CN103268471A - Vehicle illegal land occupying detection method and device - Google Patents
Vehicle illegal land occupying detection method and device Download PDFInfo
- Publication number
- CN103268471A CN103268471A CN2013101332137A CN201310133213A CN103268471A CN 103268471 A CN103268471 A CN 103268471A CN 2013101332137 A CN2013101332137 A CN 2013101332137A CN 201310133213 A CN201310133213 A CN 201310133213A CN 103268471 A CN103268471 A CN 103268471A
- Authority
- CN
- China
- Prior art keywords
- car plate
- track
- color
- video
- image
- 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
Links
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention is applicable to the technical field of intelligent transportation and provides a vehicle illegal land occupying detection method and device. The vehicle illegal land occupying detection method comprises the following steps: extracting an image containing a lane in a video frame; judging whether the lane in the image is a lane with a pre-set color or not; when the lane is the lane with the pre-set color, detecting a license plate in the video frame; when detecting the license plate, judging the color of the license plate and judging whether the color of the license plate is a limited color or not; when the color of the license plate is the limited color, detecting whether the license plate is in the lane with the pre-set color; and when the license plate is in the lane with the pre-set color, marking the license plate in the video frame. According to the embodiment of the invention, when the license plate is in the lane with the pre-set color, the license plate is marked in the video frame, so that a relative department can dispose social vehicles which occupy a bus lane or other special lanes according to the marked license plate; and therefore the operation efficiency and the safety of public transportation vehicles or other vehicles are improved.
Description
Technical field
The invention belongs to the intelligent transport technology field, relate in particular to the illegal road occupying detection method of a kind of vehicle and device.
Background technology
Perfect day by day along with public transportation system, government has set up bus zone, bus zone is as the dedicated Lanes of bus, guaranteed unimpeded the travelling of public transit vehicle energy, effectively having improved public bus network blocks up, improved the passenger traffic efficient in city, for the common people provide a facility, outside environment efficiently.
Yet, because prior art can't detect the public vehicles of swarming into or take bus zone, when the part public vehicles do not follow that traffic rules are swarmed into or when taking bus zone, can reduce the efficient of public transit vehicle operation, threaten the safety of public transit vehicle traffic, therefore swarm into or take the potential safety hazard that bus zone brings for public vehicles and can not be ignored.
Summary of the invention
The purpose of the embodiment of the invention is to provide a kind of vehicle illegal road occupying detection method, be intended to solve prior art and can't detect the public vehicles that take bus zone or other dedicated Lanes, cause operational efficiency and the low problem of security of public transit vehicle or other vehicle.
The embodiment of the invention is achieved in that the illegal road occupying detection method of a kind of vehicle, comprising:
Extract the image that comprises the track in the frame of video;
Judge whether the track is default color track in the described image;
When described track is when presetting the color track, to detect car plate in described frame of video;
When detecting car plate, described car plate is carried out color differentiate, judge whether the color of described car plate is to limit color;
When the color of described car plate is the restriction color, detect described car plate whether to be in the described default color track;
In the time of in described car plate is in described default color track, the described car plate of mark in described frame of video.
Another purpose of the embodiment of the invention is to provide a kind of vehicle illegal road occupying pick-up unit, comprising:
Extraction unit is used for extracting the image that frame of video comprises the track;
First judging unit is used for judging whether described image track is default color track;
First detecting unit, be used for when described track when presetting the color track, in described frame of video, detect car plate;
Second judging unit is used for when detecting car plate, described car plate is carried out color differentiate, and judges whether the color of described car plate is to limit color;
Second detecting unit, whether the color that is used for when described car plate is when limiting color, to detect described car plate to be in the described default color track;
Indexing unit, when being used in described car plate is in described default color track, the described car plate of mark in described frame of video.
