CN102963294A - 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
CN102963294A
CN102963294A CN2012104310559A CN201210431055A CN102963294A CN 102963294 A CN102963294 A CN 102963294A CN 2012104310559 A CN2012104310559 A CN 2012104310559A CN 201210431055 A CN201210431055 A CN 201210431055A CN 102963294 A CN102963294 A CN 102963294A
Authority
CN
China
Prior art keywords
vehicle
country beam
light
car light
zone
Prior art date
Legal status (The legal status 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 status listed.)
Granted
Application number
CN2012104310559A
Other languages
Chinese (zh)
Other versions
CN102963294B (en
Inventor
朱虹
潘行杰
王栋
苟荣涛
何振
张云星
杨静
张晓蕾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang harmony Intelligent Technology Co., Ltd.
Original Assignee
Xian University of Technology
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 Xian University of Technology filed Critical Xian University of Technology
Priority to CN201210431055.9A 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
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention discloses a method for judging opening and closing states of a high beam of a vehicle driving at night. The method comprises the following steps of shooting up a lamp polarizing and filtering image of the driving vehicle; extracting the lamp area from the polarizing and filtering image processed in the previous step; according to the mounting position of a camera, determining an effective monitoring area of a road; carrying out opening operation processing on the extracted and processed lamp image, and completing labeling; determining the lamps of the same vehicle; judging whether the lamps of the same vehicle are in a high beam state, wherein, when a condition 1 and a condition 2 are simultaneously met, the high beam of the vehicle is in the opening state, or else, the high beam of the vehicle is in the closing state; and repeating the previous steps, until completing all judgments of all vehicles. The method is suitable for the illumination environments of different road lamps at night, the non-light area is inhibited through the polarizing and filtering method, and the vehicle of which the high beam is opened can be accurately determined by subsequently extracting the features of the high beam.

