CN102963294B - Method for judging opening and closing states of high beam of vehicle driving at night - Google Patents

Method for judging opening and closing states of high beam of vehicle driving at night Download PDF

Info

Publication number
CN102963294B
CN102963294B CN 201210431055 CN201210431055A CN102963294B CN 102963294 B CN102963294 B CN 102963294B CN 201210431055 CN201210431055 CN 201210431055 CN 201210431055 A CN201210431055 A CN 201210431055A CN 102963294 B CN102963294 B CN 102963294B
Authority
CN
China
Prior art keywords
vehicle
high beam
step
gt
image
Prior art date
Application number
CN 201210431055
Other languages
Chinese (zh)
Other versions
CN102963294A (en
Inventor
朱虹
潘行杰
王栋
苟荣涛
何振
张云星
杨静
张晓蕾
Original Assignee
西安理工大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 西安理工大学 filed Critical 西安理工大学
Priority to CN 201210431055 priority Critical patent/CN102963294B/en
Publication of CN102963294A publication Critical patent/CN102963294A/en
Application granted granted Critical
Publication of CN102963294B publication Critical patent/CN102963294B/en

Links

Abstract

本发明公开了一种夜间行驶车辆远光灯开启状态的判别方法,包括以下步骤:拍照得到行驶车辆的车灯偏光滤波图像;在上步处理过的偏光滤波图像中提取车灯区域;根据摄像头的安放位置,划定道路的有效监控区域;对车灯提取处理后的图像进行开运算处理,完成贴标签处理;确定同一车辆的车灯;判断同一车辆的车灯是否为远光灯状态,当条件1和条件2同时满足时,表明该车辆的远光灯处于开启状态,否则,判定为处于关闭状态;重复上述步骤,直到所有的车辆全部判断完成。 The present invention discloses a high beam is turned on at night with a vehicle state determination method, comprising the steps of: filtering the image obtained by photographing polarized lights travel of the vehicle; extracting step lights on the treated area of ​​the polarization filter image; The camera setting positions, effective monitoring area designated road; lights for image extraction process for opening arithmetic processing, labeling processing is completed; determining the same headlight of the vehicle; the same vehicle is determined whether the high beam lights state, when when the condition 1 and condition 2 are simultaneously satisfied, it indicates that the vehicle high beam is turned on, otherwise, it is determined in the closed state; repeating the above steps until all the vehicles all judged complete. 本发明的方法,能够适应夜间不同路灯光照环境,通过偏光滤波的方式,抑制非灯光区域,之后提取远光灯特征,准确地确定出打开远光灯的车辆。 The method of the present invention, it is possible to adapt to the different lights at night lighting conditions, by way of a polarization filter, to suppress the non-lighting region, wherein after extracting the high beam, the vehicle to accurately determine the opening of the high beam.

Description

一种夜间行驶车辆远光灯开启状态的判别方法 One kind of night driving a vehicle high beam state of open discrimination method

技术领域 FIELD

[0001] 本发明属智能交通监控技术领域,涉及一种夜间行驶车辆远光灯开启状态的判别方法。 [0001] The present invention belongs to the technical field of intelligent traffic monitoring, to a high beam is turned on at night with a vehicle state determination method.

背景技术 Background technique

[0002] 视频监控模式在构建智能交通系统中,作为获取并分析交通状况的一种手段,目前已经被广泛应用。 [0002] video surveillance mode in the construction of intelligent transportation systems as a means to obtain and analyze traffic conditions, has been widely used. 对于夜间行驶的车辆,打开车灯照亮路面是安全行驶的基本,然而,远光灯的滥用,会导致会车时对方车辆驾驶员完全看不清前方路面而容易发生车祸,因此,限制远光灯的使用,是保证交通安全的一个必要手段。 For a vehicle traveling at night, headlights illuminate the road is open basic safe driving, however, the high beam abuse, can lead to the other car driver of the vehicle will be fully and easily see the road in front of a car accident, therefore, far from restrictions use light, it is a necessary means to ensure traffic safety.

[0003] 采用视频帧图像,分析行驶车辆是否打开远光灯,不失为一种有效的方法。 [0003] The video frame image analysis with the vehicle high beam is turned on, it is an effective method. 然而,由于夜间车辆行驶过程中,周围环境有很大的不同,例如,有些路段的路灯比较密集,光线很亮,有些路段路灯比较稀疏,光线很暗,这时,拍摄到的车辆远近光灯的形态有比较大的差异,很难进行准确的识别。 However, since the vehicle is running at night, the surrounding environment is very different, e.g., some sections of the street more intensive, very bright light, some sections sparse lights, light is dark, then, the captured light vehicle distance there are relatively large differences in shape, it is difficult to accurately identify.

发明内容 SUMMARY

[0004] 本发明的目的是提供一种夜间行驶车辆远光灯开启状态的判别方法,解决了现有技术受周围环境影响,导致拍摄到的车辆远近光灯的形态有比较大的差异,很难进行准确识别的问题。 [0004] The object of the present invention is to provide a high beam is turned on at night with a vehicle state determination method solves the prior art by the surrounding environment, resulting in a distance form light of a relatively large difference in the vehicle captured, it It is difficult to accurately identify the problem.

[0005] 本发明所采用的技术方案是,一种夜间行驶车辆远光灯开启状态的判别方法,按照以下步骤实施: [0005] The technical proposal of the present invention, a method for determining a vehicle at night with a high beam ON state, the following step of:

[0006] 步骤1、拍照得到行驶车辆的车灯偏光滤波图像 [0006] Step 1, obtained by photographing the vehicle traveling polarized lights filtered image

[0007] 首先,在监控摄像头的镜头前设置水平偏振片和垂直偏振片,通过调节两个偏振片之间的夹角,控制摄像头的进光量,直到捕获的画面仅保留行驶车辆的车灯部分的亮光后,固定调整好这两个偏振片,实现对行驶车辆的偏光滤波拍照; [0007] First, set the horizontal and vertical polarizing polarizer in front of the camera lens to monitor, by adjusting the angle between two polarizing plates, controls the amount of light of the camera, the picture captured until remain only partially traveling vehicle headlights after the light, fixed adjust the two polarizers, achieve travel of the vehicle polarizing filter pictures;

[0008] 步骤2、在上步处理过的偏光滤波图像中提取车灯区域 [0008] Step 2, the step in the treated polarized lights filtered image extracting area

[0009] 2.1)去除转向灯区域 [0009] 2.1) removing flasher region

[0010] 对读入的当前时刻,即t时刻的经过偏光滤波后的视频帧彩色图像的红、绿、蓝三个分量为Drt (χ, y)]mXn> [gt (x, y)]mXn> [bt (x, y) ]mXn,其中,mXn表示巾贞图像的大小为m行η列,(χ, y)表示像素点的坐标,[.]表示整个图像; [0010] read the current time, i.e. t red color image is a video frame after the timing of the polarization filter, green, and blue components of Drt (χ, y)] mXn> [gt (x, y)] [.] mXn> [bt (x, y)] mXn, wherein, MXN image represents the size of the towel Zhen η m rows and columns, (χ, y) represents the coordinates of the pixel, the entire image representation;

