicle, The processing target area 22 is determined by adding a predetermined margin M to the area 20 where the preceding vehicle 100 exists in the vehicle width direction, and In parallel with the detection of the preceding vehicle 100 , in the image obtained from the camera 4 , the light shielding portion 24 is identified based on the pixel values ​​of the pairs 102 of light spots arranged in the vehicle width direction included in the processing target region 22 , and the light distribution pattern PTN including the light shielding portion 24 is determined.

[0074] (Item 2) In the light distribution control method of item 1, when determining the light distribution pattern PTN, In the image, the left light spot pixel 26 overlapping the left light spot 102a of the light spot pair 102 and a predetermined number of left upper pixels 28a arranged above the left light spot pixel 26a are determined as a left upper expansion group 30a, and the right light spot pixel 26b overlapping the right light spot 102b and a predetermined number of right upper pixels 28b arranged above the right light spot pixel 26b are determined as a right upper expansion group 30b. The upper left expansion group 30a and a predetermined number of right-column pixels 34 arranged to the right from the upper left expansion group 30a are determined as a right expansion group 36. The right upper expansion group 30b and a predetermined number of left-column pixels 40 arranged to the left from the right upper expansion group 30b are determined as a left expansion group 42. A pixel region 46 where the right expansion group 36 and the left expansion group 42 overlap is determined as the light shielding portion 24 .

[0075] (Item 3) The light distribution control method described in item 2, When determining the light distribution pattern PTN, In the image, by converting the pixel value of the upper left pixel 28a into the pixel value of the left light spot pixel 26a, and converting the pixel value of the upper right pixel 28b into the pixel value of the right light spot pixel 26b, an upper expansion pattern image IMG1a including the upper left expansion group 30a and the upper right expansion group 30b is created. In the upper expansion pattern image IMG1a, by converting the pixel values ​​of the right column pixels 34 into the pixel values ​​of the left upper expansion group 30a, a right expansion pattern image IMG1b including the right expansion group 36 is generated. In the upper expansion pattern image IMG1a, by converting the pixel values ​​of the left column pixels 40 into the pixel values ​​of the right upper expansion group 30b, a left expansion pattern image IMG1c including the left expansion group 42 is generated. The light shielding portion 24 is determined by synthesizing the right expansion pattern image IMG1 b and the left expansion pattern image IMG1 c .

[0076] (Implementation method 2) Figure 5 FIG. 2 is a block diagram of a vehicle lighting system according to Embodiment 2. Figure 5 In the embodiment, some of the components of the vehicle lighting system 1001 are depicted as functional blocks. These functional blocks are implemented as hardware components by components and circuits such as a CPU and a memory of a computer, and as software components by computer programs, etc. It is obvious to those skilled in the art that these functional blocks can be implemented in various forms by combining hardware and software.

[0077] The vehicle lighting system 1001 includes a variable light distribution lamp 1002, a camera 1004, a light distribution control device 1006, and a light control device 1008. All of them may be built into the same housing, or some of them may be provided outside the housing, in other words, on the vehicle side.

[0078] The variable light distribution lamp 1002 is a white light source capable of irradiating a visible light beam L1 with a variable intensity distribution to the area in front of the vehicle. The variable light distribution lamp 1002 receives data indicating a light distribution pattern PTN from the lamp control device 1008, emits a visible light beam L1 with an intensity distribution corresponding to the light distribution pattern PTN, and forms the light distribution pattern PTN in front of the vehicle. The configuration of the variable light distribution lamp 1002 is not particularly limited, and for example, it may include a semiconductor light source such as an LD (laser diode) or an LED (light emitting diode), and a lighting circuit that drives the semiconductor light source to light it.

[0079] In order to form an illumination distribution corresponding to the light distribution pattern PTN, the variable light distribution lamp 1002 may also include a matrix-type pattern forming device such as a DMD (Digital Mirror Device) or a liquid crystal device, or a scanning optical pattern forming device that scans the front of the vehicle with light from a light source. The resolution of the variable light distribution lamp 1002 is, for example, 1000 to 300,000 pixels.

[0080] The camera 1004 has sensitivity in the visible light region and captures the area in front of the vehicle. The camera 1004 captures the reflected light L2 of the object in front of the vehicle to the visible light beam L1. The camera 1004 only needs to have sensitivity in at least the wavelength range of the visible light beam L1.

[0081] The light distribution control device 1006 performs ADB control to dynamically and adaptively control the light distribution pattern PTN supplied to the light distribution variable lamp 1002 based on the image IMG obtained from the camera 1004. The light distribution pattern PTN is understood as a two-dimensional illuminance distribution of an illumination pattern 902 formed by the light distribution variable lamp 1002 on the virtual vertical screen 900 in front of the vehicle. The light distribution control device 1006 can be composed of a digital processor, for example, a combination of a microcomputer including a CPU and a software program, or an FPGA (Field Programmable Gate Array), an ASIC (Application Specified IC), etc.