Description

A kind of method of discrimination of night running vehicle country beam opening
Technical field
The invention belongs to the intelligent traffic monitoring technical field, relate to a kind of method of discrimination of night running vehicle country beam opening.
Background technology
Video monitoring mode is in making up intelligent transportation system, and a kind of means as obtaining and analyze traffic are widely used at present.Vehicle for night running, opening car light, to illuminate the road surface be the basic of safety traffic, yet, the abuse of country beam, the other side vehicle driver does not see road surface, the place ahead fully and gets into an accident easily in the time of can causing meeting, therefore, the use of restriction country beam is a necessary means that guarantees safety traffic.
Adopt video frame images, analyze driving vehicle and whether open country beam, a kind of effective method of can yet be regarded as.Yet, because in the vehicle at night driving process, surrounding environment is very different, and for example, the street lamp in some highway section compares comparatively dense, light is very bright, some highway section street lamp is more sparse, and light is very dark, at this moment, the form of the headlight for vehicle light modulation that photographs has larger difference, is difficult to identify accurately.
Summary of the invention
The purpose of this invention is to provide a kind of method of discrimination of night running vehicle country beam opening, solved prior art and be subjected to surrounding environment influence, cause the form of the headlight for vehicle light modulation that photographs larger difference to be arranged, the problem that is difficult to accurately identify.
The technical solution adopted in the present invention is, a kind of method of discrimination of night running vehicle country beam opening is implemented according to following steps:
Step 1, the car light polarisation filtering image of taking pictures and obtaining driving vehicle
At first, horizontal polarization sheet and vertical polarization sheet are set before the camera lens of monitoring camera, by regulating two angles between the polaroid, the light-inletting quantity of control camera, until after the picture of catching only keeps the light of car light part of driving vehicle, good these two polaroids of fixed adjustment are realized the polarisation filtering of driving vehicle is taken pictures;
Extract the car light zone in step 2, the polarisation filtering image processed in the upper step
2.1) removal steering indicating light zone
To the current time that reads in, namely t three components of red, green, blue through the filtered frame of video coloured image of polarisation constantly are [r t(x, y)] M * n, [g t(x, y)] M * n, [b t(x, y)] M * n, wherein, m * n represents that the size of two field picture is the capable n row of m, the coordinate of (x, y) expression pixel, and [] represents whole image;
If the steering indicating light judgment matrix is [L t(x, y)] M * n, the computing formula of then judging the steering indicating light zone is as follows:
Figure BDA00002348261500021
Wherein, x=1,2 ..., m, y=1,2 ..., n, δ are adjustment amount, Th rIt is the red component threshold value of differentiating vehicle lamp light source;
2.2) the distance-light car light zone of extracting all vehicles in the image
Brightness value is made as [f t(x, y)] M * n, computing formula is as follows:
f t(x,y)=0.299·r t(x,y)+0.587·g t(x,y)+0.114·b t(x,y), (2)
This brightness value is carried out binary conversion treatment, extract distance-light car light zone, i.e. binary image [g t(x, y)] M * nComputing formula is:
Wherein, the range of choice of threshold value Th is that [ε+Δ ε, ξ-Δ ξ, ε are the brightness in darker non-car light zone to Th ∈, and Δ ε is an increment, and ξ is the brightness in brighter car light zone, and Δ ξ is an increment;
Step 3, according to the riding position of camera, delimit the effective monitoring zone of road
Set effective guarded region, only the connected domain that falls into the effective monitoring area part analyzed, namely carry out following processing:
Figure BDA00002348261500031
Wherein, Ω is the effective monitoring zone of delimiting;
Step 4, the image after step 3 processed carry out opening operation and process
4.1) the upper step was obtained
Figure BDA00002348261500032
Carry out opening operation and process, during opening operation, the structural element of selection is N * N, and N is odd number, and initial point is the structural element center, and it is right to establish
Figure BDA00002348261500033
The as a result figure that carries out after opening operation is processed is
Figure BDA00002348261500034
4.2) split the as a result figure after the calculation process