[0011] 设转向灯判断矩阵为[Lt(x,y)]mXn,则判断出转向灯区域的计算公式如下: [0011] provided for the turn signal is determined matrix [Lt (x, y)] mXn, it is judged that the turn region of light is calculated as follows:

[0012] [0012]

r, 、fI > 瓦(毛.);)力..I Ug,u,>') >ν) + J)J1: 11./;(.ν, ν) > Thr { s (Mao) r,, fI> W;) force ..I Ug, u,> ')> ν) + J) J1: 11./;(.ν, ν)> Thr {s

Lt{x,y) = < ,ViJ Lt {x, y) = <, ViJ

Λ…丨ο 其它 Λ ... Shu ο Other

[0013] 其中,χ = 1,2,...,m,y = 1,2,...,η,δ为调整量,Ί\是判别车灯光源的红色分量阈值; [0013] where, χ = 1,2, ..., m, y = 1,2, ..., η, δ is the adjustment amount, Ί \ is the red component of the light source lamp determination threshold;

[0014] 2.2)提取图像中所有车辆的远近光车灯区域 [0014] 2.2) the distance light lamp in the extracted image areas of all vehicles

[0015] 将亮度值设为[ft(x,y)]mXn,计算公式如下: [0015] The luminance value is set to [ft (x, y)] mXn, calculated as follows:

[0016] ft (x, y) = 0.299.rt (x, y) +0.587.gt (χ, y) +0.114.bt (χ, y), (2) [0016] ft (x, y) = 0.299.rt (x, y) + 0.587.gt (χ, y) + 0.114.bt (χ, y), (2)

[0017] 对该亮度值进行二值化处理,提取出远近光车灯区域,即二值化图像[gt(x,y)]mXn [0017] The binarizing process on the luminance value, the distance light lamp extracted region, i.e., the binarized image [gt (x, y)] mXn

计算公式为: The formula is:

[0018] [0018]

gi(x^) = |o 其它 ' (3) gi (x ^) = | o other '(3)

[0019] 其中,阈值Th的选择范围为Th e [ε+Δ ε, ξ-Δ ξ], ε为较暗的非车灯区域的亮度,Λ ε为一增量,ξ为较亮的车灯区域的亮度,Λ ξ为一增量; [0019] wherein the selection range of the threshold value Th Th e [ε + Δ ε, ξ-Δ ξ], ε is a non-lights are dark regions, Λ ε is an increment, [xi] for the lighter vehicles brightness of the lamp area, Λ ξ is an increment;

[0020] 步骤3、根据摄像头的安放位置,划定道路的有效监控区域 Monitoring active region [0020] Step 3. The mounting position of the camera, a road designated

[0021] 设定有效监控区域,仅对落入有效监控区域部分的连通域进行分析,即进行以下处理: [0021] monitoring area setting is effective only for the effective monitoring area falls within the communication domain analysis portion, i.e., the following process:

[0022] [0022]

~(XV) = { 0 A(Λ%电Ω,j卜1 丨品(ΛΓ,少)=1, (4) ~ (XV) = {0 A (Λ% electrically Ω, j 1 Shu Bu product (ΛΓ, low) = 1, (4)

[0023] 其中,Ω为所划定的有效监控区域; [0023] wherein, Ω is the area delimited by the effective monitoring;

[0024] 步骤4、对步骤3处理后的图像进行开运算处理 [0024] Step 4, Step 3 after the image processing arithmetic processing opening

[0025] 4.1)对上步得到的豆,(.U)进行开运算处理,开运算时,选择的结构元素为ΝΧΝ,N为奇数,原点为结构元素中心位置,设对&(U)进行开运算处理后的结果图为象 [0025] 4.1) on beans on obtained in step, (. U) opening arithmetic processing, when the opening operation, structural elements selected for ΝΧΝ, N is an odd number, the origin of the structural elements of the center position, located on & (U) for after the opening operation processing result as Pictured

[0026] 4.2)对开运算处理后的结果图乏(U)进行连通域合并处理 [0026] 4.2) after the arithmetic processing result of FIG off spent (U) for processing communication domain consolidation

[0027] 为了在车辆远光灯打开时,将同一侧同时点亮的远光灯区域与近光灯区域合并为同一个区域,连通域合并处理的计算公式如下: [0027] In order to open the vehicle high beam, the high beam light up simultaneously the same side of the near region and light areas belong to a single area, calculated communication domain consolidation process is as follows:

[0028] [0028]

_ { I fA I (.Y + /<, y) = \,k^[a,h] ,,χ _ {I fA I (.Y + / <, y) = \, k ^ [a, h] ,, χ

MM = U(U) 其他 , MM = U (U) Other,

[0029] 其中,[a, b]为连通域合并的距离,a < 0,b > O均为整数; [0029] wherein, [a, b] is the combined distance communication domain, a <0, b> O are integers;

[0030] 4.3)对连通域合并运算处理后的结果图瓦(U)进行贴标签处理; [0030] 4.3) the result Tuva (U) after the arithmetic processing communication domain consolidation labeling process;

[0031] 步骤5、确定同一车辆的车灯 [0031] Step 5, it is determined in the same vehicle headlights

[0032] 按照从上到下,从左到右的顺序,首先选择左侧的连通域,求其质心(xc, yc),质心的计算公式如下: [0032] from top to bottom, left to right, left first selected communication field, find the centroid (xc, yc), the centroid is calculated as follows:

[0033] 4=士.Σ.γ, (6) [0033] 4 = disabilities .Σ.γ, (6)

/VQ, (v,i)eQA / VQ, (v, i) eQA

[0034] >v =.Σ.1'' (7) [0034]> v = .Σ.1 '' (7)

1^nk (.V-Vieni 1 ^ nk (.V-Vieni

[0035] 其中,Ω,为找到的连通域; [0035] wherein, Ω, communication domain is found;

[0036] 之后,按照质心的位置(Xc;,y。)画一条水平直线,即在χ = X。 After [0036], according to the position of the centroid (Xc;., Y) draw a straight horizontal line, i.e., χ = X. 上判断是否有穿越的连通域, A communication domain is determined whether there through,

[0037] 如果有一个,则视其为同一车辆的车灯; [0037] If there is one, the lights them as the same vehicle;

[0038] 如果有若干个,则将最左侧的视为其同一车辆的车灯; [0038] If there are several, will be seen as the far left headlight its same vehicle;

[0039] 步骤6、判断同一车辆的车灯是否为远光灯状态 [0039] Step 6, it is determined whether the same vehicle high beam lights state