[0082] The light distribution control device 1006 includes a vehicle position detection device 1010 and a pattern determination unit 1012. Each unit operates by executing a program stored in a memory as an integrated circuit constituting the unit. Figure 6 1004 is a diagram for explaining the flow of control executed by the light distribution control device 1006 . Figure 7 (A)~ Figure 7 (C) Figure 8 (A)~ Figure 8 (D) Fig. 9 (A)~ Fig. 9 (C) Fig.10 (A)~ Fig.10 (F) Fig.11 (A)~ Fig.11 (F) and Fig.12 (A)~ Fig.12 (D) is a diagram showing structural elements used in the control executed by the light distribution control device 1006 or an image generated.

[0083] The vehicle position detection device 1010 detects the position of the front vehicle including the oncoming vehicle and the preceding vehicle by performing a prescribed image processing on the image based on the camera 1004. The image based on the camera 1004 includes not only the image IMG directly obtained from the camera 1004, but also an image on which a prescribed image processing is performed on the image IMG, that is, an image derived from the directly obtained image IMG.

[0084] First, the vehicle position detection device 1010 obtains an image IMG (camera image) from the imaging device 1004 (S1101). Figure 7 As shown in (A), the image IMG obtained from the camera 1004 includes light spots corresponding to the rear lights RL of the preceding vehicle (hereinafter referred to as rear lights RL) and light spots corresponding to the headlights HL of the opposite vehicle (hereinafter referred to as headlights HL). The rear lights RL include parking lights and tail lights.

[0085] The vehicle position detection device 1010 performs HSV conversion processing on the image IMG, and performs binarization processing on the HSV image using a predetermined threshold value for the rear lights RL (S1102). Figure 7 As shown in (B), a rear light image IMGa in which the rear light RL is extracted is obtained. In the rear light image IMGa, pixels corresponding to the rear light RL have high pixel values, and other pixels have low pixel values.

[0086] In addition, the vehicle position detection device 1010 performs HSV conversion processing on the image IMG, and performs binarization processing on the HSV image using a predetermined threshold value for the headlight HL (S1103). Figure 7 As shown in (C), a headlight image IMGb in which the headlight HL is extracted is obtained. In the headlight image IMGb, pixels corresponding to the headlight HL have high pixel values, and other pixels have low pixel values.

[0087] The extraction process of the rear lamps RL (step S1102 ) and the extraction process of the headlights HL (step S1103 ) may be performed in different orders and in parallel. The method of each extraction process is not particularly limited, and a known method may be used.

[0088] Vehicle position detection device 1010 uses Figure 8 The second structural element 1014 of the predetermined shape shown in (A) is subjected to a second contraction process a predetermined number of times on the image based on the imaging device 1004, specifically, on the rear light image IMGa (S1104). Figure 6 In the example, the number of times the second contraction process is performed on the taillight image IMGa is set to 3 times, but the number is not limited to this and the number can be set appropriately.

[0089] The second structural element 1014 of this embodiment has a cross shape. The vehicle position detection device 1010 associates the center pixel of the second structural element 1014 with the focus pixel 1014a and performs the second contraction processing on the taillight image IMGa. Figure 8 In (A), the second structural element 1014 is schematically shown, and the number of pixels constituting the portion extending upward, downward, leftward, and rightward from the center pixel of the second structural element 1014 is not limited to the number shown in the figure.

[0090] In the second contraction process, when any pixel overlapped by the second structural element 1014 includes a pixel with a low pixel value, the pixel value of the focus pixel 1014a is converted to the low pixel value. Therefore, as the number of times the second contraction process is performed increases, the light spots with smaller sizes are deleted in sequence. Generally, the size of the headlight HL or the taillight RL of the vehicle in front increases as it is closer to the vehicle. Therefore, as the number of times the second contraction process is performed increases, the headlight HL or the taillight RL of the vehicle in front that is farther away from the vehicle is deleted in sequence.

[0091] The vehicle position detection device 1010 relatively reduces the number of times the second shrinking process is applied to the rear light image IMGa, and generates a first image including a pair of rear lights RL located at a predetermined first distance and a pair of rear lights RL located at a second distance that is longer than the first distance. In addition, the vehicle position detection device 1010 relatively increases the number of times the second shrinking process is applied, and generates a second image including a pair of rear lights RL located at the first distance and deleting the pair of rear lights RL located at the second distance. That is, the number of times the second shrinking process is applied when the first image is generated is less than the number of times the second shrinking process is applied when the second image is generated, and the number of times the second shrinking process is applied when the second image is generated is more than the number of times the second shrinking process is applied when the first image is generated.