Figure BDA00002348261500035
Carry out connected domain and merge processing
For when the vehicle country beam is opened, same zone is merged into the passing light zone in the country beam zone that light simultaneously the same side, it is as follows that connected domain merges the computing formula of processing:
Figure BDA00002348261500036
Wherein, the distance that [a, b] merges for connected domain, a<0, b>0 is integer;
4.3) to the as a result figure after the processing of connected domain union operation The processing of labelling;
Step 5, determine the car light of same vehicle
According to from top to bottom, order is from left to right at first selected the connected domain in left side, asks its barycenter (xc, yc), and the computing formula of barycenter is as follows:
x c = 1 N Ω k · Σ ( x , y ) ∈ Ω k x , - - - ( 6 )
y c = 1 N Ω k · Σ ( x , y ) ∈ Ω k y , - - - ( 7 )
Wherein, Ω kBe the connected domain that finds;
Afterwards, according to the position (x of barycenter c, y c) draw a horizontal linear, namely at x=x cOn judge whether the connected domain of passing through,
If there is one, then regard it as the car light of same vehicle;
If several are arranged, then with the car light that is considered as its same vehicle of the leftmost side;
Step 6, judge whether the car light of same vehicle is the country beam state
To being judged as a pair of car light zone that belongs to same vehicle, be made as Ω LkAnd Ω Rk, judge area features and the shape facility in its UNICOM territory, left side lamp area S LWith starboard light area S RThe feature calculation formula be respectively:
S L = Σ ( x , y ) ∈ Ω Lk g ‾ ( x , y ) - - - ( 8 )
S R = Σ ( x , y ) ∈ Ω Rk g ‾ ( x , y ) - - - ( 9 )
Shape facility is the aspect ratio of boundary rectangle, finds respectively Ω LkAnd Ω RkThe leftmost side, the rightmost side, top side, four points of lower side, these four points are consisted of boundary rectangle as the left side, the right, top, the following starting point of boundary rectangle, be made as ReC LkAnd ReC Rk, its length and width is respectively L Lk, L Rk, H Lk, H Rk, the computing formula of shape facility is as follows:
ρ Lk = H Lk L Lk - - - ( 10 )
ρ Rk = H Rk L Rk - - - ( 11 )
Calculating country beam decision condition 1 is: ρ Lk+ ρ Rk<th ρ, (12)
Wherein, th ρJudgment threshold for country beam;
Calculating country beam decision condition 2 is: S R+ S L>2Th S, (13)
Wherein, th SBe the judgment threshold of country beam, be the average area of all connected domains on the same level position;
When condition 1 and condition 2 satisfy simultaneously, show that the country beam of this vehicle is in opening, otherwise, be judged to be and be in closed condition;
Repeating step 5, step 6 are finished until all vehicles are all judged.
The invention has the beneficial effects as follows, can adapt to different street lamp photoenvironments at night, by the mode of polarisation filtering, suppress non-light zone, extract afterwards the country beam feature, determine exactly the vehicle of opening country beam.
Description of drawings
Fig. 1 is the polarisation filtering image of driving vehicle on the road that photographed;
Fig. 2 is polarisation filter effect scheme drawing, is the image that vehicle has been opened the passing light state;
Fig. 3 is polarisation filter effect scheme drawing, is the image that vehicle has been opened the country beam state;
Fig. 4 is the car light extracted region effect scheme drawing in the inventive method, is the binary image to Fig. 1, and what dotted line connected is same vehicle two side lamps;
Fig. 5 is the car light extracted region effect scheme drawing in the inventive method, is the binary image to Fig. 2;
Fig. 6 is the car light extracted region effect scheme drawing in the inventive method, is the binary image to Fig. 3;
Fig. 7 is the road effective monitoring scope scheme drawing in the inventive method;
Fig. 8 is the effect scheme drawing of labelling in the inventive method, is Fig. 5 is carried out image after the tag processes;
Fig. 9 is the effect scheme drawing of labelling in the inventive method, is Fig. 6 is carried out image after the tag processes.
The specific embodiment
The method of discrimination of night running vehicle country beam opening of the present invention, implement according to following steps:
Step 1, the car light polarisation filtering image of taking pictures and obtaining driving vehicle
At first, horizontal polarization sheet and vertical polarization sheet are set before the camera lens of monitoring camera, by regulating two angles between the polaroid, the light-inletting quantity of control camera, until after the picture of catching only keeps the light of car light part of driving vehicle, good these two polaroids of fixed adjustment are realized the polarisation filtering of driving vehicle is taken pictures.
