Light distribution control device, vehicle lamp system, and light distribution control method

Abstract

A vehicle lighting system (1) includes: an arithmetic unit (10) that extracts an index value from a specified processing target area in an image (IMG); a pattern determination unit (12) that determines a light distribution pattern (PTN) so that the index value approaches a maximum value; and a lamp control unit (14) that controls a light distribution variable lamp (2) to form the light distribution pattern (PTN). The index value is at least one of the average intensity of the HOG feature amount in a plurality of pixels in the processing target area and the edge gray scale ratio, which is the ratio of the gray scale pixels belonging to the edge portion in the LBP (Local Binary Pattern) histogram of the plurality of pixels to the whole of the plurality of pixels.

Description

Technical Field

[0001] The present invention relates to a light distribution control device, a vehicle lamp system, and a light distribution control method. Background Art

[0002] In recent years, as another technology for assisting a driver's driving operation, research and development of advanced driver-assistance systems (ADAS) and autonomous driving technologies have been promoted (for example, refer to Patent Document 1). In ADAS and autonomous driving technologies, the situation around the own vehicle is grasped by an imaging device such as a camera which is the mechanical eye of the vehicle, and vehicle control corresponding to the situation is executed.

[0003] [Prior Art Documents]

[0004] [Patent Documents]

[0005] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2016-95831 Summary of the Invention

[0006] [Problems to be Solved by the Invention]

[0007] In order to accurately achieve the above-described ADAS and autonomous driving, it is required to accurately recognize a target existing in front of the own vehicle from an image generated by an imaging device.

[0008] The present invention has been completed in view of the above circumstances, and one of its objects is to provide a technology for improving the recognition accuracy of a target existing in front of the own vehicle.

[0009] [Technical Means for Solving the Technical Problem]

[0010] One aspect of the present invention is a light distribution control device that controls a light distribution variable lamp based on an image obtained from an imaging device that captures a front area of a vehicle, the light distribution variable lamp being capable of irradiating a visible light beam with a variable intensity distribution to the front area. The device includes: an arithmetic unit that extracts an index value from a specified processing target area in the image; a pattern determination unit that determines a light distribution pattern so that the index value approaches a maximum value; and a lamp control unit that controls the light distribution variable lamp to form the light distribution pattern. The index value is at least one of an average intensity of HOG (Histograms of Oriented Gradients) feature amounts of a plurality of pixels in the processing target area and a ratio of the number of pixels of gray levels belonging to an edge portion in a histogram of LBP (Local Binary Pattern) of the plurality of pixels to the whole of the plurality of pixels, that is, an edge gray level ratio.

[0011] Another aspect of the present invention is a vehicle lighting system. The system includes: a light distribution variable lamp capable of irradiating a visible light beam with a variable intensity distribution to the front area of the vehicle; an imaging device for photographing the front area; and a light distribution control device of the above aspect.

[0012] Another aspect of the present invention is a light distribution control method for controlling a light distribution variable lamp based on images repeatedly obtained from an imaging device that photographs the front area of the vehicle. The light distribution variable lamp is capable of irradiating a visible light beam with a variable intensity distribution to the front area. The control method includes the following steps: extracting an index value from a specified processing target area in the image; determining a light distribution pattern so that the index value approaches the maximum value; and controlling the light distribution variable lamp to form the light distribution pattern. The index value is at least one of the average intensity of the HOG (Histograms of Oriented Gradients) feature amounts of a plurality of pixels in the processing target area and the ratio of the gray-scale pixels belonging to the edge part in the LBP (Local Binary Pattern) histogram of the plurality of pixels to the whole of the plurality of pixels, that is, the edge gray-scale ratio.

[0013] Another aspect of the present invention is a light distribution control device for controlling a light distribution variable lamp based on an image obtained from an imaging device that photographs the front area of the vehicle. The light distribution variable lamp is capable of irradiating a visible light beam with a variable intensity distribution to the front area. The device includes: an operation unit that extracts an LBP (Local Binary Pattern) histogram from a specified processing target area in the image and calculates the similarity between the extracted LBP histogram and a pre-prepared template LBP histogram; a pattern determination unit that determines a light distribution pattern so that the similarity approaches the maximum value; and a lamp control unit that controls the light distribution variable lamp to form the light distribution pattern.

[0014] Another aspect of the present invention is a vehicle lighting system. The system includes: a light distribution variable lamp capable of irradiating a visible light beam with a variable intensity distribution to the front area of the vehicle; an imaging device for photographing the front area; and a light distribution control device of the above aspect.

[0015] Another aspect of the present invention is a light distribution control method for controlling a light distribution variable lamp based on images repeatedly obtained from an imaging device that photographs the front area of the vehicle. The light distribution variable lamp is capable of irradiating a visible light beam with a variable intensity distribution to the front area. The control method includes the following steps: extracting an LBP (Local Binary Pattern) histogram from a specified processing target area in the image and calculating the similarity between the extracted LBP histogram and a pre-prepared template LBP histogram; determining a light distribution pattern so that the similarity approaches the maximum value; and controlling the light distribution variable lamp to form the light distribution pattern.

[0016] In addition, any combination of the above components and a solution obtained by converting the expression of the present invention between methods, apparatuses, systems, etc. are also effective as the solution of the present invention.

[0017] [Advantages of the Invention]

[0018] According to the present invention, it is possible to improve the recognition accuracy of a target present in front of the host vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a block diagram of a vehicle lamp system according to Embodiment 1.

[0020] Figure 2 is a schematic diagram of an image generated when forming a reference light distribution pattern.

[0021] Figure 3 (A) of is a graph showing the relationship between the illuminance of the light distribution pattern and the average intensity of the HOG feature amount. Figure 3 (B) of is a graph showing the relationship between the illuminance of the light distribution pattern and the edge gray scale ratio in the LBP histogram. Figure 3 (C) of is a graph showing the relationship between the illuminance of the light distribution pattern and the recognition score of the target.

[0022] Figure 4 is a flowchart showing an example of the light distribution control executed by the light distribution control device.

[0023] Figure 5 is a block diagram of a vehicle lamp system according to Embodiment 2.

[0024] Figure 6 is a schematic diagram of an image generated when forming a reference light distribution pattern.

[0025] Figure 7 (A) of is a graph schematically showing an example of the LBP histogram extracted from the processing target area. Figure 7 (B) of is a graph schematically showing an example of the template LBP histogram.

[0026] Figure 8 is a flowchart showing an example of the light distribution control. DETAILED DESCRIPTION OF THE INVENTION

[0027] Hereinafter, the present invention will be described based on preferred embodiments with reference to the accompanying drawings. The embodiments are merely illustrative and do not limit the invention. All features and combinations thereof described in the embodiments are not necessarily essential to the invention. The same or equivalent components, members, and processes shown in the respective drawings are denoted by the same reference numerals, and repeated explanations are appropriately omitted. In addition, for ease of explanation, the scales and shapes of the respective parts shown in the drawings are conveniently set and are not to be construed restrictively unless otherwise specified. In addition, when terms such as "first" and "second" are used in this specification or claims, unless otherwise specified, these terms do not indicate any order or importance, but are used to distinguish one component from another. In addition, in the respective drawings, a part of the members that are not important in explaining the embodiments is omitted and shown.

[0028] (Embodiment 1)

[0029] Figure 1 is a block diagram of a vehicle lighting system according to Embodiment 1. In Figure 1 part of the components of the vehicle lighting system 1 are depicted as functional blocks. These functional blocks are implemented as hardware components by elements or circuits represented by a computer's CPU and memory, and as software components by computer programs and the like. Those skilled in the art should understand that these functional blocks can be implemented in various forms by a combination of hardware and software.

[0030] The vehicle lighting system 1 includes a light distribution variable lamp 2, a camera device 4, and a light distribution control device 6. These can all be built into the same housing, and several components can also be provided outside the housing. For example, the camera device 4 can be an in-vehicle camera mounted inside the vehicle compartment, or a lamp-integrated camera housed in the lamp chamber together with the light distribution variable lamp 2. The light distribution control device 6 can be assembled into the vehicle ECU or the lamp ECU. In addition, as will be described later, the light distribution control device 6 includes a region setting unit 8, an arithmetic unit 10, a pattern determination unit 12, and a lamp control unit 14, but a part of these can be assembled into the vehicle ECU and another part can be assembled into the lamp ECU.

[0031] The light distribution variable lamp 2 is a lamp that can irradiate a visible light beam L1 with a variable intensity distribution to the front area of the vehicle. The light distribution variable lamp 2 can change the illuminance of the light irradiated to a plurality of individual regions R arranged in the front area respectively. The plurality of individual regions R are arranged in a matrix shape, for example. The light distribution variable lamp 2 receives information indicating the light distribution pattern PTN from the light distribution control device 6 and emits a visible light beam L1 having an intensity distribution corresponding to the light distribution pattern PTN. Thereby, a light distribution pattern PTN is formed in front of the own vehicle. The light distribution pattern PTN is understood as the two-dimensional illuminance distribution of the irradiation pattern 902 formed by the light distribution variable lamp 2 on the virtual vertical screen 900 in front of the own vehicle.