[0092] Figure 8 (B) is a rear light image IMGa0 in which the number of times of the second contraction process is 0. Figure 8 (C) is the taillight image IMGa1 after the second contraction process is performed once. Figure 8 (D) is a taillight image IMGa3 in which the second contraction process is performed three times. The vehicle position detection device 1010 also generates a taillight image IMGa in which the second contraction process is performed twice, but this is not shown in the figure.

[0093] The rear light image IMGa0 includes a pair RL1 of rear lights RL located at a predetermined first distance and a pair RL2 of rear lights RL located at a second distance greater than the first distance. On the other hand, the rear light image IMGa1 includes the pair RL1 of rear lights RL located at the first distance, but the pair RL2 of rear lights RL located at the second distance is deleted. Therefore, the rear light image IMGa0 is equivalent to the first image, and the rear light image IMGa1 is equivalent to the second image. In addition, the rear light images IMGa0 and IMGa1 also include a pair RL3 of rear lights RL located closer than the pair RL1 of rear lights RL.

[0094] In addition, the rear light image IMGa1 includes a pair RL3 of rear lights RL located at a predetermined first distance and a pair RL1 of rear lights RL located at a second distance greater than the first distance. On the other hand, the rear light image IMGa3 includes a pair RL3 of rear lights RL located at the first distance, but the pair RL1 of rear lights RL located at the second distance is deleted. Therefore, the rear light image IMGa1 is equivalent to the first image, and the rear light image IMGa3 is equivalent to the second image.

[0095] The vehicle position detection device 1010 also performs the same processing on the headlight image IMGb. Figure 8 The second structural element 1014 shown in (A) of FIG. 11 performs a second contraction process on the headlight image IMGb a predetermined number of times ( S1105 ). Figure 6 In FIG. 5 , the number of times the second contraction process is performed on the headlight image IMGb is set to 5, but the present invention is not limited thereto and the number of times can be set as appropriate.

[0096] The vehicle position detection device 1010 relatively reduces the number of times the second shrinking process is applied to the headlamp image IMGb, and generates a first image including a pair of headlamps HL located at a predetermined first distance and a pair of headlamps HL located at a second distance greater than the first distance. In addition, the vehicle position detection device 1010 relatively increases the number of times the second shrinking process is applied, and generates a second image including a pair of headlamps HL located at the first distance and deleting a pair of headlamps HL located at the second distance. That is, the number of times the second shrinking process is applied when the first image is generated is less than the number of times the second shrinking process is applied when the second image is generated, and the number of times the second shrinking process is applied when the second image is generated is greater than the number of times the second shrinking process is applied when the first image is generated.

[0097] Fig. 9 (A) is a headlight image IMGb0 in which the number of times of the second contraction process is 0. Fig. 9 (B) is a headlight image IMGb1 in which the second contraction process is performed once. Fig. 9 (C) is a headlight image IMGb5 that has been subjected to the second contraction process five times. The vehicle position detection device 1010 also generates a headlight image IMGb that has been subjected to the second contraction process two to four times, but this is not shown in the figure.

[0098] The headlight image IMGb0 includes a pair HL1 of headlights HL located at a predetermined first distance and a pair HL2 of headlights HL located at a second distance greater than the first distance. On the other hand, the headlight image IMGb1 includes the pair HL1 of headlights HL located at the first distance, but the pair HL2 of headlights HL located at the second distance is deleted. Therefore, the headlight image IMGb0 is equivalent to the first image, and the headlight image IMGb1 is equivalent to the second image. In addition, the headlight images IMGb0 and IMGb1 also include a pair HL3 of headlights HL located closer than the pair HL1 of headlights HL.

[0099] In addition, the headlight image IMGb1 includes the pair HL3 of the headlight HL located at a predetermined first distance and the pair HL1 of the headlight HL located at a second distance longer than the first distance. On the other hand, the headlight image IMGb5 includes the pair HL3 of the headlight HL located at the first distance, but the pair HL1 of the headlight HL located at the second distance is deleted. Therefore, the headlight image IMGb1 is equivalent to the first image, and the headlight image IMGb5 is equivalent to the second image.

[0100] Furthermore, the second contraction process (step S1104 ) on the rear light image IMGa and the second contraction process (step S1105 ) on the headlight image IMGb may be performed in different orders and may be processed in parallel.

[0101] Vehicle position detection device 1010 uses Fig.10 (A) Fig.10 (C) and Fig.10 The first structural element 1016 of the predetermined shape shown in (E) of FIG. 1106 performs the first expansion process and the first contraction process on the image based on the imaging device 1004 , specifically, on the taillight images IMGa0 to IMGa3 ( S1106 ).

[0102] The first structural element 1016 of this embodiment has a shape that is long in the vehicle width direction (lateral direction). The vehicle position detection device 1010 associates the center pixel of the first structural element 1016 with the focus pixel 1016a, performs a first expansion process on the rear light images IMGa0 to IMGa3, and then performs a first contraction process. Fig.10 (A) Fig.10 (C) and Fig.10 In (E), the first structural element 1016 is schematically shown, and the number of pixels constituting the portion extending to the left and right from the center pixel of the first structural element 1016 is not limited to the number shown in the figure.