Such as Fig. 1, Fig. 2, shown in Figure 3, the car light polarisation filtering of a plurality of vehicles of travelling on the road among the embodiment, and the car light filter effect under two states of a vehicle, wherein, Fig. 1 is the polarisation filtering image of a plurality of vehicles of travelling on the road, be steering indicating light in the rectangle frame, Fig. 2 vehicle has been opened the image of passing light state, and Fig. 3 is the image that vehicle has been opened the country beam state.
Extract the car light zone in step 2, the polarisation filtering image processed in the upper step
2.1) removal steering indicating light zone
To the current time that reads in, namely t three components of RGB through the filtered frame of video coloured image of polarisation constantly are [r t(x, y)] M * n, [g t(x, y)] M * n, [b t(x, y)] M * n, wherein, m * n represents that the size of two field picture is the capable n row of m, the coordinate of (x, y) expression pixel, and [] represents whole image;
Since steering indicating light different from long-and-short distant light be, steering indicating light is red, at this, establishing the steering indicating light judgment matrix is [L t(x, y)] M * n, the computing formula of then judging the steering indicating light zone is as follows:
Figure BDA00002348261500061
Wherein, x=1,2 ..., m, y=1,2 ..., n, δ are adjustment amount, according to the place value of erecting equipment, preferred span is [10,40], Th rBe the red component threshold value of differentiating vehicle lamp light source, preferred span is [80,150];
2.2) the distance-light car light zone of extracting all vehicles in the image
Brightness value is made as [f t(x, y)] M * n, computing formula is as follows:
f t(x,y)=0.299·r t(x,y)+0.587·g t(x,y)+0.114·b t(x,y), (2)
This brightness value is carried out binary conversion treatment, extracts distance-light car light zone,
Because the car light part is very large with the luminance difference of other parts, therefore, adopt the method for global threshold to realize at this, i.e. binary image [g t(x, y)] M * nComputing formula is:
Figure BDA00002348261500071
Wherein, the range of choice of threshold value Th is Th ∈ [ε+Δ ε, ξ-Δ ξ], ε is the brightness in darker non-car light zone, Δ ε is an increment, preferred span is Δ ε ∈ [0.5 ε, ε], ξ is the brightness in brighter car light zone, Δ ξ is an increment, preferred span is Δ ξ ∈ [0.2 ξ, 0.8 ξ]; ε, the value of ξ depends on polarisation filtering result.
For polarisation filtering ideal situation comparatively, namely such as the situation of Fig. 1, suggestion preferably the most simply definite threshold be 7L/8, wherein L is the peak of the given brightness of image value scope [0, L] of camera.
Such as Fig. 4, Fig. 5, shown in Figure 6, it is the car light extracted region effect among the embodiment, wherein, Fig. 4 is the car light extracted region images of all vehicles in travelling on Fig. 1 road, wherein be the steering indicating light zone in the rectangle frame, after formula (1) computing, be eliminated, Fig. 5 is the car light zone of Fig. 2, and Fig. 6 is the car light zone of Fig. 3.
Step 3, according to the riding position of camera, delimit the effective monitoring zone of road
Consider that night, may there be illumination on shop or the building road both sides, when binaryzation, can be mistaken for car light, therefore, according to scheme drawing shown in Figure 7, set effective guarded region, only the connected domain that falls into the effective monitoring area part is analyzed, is namely carried out following processing:
Figure BDA00002348261500072
Wherein, the effective monitoring zone of Ω for being delimited by Fig. 7,
As shown in Figure 7, it is the effective monitoring area schematic among the embodiment, wherein, the trapezoidal part of confining is the effective monitoring zone, the border of both sides is borders of the road both sides in the monitoring camera monitoring image, the front and back boundary is in the Vehicle Driving Cycle process, and camera can capture the scope that is enough to effectively judge the car light size.If (rear boundary is too far away, and car light is too little, is taken as easily noise remove, if prezone is too near, then car light is not in picture.)
Step 4, the image after step 3 processed carry out opening operation and process
4.1) the upper step was obtained
Figure BDA00002348261500081
Carry out opening operation and process, in order to eliminate the impact of noise, the opening operation method all is described in detail on existing digital image processing relevant speciality teaching material, no longer describes in detail here.During opening operation, the structural element of selection is N * N, and N is odd number, and initial point is the structural element center, and it is right to establish