[0032] The configuration of the light distribution variable lamp 2 is not particularly limited. For example, it includes a plurality of light sources arranged in a matrix and a lighting circuit that independently drives each light source to turn it on. Preferred examples of the light source include semiconductor light sources such as LEDs (light-emitting diodes), LDs (laser diodes), and organic or inorganic ELs (electroluminescence). Each individual region R corresponds to each light source, and light is irradiated from each light source to each individual region R respectively. In addition, in order to form an illuminance distribution corresponding to the light distribution pattern PTN, the light distribution variable lamp 2 may also include a matrix-type pattern forming device such as a DMD (Digital Mirror Device) or a liquid crystal device, a scanning optical-type pattern forming device that scans the front of the vehicle with the light from the light source, and the like.

[0033] The time required for the light distribution variable lamp 2 to form one light distribution pattern PTN is, for example, 0.1 to 5 ms. In addition, the resolution of the light distribution variable lamp 2, in other words, the light distribution resolution, is, for example, 1000 to 2 million pixels. The resolution (light distribution resolution) of the light distribution variable lamp 2 means the number of unit regions in the light distribution pattern PTN where the illuminance can be independently changed. As an example, each unit region corresponds to each individual region R.

[0034] The imaging device 4 has sensitivity in the visible light region and repeatedly captures the front region of the vehicle. The imaging device 4 captures the reflected light L2 of the visible light beam L1 by an object in front of the vehicle. The imaging device 4 only needs to have sensitivity at least in the wavelength range of the visible light beam L1. The image IMG generated by the imaging device 4 is sent to the light distribution control device 6. In addition, the image IMG is also sent to the vehicle ECU. The vehicle ECU can use the acquired image IMG for object recognition in ADAS or autonomous driving. The frame rate of the imaging device 4 is, for example, 200 fps to 10,000 fps (each frame is 0.1 to 5 ms). In addition, the resolution of the imaging device 4 is, for example, 300,000 pixels to 5 million pixels.

[0035] Based on the image IMG obtained from the imaging device 4, the light distribution control device 6 controls the light irradiation from the light distribution variable lamp 2 and dynamically and adaptively controls the light distribution pattern PTN. The light distribution control device 6 may be composed of a digital processor. For example, it may be composed of 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), and the like.

[0036] The light distribution control device 6 includes a region setting unit 8, an arithmetic unit 10, a pattern determination unit 12, and a lamp control unit 14. Each unit operates by executing a program held in a memory through an integrated circuit that constitutes itself. Hereinafter, the operations of each unit will be described in detail.

[0037] Figure 2 It is a schematic diagram of an image IMG generated under the formation of a reference light distribution pattern PTNa. The region setting unit 8 determines a prescribed processing target region 18 based on a pixel region 16 having a pixel value equal to or greater than a prescribed value in the image IMG generated in a state where a prescribed reference light distribution pattern PTNa is formed in the front region. The pixel value is, for example, a luminance value.

[0038] The reference light distribution pattern PTNa is composed of light having an illuminance of 50% or more of the maximum illuminance of the light distribution variable lamp 2. That is, the reference light distribution pattern PTNa is a light distribution pattern PTN having an illuminance of 50% or more as a whole. Preferably, the reference light distribution pattern PTNa is a light distribution pattern PTN having the maximum illuminance as a whole. The maximum illuminance of the light distribution variable lamp 2 is determined by regulations, for example. Information on the reference light distribution pattern PTNa is held in advance by the pattern determination unit 12. For example, the region setting unit 8 can cause the light distribution variable lamp 2 to form the reference light distribution pattern PTNa via the pattern determination unit 12 and the lamp control unit 14.

[0039] The region setting unit 8 can determine the processing target region 18 within an irradiable range 20 in the image IMG where a visible light beam L1 can be irradiated from the light distribution variable lamp 2. As an example, in Figure 2 it is illustrated that substantially the whole of the image IMG is the irradiable range 20.

[0040] For example, the region setting unit 8 holds a threshold value related to the luminance value in advance. And, in the image IMG generated in a state where the reference light distribution pattern PTNa is formed, a set of pixels having a luminance value equal to or greater than the threshold value is determined as the pixel region. The “prescribed value” and “threshold value” related to the pixel value (luminance value) can be appropriately set based on the experiments and simulations of the designer. In addition, the region setting unit 8 can also use a set of pixels having the maximum luminance value as the pixel region 16. In this case, setting of the threshold value can be omitted.

[0041] The high-brightness pixel region 16 is generated by a light reflector 22 or a pedestrian 24 in front of the vehicle. The light reflector 22 and the pedestrian 24 are targets whose presence should be recognized in ADAS or autonomous driving. For example, the light reflector 22 is at least one selected from the group consisting of road signs, delineators, and billboards. Alternatively, the light reflector 22 is an object having a retroreflective surface in a portion irradiated with the light of the variable-beam lamp 2. In addition, the high-brightness pixel region 16 is also generated by the lamps 26 of the vehicle ahead, such as headlights or taillights. Therefore, the processing target region 18 overlaps with the light reflector 22, the pedestrian 24, and the lamps 26 of the vehicle ahead.

[0042] Then, the region setting unit 8 determines the processing target region 18 based on the specific high-brightness pixel region 16. The minimum unit of the size of the processing target region 18 corresponds to the light distribution resolution of the variable-beam lamp 2. For example, the processing target region 18 corresponds one-to-one with the unit region of the light distribution pattern PTN. That is, the minimum unit of the processing target region 18 corresponds to one individual region R.

[0043] When each pixel of the image IMG corresponds one-to-one with each unit region of the light distribution pattern PTN, the region setting unit 8 can determine the set of a plurality of unit regions overlapping with one pixel region 16 as the processing target region 18. In addition, when a plurality of unit regions correspond to one pixel, the set of a plurality of unit regions overlapping with one pixel region 16 can also be determined as the processing target region 18. Therefore, in these cases, the processing target region 18 having substantially the same shape as the light reflector, the pedestrian, and the lamps in the image IMG can be set. In addition, a predetermined margin may be provided around the pixel region 16 so that the processing target region 18 is larger than the pixel region 16.

[0044] On the other hand, when one unit region corresponds to a plurality of pixels, the region setting unit 8 can determine the processing target region 18 as follows. That is, when one pixel region 16 straddles a plurality of unit regions, all the unit regions overlapping with the pixel region 16 can be determined as the processing target region 18. Alternatively, the unit region having the largest overlapping partial area with the pixel region 16 may be set as the processing target region 18 for the pixel region 16.

[0045] Regardless of which case the correspondence between pixels and the unit area is, when the image IMG includes a plurality of pixel regions 16, it is preferable to determine the processing target region 18 for each pixel region 16. That is, it is preferable that there is no upper limit to the number of processing target regions 18 that can be set in the image IMG. However, an upper limit may also be set for the number of processing target regions 18. When an upper limit is set for the number of processing target regions 18, it is preferable to sequentially allocate the processing target regions 18 starting from the pixel region 16 with a larger area. The region setting unit 8 sends the information indicating the processing target region 18 to the arithmetic unit 10.

[0046] The arithmetic unit 10 extracts a prescribed index value from the processing target region 18 by applying known image processing to the processing target region 18 in the image IMG. The index value is at least one of the average intensity of the HOG (Histogram of Oriented Gradients) feature amounts among the plurality of pixels in the processing target region 18 and the edge gray scale ratio in the LBP (Local Binary Pattern) histogram of the plurality of pixels.

[0047] The average intensity of the HOG feature amount is a value obtained by averaging the magnitudes of the vectors of the HOG feature amounts in each pixel of the processing target region 18, and one average intensity is determined for one processing target region 18. The edge gray scale ratio is the ratio of the pixels of the gray scale belonging to the edge portion to the whole of the plurality of pixels in the LBP histogram. The gray scale of the edge portion can be appropriately set based on the experiments and simulations of the designer and is stored in the memory in advance.

[0048] When the processing target region 18 is included in the first region 28 in the image IMG, the arithmetic unit 10 of the present embodiment extracts the edge gray scale ratio from the processing target region 18. In addition, when the processing target region 18 is included in the second region 30 in the image IMG, the arithmetic unit 10 extracts the average intensity of the HHOG feature amount from the processing target region 18.