[0040] 对判断为属于同一车辆的一对车灯区域,设为Ω&和Ωκ1ί,判断其联通域的面积特征与形状特征,左侧灯面积&与右侧灯面积Sk的特征计算公式分别为: [0040] Analyzing a pair of headlights of the vehicle belonging to the same area, and set [Omega] & Ωκ1ί, wherein the shape of the area is determined characterized in that the link region, the area of ​​the left lamp & lamp calculating characterized in right area respectively Sk formula :

Sl = Σ 豆(W) Beans Sl = Σ (W)

{.r,_r)£QiA {.R, _r) £ QiA

[0041] Σ 抑,.)') (9) [0041] Σ suppression.) ') (9)

[0042] 形状特征为外接矩形的长宽比,分别找到QLk及Ωκ1ί的最左侧、最右侧、最上侧、最下侧的四个点,将该四个点作为外接矩形的左边、右边、上边、下边的起点构成外接矩形,设为Recut和RecKk,其长、宽分别为Luo LKk、HLk, HEk,形状特征的计算公式如下: [0042] wherein the aspect ratio of the circumscribed rectangle shape, and are left to find the most QLk and Ωκ1ί, rightmost, uppermost, lowermost side of the four points, four points as the left side of the circumscribed rectangle, the right , upper, lower starting point of the circumscribed rectangular configuration, and is set Recut RecKk, length, width, respectively Luo LKk, HLk, HEk, wherein the shape is calculated as follows:

[0043] Pit=Yu1- (10) [0043] Pit = Yu1- (10)

^lk ^ Lk

[0044] (11) [0044] (11)

Lrjc Lrjc

[0045] 计算远光灯判定条件I为:P Lk+ P nk < thp , (12) [0045] The determination condition high beam I is calculated: P Lk + P nk <thp, (12)

[0046] 其中,thp为远光灯的判断阈值; [0046] wherein, thp is the threshold value determination high beam;

[0047] 计算远光灯判定条件2为:Sk+Sl > 2.Ths, (13) [0047] Calculation of the high beam determination condition 2: Sk + Sl> 2.Ths, (13)

[0048] 其中,ths为远光灯的判断阈值,为同一水平位置上所有连通域的平均面积; [0048] wherein, for the high beam THS determination threshold value, the same level as the average area of ​​all the communicating position domain;

[0049] 当条件I和条件2同时满足时,表明该车辆的远光灯处于开启状态,否则,判定为处于关闭状态; [0049] When the conditions I and 2 satisfy the condition, indicating that the vehicle high beam is turned on, otherwise, it is determined in a closed state;

[0050] 重复步骤5、步骤6,直到所有的车辆全部判断完成。 [0050] Repeat Step 5, Step 6, until all the vehicles all judged complete.

[0051] 本发明的有益效果是,能够适应夜间不同路灯光照环境,通过偏光滤波的方式,抑制非灯光区域,之后提取远光灯特征,准确地确定出打开远光灯的车辆。 [0051] Advantageous effects of the present invention is able to adapt to different lights at night lighting conditions, by way of a polarization filter, to suppress the non-lighting region, wherein after extracting the high beam, the vehicle to accurately determine the opening of the high beam.

附图说明 BRIEF DESCRIPTION

[0052] 图1为拍到的道路上行驶车辆的偏光滤波图像; [0052] FIG. 1 is a polarization filter on the photographed image of the vehicle traveling road;

[0053] 图2为偏光滤波效果示意图,是车辆打开了近光灯状态的图像; [0053] FIG. 2 is a schematic view of a polarization filtering effect is nearly the vehicle opened state image light;

[0054] 图3为偏光滤波效果示意图,是车辆打开了远光灯状态的图像; [0054] FIG. 3 is a schematic view of a polarization filtering effect is an image of the vehicle to open the high beam state;

[0055] 图4为本发明方法中的车灯区域提取效果示意图,是对图1的二值化图像,虚线连接的是同一个车辆两侧灯; Method lamp schematic region [0055] FIG. 4 of the present invention results in the extraction, is a binarized image of FIG. 1, a dashed connection is the same on both sides of the vehicle lamp;

[0056] 图5为本发明方法中的车灯区域提取效果示意图,是对图2的二值化图像; Lamp schematic region [0056] FIG. 5 shows the effect of extraction process of the invention, FIG 2 is a binarized image;

[0057] 图6为本发明方法中的车灯区域提取效果示意图,是对图3的二值化图像; Lights effect region extraction method [0057] FIG. 6 is a schematic view of the present invention, FIG. 3 is a binarized image;

[0058] 图7为本发明方法中的道路有效监控范围示意图; Effective monitoring range schematic way the method [0058] FIG. 7 of the present invention;

[0059] 图8为本发明方法中的贴标签效果示意图,是对图5进行标签处理后的图像; Labeling effect of the method [0059] Figure 8 is a schematic view of the invention, FIG 5 is an image after the labeling process;

[0060] 图9为本发明方法中的贴标签效果示意图,是对图6进行标签处理后的图像。 Labeling results [0060] FIG. 9 is a schematic diagram of the process of the invention, FIG 6 is an image of the label after the treatment.

具体实施方式 Detailed ways

[0061] 本发明的夜间行驶车辆远光灯开启状态的判别方法,按照以下步骤实施: [0061] The present invention night traveling vehicle high beam ON state determination method, according to the embodiment of the following steps:

[0062] 步骤1、拍照得到行驶车辆的车灯偏光滤波图像 [0062] Step 1, obtained by photographing the vehicle traveling polarized lights filtered image

[0063] 首先,在监控摄像头的镜头前设置水平偏振片和垂直偏振片,通过调节两个偏振片之间的夹角,控制摄像头的进光量,直到捕获的画面仅保留行驶车辆的车灯部分的亮光后,固定调整好这两个偏振片,实现对行驶车辆的偏光滤波拍照。 [0063] First, set the horizontal and vertical polarizing polarizer in front of the camera lens to monitor, by adjusting the angle between two polarizing plates, controls the amount of light of the camera, the picture captured until remain only partially traveling vehicle headlights after the light, fixed adjust the polarizing plates, a polarizing filter to achieve pictures travel of the vehicle.

[0064] 如图1、图2、图3所示,是实施例中的道路上行驶多个车辆的车灯偏光滤波,以及一个车辆的两个状态下的车灯滤波效果,其中,图1是道路上行驶多个车辆的偏光滤波图像,矩形框内是转向灯,图2车辆打开了近光灯状态的图像,图3是车辆打开了远光灯状态的图像。 [0064] As shown in FIG 1, FIG 2, FIG. 3, a plurality of traveling of the vehicle headlights in a polarization filter on the road embodiment, the filter effect and the lights in the two states of a vehicle embodiment, wherein FIG. 1 filtered image is polarized with a plurality of vehicles on the road, the rectangular frame is a turn signal lamp, opening the vehicle 2 near the image light state, FIG. 3 is an image of the vehicle high beam opened state.