In embodiments of the present invention, when detecting car plate, described car plate is carried out color differentiate, judge whether the color of described car plate is to limit color; When the color of described car plate is the restriction color, detect described car plate whether to be in the described default color track; In the time of in described car plate is in described default color track, the described car plate of mark in described frame of video.So that relevant department can be according to the car plate of mark, disposal takies the public vehicles of bus zone or other dedicated Lanes, thereby solved and to have detected the public vehicles that take bus zone or other dedicated Lanes, the operational efficiency and the low problem of security that cause public transit vehicle or other vehicle, the operational efficiency and the security that have improved public transit vehicle or other vehicle.
Description of drawings
Fig. 1 is the realization flow figure of the illegal road occupying detection method of vehicle that provides of the embodiment of the invention;
Fig. 2 is the illegal road occupying detection method of the vehicle step S102 specific implementation process flow diagram that the embodiment of the invention provides;
Fig. 3 is that the judgement straight line that the embodiment of the invention provides is the preferable sample figure in yellow track;
Fig. 4 is the illegal road occupying detection method of the vehicle step S103 specific implementation process flow diagram that the embodiment of the invention provides;
Fig. 5 is the preferable sample figure that effective coverage that the embodiment of the invention provides is divided into a plurality of zones;
Fig. 6 be the embodiment of the invention provide judge whether public vehicles are swarmed into, take the preferable sample figure of bus zone or other dedicated Lanes, wherein Fig. 6 A represents that public vehicles swarm into bus zone or other dedicated Lanes, and Fig. 6 B represents that public vehicles take bus zone or other dedicated Lanes;
Fig. 7 is the preferable implementing procedure figure in actual applications that the embodiment of the invention provides;
Fig. 8 is the structured flowchart of the illegal road occupying pick-up unit of vehicle that provides of the embodiment of the invention.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explaining the present invention, and be not used in restriction the present invention.
In embodiments of the present invention, when detecting car plate, described car plate is carried out color differentiate, judge whether the color of described car plate is to limit color; When the color of described car plate is the restriction color, detect described car plate whether to be in the described default color track; In the time of in described car plate is in described default color track, the described car plate of mark in described frame of video.So that relevant department can be according to the car plate of mark, disposal takies the public vehicles of bus zone or other dedicated Lanes, thereby solved and to have detected the public vehicles that take bus zone or other dedicated Lanes, the operational efficiency and the low problem of security that cause public transit vehicle or other vehicle, the operational efficiency and the security that have improved public transit vehicle or other vehicle.
Fig. 1 shows the realization flow of the illegal road occupying detection method of a kind of vehicle that the embodiment of the invention provides, and details are as follows:
In step S101, extract the image that comprises the track in the frame of video;
In the present embodiment, by being installed in the video acquisition device of roadside or vehicle mirrors, image is gathered, to obtain the image that comprises the track in real time.
In the present embodiment, extract the image that comprises the track in the image, specifically can filter image, obtain comprising the effective coverage in track after the filtration, again the effective coverage is narrowed down to a certain size, obtain image to be detected, image to be detected is converted into gray-scale map, and carry out Gauss and smoothly carry out rim detection with the default operator that detects of employing, obtain comprising the image in track.The default factor that detects includes but not limited to gradient operator, Roberts gradient operator, Prewitt operator, Sobel operator, directional operator, Laplacian operator, Ma Er operator, Canny edge detection operator.
Preferably, adopt the Canny edge detection operator in the present embodiment.
In step S102, judge whether the track is default color track in the described image;
In the present embodiment, default color track can be established certainly for the user, also can be system default.
In the present embodiment, default color track is yellow track, one or more in blue track, black track, the white track.Preferably, because domestic bus zone adopts yellow, therefore default color track can be yellow track.
As the preferred embodiments of the present invention, Fig. 2 shows the concrete implementing procedure of step S102, and details are as follows:
In step S201, obtain the central point of described image, generate zone, an image left side and the right zone of image according to divide described image by the straight line of described central point.
In the present embodiment, the image that will comprise the track is divided into two parts, particularly, can draw central point according to pixel coordinate by the pixel coordinate of image, according to the straight line by central point image is divided into left and right sides two parts.
In step S202, obtain the straight line of weight maximum in the right zone of image zone, a left side and image.