[0049] The arithmetic unit 10 stores the information related to the first region 28 and the second region 30 in the memory in advance. The first region is a region where a prescribed light reflector 22 is predicted to exist in the image IMG. The second region 30 is a region where a pedestrian 24 is predicted to exist in the image IMG. For example, the first region 28 extends in the vehicle width direction above in the image IMG. In addition, the second region 30 is located at both ends in the vehicle width direction below the first region 28. The positions and shapes of the first region 28 and the second region 30 can be appropriately set based on the experiments and simulations of the designer.

[0050] In addition, the positions and shapes of the first region 28 and the second region 30 can be fixed regardless of the posture change of the vehicle or the change in the driving environment, or can change according to the posture change of the vehicle or the change in the driving environment. As the posture change of the vehicle, posture changes in the pitch direction and yaw direction can be cited. As the change in the driving environment of the vehicle, cases such as the driving road of the vehicle changing from a straight road to a curved road, or changing from a horizontal road to an inclined road can be cited. The displacement and deformation of the first region 28 and the second region 30 are, for example, executed by the arithmetic unit 10.

[0051] By receiving signals from various sensors such as a vehicle height sensor, a yaw sensor, an acceleration sensor, and a steering sensor mounted on the vehicle, the arithmetic unit 10 can detect the posture change and the change in the form environment. In addition, for example, the arithmetic unit 10 pre-holds a conversion table that correlates the positions or shapes of the regions to be displaced or deformed with various sensor values. The arithmetic unit 10 can displace or deform the target region using this conversion table.

[0052] The LBP feature amount is more suitable for identifying the light reflector 22 than the HOG feature amount. Therefore, by extracting the edge gray scale ratio in the LBP histogram from the processing target region 18 included in the first region 28 where the light reflector 22 is presumed to exist, a lighting pattern PTN more suitable for identifying the light reflector 22 can be formed. On the other hand, the HOG feature amount is more suitable for identifying the pedestrian 24 than the LBP feature amount. Therefore, by extracting the average intensity of the HOG feature amount from the processing target region 18 included in the second region 30 where the pedestrian 24 is presumed to exist, a lighting pattern PTN more suitable for identifying the pedestrian 24 can be formed.

[0053] In addition, when the processing target region 18 is included in a region other than the first region 28 and the second region 30 in the image IMG, the average intensity of the HOG feature amount can be extracted from the processing target region 18, or the edge gray scale ratio can be extracted. For example, by extracting the average intensity of the HOG feature amount from the processing target region 18 included in this region, a lighting pattern PTN that prioritizes the safety of the pedestrian 24 can be formed.

[0054] In addition, for all the processing target regions 18 in the image IMG, regardless of whether they are included in any region of the image IMG, the arithmetic unit 10 can extract only the average intensity of the HOG feature amount, or can extract only the edge gray scale ratio in the LBP histogram. In addition, both the average intensity of the HOG feature amount and the edge gray scale ratio in the LBP histogram can be extracted from the same processing target region 18. The arithmetic unit 10 sends the information indicating the index value to the pattern determination unit 12.

[0055] The pattern determination unit 12 determines the light distribution pattern PTN so that the index value obtained from the arithmetic unit 10 approaches the maximum value. The average intensity of the HOG feature amount and the edge gray scale ratio in the LBP histogram as the index values can be adjusted by changing the illuminance (intensity) of the light irradiated to the processing target area 18. Therefore, the pattern determination unit 12 can make the index value of the processing target area 18 in the image IMG obtained under the formation of the light distribution pattern PTN approach the maximum value by changing the illuminance of the portion of the light distribution pattern PTN that overlaps with the processing target area 18 (hereinafter appropriately referred to as the overlapping portion).

[0056] In addition, when the arithmetic unit 10 extracts both the average intensity of the HOG feature amount and the edge gray scale ratio in the LBP histogram from each processing target area 18, for example, the pattern determination unit 12 uses the average value of the illuminance determined based on each index value as the illuminance of the overlapping area. In addition, the illuminance of the portion of the light distribution pattern PTN other than the overlapping portion is set based on other light distribution controls. The pattern determination unit 12 sends the information indicating the determined light distribution pattern PTN to the lamp control unit 14.

[0057] Figure 3 (A) of is a graph showing the relationship between the illuminance of the light distribution pattern PTN and the average intensity of the HOG feature amount. Figure 3 (B) of is a graph showing the relationship between the illuminance of the light distribution pattern PTN and the edge gray scale ratio in the LBP histogram. Figure 3 (C) of is a graph showing the relationship between the illuminance of the light distribution pattern PTN and the recognition score of the target. In addition, in Figure 3 (A) to Figure 3 (C) of, the illuminance (horizontal axis) of the light distribution pattern PTN is expressed as a percentage when the maximum illuminance is set to 10%.

[0058] As Figure 3 (A) of shows, regarding the average intensity of the HOG feature amount, if the illuminance is decreased from the maximum illuminance, there is almost no change until the illuminance reaches 50% of the maximum illuminance, and if it is lower than 50%, it gradually increases. And, as an example, it is maximum between about 3.5% and about 30% of the maximum illuminance, and between about 12.5% and about 25% in particular.

[0059] As Figure 3 (B) of shows, regarding the edge gray scale ratio in the LBP histogram, if the illuminance is decreased from the maximum illuminance, there is almost no change until the illuminance reaches 50% of the maximum illuminance, and if it is lower than 50%, it gradually increases. And, as an example, it is maximum between about 3.5% and about 30% of the maximum illuminance, and between about 3.5% and about 12.5% in particular.

[0060] As Figure 3As shown in (C), regarding the recognition score, if the illuminance is reduced from the maximum illuminance, there is almost no change until the illuminance reaches at least 50% of the maximum illuminance, and if it is lower than 50%, it gradually increases. Also, between approximately 3.5% and approximately 30% of the maximum illuminance, as an example, it is maximum between approximately 3.5% and approximately 12.5%. In addition, this recognition score is obtained using an AI (Artificial Intelligence) engine constructed in a manner that recognizes the target in the image IMG based on a prescribed target recognition algorithm.

[0061] Therefore, for either the average intensity of the HOG feature amount or the edge gray scale ratio in the LBP histogram, by determining the light distribution pattern PTN in a way close to the maximum value, the accuracy of recognizing the target using the imaging device 4 can be improved. In particular, by forming a reference light distribution pattern PTNa with an illuminance of 50% or more to set the processing target area 18 and gradually reducing the illuminance of the overlapping part to make each index value approach the maximum value, it is possible to suppress the omission of setting the processing target area 18 (omission of detecting the target), and to more reliably improve the recognition accuracy of the target. In addition, by adjusting the illuminance of the overlapping part in the light distribution pattern PTN to a range of 3.5% or more and 30% or less of the maximum illuminance, the average intensity of the HOG feature amount and the edge gray scale ratio can be adjusted to values that can obtain good recognition scores.

[0062] The lamp control unit 14 sends information indicating the determined light distribution pattern PTN to the light distribution variable lamp 2 and controls the light distribution variable lamp 2 to form the light distribution pattern PTN. For example, when the dimming method of the light source is analog dimming, the lamp control unit 14 adjusts the DC level of the drive current flowing through the light source. In addition, when the dimming method of the light source is PWM (Pulse Width Modulation) dimming, the lamp control unit 14 switches the current flowing through the light source and adjusts the average level of the drive current by adjusting the ratio of the conduction period. In addition, when the light distribution variable lamp 2 has a DMD, the lamp control unit 14 can also control the on / off switching of each mirror element constituting the DMD. When the light distribution variable lamp 2 has a liquid crystal device, the lamp control unit 14 can also control the light transmittance of the liquid crystal device.

[0063] Thus, a light distribution pattern PTN for improving the recognition accuracy of a target is formed in front of the host vehicle. And, with the formation of the light distribution pattern PTN, an image IMG is generated by the imaging device 4. For example, until the reset process described later is executed, the region setting unit 8 also assigns the processing target region 18 determined on the image IMG generated with the formation of the reference light distribution pattern PTNa to the newly acquired image IMG. That is, until the reset process is implemented, the processing target region 18 is fixed. After that, the extraction of the index value, the determination of the light distribution pattern PTN, and the formation of the light distribution pattern PTN are executed again.

[0064] As an example, if a series of operations from the acquisition of the image IMG to the formation of the light distribution pattern PTN are repeated a specified number of times, the light distribution control device 6 executes a reset process. In the reset process, the lamp control unit 14 controls the light distribution variable lamp 2 so as to form the reference light distribution pattern PTNa. The region setting unit 8 determines the processing target region 18 in the new image IMG generated with the formation of the reference light distribution pattern PTNa. Thus, the processing target region 18 can be made to follow the movement of the pixel region 16.