[0103] In the first expansion process, when any pixels overlapped by the first structural element 1016 include pixels with high pixel values, the pixel value of the pixel of interest 1016a is converted to the high pixel value. In addition, in the first contraction process, when any pixels overlapped by the first structural element 1016 include pixels with low pixel values, the pixel value of the pixel of interest 1016a is converted to the low pixel value. Thus, a lateral expansion region 1018 is generated in which pairs of rear lamps RL arranged in the vehicle width direction included in the rear lamp images IMGa0 to IMGa3 are connected to each other.

[0104] The vehicle position detection device 1010 uses the first structural element 1016 with a relatively short length in the vehicle width direction in the first image and uses the first structural element 1016 with a relatively long length in the vehicle width direction in the second image, and performs the first expansion process and the first contraction process on each image. That is, the first structural element 1016 used in the first image is shorter in the vehicle width direction than the first structural element 1016 used in the second image, and the first structural element 1016 used in the second image is longer in the vehicle width direction than the first structural element 1016 used in the first image. In other words, the vehicle position detection device 1010 performs the first expansion process and the first contraction process on the image in which the number of times the second contraction process is performed is smaller, using the first structural element 1016 with a shorter length in the vehicle width direction.

[0105] Fig.10 (A) is a first structural element 1016 b used in the first expansion process and the first contraction process of the taillight image IMGa0 . Fig.10 (C) is a first structural element 1016 c used in the first expansion process and the first contraction process of the taillight image IMGa1 . Fig.10 (E) is a first structural element 1016d used in the first expansion process and the first contraction process of the taillight image IMGa3.

[0106] Generally, the distance between the pair of rear lights RL gradually increases as they approach the vehicle. Therefore, in the first expansion process using the first structural element 1016 having a shorter length in the vehicle width direction, the pair of rear lights RL far from the vehicle can be connected, but the pair of rear lights RL close to the vehicle cannot be connected. Therefore, by performing the first expansion process using the first structural element 1016 having different lengths, the pair to be generated in the lateral expansion region 1018 can be selected according to the distance from the vehicle.

[0107] The first structural element 1016b used in the rear light image IMGa0 as the first image is shorter in the left-right direction than the first structural element 1016c used in the rear light image IMGa1 as the second image. In addition, the number of times of the second expansion process of the rear light image IMGa0 is 0, and it includes not only the pairs RL1 and RL3 of the rear lights RL closer to the vehicle but also the pair RL2 of the rear lights RL farther from the vehicle. Therefore, when the first expansion process is performed on the rear light image IMGa0 using the first structural element 1016b, as shown in FIG. Fig.10 As shown in (B), the pair RL2 of the rear lamps RL are connected to form the lateral expansion region 1018, but the pairs RL1 and RL3 are not connected and remain separated.

[0108] On the other hand, the first structural element 1016c used in the rear light image IMGa1 as the second image is longer in the left and right directions than the first structural element 1016b. Therefore, when the first dilation process is performed on the rear light image IMGa1 using the first structural element 1016c, Fig.10 As shown in (D), the pair of rear lamps RL1 closer to the host vehicle is connected to form the lateral expansion region 1018. In addition, the pair of rear lamps RL3 closer to the host vehicle than the pair of rear lamps RL1 is kept separated.

[0109] In addition, the first structural element 1016c used in the rear light image IMGa1 as the first image is shorter in the left-right direction than the first structural element 1016d used in the rear light image IMGa3 as the second image. Therefore, when the first dilation process is performed on the rear light image IMGa1 using the first structural element 1016c, Fig.10 As shown in (D), the pair RL1 of the rear lamps RL are connected to form the lateral expansion region 1018, but the pair RL3 are not connected and remain separated.

[0110] On the other hand, the first structural element 1016d used in the rear light image IMGa3 as the second image is longer in the left and right directions than the first structural element 1016c. Therefore, when the first dilation process is performed on the rear light image IMGa3 using the first structural element 1016d, Fig.10 As shown in (F), the pair of rear lights RL3 closer to the vehicle are connected to form a lateral expansion region 1018 .

[0111] The length of each first structural element 1016 can be appropriately set according to the interval of the pair of light spots that are the generation target of the lateral expansion region 1018. In addition, the correspondence between the number of times of the second contraction process and the length of the first structural element 1016 can also be appropriately set. For example, the length of the first structural element 1016 used in each of the taillight images IMGa0 to IMGa3 is set corresponding to the pair of light spots with the shortest interval among the pairs of light spots remaining after each number of the second contraction process.