[0065] 步骤2、在上步处理过的偏光滤波图像中提取车灯区域 [0065] Step 2, the step in the treated polarized lights filtered image extracting area

[0066] 2.1)去除转向灯区域 [0066] 2.1) removing flasher region

[0067] 对读入的当前时刻,即t时刻的经过偏光滤波后的视频帧彩色图像的红绿蓝三个分量为[rt(x, y)]mXn、[gt(x, y)]mXn、[bt (x, y) ]mXn,其中,mXn表示巾贞图像的大小为m行η列,(X,y)表示像素点的坐标,[.]表示整个图像; [0067] read the current time, i.e., after the polarization filter component video RGB three color image frame time t is [rt (x, y)] mXn, [gt (x, y)] mXn [.], [bt (x, y)] mXn, wherein, MXN image represents the size of the towel Zhen η m rows and columns, (X, y) represents the coordinates of the pixel, the entire image representation;

[0068] 由于转向灯与远近光灯不同的是,转向灯是红色的,在此,设转向灯判断矩阵为[Lt (x, y)]mXn,则判断出转向灯区域的计算公式如下: [0068] Since the turn signal lamp and the distance light is different, turn signal lamp is red, here, it is assumed flasher is determined matrix [Lt (x, y)] mXn, it is determined that the steering is calculated light region is as follows:

[0069] [0069]

f I /; (.χ,.V) > g, (.ϊ, ν) j I'-1 KlCf, (.\\ ν) > k (.Y, ν) + δ) j I'- j L/;(.r, v) > Thr {、 f I /; (.χ, .V)> g, (.ϊ, ν) j I'-1 KlCf, (\\ ν.)> k (.Y, ν) + δ) j I'- j L /;(.r, v)> Thr {,

= 其他 '⑴ = Other '⑴

[0070] 其中,χ = l,2,...,m,y = 1,2,...,η,δ为调整量,根据安装设备的地点取值,优选取值范围为[10,40],Ί\是判别车灯光源的红色分量阈值,优选取值范围为[80,150]; [0070] where, χ = l, 2, ..., m, y = 1,2, ..., η, δ is the adjustment amount, depending on the location of the mounting device value, preferably in the range [10, 40], Ί \ is the red component of the light source lamp is determined threshold, preferably in the range [80, 150];

[0071] 2.2)提取图像中所有车辆的远近光车灯区域 [0071] 2.2) the distance light lamp in the extracted image areas of all vehicles

[0072] 将亮度值设为[ft (x, y)]fflXn,计算公式如下: [0072] The luminance value is set to [ft (x, y)] fflXn, calculated as follows:

[0073] ft (x, y) = 0.299.rt (x, y) +0.587.gt (x, y) +0.114.bt (x, y), (2) [0073] ft (x, y) = 0.299.rt (x, y) + 0.587.gt (x, y) + 0.114.bt (x, y), (2)

[0074] 对该亮度值进行二值化处理,提取出远近光车灯区域, [0074] The binarizing process on the luminance value, the distance light lamp extracted area,

[0075] 因为车灯部分与其他部分的亮度差异很大,因此,在此采用全局阈值的方法来实现,即二值化图像[gt(X,y)]mXn计算公式为: [0075] Since a large portion of the lamp the difference in brightness of other portions, and therefore, this method achieved a global threshold value, i.e., the binarized image [gt (X, y)] mXn calculated as:

[0076] [0076]

σ (χ./: U ν) > Th j ί-1 IXi (χ, y)-0 r 3、 σ (χ /:. U ν)> Th j ί-1 IXi (χ, y) -0 r 3,

[0 其它 , [Other 0,

[0077] 其中,阈值Th的选择范围为Th e [ε+Δ ε, ξ-Δ ξ], ε为较暗的非车灯区域的亮度,Λ ε为一增量,优选取值范围为Λ ε e [0.5 ε , ε ], ξ为较亮的车灯区域的亮度,Δ ξ为一增量,优选取值范围为Λ ξ e [0.2 ξ , 0.8 ξ ] ; ε , ξ的取值,取决于偏光滤波结果O [0077] wherein the selection range of the threshold value Th Th e [ε + Δ ε, ξ-Δ ξ], ε is a non-lights are dark regions, Λ ε is an increment, preferably in the range of Lambda ε e [0.5 ε, ε], ξ lights brighter luminance region, Δ ξ is an increment, preferably in the range of Λ ξ e [0.2 ξ, 0.8 ξ]; value ε, ξ, and O depending on polarization filtering result

[0078] 对于偏光滤波较为理想的情况,即如图1的情况,建议优选最简单地确定阈值为7L/8,其中L为摄像头给定的图像亮度值范围[0,L]的最高值。 [0078] For polarization filtering ideal case, i.e. the case of FIG. 1, is recommended preferably simplest determination threshold value 7L / 8, where L is a camera of a given image luminance value range [0, L] the highest value.

[0079] 如图4、图5、图6所示,是实施例中的车灯区域提取效果,其中,图4是图1道路上行驶中所有车辆的车灯区域提取图像,其中矩形框内是转向灯区域,按照公式(I)计算处理后,被消除,图5是图2的车灯区域,图6是图3的车灯区域。 [0079] As shown in FIG 4, FIG 5, FIG. 6, a region extraction efficiency lamp in the embodiment, wherein FIG. 4 is a lamp of the extracted image areas of all vehicles traveling on a road in FIG. 1, wherein the rectangular frame turn signal region is post-processing according to formula (I) is calculated, is eliminated, FIG. 5 is a view of the lamp area 2, FIG. 6 is a view of the lamp 3 region.

[0080] 步骤3、根据摄像头的安放位置,划定道路的有效监控区域 Monitoring active region [0080] Step 3. The mounting position of the camera, a road designated

[0081] 考虑到夜间道路两旁可能有店铺或者建筑物上的照明,在二值化时,会被误判为车灯,因此,按照图7所示的示意图,设定有效监控区域,仅对落入有效监控区域部分的连通域进行分析,即进行以下处理: [0081] Considering the roads at night, or shop lighting may have on the building, at the time of binarization, will be mistaken for the lights, therefore, according to the diagram shown in FIG. 7, is set to monitor the effective area, only field falls within the communication area effective monitoring portion is analyzed, i.e., the following process:

[0082] [0082]

,JO g,(.v, ν)ίΩ,^ί.| !.g,(.Y, v)-l /4x , JO g, (v, ν.) ΊΩ, ^ ί |.! .G, (. Y, v) -l / 4x

gt X,y Ur,(-Y,.r) ' gt X, y Ur, (- Y, .r) '

[0083] 其中,Ω为由图7所划定的有效监控区域, [0083] wherein, [Omega] FIG delineated by the effective monitoring region,

[0084] 如图7所示,是实施例中的有效监控区域示意图,其中,梯形所框定的部分为监控有效区域,两侧的边界是监控摄像头监视画面中的道路两侧的边界,前后界限是车辆行驶过程中,摄像头能够捕获到足以用来有效判断车灯大小的范围。 [0084] As shown in FIG. 7 is a schematic view of the effective area of ​​the monitoring in the embodiment, wherein the trapezoidal portion of the frame to monitor a predetermined active area on both sides of the border surveillance camera monitor screen boundaries on both sides of the road, the front and rear boundaries the vehicle is traveling, the camera can be used to effectively capture sufficient size judgment lamp. (如果后界太远,车灯太小,容易被当作噪声去除,如果前界太近,则车灯已不在画面中。) (If the boundary is too far, too small lamp, is easily removed as a noise, if the front sector too, the screen no longer lights.)