[0065] The timing for starting the reset process can be grasped, for example, by the region setting unit 8 counting the number of images IMG acquired after the formation of the reference light distribution pattern PTNa or after the setting of the processing target region 18. Or, it can also be grasped by the pattern determination unit 12 counting the number of determinations of the light distribution pattern PTN, or the lamp control unit 14 counting the number of formations of the light distribution pattern PTN. The "specified number" related to the execution of the reset process, the number of images IMG, the number of determinations of the light distribution pattern PTN, or the number of formations can be appropriately set based on the experiments and simulations of the designer. By executing the reset process, the illuminance of the light irradiated to the region deviated from the processing target region 18 preferably gradually increases. Thus, the discomfort caused to the driver can be reduced.

[0066] The pattern determination unit 12 of the present embodiment gradually changes the illuminance of the part of the light distribution pattern that overlaps with the processing target region, and decreases the amount of change in illuminance as the number of times of change in the illuminance of the light irradiated to the processing target region 18 increases. That is, the pattern determination unit 12 decreases the change in the illuminance of the light irradiated to the processing target region 18 as the number of determinations of the light distribution pattern PTN after the processing target region 18 is set increases.

[0067] For example, the illuminance of the overlapping portion of the next formed light distribution pattern PTN with respect to the reference light distribution pattern PTNa is set to a value obtained by multiplying the maximum illuminance value by the light reduction rate α. The illuminance of the overlapping portion of the light distribution pattern PTN determined thereafter is set to a value obtained by multiplying the illuminance of the overlapping portion of the light distribution pattern PTN formed at this time by a light reduction rate α smaller than the light reduction rate α used last time. Thus, the index value gradually increases and approaches the maximum value. If the illuminance of the overlapping portion is gradually decreased, the index value may turn to decrease at a certain point in time.

[0068] If the index value turns to decrease, the pattern determination unit 12 increases the illuminance of the overlapping portion of the light distribution pattern PTN to be formed thereafter. That is, the illuminance of the overlapping portion in the light distribution pattern PTN determined immediately after the index value turns to decrease is set to a value obtained by multiplying the illuminance of the overlapping portion of the light distribution pattern PTN formed at this time by the light increase rate β. Thus, the index value turns to increase again and approaches the maximum value. Preferably, the light increase rate β used immediately after the index value turns to decrease is smaller than the light reduction rate α used before the turn to decrease.

[0069] The illuminance of the overlapping portion of the light distribution pattern PTN determined thereafter is set to a value obtained by multiplying the illuminance of the overlapping portion of the light distribution pattern PTN formed at this time by a light increase rate β smaller than the light increase rate β used last time. If the index value turns to decrease again, the illuminance of the overlapping portion is decreased. Thereafter, until the reset process is executed, the decrease and increase of the illuminance of the overlapping portion are alternately repeated.

[0070] By gradually decreasing the light reduction rate α and the light increase rate β, it is possible to suppress an increase in the amount of decrease in the index value when the index value changes from increasing to decreasing. Thus, it is possible to easily stabilize the index value and the illuminance of the overlapping portion. The change in the illuminance of the overlapping portion can be linear or non-linear. That is, the light reduction rate α can change for each determination of the light distribution pattern PTN, or can change after determining the light distribution pattern PTN a specified number of times with the same light reduction rate α. The same applies to the light increase rate β.

[0071] In the above control, the illuminance adjustment adapted to the light reflector 22 or the pedestrian 24 is also performed on the lamp 26 of the preceding vehicle. However, since the lamp 26 is a self-luminous body, even if the illuminance of the light irradiated onto the lamp 26 changes, the change in the shooting state in the image IMG is small. Therefore, there is no problem in regarding the pixel region 16 from the lamp 26 included in the image IMG as the light reflector 22 or the pedestrian 24.

[0072] Figure 4This is a flowchart showing an example of the light distribution control executed by the light distribution control device 6. This process is, for example, instructed to execute the light distribution control by a lamp switch (not shown), and is repeatedly executed at a predetermined timing when the ignition is turned on. In addition, during the execution of the light distribution control, the imaging device 4 repeatedly captures the front area and sends the image IMG to the light distribution control device 6.

[0073] First, the light distribution control device 6 controls the light distribution variable lamp 2 to form the reference light distribution pattern PTNa (S101). Then, the light distribution control device 6 determines whether there is a high-brightness pixel region 16 in the image IMG generated under the formation of the reference light distribution pattern PTNa (S102). If there is a high-brightness pixel region 16 (Yes in S102), the light distribution control device 6 determines the processing target region 18 based on the pixel region 16 (S103). If there is no high-brightness pixel region 16 (No in S102), the light distribution control device 6 repeatedly determines whether there is a pixel region 16 based on the sequentially acquired images IMG (S102).

[0074] Next, the light distribution control device 6 determines the light distribution pattern PTN that reduces the part overlapping with the processing target region 18 and forms this light distribution pattern PTN (S104). After that, the light distribution control device 6 determines whether the number of images IMG acquired after the formation of the reference light distribution pattern PTNa exceeds a predetermined value (S105). If the number of acquired images IMG exceeds the predetermined value (Yes in S105), the light distribution control device 6 ends this routine. When this routine ends, the next routine starts and the reference light distribution pattern PTNa is formed (S101). From the end of this routine to the formation of the reference light distribution pattern PTNa in the next routine, it is equivalent to executing the reset process.

[0075] If the number of acquired images IMG does not exceed the predetermined value (No in S105), the light distribution control device 6 determines whether the index value has changed to decrease based on the image IMG generated by the imaging device 4 in the state where the light distribution pattern PTN is formed in step S104 (S106). When multiple processing target regions 18 are set, the light distribution control device 6 determines whether the index value has changed to decrease for each processing target region 18. If the index value has not changed to decrease (No in S106), the light distribution control device 6 forms the light distribution pattern PTN that further reduces the part overlapping with the processing target region 18 (S104).

[0076] When the index value turns to decrease (Yes in S106), the light distribution control device 6 determines a light distribution pattern PTN for increasing the light intensity of the portion overlapping with the processing target area 18, and forms the light distribution pattern PTN (S107). After that, the light distribution control device 6 determines whether the number of images IMG obtained after the formation of the reference light distribution pattern PTNa exceeds a specified value (S108). When the number of obtained images IMG exceeds the specified value (Yes in S108), the light distribution control device 6 ends this routine and executes a reset process.

[0077] When the number of obtained images IMG does not exceed the specified value (No in S108), the light distribution control device 6 determines whether the index value turns to decrease based on the image IMG generated by the imaging device 4 in the situation of forming the light distribution pattern PTN in step S107 (S109). When the index value does not turn to decrease (No in S109), the light distribution control device 6 forms a light distribution pattern PTN for further increasing the light intensity of the portion overlapping with the processing target area 18 (S107). When the index value turns to decrease (Yes in S109), the light distribution control device 6 determines a light distribution pattern PTN for decreasing the light intensity of the portion overlapping with the processing target area 18, and forms the light distribution pattern PTN (S104).

[0078] As described above, the light distribution control device 6 of the present embodiment includes: an arithmetic unit 10 that extracts an index value from a specified processing target area 18 in the image IMG generated by the imaging device 4; a pattern determination unit 12 that determines the light distribution pattern PTN so as to make the index value approach the maximum value; and a lamp control unit 14 that controls the light distribution variable lamp 2 so as to form the light distribution pattern PTN. In addition, the index value extracted by the arithmetic unit 10 is at least one of the average intensity of the HOG feature amounts in a plurality of pixels of the processing target area 18 and the ratio of the pixels of the gray levels belonging to the edge portion in the LBP histogram of the plurality of pixels to the whole of the plurality of pixels, that is, the edge gray level ratio. Further, the vehicle lamp system 1 of the present Embodiment 1 includes: a light distribution variable lamp 2 that can irradiate a visible light beam L1 with a variable intensity distribution to the front area of the vehicle; an imaging device 4 that captures the front area; and the light distribution control device 6 of the present embodiment.

[0079] In this way, by increasing at least one of the average intensity of the HOG feature amount and the edge gray-scale ratio of the LBP histogram extracted from the processing target region 18 to control the light distribution relative to the processing target region 18, the accuracy of target recognition using the image IMG can be improved in ADAS or autonomous driving. In addition, when the target is the light reflector 22, the illuminance of the light irradiated onto the light reflector 22 can be appropriately reduced, thereby suppressing lens flare or blooming that may occur around the light reflector 22. Thus, the recognition accuracy of other targets and traffic participants existing around the light reflector 22 can also be improved.