[0112] The rear light image IMGa1 does not include the pair RL2 of the rear light RL due to the second contraction process. Therefore, by using the first expansion process of the first structural element 1016c, the generation of the lateral expansion region 1018 originating from the pair RL2 of the rear light RL can be avoided. Similarly, the rear light image IMGa3 does not include the pair RL1 and RL2 of the rear light RL due to the second contraction process. Therefore, by using the first expansion process of the first structural element 1016d, the generation of the lateral expansion region 1018 originating from the pair RL1 and RL2 of the rear light RL can be avoided.

[0113] If the first expansion treatment is performed using the first structural elements 1016c and 1016d that are relatively longer than the rear light RL2, the resulting lateral expansion region 1018 is likely to be longer in the vehicle width direction than the lateral expansion region 1018 obtained using the relatively shorter first structural element 1016b.

[0114] For example, when the above-mentioned extraction process of each lamp is not performed, that is, when the headlamp HL and the rear lamp RL are extracted, Figure 7 When the first expansion process is performed on the image IMG of (A), if relatively long first structural elements 1016c and 1016d are used, a lateral expansion region 1018 may be generated by the pair RL2 of rear lights RL and the pair HL2 of head lights HL arranged in the left and right directions.

[0115] Alternatively, when the vehicle is traveling on a road with multiple lanes on one side, Figure 7 All the light spots in the image IMG of (A) may become the rear lights RL. In this case, even when the above-mentioned rear light RL extraction process is performed, that is, even when the first expansion process is performed on the rear light image IMGa, by using the relatively long first structural elements 1016c and 1016d, it is possible that a lateral expansion area 1018 is generated by the pair RL2 of the two groups of rear lights RL arranged in the left and right directions.

[0116] In these cases, regardless of the fact that two vehicles are actually aligned in front, it is determined that there is one vehicle in the vehicle position detection based on the lateral expansion area 1018. In contrast, by making the length of the first structural element 1016 different according to the interval between the pairs of light spots, the lateral expansion area 1018 corresponding to each front vehicle can be generated more reliably. Therefore, the position of the vehicle can be detected more accurately.

[0117] The vehicle position detection device 1010 also performs the same processing on the headlight image IMGb. Fig.11 (A) Fig.11 (C) and Fig.11 The first structural element 1016 of a predetermined shape that is longer in the vehicle width direction and schematically shown in (E) of the embodiment of the present invention performs the first expansion process and the first contraction process on the headlamp images IMGb0 to IMGb5 (S1107). The vehicle position detection device 1010 associates the center pixel of the first structural element 1016 with the focus pixel 1016a, performs the first expansion process on the headlamp images IMGb0 to IMGb5, and then performs the first contraction process. Through the first expansion process and the first contraction process, a lateral expansion region 1018 is generated in which the pairs of headlamps HL arranged in the vehicle width direction included in the headlamp image IMGb are connected to each other.

[0118] The vehicle position detection device 1010 uses the first structural element 1016 with a relatively short length in the vehicle width direction in the first image and uses the first structural element 1016 with a relatively long length in the vehicle width direction in the second image, and performs the first expansion process and the first contraction process on each image. That is, the first structural element 1016 used in the first image is shorter in the vehicle width direction than the first structural element 1016 used in the second image, and the first structural element 1016 used in the second image is longer in the vehicle width direction than the first structural element 1016 used in the first image.

[0119] Fig.11 (A) is a first structural element 1016 e used in the first expansion process and the first contraction process of the headlight image IMGb0 . Fig.11 (C) is a first structural element 1016f used in the first expansion process and the first contraction process of the headlight image IMGb1. Fig.11 (E) is a first structural element 1016g used in the first expansion process and the first contraction process of the headlight image IMGb5.

[0120] The first structural element 1016e used in the headlight image IMGb0 as the first image is shorter in the left-right direction than the first structural element 1016f used in the headlight image IMGb1 as the second image. Therefore, when the first dilation process is performed on the headlight image IMGb0 using the first structural element 1016e, Fig.11 As shown in (B), the pair HL2 of the headlight HL is connected to form the lateral expansion region 1018, but the pairs HL1 and HL3 are not connected and remain separated.

[0121] On the other hand, the first structural element 1016f used in the headlight image IMGb1 as the second image is longer in the left-right direction than the first structural element 1016e. Therefore, when the first dilation process is performed on the headlight image IMGb1 using the first structural element 1016f, Fig.11 As shown in (D), the pair of headlights HL1 closer to the host vehicle is connected to form the lateral expansion region 1018. In addition, the pair of headlights HL3 closer to the host vehicle than the pair of headlights HL1 is maintained in a separated state.

[0122] In addition, the first structural element 1016f used in the headlight image IMGb1 as the first image is shorter in the left-right direction than the first structural element 1016g used in the headlight image IMGb5 as the second image. Therefore, when the first dilation process is performed on the headlight image IMGb1 using the first structural element 1016f, Fig.11 As shown in (D), the pair HL1 of the headlight HL is connected to form the lateral expansion region 1018, but the pair HL3 is not connected and remains separated.