[0085] 步骤4、对步骤3处理后的图像进行开运算处理 [0085] Step 4, Step 3 after the image processing arithmetic processing opening

[0086] 4.1)对上步得到的泛(X,.).)进行开运算处理,以便消除噪声的影响,开运算方法在现有数字图像处理相关专业教材上均有详细说明,这里不再详述。 [0086] 4.1) of the pan (X ,.) obtained in the previous step.) Opening calculation process, in order to eliminate the effect of noise, calculation methods are described in the open on a conventional digital image processing details related professional materials, here not detail. 开运算时,选择的结构元素为NXN,N为奇数,原点为结构元素中心位置,设对裒(x,.y)进行开运算处理后的结果图为幺⑷.); When the opening operation, as the structure element selected NXN, N is an odd number, the center position of the origin of the structural elements, arranged on the take out (x, .y) picture shows the result of the opening operation ⑷ unitary process).;

[0087] 4.2)对开运算处理后的结果图殳UJ)进行连通域合并处理 [0087] 4.2) after the arithmetic processing result of FIG Shu off the UJ) for communication domain consolidation process

[0088] 因为车辆的远光灯和近光灯处于同一水平位置上,当远光灯打开时,近光灯同时亮,这里所述的连通域合并处理,是为了在车辆远光灯打开时,将同一侧同时点亮的远光灯区域与近光灯区域合并为同一个区域,连通域合并处理的计算公式如下: [0088] Because the high and low beam headlights of the vehicle at the same horizontal position, when the high beam is turned on, while the light beam lights, herein communication domain consolidation process, in order to open the vehicle high beam , the same side high beam simultaneously lit light region and near region were combined for the same area, calculated communication domain consolidation process is as follows:

[0089] [0089]

_ JI 如呢(x + /c, y) = 1,/( e[a,/)] ⑷ As it JI _ (x + / c, y) = 1, / (e [a, /)] ⑷

JU(u) 難 , JU (u) is difficult,

[0090] 其中,[a,b]为连通域合并的距离,a <0,b> O均为整数,根据经验取值,建议优选a G [-8, -4], b G [4, 8]; [0090] wherein, [a, b] is the combined distance communication domain, a <0, b> O are integers, the value based on experience, it is recommended preferably a G [-8, -4], b G [4, 8];

[0091] 4.3)对连通域合并运算处理后的结果图瓦(U)进行贴标签处理 [0091] 4.3) the result Tuva (U) after the arithmetic processing communication domain consolidation labeling process

[0092] 贴标签方法在数字图像处理方面的专业教材上均有详细说明,这里不再详述。 [0092] The labeling method of teaching material in the digital image processing are described in detail, not described in detail here. 对图3、图4、图5进行贴标签之后,便可对不同的连通域进行标识,贴标签处理结果如图8、图9所示,其中,图8是对图5进行标签处理后的图像,图9是对图6进行标签处理后的图像,附图中的浅色表示I号标签,深色表示2号标签; After 3, 4, 5 for labeling different communication domains can be identified, the labeling processing result shown in FIG. 8, FIG. 9, wherein FIG 8 is an image of the label of FIG. 5 after the treatment, FIG 9 is an image of FIG. 6 after the labeling process, represented in the drawings light I, labels, label 2 indicates the dark;

[0093] 步骤5、确定同一车辆的车灯 [0093] Step 5, it is determined in the same vehicle headlights

[0094] 在汽车结构布局设计上,将车灯设置于车头两侧同一高度位置,所以,按照从上到下,从左到右的顺序,首先选择左侧的连通域,求其质心(xc, yc),质心的计算公式如下: [0094] The structural layout design on an automobile, the lamp is provided at the same height on both sides of the front, so that, from top to bottom, left to right, left first select the communication field, find the centroid (xc , yc), the centroid is calculated as follows:

[0095] xc= 士.Σχ, (6) [0095] xc = disabilities .Σχ, (6)

^Qk (.v,v)eQA ^ Qk (.v, v) eQA

[0096] y^=^—' Σ.ν' (7) [0096] y ^ = ^ - 'Σ.ν' (7)

(-v,v)eQA (-V, v) eQA

[0097] 其中,Ω,为找到的连通域; [0097] wherein, Ω, communication domain is found;

[0098] 之后,按照质心的位置(Xc;,y。)画一条水平直线,即在χ = X。 After [0098], according to the position of the centroid (Xc;., Y) draw a straight horizontal line, i.e., χ = X. 上判断是否有穿越的连通域,如果有一个,则视其为同一车辆的车灯;如果有若干个,则将最左侧的视为其同一车辆的车灯; A communication domain is determined whether there through, if there is one, them as the same vehicle lamp; if there are several, of the leftmost lamp which is regarded as the same vehicle;

[0099] 如图4所示,是实施例中的同一车辆车灯检测示意图,其中的虚线给出的直线,连接的就是按照公式出),(7)计算出质心位置后,测得的同一车辆的两侧车灯。 [0099] As shown in FIG 4, a vehicle lamp is the same schematic embodiment examples of the detection, wherein the straight line given by the dotted line, the connection is in accordance with equation), (7) the calculated centroid position, measured in the same lights on both sides of the vehicle.

[0100] 步骤6、判断同一车辆的车灯是否为远光灯状态 [0100] Step 6, it is determined whether the same vehicle high beam lights state

[0101] 对判断为属于同一车辆的一对车灯区域,设为Qut和Ωκ1ί,判断其联通域的面积特征与形状特征,左侧灯面积&与右侧灯面积Sk的特征计算公式分别为:S - Xν) 8) [0101] Analyzing a pair of headlights of the vehicle belonging to the same area, and to Qut Ωκ1ί, wherein the shape of the area is determined characterized in that the link region, the area of ​​the left lamp & lamp calculating characterized in right area respectively Sk formula : S - Xν) 8)

[0102] sH= YjS(x^y) 〔9) [0102] sH = YjS (x ^ y) [9)

(Λ-,ν)εΩ^ (Λ-, ν) εΩ ^

[0103] 形状特征为外接矩形的长宽比,分别找到QLk及Ωκ1ί的最左侧、最右侧、最上侧、最下侧的四个点,将该四个点作为外接矩形的左边、右边、上边、下边的起点构成外接矩形,设为Recut和RecKk,其长、宽分别为Luo LKk、HLk, HEk,形状特征的计算公式如下: [0103] wherein the shape of an aspect ratio of the circumscribed rectangle, and were found QLk Ωκ1ί the leftmost, rightmost, uppermost, lowermost side of the four points, four points as the left side of the circumscribed rectangle, the right , upper, lower starting point of the circumscribed rectangular configuration, and is set Recut RecKk, length, width, respectively Luo LKk, HLk, HEk, wherein the shape is calculated as follows:

[0104] PLk ^ (10) [0104] PLk ^ (10)

Llk Llk

[0105] Pm =^l (11) [0105] Pm = ^ l (11)

[0106] 计算远光灯判定条件I为:P Lk+ P Kk < thp , (12) [0106] Calculation of the high beam determination condition I: P Lk + P Kk <thp, (12)

[0107] 其中,thp为远光灯的判断阈值,远光灯打开时,由于一侧两个灯都点亮,为经验值,所以建议thp的优选取值范围为thp e [I, 1.6]; [0107] wherein, thp is the threshold value determination high beam, the high beam when open, since the side of the two lamps are lit, an empirical value, it is proposed preferably in the range of thp thp e [I, 1.6] ;

[0108] 计算远光灯判定条件2为:Sk+Sl > 2.Ths, (13) [0108] Calculation of the high beam determination condition 2: Sk + Sl> 2.Ths, (13)

[0109] 其中,ths为远光灯的判断阈值,为同一水平位置上所有连通域的平均面积; [0109] wherein, for the high beam THS determination threshold value, the same level as the average area of ​​all the communicating position domain;

[0110] 当条件I和条件2同时满足时,表明该车辆的远光灯处于开启状态,否则,判定为处于关闭状态。 [0110] When the conditions I and 2 satisfy the condition, indicating that the vehicle high beam is turned on, otherwise, it is determined in the closed state.

[0111] 重复步骤5、步骤6,直到所有的车辆全部判断完成。 [0111] Repeat Step 5, Step 6, until all the vehicles all judged complete.

[0112] 本发明的方法通过以上的步骤,便可在行驶车辆进入有效监控区域内时,判断其远光灯是否开启,如果属于远光灯开启状态,再利用运动目标跟踪技术,再计算出其在监控区域内持续开启远光灯的时间,为智能交通系统提供一种保障安全的有效检测方法。 When [0112] The method of the present invention by the above procedure, can be effective in the traveling vehicle enters the monitored region, it is determined whether the high beam is turned on, if they are high beam ON state, and then using motion tracking techniques, and then calculate the its duration turn-on time of the high beam in the monitored area, to provide an effective method for detecting a guarantee for the safety of intelligent transportation systems.

Claims (6)