[0080] In addition, the light distribution control device 6 of the present embodiment includes a region setting unit 8. The region setting unit 8 determines the processing target region 18 based on a pixel region 16 having a pixel value equal to or greater than a specified value in the image IMG generated in a state where the reference light distribution pattern PTNa is formed in the front region. The reference light distribution pattern PTNa is composed of light having an illuminance of 50% or more of the maximum illuminance of the light distribution variable lamp 2. In this way, on the basis of forming the high-illuminance reference light distribution pattern PTNa to determine the processing target region 18, the illuminance of the portion overlapping with the processing target region 18 is gradually reduced, thereby reducing the detection omission of targets whose existence should be grasped in ADAS or autonomous driving and further reliably improving the recognition accuracy of the targets.

[0081] In addition, when the processing target region 18 is included in the first region 28 where a specified light reflector 22 is predicted to exist in the image IMG, the operation unit 10 of the present embodiment extracts the edge gray-scale ratio from the processing target region 18. When the processing target region 18 is included in the second region 30 where a pedestrian 24 is predicted to exist in the image IMG, the average intensity of the HOG feature amount is extracted from the processing target region. Thus, the recognition accuracy of the target can be further improved.

[0082] In addition, the pattern determination unit 12 of the present embodiment gradually changes the illuminance of the portion of the light distribution pattern PTN overlapping with the processing target region 18 and reduces the change amount as the number of times of the change increases. Thus, the illuminance of the overlapping portion in the light distribution pattern PTN can be easily stabilized, and the recognition accuracy of the target can be further improved.

[0083] As described above, Embodiment 1 of the present invention has been described in detail. The above-described Embodiment 1 merely represents a specific example when implementing the present invention. The content of Embodiment 1 does not limit the technical scope of the present invention. Within the scope not departing from the idea of the invention defined in the claims, various design changes such as changes, additions, and deletions of components can be made. The new embodiments with design changes have the effects of both the combined embodiments and the deformations. In the above-described Embodiment 1, for the content that can be subject to such design changes, expressions such as "in this embodiment" and "in the present embodiment" are marked for emphasis, but design changes are also allowed for the content without such expressions. Any combination of the above components is also effective as a solution of the present invention. The hatching marked on the cross-section of the drawings does not limit the material of the object marked with hatching.

[0084] The invention of the above-described Embodiment 1 can also be determined by the items described below.

[0085] (Item 1)

[0086] A light distribution control method is a light distribution control method for controlling a light distribution variable lamp (2) based on an image (IMG) repeatedly obtained from an imaging device (4) that captures the front area of a vehicle. The light distribution variable lamp (2) can irradiate a visible light beam (L1) with a variable intensity distribution to the front area, and includes the following steps:

[0087] Extract an index value from a specified processing target area (18) in the image (IMG);

[0088] Determine a light distribution pattern (PTN) in such a way that the index value approaches the maximum value; and

[0089] Control the light distribution variable lamp (2) to form the light distribution pattern (PTN),

[0090] The index value is at least one of the average intensity of the HOG (Histograms of Oriented Gradients) feature amounts of multiple pixels in the processing target area (18) and the edge gray-scale ratio, which is the ratio of the gray-scale pixels belonging to the edge part to the whole of the multiple pixels, in the LBP (Local Binary Pattern) histogram of the multiple pixels.

[0091] (Embodiment 2)

[0092] Figure 5 is a block diagram of a vehicle lighting system according to Embodiment 2. In Figure 5In [the figure], a part of the components of the vehicle lighting system 1 is depicted as functional blocks. These functional blocks are implemented as hardware components by elements or circuits represented by a computer's CPU and memory, and as software components by computer programs and the like. Those skilled in the art should understand that these functional blocks can be implemented in various forms through combinations of hardware and software.

[0093] The vehicle lighting system 1 includes a light distribution variable lamp 2, an imaging device 4, and a light distribution control device 6. These can all be built into the same housing, and several components can also be provided outside the housing. For example, the imaging device 4 can be an in-vehicle camera mounted inside the vehicle compartment, or a lamp-integrated camera housed in the lamp chamber together with the light distribution variable lamp 2. The light distribution control device 6 can be assembled into the vehicle ECU or the lamp ECU. Additionally, as will be described later, the light distribution control device 6 includes a region setting unit 8, an arithmetic unit 10, a pattern determination unit 12, and a lamp control unit 14, but a part of these can be assembled into the vehicle ECU and another part can be assembled into the lamp ECU.

[0094] The light distribution variable lamp 2 is a lamp capable of irradiating a visible light beam L1 with a variable intensity distribution to the front area of the vehicle. The light distribution variable lamp 2 can separately change the illuminance of the light irradiated to a plurality of individual regions R arranged in the front area. The plurality of individual regions R are arranged in a matrix shape, for example. The light distribution variable lamp 2 receives information indicating the light distribution pattern PTN from the light distribution control device 6 and emits a visible light beam L1 having an intensity distribution corresponding to the light distribution pattern PTN. Thereby, a light distribution pattern PTN is formed in front of the host vehicle. The light distribution pattern PTN is understood as the two-dimensional illuminance distribution of the irradiation pattern 902 formed by the light distribution variable lamp 2 on the virtual vertical screen 900 in front of the host vehicle.

[0095] The configuration of the light distribution variable lamp 2 is not particularly limited. For example, it includes a plurality of light sources arranged in a matrix and a lighting circuit that independently drives each light source to turn it on. Preferred examples of the light source include semiconductor light sources such as LEDs (light-emitting diodes), LDs (laser diodes), and organic or inorganic ELs (electroluminescence). Each individual region R corresponds to each light source, and light is irradiated from each light source to each individual region R respectively. In addition, in order to form an illuminance distribution corresponding to the light distribution pattern PTN, the light distribution variable lamp 2 may include a matrix-type pattern forming device such as a DMD (Digital Mirror Device) or a liquid crystal device, or a scanning optical-type pattern forming device that scans the front of the host vehicle with the light from the light source.

[0096] The time required for the light distribution variable lamp 2 to form one light distribution pattern PTN is, for example, 0.1 to 5 ms. Additionally, the resolution of the light distribution variable lamp 2, in other words, the light distribution resolution, is, for example, 1000 to 2 million pixels. The resolution of the light distribution variable lamp 2 (light distribution resolution) means the number of unit areas in the light distribution pattern PTN where the illuminance can be independently changed. As an example, each unit area corresponds to each individual area R.

[0097] The imaging device 4 has sensitivity in the visible light region and repeatedly captures the front area of the vehicle. The imaging device 4 captures the reflected light L2 of the visible light beam L1 by an object in front of the vehicle. The imaging device 4 only needs to have sensitivity at least in the wavelength range of the visible light beam L1. The image IMG generated by the imaging device 4 is sent to the light distribution control device 6. Additionally, the image IMG is also sent to the vehicle ECU. The vehicle ECU can use the acquired image IMG for object recognition in ADAS or autonomous driving. The frame rate of the imaging device 4 is, for example, 200 fps to 10000 fps (each frame is 0.1 to 5 ms). Additionally, the resolution of the imaging device 4 is, for example, 300,000 pixels to 5 million pixels.

[0098] Based on the image IMG obtained from the imaging device 4, the light distribution control device 6 controls the light irradiation from the light distribution variable lamp 2 and dynamically and adaptively controls the light distribution pattern PTN. The light distribution control device 6 can be composed of a digital processor. For example, it can be composed of a combination of a microcomputer including a CPU and a software program, or can be composed of an FPGA (Field Programmable Gate Array), an ASIC (Application Specified IC), etc.

[0099] The light distribution control device 6 includes a region setting unit 8, an arithmetic unit 10, a pattern determination unit 12, and a lamp control unit 14. Each unit operates by executing a program held in a memory through an integrated circuit that constitutes itself. Hereinafter, the operations of each unit will be described in detail.

[0100] Figure 6 It is a schematic diagram of the image IMG generated under the formation of the reference light distribution pattern PTNa. The region setting unit 8 determines a specified processing target region 18 based on the pixel region 16 in the image IMG having a pixel value equal to or greater than a specified value. The pixel value is, for example, a brightness value. The region setting unit 8 can determine the processing target region 18 within the irradiable range 20 where the visible light beam L1 can be irradiated from the light distribution variable lamp 2 in the image IMG. As an example, in Figure 6 it shows a case where substantially the entire image IMG is the irradiable range 20.

[0101] The region setting unit 8 determines, in the image IMG generated in a state where the prescribed reference light distribution pattern PTNa is formed in front of the host vehicle, the set of pixels having the maximum luminance value as the pixel region 16. The reference light distribution pattern PTNa is, for example, a light distribution pattern PTN having the maximum illuminance as a whole. Information on the reference light distribution pattern PTNa is held in advance by the pattern determination unit 12. For example, the region setting unit 8 can cause the light distribution variable lamp 2 to form the reference light distribution pattern PTNa via the pattern determination unit 12 and the lamp control unit 14. Further, the region setting unit 8 may hold in advance a threshold value related to the pixel value, and use the set of pixels having a pixel value equal to or greater than the threshold value as the pixel region 16. The "prescribed value" and "threshold value" related to the pixel value (luminance value) can be appropriately set based on the experiments and simulations of the designer.