Figure BDA00002348261500082
The as a result figure that carries out after opening operation is processed is
Figure BDA00002348261500083
4.2) split the as a result figure after the calculation process
Figure BDA00002348261500084
Carry out connected domain and merge processing
Because the country beam of vehicle and passing light are on the same level position, when country beam is opened, passing light is simultaneously bright, connected domain described here merges to be processed, for when the vehicle country beam is opened, same zone is merged into the passing light zone in the country beam zone that light simultaneously the same side, and it is as follows that connected domain merges the computing formula of processing:
Figure BDA00002348261500085
Wherein, the distance that [a, b] merges for connected domain, a<0, b>0 is integer, and rule of thumb value is advised preferred a ∈ [8 ,-4], b ∈ [4,8];
4.3) to the as a result figure after the processing of connected domain union operation
Figure BDA00002348261500086
The processing of labelling
Labeling method all is described in detail on the specialized teaching material aspect the digital image processing, no longer describes in detail here.After Fig. 3, Fig. 4, Fig. 5 labelled, just can different connected domains be identified, the result of labelling such as Fig. 8, shown in Figure 9, wherein, Fig. 8 carries out image after the tag processes to Fig. 5, Fig. 9 carries out image after the tag processes to Fig. 6, and the light color in the accompanying drawing represents label No. 1, No. 2 labels of dark expression;
Step 5, determine the car light of same vehicle
On the vehicle structure topological design, car light is arranged at sustained height position, headstock both sides, so according to from top to bottom, order is from left to right at first selected the connected domain in left side, asks its barycenter (x c, y c), the computing formula of barycenter is as follows:
x c = 1 N Ω k · Σ ( x , y ) ∈ Ω k x , - - - ( 6 )
y c = 1 N Ω k · Σ ( x , y ) ∈ Ω k y , - - - ( 7 )
Wherein, Ω kBe the connected domain that finds;
Afterwards, according to the position (x of barycenter c, y c) draw a horizontal linear, namely at x=x cOn judge whether the connected domain of passing through, if there is one, then regard it as the car light of same vehicle; If several are arranged, then with the car light that is considered as its same vehicle of the leftmost side;
As shown in Figure 4, be that the same vehicle car light among the embodiment detects scheme drawing, the straight line that dotted line wherein provides, connection be exactly according to formula (6), after (7) calculate centroid position, two side lights of the same vehicle that records.
Step 6, judge whether the car light of same vehicle is the country beam state
To being judged as a pair of car light zone that belongs to same vehicle, be made as Ω LkAnd Ω Rk, judge area features and the shape facility in its UNICOM territory, left side lamp area S LWith starboard light area S RThe feature calculation formula be respectively:
S L = Σ ( x , y ) ∈ Ω Lk g ‾ ( x , y ) - - - ( 8 )
S R = Σ ( x , y ) ∈ Ω Rk g ‾ ( x , y ) - - - ( 9 )
Shape facility is the aspect ratio of boundary rectangle, finds respectively Ω LkAnd Ω RkThe leftmost side, the rightmost side, top side, four points of lower side, these four points are consisted of boundary rectangle as the left side, the right, top, the following starting point of boundary rectangle, be made as ReC LkAnd ReC Rk, its length and width is respectively L Lk, L Rk, H Lk, H Rk, the computing formula of shape facility is as follows:
ρ Lk = H Lk L Lk - - - ( 10 )
ρ Rk = H Rk L Rk - - - ( 11 )
Calculating country beam decision condition 1 is: ρ Lk+ ρ Rk<th ρ, (12)
Wherein, th ρBe the judgment threshold of country beam, when country beam is opened, because two lamps of a side are all lighted, be empirical value, so suggestion th ρPreferred span be th ρ∈ [1,1.6];
Calculating country beam decision condition 2 is: S R+ S L>2Th S, (13)
Wherein, th SBe the judgment threshold of country beam, be the average area of all connected domains on the same level position;
When condition 1 and condition 2 satisfy simultaneously, show that the country beam of this vehicle is in opening, otherwise, be judged to be and be in closed condition.
Repeating step 5, step 6 are finished until all vehicles are all judged.
Method of the present invention is by above step, just in the time of can be in driving vehicle enters the effective monitoring zone, judge whether its country beam is opened, if belong to the country beam opening, recycling motion target tracking technology, calculate again it continues to open country beam in guarded region time, for intelligent transportation system provides a kind of safing effective method of inspection.