1.一种夜间行驶车辆远光灯开启状态的判别方法,其特征在于,按照以下步骤实施: 步骤1、拍照得到行驶车辆的车灯偏光滤波图像首先,在监控摄像头的镜头前设置水平偏振片和垂直偏振片,通过调节两个偏振片之间的夹角,控制摄像头的进光量,直到捕获的画面仅保留行驶车辆的车灯部分的亮光后,固定调整好这两个偏振片,实现对行驶车辆的偏光滤波拍照; 步骤2、在上步处理过的偏光滤波图像中提取车灯区域2.1)去除转向灯区域对读入的当前时刻,即t时刻的经过偏光滤波后的视频帧彩色图像的红绿蓝三个分量为[rt(x, y)]mXn、[gt (x, y)]mXn> [bt(x, y)]mXn,其中,mXn 表示巾贞图像的大小为m行n 列,(X,y)表示像素点的坐标,[.]表示整个图像; 设转向灯判断矩阵为[Lt(x,y)]mXn,则判断出转向灯区域的计算公式如下: r , 、Jl rt(x,y)>gt(x,j,)并且(g,(.τ,y) > bt(x, v) + J A high beam is turned on at night with a vehicle state determination method, wherein the following steps: Step 1, the filtered image obtained by photographing polarized lights traveling of the vehicle First, the level of the polarizing plate disposed in front of the camera lens to monitor and the vertical polarizer by adjusting the angle between two polarizing plates, controls the amount of light of the camera, the picture captured until reservation only light traveling vehicle headlights portion fixed adjust the polarizing plates, to achieve polarization filter photographed running of the vehicle; step 2, step on the treated polarized lights filtered image extracting area 2.1) removing flasher zone the current time read, i.e., a color video image frame time t after the polarizing filter red, green and blue components of the three [rt (x, y)] mXn, [gt (x, y)] mXn> [bt (x, y)] mXn, wherein, mXn image represents the size of the towel m rows Zhen n columns, (X, y) represents coordinates of a pixel point, represents the entire image; provided flasher is determined matrix [Lt (x, y)] mXn, it is determined that the steering is calculated light region below [.]: r, , Jl rt (x, y)> gt (x, j,) and (g, (. τ, y)> bt (x, v) + J )并且r,(x,y)>Thr 、A(叫o 其他 '⑴ 其中,X = 1,2,...,m,y = 1,2,...,n,δ为调整量,Thr是判别车灯光源的红色分量阈值; 2.2)提取图像中所有车辆的远近光车灯区域将亮度值设为[ft(x,y)]mXn,计算公式如下: ft (χ, y) = 0.299.rt (x, y) +0.587.gt (x, y)+0.114.bt (x, y), (2 ) 对该亮度值进行二值化处理,提取出远近光车灯区域,即二值化图像[gt(x,y)]mXn计算公式为: 彳侧I軸>') = (), ο) A [O 其它其中,阈值Th的选择范围为Th e [ε+Δ ε, ξ-Δ ξ], ε为较暗的非车灯区域的亮度,Λ ε为一增量,ξ为较亮的车灯区域的亮度,Λ ξ为一增量; 步骤3、根据摄像头的安放位置,划定道路的有效监控区域设定有效监控区域,仅对落入有效监控区域部分的连通域进行分析,即进行以下处理:~ OV) = { 0 8ι (.γ,^ Ώ,并目品(χ,-ν) = I, ( 4) 其中,Ω为所划定的有效监控区域; 步骤4、对步骤3处 ) And r, (x, y)> Thr, A (o Other called '⑴ wherein, X = 1,2, ..., m, y = 1,2, ..., n, δ is the adjustment amount, Thr is the red component of the light source lamp is determined threshold; lamp distance light areas of all vehicles luminance value 2.2) to extract the image [ft (x, y)] mXn, calculated as follows: ft (χ, y) = 0.299.rt (x, y) + 0.587.gt (x, y) + 0.114.bt (x, y), (2) binarizing process on the luminance value, the distance light lamp extracted region, i.e., two binarized image [gt (x, y)] mXn calculated as follows: left foot side of I-axis> ') = (), ο) a [O other wherein the selected threshold range of Th is Th e [ε + Δ ε, ξ -Δ ξ], ε is a non-lights are dark regions, Λ ε is an increment, [xi] is the brightness of brighter areas of lights, Λ ξ is an increment; step 3 the mounting position of the camera , effectively monitor designated areas of the road is set to monitor the effective area, only the effective monitoring area falls within the communication domain analysis portion, i.e., the following process: ~ OV) = {0 8ι (.γ, ^ Ώ, mesh goods and (χ, -ν) = I, (4) wherein, [Omega] to limit the area delineated; step 4, at step 3 后的图像进行开运算处理4.1)对上步得到的孟(U)进行开运算处理,开运算时,选择的结构元素为NXN,N为奇数,原点为结构元素中心位置,设对泛(^)进行开运算处理后的结果图为么(Xj); 4.2)对开运算处理后的结果图幺(x,_F)进行连通域合并处理为了在车辆远光灯打开时,将同一侧同时点亮的远光灯区域与近光灯区域合并为同一个区域,连通域合并处理的计算公式如下: _., JI 如果#办+ *,3;) = 1,是£[(7,6] UrC-Y, V) 具匕其中,[a,b]为连通域合并的距离,a <0,b> O均为整数; 4.3)对连通域合并运算处理后的结果图艮U,>.)进行贴标签处理; 步骤5、确定同一车辆的车灯按照从上到下,从左到右的顺序,首先选择左侧的连通域,求其质心(xc, yc),质心的计算公式如下: Σχ, ⑷ ivOfr (-ϊ,ι yc=士.Ty> (7) Qk U\y)Gnk 其中,Qk为找到的连通域; 之后,按照质心的位置(U。 Image opening operation after treatment 4.1) of the obtained in step Meng (U) opening arithmetic processing, when the opening operation, structural elements selected for NXN, N is an odd number, the origin of the structural elements of the center position, is provided on the pan (^ ) the picture shows the opening operation results after the treatment it (Xj of); after 4.2) split arithmetic processing result of FIG unitary (x, _F) for communication domain consolidation process to high beam when the vehicle is turned on, while the same side of point high beam bright light region and the near region were combined for the same area, communication domain is calculated as follows merging process: _, JI do if # + *, 3;) = 1, a £ [(7,6]. UrC-Y, V) with a dagger wherein, [a, b] is the combined distance communication domain, a <0, b> O are integers; 4.3) of the result of the arithmetic processing communication domain consolidation FIG Gen U,>. ) for the labeling process; step 5, is determined according to the vehicle lamp of the same from top to bottom, left to right, left first select the communication field, find the centroid (xc, yc), the centroid is calculated as follows: Σχ, ⑷ ivOfr (-ϊ, ι yc = Disabled .Ty> (7) Qk U \ y) Gnk wherein, communication domain is found Qk; then, according to the position of the centroid (U. the )画一条水平直线,即在χ = χ。 ) Draw a straight horizontal line, i.e. at = χ χ. 上判断是否有穿越的连通域, 如果有一个,则视其为同一车辆的车灯; 如果有若干个,则将最左侧的视为其同一车辆的车灯; 步骤6、判断同一车辆的车灯是否为远光灯状态对判断为属于同一车辆的一对车灯区域,设为QuJP Ωκ1ί,判断其联通域的面积特征与形状特征,左侧灯面积&与右侧灯面积Sk的特征汁算公式分别为: Sl = Σ 取>,) 8 1 s- = J^g(x,y) (9 > 形状特征为外接矩形的长宽比,分别找到Qut及Ωκ1ί的最左侧、最右侧、最上侧、最下侧的四个点,将该四个点作为外接矩形的左边、右边、上边、下边的起点构成外接矩形,设为ReCut和ReCKk,其长、宽分别为Lu^ LKk、HLk, HEk,形状特征的计算公式如下:P (10) ^LkP ( 11 ….LRk 计算远光灯判定条件I为:P Lk+ P H5 < thp, (12) 其中,thp为远光灯的判断阈值; 计算远光灯判定条件2为:Sk+Sl > 2.Ths, (13) 其中,ths为远光灯的判断 A communication domain is determined whether there through, if there is one, them as the same vehicle lamp; if there are several, of the leftmost lamp which is regarded as the same vehicle; step 6, the same vehicle is determined high beam lights state determines whether a pair of headlights of the vehicle belonging to the same region, to QuJP Ωκ1ί, wherein the shape feature is determined which area Unicom domain, wherein the left lamp area of ​​the lamp & right side area Sk juice calculation formula are: Sl = Σ take>,) 8 1 s- = (9> shaped feature J ^ g (x, y) is the aspect ratio of the circumscribed rectangle, respectively, to find the most left and Qut Ωκ1ί most right, uppermost, lowermost side of the four points, four points as the left side of the circumscribed rectangle, starting right, top, bottom rectangular external configuration, and to ReCut ReCKk, length, width, respectively Lu ^ LKk, HLk, HEk, wherein the shape is calculated as follows: P (10) ^ LkP (11 ... .LRk high beam determination condition I is calculated: P Lk + P H5 <thp, (12) wherein, thp is the high beam the threshold value determination; calculated as the high-beam determination condition 2: Sk + Sl> 2.Ths, (13) wherein, ths is the determination of the high beam 阈值,为同一水平位置上所有连通域的平均面积; 当条件I和条件2同时满足时,表明该车辆的远光灯处于开启状态,否则,判定为处于关闭状态; 重复步骤5、步骤6,直到所有的车辆全部判断完成。 Threshold value, for the same horizontal position on the average area of ​​all communication domain; and when the condition I 2 satisfy the condition, indicating that the vehicle high beam is turned on, otherwise, it is determined in the closed state; repeat step 5, step 6, all judgment until all of the vehicles completed.
2.根据权利要求1所述的夜间行驶车辆远光灯开启状态的判别方法,其特征在于:所述的步骤2.1)中,调整量δ的优选取值范围为[10,40],红色分量阈值Ί\的优选范围为[80, 150]。 The night according to claim 1 with the vehicle high beam open state determination method, wherein: said step 2.1), the adjustment amount δ is preferably in the range [10, 40], the red component threshold Ί \ is preferably in the range [80, 150].
3.根据权利要求1所述的夜间行驶车辆远光灯开启状态的判别方法,其特征在于:所述的步骤2.2)中, 一般情况下,增量Λ ε的优选取值范围为Λ ε e [0.5 ε , ε ],增量Λ ξ的优选取值范围为Λ ξ e [0.2ξ ,0.8ξ], 对于偏光滤波较为理想的情况,优选阈值Th为7L/8,其中L为摄像头给定的图像亮度值范围[0,L]的最高值。 The night according to claim 1 with the vehicle high beam open state determination method, wherein: said step 2.2) and, under normal circumstances, the increment Λ ε is preferably in the range of Λ ε e [0.5 ε, ε], Λ ξ increment is preferably in the range of Λ ξ e [0.2ξ, 0.8ξ], it is preferable for the polarizing filter, the threshold value Th is preferably 7L / 8, where L is given camera image luminance value range [0, L] the highest value.
4.根据权利要求1所述的夜间行驶车辆远光灯开启状态的判别方法,其特征在于:所述的步骤4.2)中,优选ae [8,-4],be [4,8]。 The night according to claim 1 with the vehicle high beam open state determination method, characterized in that: in said step 4.2), preferably ae [8, -4], be [4,8].
5.根据权利要求1所述的夜间行驶车辆远光灯开启状态的判别方法,其特征在于:所述的步骤6中,远光灯的判断阈值thp的优选取值范围为thp e [I, 1.6]。 The night according to claim 1 with the vehicle high beam open state determination method, wherein: in the step 6, the threshold value is determined preferably in the range of the high beam is thp thp e [I, 1.6].
6.根据权利要求1到5任一所述的夜间行驶车辆远光灯开启状态的判别方法,其特征在于:所述的步骤6之后,再利用运动目标跟踪技术,计算出该车辆的远光灯在监控区域内持续开启的时间。 6. The night 1-1 according to any one of claims 5 running of the vehicle high beam open state determination method, characterized in that: after said step 6, reuse moving target tracking, the beam is calculated vehicle lamp on continuously in time within the monitoring area.
CN 201210431055 2012-11-02 2012-11-02 Method for judging opening and closing states of high beam of vehicle driving at night CN102963294B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201210431055 CN102963294B (en) 2012-11-02 2012-11-02 Method for judging opening and closing states of high beam of vehicle driving at night