[0102] The high-luminance pixel region 16 is generated by a light reflector in front of the host vehicle. The light reflector is an object that should be recognized in ADAS or autonomous driving. For example, the light reflector is at least one selected from the group consisting of road signs, delineators, and billboards. Alternatively, the light reflector is an object having a retroreflective surface in a portion irradiated with the light of the light distribution variable lamp 2. Further, the high-luminance pixel region 16 is also generated by the lamps of the preceding vehicle, such as the headlamp or the tail lamp. Therefore, the processing target region 18 overlaps with the light reflector or the lamps of the preceding vehicle.

[0103] Then, the region setting unit 8 determines the processing target region 18 based on the specific high-luminance pixel region 16. The minimum unit of the size of the processing target region 18 corresponds to the light distribution resolution of the light distribution variable lamp 2. For example, the processing target region 18 corresponds one-to-one to the unit region of the light distribution pattern PTN. That is, the minimum unit of the processing target region 18 corresponds to one individual region R.

[0104] In a case where each pixel of the image IMG corresponds one-to-one to each unit region of the light distribution pattern PTN, the region setting unit 8 can determine the set of a plurality of unit regions overlapping with one pixel region 16 as the processing target region 18. Further, in a case where a plurality of unit regions correspond to one pixel, the set of a plurality of unit regions overlapping with one pixel region 16 can also be determined as the processing target region 18. Therefore, in these cases, the processing target region 18 having substantially the same shape as the light reflector and the lamps in the image IMG can be set. Further, a prescribed margin may be provided around the pixel region 16 to make the processing target region 18 larger than the pixel region 16.

[0105] On the other hand, when one unit area corresponds to multiple pixels, the area setting unit 8 can determine the processing target area 18 as follows. That is, when one pixel area 16 straddles multiple unit areas, all the unit areas overlapping with the pixel area 16 can be determined as the processing target area 18. Alternatively, the unit area with the largest overlapping area with the pixel area 16 can be set as the processing target area 18 for the pixel area 16.

[0106] Regardless of the correspondence between pixels and unit areas, when the image IMG contains multiple pixel areas 16, it is preferable to determine the processing target area 18 for each pixel area 16. That is, it is preferable that there is no upper limit to the number of processing target areas 18 that can be set in the image IMG. However, the number of processing target areas 18 can also be set with an upper limit. When the number of processing target areas 18 is set with an upper limit, it is preferable to sequentially allocate the processing target areas 18 starting from the pixel area 16 with a larger area. The area setting unit 8 sends the information indicating the processing target area 18 to the arithmetic unit 10.

[0107] Figure 7 (A) is a diagram schematically showing an example of the LBP histogram extracted from the processing target area 18. Figure 7 (B) is a diagram schematically showing an example of the template LBP histogram. As Figure 7 shown in (A), the arithmetic unit 10 extracts a (Local Binary Pattern) histogram of the processing target area 18 in the image IMG by applying known image processing to the processing target area 18.

[0108] In addition, as Figure 7 shown in (B), the arithmetic unit 10 stores the pre-prepared template LBP histogram in the memory. As an example of the template LBP histogram, it is the LBP histogram extracted from the processing target area 18 overlapping with a light reflector (such as a road sign) in the image IMG taken during the day. The template LBP histogram can also be a histogram obtained by averaging multiple LBP histograms. In addition, for example, the template LBP histogram is a histogram that has been pre-confirmed to obtain an identification score above a reference value, and more preferably the highest score, in the object recognition algorithm used in ADAS or autonomous driving.

[0109] The operation unit 10 calculates the similarity between the extracted LBP histogram and the template LBP histogram. The operation unit 10 in this embodiment calculates an index value representing the similarity through the inner product operation between the vector obtained from the extracted LBP histogram and the vector obtained from the template LBP histogram. The LBP histogram can be regarded as a vector having pixel values in the components. The index value is, for example, the cosine (cosθ) of the angle formed by two vectors obtained by the inner product operation. Alternatively, the index value is the inner product value of two vectors after being normalized respectively. Therefore, the maximum value of the index value is 1. The operation unit 10 sends the information representing the index value as the similarity to the pattern determination unit 12.

[0110] The pattern determination unit 12 determines the light distribution pattern PTN in such a way that the similarity approaches the maximum value. The pattern determination unit 12 in this embodiment determines the light distribution pattern PTN in such a way that the index value approaches the maximum value, that is, in such a way that the cosine of the angle formed by two vectors or the inner product of two vectors approaches 1. The similarity and the index value can be adjusted by changing the illuminance (intensity) of the light irradiated onto the processing target area 18. Therefore, the pattern determination unit 12 can make the LBP histogram of the processing target area 18 in the image IMG obtained under the formation of the light distribution pattern PTN approach the template LBP histogram by changing the illuminance of the part of the light distribution pattern PTN that overlaps with the processing target area 18 (hereinafter appropriately referred to as the overlapping part).

[0111] In other words, by making the similarity and the index value approach the maximum value, the shooting condition of the light reflector in the image IMG captured under the formation of the light distribution pattern PTN can be made to approach the shooting condition in the image IMG captured during the day. Thereby, the accuracy of identifying the light reflector using the imaging device 4 can be improved. In addition, the illuminance of the part of the light distribution pattern PTN other than the overlapping part is set based on other light distribution controls. The pattern determination unit 12 sends the information representing the determined light distribution pattern PTN to the lamp control unit 14.

[0112] The lamp control unit 14 sends the information indicating the determined light distribution pattern PTN to the light distribution variable lamp 2 and controls the light distribution variable lamp 2 to form the light distribution pattern PTN. For example, when the dimming method of the light source is analog dimming, the lamp control unit 14 adjusts the DC level of the drive current flowing through the light source. Additionally, when the dimming method of the light source is PWM (Pulse Width Modulation) dimming, the lamp control unit 14 switches the current flowing through the light source and adjusts the average level of the drive current by adjusting the ratio of the conduction period. Additionally, when the light distribution variable lamp 2 has a DMD, the lamp control unit 14 can also control the on / off switching of each mirror element constituting the DMD. When the light distribution variable lamp 2 has a liquid crystal device, the lamp control unit 14 can also control the light transmittance of the liquid crystal device.

[0113] Thus, a light distribution pattern PTN for improving the recognition accuracy of a target is formed in front of the host vehicle. And, with the formation of the light distribution pattern PTN, an image IMG is generated by the imaging device 4. As an example, until the reset process described later is executed, the area setting unit 8 also assigns the processing target area 18 determined on the image IMG generated with the formation of the reference light distribution pattern PTNa to the newly acquired image IMG. That is, until the reset process is implemented, the processing target area 18 is fixed. After that, the calculation of the similarity, the determination of the light distribution pattern PTN, and the formation of the light distribution pattern PTN are executed again.

[0114] As an example, if a series of operations from the acquisition of the image IMG to the formation of the light distribution pattern PTN are repeated a specified number of times, the light distribution control device 6 executes a reset process. In the reset process, the lamp control unit 14 controls the light distribution variable lamp 2 so as to form the reference light distribution pattern PTNa. The area setting unit 8 determines the processing target area 18 in the new image IMG generated with the formation of the reference light distribution pattern PTNa. Thus, it is possible to make the processing target area 18 follow the movement of the pixel area 16.

[0115] The timing for starting the reset process can be grasped, for example, by the area setting unit 8 counting the number of images IMG acquired after the formation of the reference light distribution pattern PTNa or after the setting of the processing target area 18. Or, it can also be grasped by the pattern determination unit 12 counting the number of determinations of the light distribution pattern PTN, or the lamp control unit 14 counting the number of formations of the light distribution pattern PTN. The "specified number" related to the execution of the reset process, the number of images IMG, the number of determinations of the light distribution pattern PTN, or the number of formations can be appropriately set based on the experiments and simulations of the designer. By executing the reset process, the illuminance of the light irradiated to the area deviating from the processing target area 18 preferably gradually increases. Thus, it is possible to reduce the discomfort caused to the driver.

[0116] The pattern determination unit 12 of the present embodiment decreases the amount of change in illuminance as the number of times of change in the illuminance of the light irradiated to the processing target area 18 increases. That is, the pattern determination unit 12 decreases the change in the illuminance of the light irradiated to the processing target area 18 as the number of determinations of the light distribution pattern PTN after the setting of the processing target area 18 increases.