[0123] On the other hand, the first structural element 1016g used in the headlight image IMGb5 as the second image is longer in the left-right direction than the first structural element 1016f. Therefore, when the first dilation process is performed on the headlight image IMGb5 using the first structural element 1016g, Fig.11 As shown in (F), the pair of headlights HL3 closer to the vehicle is connected to form a lateral expansion region 1018 .

[0124] The headlight image IMGb1 does not include the pair HL2 of the headlight HL due to the second contraction process. Therefore, by using the first expansion process of the first structural element 1016f, it is possible to avoid generating the lateral expansion region 1018 originating from the pair HL2 of the headlight HL. Similarly, the headlight image IMGa5 does not include the pair HL1 and HL2 of the headlight HL due to the second contraction process. Therefore, by using the first expansion process of the first structural element 1016g, it is possible to avoid generating the lateral expansion region 1018 originating from the pair HL1 and HL2 of the headlight HL. Based on this, the position of the vehicle can be detected more accurately.

[0125] Furthermore, the first expansion process and the first contraction process (step S1106 ) on the rear light image IMGa and the first expansion process and the first contraction process (step S1107 ) on the headlight image IMGb may be performed in different orders and may be processed in parallel.

[0126] The vehicle position detection device 1010 synthesizes the rear light images IMGa0 to IMGa3, that is, performs an OR operation on each rear light image IMGa, and generates a rear light transverse expansion image including a transverse expansion region 1018 based on a pair of rear lights RL (S1108). In addition, the vehicle position detection device 1010 synthesizes the headlight images IMGb0 to IMGb5, that is, performs an OR operation on each headlight image IMGb, and generates a headlight transverse expansion image including a transverse expansion region 1018 based on a pair of headlights HL (S1109). In addition, the synthesis process of the rear light image IMGa (step S1108) and the synthesis process of the headlight image IMGb (step S1109) are different in order, and can also be processed in parallel.

[0127] Then, the vehicle position detection device 1010 synthesizes the rear light transverse expansion image and the headlight transverse expansion image, that is, performs an OR operation on the two images to generate Fig.12 The vehicle position detection device 1010 can detect the position of the front vehicle based on each lateral expansion area 1018 included in the collective lateral expansion image IMGc (S1110). For example, the vehicle position detection device 1010 detects the position of each lateral expansion area 1018 in the collective lateral expansion image IMGc as the position of the front vehicle in the area in front of the vehicle.

[0128] The vehicle position detection device 1010 transmits the collective transverse expansion image IMGc as detection result information to the pattern determination unit 1012. In addition, the vehicle position detection device 1010 may extract angle information indicating the position of the front vehicle from the collective transverse expansion image IMGc, and transmit the extraction result to the pattern determination unit 1012. In addition, the vehicle position detection device 1010 may transmit the detection result information to, for example, an ECU that manages automatic driving.

[0129] The pattern determination unit 1012 determines the light distribution pattern PTN in which the portion corresponding to the vehicle ahead is shielded based on the detection result of the vehicle position detection device 1010. "Shielding a certain portion" means not only making the brightness (illuminance) of the portion completely zero, but also including reducing the brightness (illuminance) of the portion.

[0130] The pattern determination unit 1012 of this embodiment generates an image of the transverse expansion region 1018, that is, the pixel value of each pixel of the collective transverse expansion image IMGc, and generates Fig.12 In the reverse image IMGd shown in (B) ( S1111 ), in the reverse image IMGd, the lateral expansion region 1018 has a low pixel value, and the region other than the lateral expansion region 1018 has a high pixel value.

[0131] Next, the pattern determination unit 1012 uses Fig.12 The third structural element 1020 of the prescribed shape shown in (C) is subjected to the third contraction process on the reverse image IMGd (S1112). The third structural element 1020 of this embodiment has a shape that is long in the vertical direction. The vehicle position detection device 1010 associates the upper end pixel of the third structural element 1020 with the focus pixel 1020a and performs the third contraction process on the reverse image IMGd. In addition, Fig.12 (C) schematically shows the third structural element 1020 , and the number of pixels constituting the portion extending downward from the upper end pixel of the third structural element 1020 is not limited to the number shown in the figure.

[0132] In the third shrinking process, when any pixel overlapped by the third structural element 1020 includes a pixel with a low pixel value, the pixel value of the focus pixel 1020a is converted to the low pixel value. As a result, the high pixel value area located above the inverted lateral expansion area 1018 shrinks upward, generating Fig.12 The light distribution pattern image IMGe shown in (D) of FIG. 1 includes an upper contraction region 1022 extending upward from the lateral expansion region 1018 .