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201210431055 CN102963294B (en) 2012-11-02 2012-11-02 Method for judging opening and closing states of high beam of vehicle driving at night

Publications (2)

Publication Number Publication Date
CN102963294A CN102963294A (en) 2013-03-13
CN102963294B true CN102963294B (en) 2014-12-10

Family

ID=47793824

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201210431055 CN102963294B (en) 2012-11-02 2012-11-02 Method for judging opening and closing states of high beam of vehicle driving at night

Country Status (1)

Country Link
CN (1) CN102963294B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103295399B (en) * 2013-05-14 2015-01-21 西安理工大学 On-state judging method of headlights on full beam of night-driving cars based on morphological characteristics
CN103870809B (en) * 2014-02-27 2017-06-16 奇瑞汽车股份有限公司 Method and apparatus for detecting a vehicle
CN105575128B (en) * 2016-01-20 2018-03-13 王威立 Vehicle high beam violation detection and capture system
CN106128096B (en) * 2016-06-29 2018-10-26 安徽金赛弗信息技术有限公司 A motor vehicle lamp according to the specified turn the steering intelligent recognition method and apparatus lane change violation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2911206A1 (en) * 2006-12-21 2008-07-11 Valeo Vision Sa Method for automatic discrimination of bright areas detected by a device for assisting a vehicle driving.
DE102010048689A1 (en) * 2010-10-16 2011-05-12 Daimler Ag Method for adjusting and/or calibrating headlight of motor vehicle, involves comparing determined global position at bright-dark-limit with reference position, and calibrating and/or adjusting headlight with determined deviation
JP2011173578A (en) * 2010-01-29 2011-09-08 Imasen Electric Ind Co Ltd Dimming device
CN102298845A (en) * 2011-08-29 2011-12-28 安徽超远信息技术有限公司 One kind of video detection method and system for high beam
CN102567979A (en) * 2012-01-20 2012-07-11 南京航空航天大学 Vehicle-mounted infrared night vision system and multi-source images fusing method thereof

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5877897A (en) * 1993-02-26 1999-03-02 Donnelly Corporation Automatic rearview mirror, vehicle lighting control and vehicle interior monitoring system using a photosensor array

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2911206A1 (en) * 2006-12-21 2008-07-11 Valeo Vision Sa Method for automatic discrimination of bright areas detected by a device for assisting a vehicle driving.
JP2011173578A (en) * 2010-01-29 2011-09-08 Imasen Electric Ind Co Ltd Dimming device
DE102010048689A1 (en) * 2010-10-16 2011-05-12 Daimler Ag Method for adjusting and/or calibrating headlight of motor vehicle, involves comparing determined global position at bright-dark-limit with reference position, and calibrating and/or adjusting headlight with determined deviation
CN102298845A (en) * 2011-08-29 2011-12-28 安徽超远信息技术有限公司 One kind of video detection method and system for high beam
CN102567979A (en) * 2012-01-20 2012-07-11 南京航空航天大学 Vehicle-mounted infrared night vision system and multi-source images fusing method thereof

Also Published As

Publication number Publication date
CN102963294A (en) 2013-03-13

Similar Documents

Publication Publication Date Title
CN102834309B (en) The automatic vehicle equipment to monitor, control and alarm system
JP4544233B2 (en) Vehicle detection device and the headlamp control apparatus
US7566851B2 (en) Headlight, taillight and streetlight detection
JP5761601B2 (en) Object identification device
CN101799987B (en) Self-adaptive intelligent traffic light and control method thereof
CN103931172A (en) Systems and methods for intelligent monitoring of thoroughfares using thermal imaging
JP2007293688A (en) On-vehicle device for detecting front environment of vehicle and lighting equipment for vehicle
JP4853160B2 (en) Vehicle detection device and the headlamp control apparatus
EP1851681A1 (en) Method for identifying the activation of the brake lights of preceding vehicles
CN102782720B (en) Object recognition apparatus, the mobile main control device and the information providing apparatus
JP3759280B2 (en) Road monitoring event detection device
EP2448251A2 (en) Bundling night vision and other driver assistance systems (DAS) using near infra red (NIR) illumination and a rolling shutter
US20100265330A1 (en) Intelligent headlight control using camera sensors
CN101209207A (en) Eyelid detecting apparatus, eyelid detecting method and program thereof
JP2005301518A (en) Signal detection device and signal detecting method
JP2006193068A (en) Vehicular tunnel detector and vehicular light controller
US9459515B2 (en) Adjustable camera mount for a vehicle windshield
JP4743037B2 (en) Vehicle detection device
CN103118900B (en) Headlamp apparatus and method for brightness control
CN1751391A (en) Automatic vehicle exterior light control system
CN101577054A (en) Control method of traffic signal lamp and system
CN101142814A (en) Image processing device and method, program, and recording medium
JP3909691B2 (en) In-vehicle image processing apparatus
JP6117634B2 (en) Lens deposits sensing device, lens deposits detection method, and a vehicle system
CN102835115B (en) The mask for the color image sensor of the camera vehicle

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted
ASS Succession or assignment of patent right

Owner name: ZHEJIANG ANXIE INTELLIGENT TECHNOLOGY CO., LTD.

Free format text: FORMER OWNER: XI AN UNIV. OF SCIENCE AND ENGINEERING

Effective date: 20150227

C41 Transfer of the right of patent application or the patent right
COR Bibliographic change or correction in the description

Free format text: CORRECT: ADDRESS; FROM: 710048 XI AN, SHAANXI PROVINCE TO: 310016 HANGZHOU, ZHEJIANG PROVINCE