Claims (6)

1. the method for discrimination of a night running vehicle country beam opening is characterized in that, implements according to following steps:
Step 1, the car light polarisation filtering image of taking pictures and obtaining driving vehicle
At first, horizontal polarization sheet and vertical polarization sheet are set before the camera lens of monitoring camera, by regulating two angles between the polaroid, the light-inletting quantity of control camera, until after the picture of catching only keeps the light of car light part of driving vehicle, good these two polaroids of fixed adjustment are realized the polarisation filtering of driving vehicle is taken pictures;
Extract the car light zone in step 2, the polarisation filtering image processed in the upper step
2.1) removal steering indicating light zone
To the current time that reads in, namely t three components of RGB through the filtered frame of video coloured image of polarisation constantly are [r t(x, y)] M * n, [g t(x, y)] M * n, [b t(x, y)] M * n, wherein, m * n represents that the size of two field picture is the capable n row of m, the coordinate of (x, y) expression pixel, and [] represents whole image;
If the steering indicating light judgment matrix is [L t(x, y)] M * n, the computing formula of then judging the steering indicating light zone is as follows:
Figure FDA00002348261400011
Wherein, x=1,2 ..., m, y=1,2 ..., n, δ are adjustment amount, Th rIt is the red component threshold value of differentiating vehicle lamp light source;
2.2) the distance-light car light zone of extracting all vehicles in the image
Brightness value is made as [f t(x, y)] M * n, computing formula is as follows:
f t(x,y)=0.299·r t(x,y)+0.587·g t(x,y)+0.114·b t(x,y), (2)
This brightness value is carried out binary conversion treatment, extract distance-light car light zone, i.e. binary image [g t(x, y)] M * nComputing formula is:
Figure FDA00002348261400021
Wherein, the range of choice of threshold value Th is Th ∈ [ε+Δ ε, ξ-Δ ξ], and ε is the brightness in darker non-car light zone, and Δ ε is an increment, and ξ is the brightness in brighter car light zone, and Δ ξ is an increment;
Step 3, according to the riding position of camera, delimit the effective monitoring zone of road
Set effective guarded region, only the connected domain that falls into the effective monitoring area part analyzed, namely carry out following processing:
Figure FDA00002348261400022
Wherein, Ω is the effective monitoring zone of delimiting;
Step 4, the image after step 3 processed carry out opening operation and process
4.1) the upper step was obtained
Figure FDA00002348261400023
Carry out opening operation and process, during opening operation, the structural element of selection is N * N, and N is odd number, and initial point is the structural element center, and it is right to establish
Figure FDA00002348261400024
The as a result figure that carries out after opening operation is processed is
4.2) split the as a result figure after the calculation process
Figure FDA00002348261400026
Carry out connected domain and merge processing
For when the vehicle country beam is opened, same zone is merged into the passing light zone in the country beam zone that light simultaneously the same side, it is as follows that connected domain merges the computing formula of processing:
Wherein, the distance that [a, b] merges for connected domain, a<0, b>0 is integer;
4.3) to the as a result figure after the processing of connected domain union operation
Figure FDA00002348261400028
The processing of labelling;
Step 5, determine the car light of same vehicle
According to from top to bottom, order is from left to right at first selected the connected domain in left side, asks its barycenter (xc, yc), and the computing formula of barycenter is as follows:
x c = 1 N Ω k · Σ ( x , y ) ∈ Ω k x , - - - ( 6 )
y c = 1 N Ω k · Σ ( x , y ) ∈ Ω k y , - - - ( 7 )
Wherein, Ω kBe the connected domain that finds;
Afterwards, according to the position (x of barycenter c, y c) draw a horizontal linear, namely at x=x cOn judge whether the connected domain of passing through,
If there is one, then regard it as the car light of same vehicle;
If several are arranged, then with the car light that is considered as its same vehicle of the leftmost side;
Step 6, judge whether the car light of same vehicle is the country beam state
To being judged as a pair of car light zone that belongs to same vehicle, be made as Ω LkAnd Ω Rk, judge area features and the shape facility in its UNICOM territory, left side lamp area S LWith starboard light area S RThe feature calculation formula be respectively:
S L = Σ ( x , y ) ∈ Ω Lk g ‾ ( x , y ) - - - ( 8 )
S R = Σ ( x , y ) ∈ Ω Rk g ‾ ( x , y ) - - - ( 9 )
Shape facility is the aspect ratio of boundary rectangle, finds respectively Ω LkAnd Ω RkThe leftmost side, the rightmost side, top side, four points of lower side, these four points are consisted of boundary rectangle as the left side, the right, top, the following starting point of boundary rectangle, be made as ReC LkAnd ReC Rk, its length and width is respectively L Lk, L Rk, H Lk, H Rk, the computing formula of shape facility is as follows:
ρ Lk = H Lk L Lk - - - ( 10 )
ρ Rk = H Rk L Rk - - - ( 11 )
Calculating country beam decision condition 1 is: ρ Lk+ ρ Rk<th ρ, (12)
Wherein, th ρJudgment threshold for country beam;
Calculating country beam decision condition 2 is: S R+ S L>2Th S, (13)
Wherein, th SBe the judgment threshold of country beam, be the average area of all connected domains on the same level position;
When condition 1 and condition 2 satisfy simultaneously, show that the country beam of this vehicle is in opening, otherwise, be judged to be and be in closed condition;
Repeating step 5, step 6 are finished until all vehicles are all judged.
2. the method for discrimination of night running vehicle country beam opening according to claim 1, it is characterized in that: described step 2.1), the preferred span of adjustment amount δ is [10,40], red component threshold value Th rPreferable range be [80,150].
3. the method for discrimination of night running vehicle country beam opening according to claim 1 is characterized in that: described step 2.2),
Generally speaking, the preferred span of increment Delta ε is Δ ε ∈ [0.5 ε, ε], and the preferred span of increment Delta ξ is Δ ξ ∈ [0.2 ξ, 0.8 ξ],
For polarisation filtering ideal situation comparatively, preferred threshold value Th is 7L/8, and wherein L is the peak of the given brightness of image value scope [0, L] of camera.
4. the method for discrimination of night running vehicle country beam opening according to claim 1 is characterized in that: described step 4.2), and preferred a ∈ [8 ,-4], b ∈ [4,8].
5. the method for discrimination of night running vehicle country beam opening according to claim 1 is characterized in that: in the described step 6, and the judgment threshold th of country beam ρPreferred span be th ρ∈ [1,1.6].
6. according to claim 1 to the method for discrimination of 5 arbitrary described night running vehicle country beam openings, it is characterized in that: after the described step 6, recycling motion target tracking technology, the country beam that calculates this vehicle continues the time of unlatching in guarded region.
CN201210431055.9A 2012-11-02 2012-11-02 Method for judging opening and closing states of high beam of vehicle driving at night Active CN102963294B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210431055.9A 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
CN201210431055.9A 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 true CN102963294A (en) 2013-03-13
CN102963294B CN102963294B (en) 2014-12-10