[0117] For example, the illuminance of the overlapping portion of the next formed light distribution pattern PTN with respect to the reference light distribution pattern PTNa is set to a value obtained by multiplying the maximum illuminance value by the light reduction rate α. The illuminance of the overlapping portion of the light distribution pattern PTN determined thereafter is set to a value obtained by multiplying the illuminance of the overlapping portion of the light distribution pattern PTN formed at this time by a light reduction rate α smaller than the light reduction rate α used last time. Thus, the similarity and the index value gradually increase and approach the maximum value. If the illuminance of the overlapping portion is gradually decreased, the similarity and the index value will turn to decrease at a certain time. That is, the LBP histogram of the processing target area 18 starts to deviate from the template LBP histogram.

[0118] If the similarity and the index value turn to decrease, the pattern determination unit 12 increases the illuminance of the overlapping portion of the light distribution pattern PTN to be formed thereafter. That is, the illuminance of the overlapping portion in the light distribution pattern PTN determined immediately after the similarity and the index value turn to decrease is set to a value obtained by multiplying the illuminance of the overlapping portion of the light distribution pattern PTN formed at this time by the light increase rate β. Thus, the similarity and the index value turn to increase again and approach the maximum value. Preferably, the light increase rate β used immediately after the similarity and the index value turn to decrease is smaller than the light reduction rate α used before turning to decrease.

[0119] The illuminance of the overlapping portion of the light distribution pattern PTN determined thereafter is set to a value obtained by multiplying the illuminance of the overlapping portion of the light distribution pattern PTN formed at this time by a light increase rate β smaller than the light increase rate β used last time. If the similarity and the index value turn to decrease again, the illuminance of the overlapping portion is decreased. Thereafter, until the reset process is executed, the decrease and increase of the illuminance of the overlapping portion are repeated alternately.

[0120] By gradually decreasing the light reduction rate α and the light increase rate β, it is possible to suppress the decrease amount of the similarity and the index value when the similarity and the index value turn from increasing to decreasing from becoming large. Thus, it is possible to easily maintain the state where the LBP histogram extracted from the processing target area 18 is close to the template LBP histogram. The change in the illuminance of the overlapping portion can be linear or non-linear. That is, the light reduction rate α can change for each determination of the light distribution pattern PTN, or can change after determining the light distribution pattern PTN a specified number of times with the same light reduction rate α. The same applies to the light increase rate β.

[0121] The above-mentioned template LBP histogram corresponds to the light reflectors photographed during the day. That is, in the above control, all the high-brightness pixel regions 16 included in the image IMG are regarded as light reflectors, and the illuminance of the light irradiated onto the processing target region 18 is adjusted. In this case, the illuminance adjustment suitable for the light reflectors is also performed on the lamps of the vehicle ahead. However, since the lamps are self-luminous bodies, even if the illuminance of the light irradiated onto the lamps changes, the change in the photographed state on the image IMG is small. Therefore, it is no problem to regard all the high-brightness pixel regions 16 included in the image IMG as light reflectors.

[0122] Figure 8 It is a flowchart showing an example of the light distribution control executed by the light distribution control device 6. This process is, for example, instructed to execute the light distribution control by a lamp switch (not shown), and is repeatedly executed at a prescribed timing when the ignition is turned on. In addition, during the execution of the light distribution control, the imaging device 4 repeatedly photographs the front region and transmits the image IMG to the light distribution control device 6.

[0123] First, the light distribution control device 6 controls the light distribution variable lamp 2 so as to form the reference light distribution pattern PTNa (S201). Then, the light distribution control device 6 determines whether there is a high-brightness pixel region 16 in the image IMG generated under the formation of the reference light distribution pattern PTNa (S202). When there is a high-brightness pixel region 16 (Yes in S202), the light distribution control device 6 determines the processing target region 18 based on the pixel region 16 (S203). When there is no high-brightness pixel region 16 (No in S202), the light distribution control device 6 repeatedly determines whether there is a pixel region 16 based on the sequentially obtained images IMG (S202).

[0124] Next, the light distribution control device 6 determines the light distribution pattern PTN for reducing the light in the part overlapping with the processing target region 18, and forms this light distribution pattern PTN (S204). After that, the light distribution control device 6 determines whether the number of images IMG obtained after the formation of the reference light distribution pattern PTNa exceeds a prescribed value (S205). When the number of obtained images IMG exceeds the prescribed value (Yes in S205), the light distribution control device 6 ends this routine. When this routine ends, the next routine starts and the reference light distribution pattern PTNa is formed (S201). From the end of this routine to the formation of the reference light distribution pattern PTNa in the next routine corresponds to the execution of the reset process.

[0125] When the number of acquired images IMG does not exceed a specified value (No in S205), the light distribution control device 6 determines whether the index value has started to decrease based on the image IMG generated by the imaging device 4 in the state where the light distribution pattern PTN is formed in step S204 (S206). When a plurality of processing target regions 18 are set, the light distribution control device 6 determines for each processing target region 18 whether the index value has started to decrease. When the index value has not started to decrease (No in S206), the light distribution control device 6 forms a light distribution pattern PTN that further reduces the light intensity of the portion overlapping with the processing target region 18 (S204).

[0126] When the index value has started to decrease (Yes in S206), the light distribution control device 6 determines a light distribution pattern PTN that increases the light intensity of the portion overlapping with the processing target region 18, and forms this light distribution pattern PTN (S207). After that, the light distribution control device 6 determines whether the number of acquired images IMG after the formation of the reference light distribution pattern PTNa exceeds the specified value (S208). When the number of acquired images IMG exceeds the specified value (Yes in S208), the light distribution control device 6 ends this routine and executes a reset process.

[0127] When the number of acquired images IMG does not exceed the specified value (No in S208), the light distribution control device 6 determines whether the index value has started to decrease based on the image IMG generated by the imaging device 4 in the state where the light distribution pattern PTN is formed in step S207 (S209). When the index value has not started to decrease (No in S209), the light distribution control device 6 forms a light distribution pattern PTN that further increases the light intensity of the portion overlapping with the processing target region 18 (S207). When the index value has started to decrease (Yes in S209), the light distribution control device 6 determines a light distribution pattern PTN that reduces the light intensity of the portion overlapping with the processing target region 18, and forms this light distribution pattern PTN (S204).

[0128] In addition, in the present embodiment, a light reflector is the object of light distribution control, but even for an object other than a light reflector, by preparing a template LBP histogram corresponding to the object, it can be used as an object of light distribution control.

[0129] As described above, the light distribution control device 6 of the present embodiment includes: an arithmetic unit 10 that extracts an LBP (Local Binary Pattern) histogram from a predetermined processing target area 18 in the image IMG generated by the imaging device 4 and calculates the similarity between the extracted LBP histogram and a template LBP histogram prepared in advance; a pattern determination unit 12 that determines a light distribution pattern so that the similarity approaches the maximum value; and a lamp control unit 14 that controls the light distribution variable lamp 2 so as to form a light distribution pattern. In addition, the vehicle lamp system 1 of the present embodiment includes: a light distribution variable lamp 2 that can irradiate a visible light beam L1 with a variable intensity distribution to the front area of the vehicle; an imaging device 4 that captures the front area; and the light distribution control device 6 of the present embodiment.

[0130] Conventionally, in light distribution control for assisting a driver in recognizing an object, the illuminance of the light irradiated onto the object is changed using the brightness of the object itself as an index. For example, in conventional light distribution control, when the object is a light-reflecting object and the brightness of the light-reflecting object is a low brightness value that is assumed to be difficult for the driver to visually recognize, the illuminance of the light irradiated onto the light-reflecting object is increased. In addition, when the brightness of the light-reflecting object is a high brightness value that is assumed to cause glare to the driver, the illuminance of the light irradiated onto the light-reflecting object is decreased.

[0131] In contrast, the present inventor has found that light distribution control based on brightness itself is helpful for object recognition by a driver, but not necessarily for object recognition using the image IMG in ADAS or autonomous driving. And based on such an insight, the present inventor repeatedly conducted in-depth research and came up with using the LBP histogram as a new index.

[0132] That is, by comparing the LBP histogram extracted from the processing target area 18 with the template LBP histogram prepared in advance and performing light distribution control so that the similarity approaches the maximum value, the accuracy of object recognition using the image IMG in ADAS or autonomous driving can be improved. In addition, when the object is a light-reflecting object, the illuminance of the light irradiated onto the light-reflecting object can be appropriately reduced, thereby suppressing possible lens flare or blooming around the light-reflecting object. As a result, the recognition accuracy of objects and traffic participants existing around the light-reflecting object can also be improved.

[0133] In addition, the light distribution control device 6 of the present embodiment includes a region setting unit 8 that determines the processing target area 18 based on a pixel area 16 having a pixel value equal to or higher than a predetermined value in the image IMG. Thereby, the recognition accuracy of the light-reflecting object can be further improved.