In the present embodiment, obtain the straight line of weight maximum in image left side zone and the zone, the image right side by preset algorithm, preset algorithm includes but not limited to the conversion of Hough straight line, thereby tries to achieve the straight line of weight maximum.
In the present embodiment, the straight line of weight maximum is illustrated in crossing that the maximum point of hough space cathetus.
In the present embodiment, pixel being gone up in image left side zone is not that 0 point is done the conversion of Hough straight line, the straight line in the corresponding hough space of a point in the image, a point in the corresponding hough space of the straight line in the image.Because a point in the corresponding hough space of each bar straight line in the zone, an image left side, so the corresponding hough space cathetus of straight line of weight maximum intersects that maximum points in the zone, an image left side.In like manner, because a point in the corresponding hough space of each bar straight line in the right zone of image, so the corresponding hough space cathetus of straight line of weight maximum intersects that maximum points in the right zone of image.
In step S203, extend predetermined width left, to the right in the horizontal direction of described straight line, generate the quadrilateral track.
In the present embodiment, the quadrilateral track is the zone that comprises the straight line of weight maximum in the frame of video, and predetermined width can be established certainly for the user, also can be system default.Can extend identical predetermined width left, to the right in the horizontal direction of straight line, also can extend different predetermined width left, to the right in the horizontal direction of straight line.For example, can extend the predetermined width of 50 pixels in the horizontal direction of straight line left, extend the predetermined width of 50 pixels to the right in the horizontal direction of straight line, also can extend the predetermined width of 50 pixels in the horizontal direction of straight line left, extend the predetermined width of 80 pixels in the horizontal direction of straight line to the right.
In the present embodiment, by taking all factors into consideration the width in track on the various tracks, and width is added up the generation mean breadth, mean breadth is as predetermined width.
For ease of explanation, be example with the practical application, when the horizontal direction of straight line is extended identical predetermined width left, to the right, and predetermined width adopt 50 pixels apart from the time.The horizontal direction of straight line is extended predetermined width left, to the right, generate the parallelogram track, thereby the track that has met on the track has the actual conditions of certain width, and increased the effective statistical sample of pixel for follow-up statistics, improved and judged that this parallel edges shape track is the accuracy rate in default color track.
In step S204, obtain described in the quadrilateral track quantity of pixel and the number that classifies as default color pixel point.
In the present embodiment, the implementation process of obtaining the quantity of pixel in the frame of video is prior art, and detailed process is not done at this and to be given unnecessary details.
In the present embodiment, classify as the number of default color pixel point, specifically can be by color parameter (the Red Green Blue that obtains pixel in the quadrilateral track, RGB) value, CLUT by RGB, can obtain the color parameter scope of default color, when the color parameter of pixel belongs to the color parameter scope of default color, represent that this pixel classifies as default color pixel point.In local data base, the initial value that classifies as default color pixel point is set to 0, classifies as the number of presetting color pixel point by record, and default color pixel point is added up.One of every acquisition belongs to default color pixel point, and the number that classifies as default color pixel point namely adds 1, counts the number that belongs to default color pixel point in the quadrilateral track thus.
In step S205, generate the quadrilateral track for presetting the probability in color track according to the quantity of pixel in the described number that is categorized as default color pixel point and the quadrilateral track.
In the present embodiment,, generate the quadrilateral track and be the probability in default color track divided by the quantity of pixel in the quadrilateral track according to the described number of default color pixel point that is categorized as,
In step S206, when described probability reached pre-set threshold value, described quadrilateral track was default color track.
In the present embodiment, pre-set threshold value can be established certainly for the user, also can be system default.Preferably, pre-set threshold value adopts 1%.
In the present embodiment, for ease of explanation, be that yellow track is example with default color track, to the straight line that detects, about the certain pixel of each horizontal-extending, generate the quadrilateral track, obtain the pixel in the quadrilateral track, utilize default support vector machine (Support Vectors Machine, SVM), whether the rgb value of judging the pixel in the quadrilateral track meets the pixel rgb value distribution of yellow track, when the pixel that meets yellow bus zone RGB distribution in the quadrilateral track reaches certain proportion, judges that this straight line is yellow track.