[0133] The pattern determination unit 1012 determines the light distribution pattern PTN including the light shielding portion based on the light distribution pattern image IMGe (S1113). When determining the light distribution pattern PTN, the pattern determination unit 1012 includes the upper contraction area 1022 in the light shielding portion. For example, the pattern determination unit 1012 determines the shape of the pixel group with low pixel values ​​in the light distribution pattern image IMGe as the light shielding portion, and determines the shape of the pixel group with high pixel values ​​as the shape of the light distribution pattern PTN. In addition, the pattern determination unit 1012 determines a predetermined first illuminance for the area excluding the light shielding portion, and determines a second illuminance lower than the first illuminance for the light shielding portion. As described above, the second illuminance may be zero, or an illuminance higher than zero and lower than the first illuminance.

[0134] The pattern determination unit 1012 sends data indicating the light distribution pattern PTN to the lamp control device 1008. The lamp control device 1008 controls the variable light distribution lamp 1002 in such a manner that a visible light beam L1 having an intensity distribution corresponding to the light distribution pattern PTN determined by the pattern determination unit 1012 is emitted. For example, when the variable light distribution lamp 1002 includes a DMD, the lamp control device 1008 controls the lighting and extinguishing of the light source and the on / off switching of each reflector element constituting the DMD. Thus, a light distribution pattern PTN having a light shielding portion overlapping with the vehicle ahead is formed, which can improve the visual recognition of the driver of the vehicle ahead without causing glare to the driver of the vehicle ahead.

[0135] Furthermore, the pattern decision unit 1012 may perform the generation of the reversed image IMGd and the image processing using the third structuring element 1020 in the reverse order. That is, the pattern decision unit 1012 performs the third dilation processing on the collective transverse dilation image IMGc using the third structuring element 1020 to generate an upper dilation region extending upward from the transverse dilation region 1018. Thereafter, the pattern decision unit 1012 inverts the pixel value of each pixel of the image generating the upper dilation region to generate the reversed image IMGd, so that the upper dilation region in the reversed image IMGd is included in the light shielding portion.

[0136] As described above, the vehicle position detection device 1010 of the present embodiment performs the first expansion process and the first contraction process on the image based on the camera 1004 that captures the area in front of the vehicle using the first structural element 1016 of a predetermined shape that is long in the vehicle width direction, generates a lateral expansion region 1018 in which pairs of light spots arranged in the vehicle width direction included in the image are connected to each other, and detects the position of the vehicle in front based on the lateral expansion region 1018. Thus, for example, a new detection method can be provided that can more easily detect the position of the vehicle in front than in the case where the image IMG of the camera 1004 is detected by performing image analysis including algorithm recognition, deep learning, etc.

[0137] In addition, the vehicle position detection device 1010 relatively reduces the number of times the second contraction process using the second structural element 1014 of a predetermined shape is performed on the image based on the camera 1004, and generates a first image including a pair of light points located at a predetermined first distance and a pair of light points located at a second distance longer than the first distance, and relatively increases the number of times the second contraction process is performed on the image based on the camera 1004, and generates a second image including a pair of light points located at the first distance and deleting the pair of light points located at the second distance. Then, the first structural element 1016 having a relatively short length in the vehicle width direction is used in the first image, and the first structural element 1016 having a relatively long length in the vehicle width direction is used in the second image, and the first expansion process and the first contraction process are performed on each image. In other words, the vehicle position detection device 1010 performs the second contraction process on the image a predetermined number of times to generate the first image, and performs the second contraction process on the image a number of times greater than the number of times when the first image is generated to generate the second image. Then, the first expansion process and the first contraction process are performed in the first image using the first structural element 1016 having a predetermined length in the vehicle width direction, and the first expansion process and the first contraction process are performed in the second image using the first structural element 1016 which is longer in the vehicle width direction than the first structural element 1016 used in the first image.

[0138] That is, the vehicle position detection device 1010 distinguishes the light spot of the vehicle ahead based on the distance by making the second contraction process different in number, and performs the first expansion process and the first contraction process on each distinguished light spot using the first structural element 1016 of different sizes. Thus, the position of the vehicle ahead can be detected with higher accuracy.

[0139] In addition, the vehicle lighting system 1001 of the present embodiment includes: a camera 1004 for photographing the area in front of the vehicle; a variable light distribution lamp 1002 capable of irradiating the front area with a visible light beam L1 having a variable intensity distribution; a light distribution control device 1006 having a vehicle position detection device 1010 and a pattern determination unit 1012 for determining a light distribution pattern PTN including a light shielding portion based on the detection result of the vehicle position detection device 1010; and a light control device 1008 for controlling the variable light distribution lamp 1002 in such a manner as to form the light distribution pattern PTN. Thus, glare to the driver of the vehicle in front can be avoided and the visual recognition of the driver of the vehicle can be improved.