Family

ID=47793824

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210431055.9A Active 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)

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103295399A (en) * 2013-05-14 2013-09-11 西安理工大学 On-state judging method of headlights on full beam of night-driving cars based on morphological characteristics
CN103870809A (en) * 2014-02-27 2014-06-18 奇瑞汽车股份有限公司 Vehicle detection method and device
CN105575128A (en) * 2016-01-20 2016-05-11 王威立 High beam role breaking detection and snapshot system for motor vehicles
CN106128096A (en) * 2016-06-29 2016-11-16 安徽金赛弗信息技术有限公司 A kind of motor vehicles is not opened steering indicating light by regulation and is broken rules and regulations lane change intelligent identification Method and device thereof
CN106671881A (en) * 2017-03-07 2017-05-17 苏州大学 Automobile monitoring system
CN106778534A (en) * 2016-11-28 2017-05-31 江苏鸿信系统集成有限公司 Surrounding environment method for recognition of lamplight in a kind of vehicle traveling
CN107169422A (en) * 2017-04-21 2017-09-15 浙江安谐智能科技有限公司 The method of discrimination of high beam open and-shut mode based on headlamp radiation direction
CN108230690A (en) * 2018-02-09 2018-06-29 浙江安谐智能科技有限公司 A kind of high beam based on convolutional neural networks continues the method for discrimination of opening
CN109263538A (en) * 2018-08-02 2019-01-25 惠州市德赛西威汽车电子股份有限公司 It is a kind of can intelligent recognition close high beam preceding viewing system and control method
CN109782478A (en) * 2019-03-21 2019-05-21 南京汉旗新材料股份有限公司 A kind of TN/STN polarizing film form and aspect regulation method
CN110246337A (en) * 2018-08-10 2019-09-17 浙江大华技术股份有限公司 A kind of method, apparatus and computer storage medium detecting Misuse car light
CN110450706A (en) * 2019-08-22 2019-11-15 哈尔滨工业大学 A kind of adaptive distance light lamp control system and image processing algorithm
CN110646173A (en) * 2019-09-30 2020-01-03 青岛讯极科技有限公司 Detection method for continuous opening of high beam
CN110738158A (en) * 2019-10-11 2020-01-31 奇点汽车研发中心有限公司 Vehicle light control method and device, electronic equipment and storage medium
CN110856324A (en) * 2018-07-27 2020-02-28 东旭科技集团有限公司 Method and device for controlling street lamp, street lamp and readable storage medium
CN113246846A (en) * 2021-06-07 2021-08-13 奇瑞新能源汽车股份有限公司 Vehicle light control method and device and vehicle
CN113611111A (en) * 2021-07-29 2021-11-05 郑州高识智能科技有限公司 Vehicle distance calculation method based on vehicle high beam
CN116582980A (en) * 2023-07-06 2023-08-11 宏泰智能科技(东莞)有限公司 LED car lamp control system and method based on Internet of things

Citations (6)

* 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 LIGHT AREAS DETECTED BY A DRIVING DEVICE FOR VEHICLE.
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 Far-beam light video detection method and system
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
US20120257059A1 (en) * 1993-02-26 2012-10-11 Donnelly Corporation Vehicular vision system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120257059A1 (en) * 1993-02-26 2012-10-11 Donnelly Corporation Vehicular vision system
FR2911206A1 (en) * 2006-12-21 2008-07-11 Valeo Vision Sa METHOD FOR AUTOMATIC DISCRIMINATION OF LIGHT AREAS DETECTED BY A DRIVING DEVICE FOR VEHICLE.
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 Far-beam light video detection method and system
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

Cited By (29)

* 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
CN103295399A (en) * 2013-05-14 2013-09-11 西安理工大学 On-state judging method of headlights on full beam of night-driving cars based on morphological characteristics
CN103870809A (en) * 2014-02-27 2014-06-18 奇瑞汽车股份有限公司 Vehicle detection method and device
CN103870809B (en) * 2014-02-27 2017-06-16 奇瑞汽车股份有限公司 The detection method and device of vehicle
CN105575128A (en) * 2016-01-20 2016-05-11 王威立 High beam role breaking detection and snapshot system for motor vehicles
CN105575128B (en) * 2016-01-20 2018-03-13 王威立 Motor vehicle far-reaching lamp peccancy detection and capturing system
CN106128096B (en) * 2016-06-29 2018-10-26 安徽金赛弗信息技术有限公司 A kind of motor vehicle does not open steering indicating light by regulation and breaks rules and regulations lane change intelligent identification Method and its device
CN106128096A (en) * 2016-06-29 2016-11-16 安徽金赛弗信息技术有限公司 A kind of motor vehicles is not opened steering indicating light by regulation and is broken rules and regulations lane change intelligent identification Method and device thereof
CN106778534A (en) * 2016-11-28 2017-05-31 江苏鸿信系统集成有限公司 Surrounding environment method for recognition of lamplight in a kind of vehicle traveling
CN106671881A (en) * 2017-03-07 2017-05-17 苏州大学 Automobile monitoring system
CN107169422A (en) * 2017-04-21 2017-09-15 浙江安谐智能科技有限公司 The method of discrimination of high beam open and-shut mode based on headlamp radiation direction
CN107169422B (en) * 2017-04-21 2020-11-13 浙江安谐智能科技有限公司 Method for judging opening and closing states of high beam based on light direction of headlamp
CN108230690A (en) * 2018-02-09 2018-06-29 浙江安谐智能科技有限公司 A kind of high beam based on convolutional neural networks continues the method for discrimination of opening
CN108230690B (en) * 2018-02-09 2020-08-28 浙江安谐智能科技有限公司 Method for judging continuous opening state of high beam based on convolutional neural network
CN110856324B (en) * 2018-07-27 2021-10-15 东旭光电科技股份有限公司 Method and device for controlling street lamp, street lamp and readable storage medium
CN110856324A (en) * 2018-07-27 2020-02-28 东旭科技集团有限公司 Method and device for controlling street lamp, street lamp and readable storage medium
CN109263538A (en) * 2018-08-02 2019-01-25 惠州市德赛西威汽车电子股份有限公司 It is a kind of can intelligent recognition close high beam preceding viewing system and control method
CN110246337A (en) * 2018-08-10 2019-09-17 浙江大华技术股份有限公司 A kind of method, apparatus and computer storage medium detecting Misuse car light
CN109782478A (en) * 2019-03-21 2019-05-21 南京汉旗新材料股份有限公司 A kind of TN/STN polarizing film form and aspect regulation method
CN110450706A (en) * 2019-08-22 2019-11-15 哈尔滨工业大学 A kind of adaptive distance light lamp control system and image processing algorithm
CN110450706B (en) * 2019-08-22 2022-03-08 哈尔滨工业大学 Self-adaptive high beam control system and image processing algorithm
CN110646173A (en) * 2019-09-30 2020-01-03 青岛讯极科技有限公司 Detection method for continuous opening of high beam
CN110738158A (en) * 2019-10-11 2020-01-31 奇点汽车研发中心有限公司 Vehicle light control method and device, electronic equipment and storage medium
CN113246846A (en) * 2021-06-07 2021-08-13 奇瑞新能源汽车股份有限公司 Vehicle light control method and device and vehicle
CN113246846B (en) * 2021-06-07 2022-06-03 奇瑞新能源汽车股份有限公司 Vehicle light control method and device and vehicle
CN113611111A (en) * 2021-07-29 2021-11-05 郑州高识智能科技有限公司 Vehicle distance calculation method based on vehicle high beam
CN113611111B (en) * 2021-07-29 2023-09-08 郑州高识智能科技有限公司 Vehicle distance calculation method based on vehicle high beam
CN116582980A (en) * 2023-07-06 2023-08-11 宏泰智能科技(东莞)有限公司 LED car lamp control system and method based on Internet of things
CN116582980B (en) * 2023-07-06 2023-12-08 江苏省德懿翔宇光电科技有限公司 LED car lamp control system and method based on Internet of things