For ease of explanation, Fig. 3 shows and judges that straight line is the preferable sample figure in yellow track.
In step S103, when described track is when presetting the color track, to detect car plate in described frame of video.
In the present embodiment, the mode that detects of car plate includes but not limited to that car plate based on adaboost detects, detects based on the car plate of BP neural network.
In the present embodiment, the zone of electing in described frame of video to be detected is with the scanning window of different scale size, this surveyed area is carried out region-wide scanning, and with the Adaboost sorter that imports, calculate the corresponding Haar feature of scanning window, the processing of classifying detects car plate.
As the preferred embodiments of the present invention, Fig. 4 shows the implementing procedure of step S103, and details are as follows:
In step S401, obtain the effective coverage in the described frame of video.
In the present embodiment, obtain the effective coverage in the frame of video, can carry out pre-service to frame of video, particularly, reject frame of video the right and left fringe region, the zone, top and the bottom has kept in the frame of video about 1/2 height zone, thereby has finished obtaining of the effective coverage in the frame of video, and avoided because car plate when far away apart from the camera lens of video capture device, frame of video resolution is lower, and fringe region is fuzzyyer, causes the low situation of the detected car plate degree of reliability in the edge region.
In step S402, described effective coverage is divided into a plurality of zones.
In the present embodiment, being divided into a plurality of zones from top to bottom according to the height in effective zone, preferably, being divided into three zones, is respectively upper area, zone line, lower area.
For ease of explanation, Fig. 5 has shown the preferable sample figure that the effective coverage is divided into a plurality of zones.
In step S403, in described zone, car plate is detected in regular turn by preset order.
In the present embodiment, since in the track of reality, the car that can only travel in the track, and vehicle has the situation of certain length simultaneously, therefore if car plate is detected at zone line, upper area and lower area do not have car plate, if car plate is detected at lower area, in the first area car plate may be arranged, zone line does not have car plate, if car plate is detected at upper area, in the lower area car plate may be arranged, zone line does not have car plate.
In the present embodiment, by preset order car plate is carried out in the zone and detect, preset order includes but not limited to carry out earlier car plate and detects in zone line, when detecting less than car plate in zone line, carries out car plate and detect in upper area or lower area.
In the present embodiment, the mode that detects of car plate includes but not limited to that car plate based on adaboost detects, detects based on the car plate of BP neural network.
In the present embodiment, by by preset order, detect carrying out car plate in the zone line of appointment, reduced the zone of detecting, saved the time of detecting, improved the efficient that car plate detects.
As a preferred embodiment of the present invention, before frame of video is divided into a plurality of zones, also comprise, obtain the resolution of frame of video in the video acquisition device, with the extremely default resolution of described resolution compression.Thereby avoided because the resolution difference of frame of video, caused the size in a plurality of zones of dividing also different, influenced validity and accuracy that follow-up car plate detects.
As a preferred embodiment of the present invention, when detecting described car plate, obtain the zone that in frame of video, comprises car plate, merge described zone and generate license plate area;
In the present embodiment, in frame of video, merge the license plate area of zone to generate that detects car plate, particularly, adopt the scanning window of default a plurality of different scales, scanning comprises the zone of car plate in frame of video, calculates overlapping area in per two zones that comprise car plate according to the coordinate in the zone that comprises car plate, when the ratio that accounts for the total area in two zones that comprise car plate when overlapping area reaches preset ratio, these two zones that comprise car plate are merged, thereby generate license plate area.Avoided only comprising by the scanning of a scanning window zone of car plate, caused occurring not scanning the situation of complete car plate, thereby improved the reliability of the positional information of the follow-up car plate that gets access to.
In the present embodiment, obtain the positional information of car plate in the described license plate area, positional information comprises position coordinates.The positional information of car plate in the license plate area after merging is preserved with the form of overall array.