[0134] In addition, the arithmetic unit 10 of the present embodiment calculates an index value representing similarity by performing an inner product operation between the vector obtained from the extracted LBP histogram and the vector obtained from the template LBP histogram. The pattern determination unit 12 determines the light distribution pattern PTN in such a way that the index value approaches the maximum value. In this way, by using the index value obtained by the inner product operation as the similarity, simplification of the light distribution control can be achieved.

[0135] In addition, the pattern determination unit 12 of the present embodiment gradually changes the illuminance of the portion of the light distribution pattern PTN that overlaps with the processing target area 18, and decreases the amount of change as the number of times of the change increases. Thereby, it is possible to easily maintain a state where the similarity between the LBP histogram extracted from the processing target area 18 and the template LBP histogram is close. Therefore, the recognition accuracy of the target can be further improved.

[0136] As described above, Embodiment 2 of the present invention has been described in detail. The above-described Embodiment 2 merely represents a specific example when implementing the present invention. The content of Embodiment 2 does not limit the technical scope of the present invention, and various design changes such as changes, additions, and deletions of components can be made without departing from the idea of the invention defined in the claims. The new embodiment to which a design change is applied has the effects of both the combined embodiments and the respective deformations. In the above-described Embodiment 2, for the content that allows such design changes, expressions such as "in the present embodiment" and "in this embodiment" are marked for emphasis, but content without such expressions is also allowed to be designed and changed. Any combination of the above components is also effective as a solution of the present invention. The hatching marked on the cross-section of the drawing does not limit the material of the object marked with the hatching.

[0137] The invention of the above-described Embodiment 2 can also be determined by the items described below.

[0138] (Item 2)

[0139] A light distribution control method for controlling a light distribution variable lamp (2) based on an image (IMG) repeatedly obtained from an imaging device (4) that captures the front area of a vehicle. The light distribution variable lamp (2) can irradiate a visible light beam (L1) with a variable intensity distribution to the front area, and includes the following steps:

[0140] Extract an LBP (Local Binary Pattern) histogram for a specified processing target area (18) in the image (IMG), and calculate the similarity between the extracted LBP histogram and a pre-prepared template LBP histogram;

[0141] Determine a light distribution pattern in such a way that the similarity approaches the maximum value; and

[0142] Control the light distribution variable lamp (2) to form a light distribution pattern.

[0143] [Industrial Applicability]

[0144] The present invention can be used for a light distribution control device, a vehicle lamp system, and a light distribution control method.

[0145] [Explanation of Reference Numerals]

[0146] 1 Vehicle lamp system, 2 Light distribution variable lamp, 4 Imaging device, 6 Light distribution control device, 8 Region setting unit, 10 Arithmetic unit, 12 Pattern determination unit, 14 Lamp control unit, 16 Pixel region, 18 Processing target region, 22 Light reflector, 24 Pedestrian, 28 First region, 30 Second region.

Claims

1. A light distribution control device is a light distribution control device that controls a light distribution variable lamp based on an image obtained from an imaging device that captures the front area of a vehicle. The light distribution variable lamp can irradiate a visible light beam with a variable intensity distribution onto the front area. It includes: An arithmetic unit that extracts an index value from a specified processing target area in the image, A pattern determination unit that determines a light distribution pattern in such a way that the index value approaches a maximum value, and A lamp control unit that controls the light distribution variable lamp to form the light distribution pattern; The index value is at least one of the average intensity of the histogram of oriented gradients feature amount among a plurality of pixels in the processing target area and the ratio of the number of pixels having gray levels belonging to the edge part in the local binary pattern histogram of the plurality of pixels to the whole of the plurality of pixels, that is, the edge gray level ratio. The light distribution control device includes an area setting unit that determines the processing target area based on a pixel area having a pixel value equal to or greater than a specified value in the image generated in a state where a reference light distribution pattern composed of light having an illuminance of 50% or more of the maximum illuminance of the light distribution variable lamp is formed in the front area.

2. A light distribution control device is a light distribution control device that controls a light distribution variable lamp based on an image obtained from an imaging device that captures the front area of a vehicle. The light distribution variable lamp can irradiate a visible light beam with a variable intensity distribution onto the front area. It includes: An arithmetic unit that extracts an index value from a specified processing target area in the image, A pattern determination unit that determines a light distribution pattern in such a way that the index value approaches a maximum value, and A lamp control unit that controls the light distribution variable lamp to form the light distribution pattern; The index value is at least one of the average intensity of the histogram of oriented gradients feature amount among a plurality of pixels in the processing target area and the ratio of the number of pixels having gray levels belonging to the edge part in the local binary pattern histogram of the plurality of pixels to the whole of the plurality of pixels, that is, the edge gray level ratio. When the processing target area is included in a first area where a specified light reflector is predicted to exist in the image, the arithmetic unit extracts the edge gray level ratio from the processing target area. And when the processing target area is included in a second area where a pedestrian is predicted to exist in the image, the arithmetic unit extracts the average intensity of the histogram of oriented gradients feature amount from the processing target area.

3. The light distribution control device according to claim 1, The pattern determination unit gradually changes the illuminance of a part of the light distribution pattern that overlaps with the processing target area, and decreases the amount of change as the number of times of the change increases.

4. The light distribution control device according to claim 2, The pattern determination unit gradually changes the illuminance of a part of the light distribution pattern that overlaps with the processing target area, and decreases the amount of change as the number of times of the change increases.

5. A vehicle lighting system includes: A light distribution variable lamp that can irradiate a visible light beam with a variable intensity distribution onto the front area of the vehicle, An imaging device that captures the front area, and The light distribution control device according to claim 1.

6. A light distribution control method is a light distribution control method for controlling a light distribution variable lamp based on images repeatedly obtained from an imaging device that captures the front area of a vehicle. The light distribution variable lamp can irradiate a visible light beam with a variable intensity distribution onto the front area. It includes the following steps: extracting an index value from a specified processing target area in the image, determining a light distribution pattern in such a way that the index value approaches the maximum value, and controlling the light distribution variable lamp to form the light distribution pattern; The index value is at least one of the average intensity of the histogram of oriented gradients feature amounts of multiple pixels in the processing target area and the ratio of the number of pixels with gray levels belonging to the edge part in the local binary pattern histogram of the multiple pixels to the total number of the multiple pixels, that is, the edge gray level ratio. Based on a pixel area having a pixel value equal to or greater than a specified value in the image generated in a state where a reference light distribution pattern composed of light with an illuminance of 50% or more of the maximum illuminance of the light distribution variable lamp is formed in the front area, the processing target area is determined.

7. A light distribution control device is a light distribution control device for controlling a light distribution variable lamp based on an image obtained from an imaging device that captures the front area of a vehicle. The light distribution variable lamp can irradiate a visible light beam with a variable intensity distribution onto the front area. It includes: An arithmetic unit that extracts a local binary pattern histogram from a specified processing target area in the image and calculates the similarity between the extracted local binary pattern histogram and a template local binary pattern histogram prepared in advance, a pattern determination unit that determines a light distribution pattern in such a way that the similarity approaches the maximum value, and a lamp control unit that controls the light distribution variable lamp to form the light distribution pattern, The arithmetic unit calculates an index value representing the similarity through an inner product operation between a vector obtained from the extracted local binary pattern histogram and a vector obtained from the template local binary pattern histogram; The pattern determination unit determines the light distribution pattern in such a way that the index value approaches the maximum value.

8. The light distribution control device according to claim 7, including a region setting unit that determines the processing target area based on a pixel area having a pixel value equal to or greater than a specified value in the image.

9. The light distribution control device according to claim 7 or 8, The pattern determination unit gradually changes the illuminance of a part of the light distribution pattern that overlaps with the processing target area, and reduces the change amount as the number of times of the change increases.

10. The light distribution control device according to claim 7, The pattern determination unit gradually changes the illuminance of a part of the light distribution pattern that overlaps with the processing target area, and reduces the change amount as the number of times of the change increases.

11. A vehicle lighting system includes: a light distribution variable lamp that can irradiate a visible light beam with a variable intensity distribution onto the front area of the vehicle, an imaging device that captures the front area, and the light distribution control device according to claim 7 or 8.

12. A light distribution control method is a light distribution control method for controlling a light distribution variable lamp based on images repeatedly obtained from an imaging device in the front area of a photographed vehicle. The light distribution variable lamp can irradiate a visible light beam with a variable intensity distribution to the front area. It includes the following steps: Extract a local binary pattern histogram for a specified processing target area in the image, and calculate the similarity between the extracted local binary pattern histogram and a pre-prepared template local binary pattern histogram. Determine a light distribution pattern in such a way that the similarity approaches the maximum value, and Control the light distribution variable lamp to form the light distribution pattern. Calculate an index value representing the similarity through an inner product operation of a vector obtained from the extracted local binary pattern histogram and a vector obtained from the template local binary pattern histogram. Determine the light distribution pattern in such a way that the index value approaches the maximum value.

Images