Also Published As

Publication number Publication date
CN102963294B (en) 2014-12-10

Similar Documents

Publication Publication Date Title
CN102963294B (en) Method for judging opening and closing states of high beam of vehicle driving at night
O'Malley et al. Vehicle detection at night based on tail-light detection
CN105981042B (en) Vehicle detection system and method
US9704060B2 (en) Method for detecting traffic violation
CN108357418B (en) Preceding vehicle driving intention analysis method based on tail lamp identification
CN104508727B (en) Three-dimensional body detection device and three-dimensional body detection method
CN104021378B (en) Traffic lights real-time identification method based on space time correlation Yu priori
TWI302879B (en) Real-time nighttime vehicle detection and recognition system based on computer vision
CN109299674B (en) Tunnel illegal lane change detection method based on car lamp
CN103295399B (en) On-state judging method of headlights on full beam of night-driving cars based on morphological characteristics
Li et al. Nighttime lane markings recognition based on Canny detection and Hough transform
JP6723328B2 (en) Vehicle detection method, night-time vehicle detection method and system based on dynamic light intensity
CN104050450A (en) Vehicle license plate recognition method based on video
CN102556021B (en) Control device for preventing cars from running red light
JP2012155612A (en) Lane detection apparatus
CN101872546A (en) Video-based method for rapidly detecting transit vehicles
CN107891808A (en) Driving based reminding method, device and vehicle
CN104978746B (en) A kind of body color recognition methods of driving vehicle
CN102496281A (en) Vehicle red-light violation detection method based on combination of tracking and virtual loop
CN110450706A (en) A kind of adaptive distance light lamp control system and image processing algorithm
CN103050008B (en) Method for detecting vehicles in night complex traffic videos
Lin et al. Adaptive IPM-based lane filtering for night forward vehicle detection
CN103020930A (en) Nighttime monitoring video enhancing method
CN111649918A (en) Method for monitoring turning on of high beam and continuous tracking of vehicle in real time
CN107169422A (en) The method of discrimination of high beam open and-shut mode based on headlamp radiation direction

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
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 patent application or patent right or utility model
COR Change of bibliographic data

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

TR01 Transfer of patent right

Effective date of registration: 20150227

Address after: Hangzhou City, Zhejiang province 310016 Jianggan District Hangzhou sea 78-8 China branch building (Taihe International) room 1101

Patentee after: Zhejiang harmony Intelligent Technology Co., Ltd.

Address before: 710048 Shaanxi city of Xi'an Province Jinhua Road No. 5

Patentee before: Xi'an University of Technology