[0140] In addition, the pattern determination unit 1012 generates an inverted image IMGd by inverting the pixel value of each pixel of the image generating the lateral expansion region 1018, and performs a third contraction process on the inverted image IMGd using a third structural element 1020 of a predetermined shape that is long in the vertical direction, thereby generating an upper contraction region 1022 extending upward from the lateral expansion region 1018, and including the upper contraction region 1022 in the light shielding portion. Thus, glare to the driver of the vehicle ahead can be further avoided.

[0141] The above describes the second embodiment of the present invention in detail. The above-mentioned second embodiment only represents a specific example when the present invention is implemented. The content of the second embodiment does not limit the technical scope of the present invention. Various design changes such as changes, additions, and deletions of constituent elements can be made without departing from the scope of the idea of ​​the invention specified in the claims. The new embodiment with the design changes has the respective effects of the combined embodiment and the deformation. In the above-mentioned second embodiment, the contents that can be subjected to such design changes are marked with expressions such as "this embodiment" and "in this embodiment" to emphasize, but the contents without such expressions also allow design changes. Any combination of the above constituent elements is also valid as a scheme of the present invention. The hatching marked on the cross section of the accompanying drawings does not limit the material of the object marked with the hatching.

[0142] The invention of the above-mentioned second embodiment can also be specified by the items described below.

[0143] (Item 4) A vehicle position detection method, which performs a first expansion process and a first contraction process on an image based on an image capturing device 1004 capturing a front area of ​​a vehicle using a first structural element 1016 of a predetermined shape that is long in a vehicle width direction, thereby generating a lateral expansion area 1018 in which pairs of light spots arranged in the vehicle width direction included in the image are connected to each other, The position of the preceding vehicle is detected based on the lateral expansion area 1018 .

[0144] [Industrial Applicability] The present invention can be used in a light distribution control device, a vehicle position detection device, a vehicle lighting system, a light distribution control method, and a vehicle position detection method.

[0145] [Explanation of Reference Numerals] 1 vehicle lighting system, 2 variable light distribution lamp, 4 camera, 6 light distribution control device, 8 lamp control device, 10 high-speed camera, 12 low-speed camera, 14 vehicle detection unit, 16 area determination unit, 18 pattern determination unit, 20 existence area, 22 processing target area, 24 shading unit, 1001 vehicle lighting system, 1002 variable light distribution lamp, 1004 camera, 1006 light distribution control device, 1008 lamp control device, 1010 vehicle position detection device, 1012 pattern determination unit, 1014 second structural element, 1016 first structural element, 1018 lateral expansion area, 1020 third structural element, 1022 upper contraction area.

Claims

1. A vehicle position detection device, wherein: A first expansion process and a first contraction process are performed on an image based on an image capturing device capturing an area in front of the vehicle using a first structural element of a predetermined shape that is long in a vehicle width direction, thereby generating a lateral expansion area in which pairs of light spots arranged in the vehicle width direction included in the image are connected to each other; The position of the preceding vehicle is detected based on the lateral expansion area.

2. The vehicle position detection device according to claim 1, wherein: Relatively reducing the number of times of second contraction processing using a second structural element of a predetermined shape for the image based on the imaging device, and generating a first image including the pair of light spots located at a predetermined first distance and the pair of light spots located at a second distance greater than the first distance; For the image based on the imaging device, the number of times of the second contraction process is relatively increased to generate a second image including the pair of light spots located at the first distance and deleting the pair of light spots located at the second distance; The first structural element having a relatively short length in the vehicle width direction is used in the first image, and the first structural element having a relatively long length in the vehicle width direction is used in the second image, and the first expansion processing and the first contraction processing are performed on each image.

3. A vehicle lighting system, comprising: a camera that captures the area in front of the vehicle, a variable light distribution lamp capable of irradiating a visible light beam with a variable intensity distribution to the front area, A light distribution control device comprising the vehicle position detection device according to claim 10 or 11 and a pattern determination unit for determining a light distribution pattern including a light shielding unit based on a detection result of the vehicle position detection device, and A lamp control device controls the variable light distribution lamp so as to form the light distribution pattern.

4. The vehicle lighting system according to claim 3, wherein: The pattern determination unit generates an inverted image by inverting the pixel value of each pixel of the image in which the lateral expansion region is generated; A third shrinking process is performed on the reverse image using a third structural element of a predetermined shape that is long in the vertical direction, thereby generating an upper shrinking region that extends upward from the lateral expansion region and including the upper shrinking region in the light shielding portion.

5. A vehicle position detection method, comprising: Performing a first expansion process and a first contraction process on an image based on an imaging device that images a front area of ​​the vehicle using a first structural element of a predetermined shape that is long in a vehicle width direction, thereby generating a lateral expansion area in which pairs of light spots arranged in the vehicle width direction included in the image are connected to each other; and The position of the preceding vehicle is detected based on the lateral expansion area.

Citation Information

Patent Citations

  • Vehicular headlamp

    JP2015064964A