As the preferred embodiments of the present invention, judge whether described car plate is effective car plate, keeps when described car plate is effective car plate, otherwise rejects.
In the present embodiment, judge whether described car plate is effective car plate, whether the marginal density that can judge license plate area meets greater than predetermined threshold value, particularly, positional information according to car plate navigates to car plate, car plate is carried out gray processing handle, and again the car plate after the gray processing is carried out sharpening and handles, carrying out histogram equalization then strengthens, the default factor that detects of recycling is extracted frame of video vertical edge information, and the default factor that detects includes but not limited to gradient operator, the Roberts gradient operator, the Prewitt operator, the Sobel operator, directional operator, Laplacian operator, the Ma Er operator, Canny edge detection operator etc.Generate the marginal density of car plate again according to Marginal density function.Threshold value can arrange predetermined threshold value according to test video herein, if marginal density greater than this predetermined threshold value, judges that then car plate is effective car plate.Thereby avoided mistake to identify vehicle body as the situation of car plate, improved follow-up positional information according to car plate, judged whether car plate is in the accuracy rate in the described yellow track.
As another embodiment of the present invention, judge whether described car plate is effective car plate, the maximal value of difference that can be by obtaining the RGB average, judge that maximal value is whether in the predetermined threshold value interval, particularly, the positional information according to car plate navigates to car plate, the RGB average of calculating car plate pixel, calculate the three averages absolute value of difference in twos then, choose three maximal values in the absolute difference at last.Threshold interval according to determining carries out threshold decision.Therefore wherein, threshold interval can obtain according to the statistical sample data of ground region, represents its feature in the time of in maximal value is in threshold interval near ground region, and the car plate in this threshold interval not is judged to be effective car plate.Car plate in maximal value is in this threshold interval is judged to be invalid car plate.Thereby avoided mistake to identify near the situation on the ground of car plate, improved follow-up positional information according to car plate, judged whether car plate is in the accuracy rate in the described yellow track.
As a preferred embodiment of the present invention, judge the validity of described car plate, the maximal value of marginal density that can be by obtaining license plate area and the difference of RGB average is carried out threshold decision to the maximal value of the difference of marginal density and RGB average.When marginal density greater than default marginal density threshold value, and when the maximal value of the difference of RGB average does not belong to the preset threshold value interval, judge that then car plate is effective car plate, thereby avoided identifying near the ground of vehicle body and car plate as the situation of car plate by mistake, improved follow-up positional information according to car plate, judged whether car plate is in the accuracy rate in the described color track.
In step S104, when detecting car plate, described car plate is carried out color differentiate, judge whether the color of described car plate is to limit color.
In the present embodiment, limiting color is the color except default color in the color classification.Preferably, the restriction color comprises one or more in blueness, black, the white.When default color was yellow, limiting color was the color except yellow in the color classification.
In the present embodiment, car plate is carried out color to be differentiated, particularly, the rgb value in the frame of video of car plate is mapped to HSV (Hue, Saturation, Value) in the space, and obtain the mean value of the tone value of frame of video in the HSV space, and because the tone value difference of different colours, therefore can be with the mean value of tone as eigenwert, default svm classifier device carries out color classification according to this eigenwert, to pick out the car plate that limits color.
In step S105, when the color of described car plate is the restriction color, detect described car plate whether to be in the described default color track;
In the present embodiment, detect described car plate and whether be in the described default color track, comprising:
Obtain the centre coordinate of described car plate;
When car plate central spot during in the right in left-hand lane of frame of video left side one side of something and described car plate upper left corner coordinate, expression detects described car plate and is in the described default color track; Or
When car plate central spot during in the left side in right-hand lane of the right one side of something of frame of video and described car plate upper right corner coordinate, expression detects described car plate and is in the described default color track.
In the present embodiment, according to the position coordinates in the positional information of car plate, generate the centre coordinate of car plate, in frame of video, determine the car plate central point according to centre coordinate, judge that the car plate central point is to be in frame of video left side one side of something, still frame of video is right half of, when the car plate central spot is half of in a frame of video left side, whether judge car plate upper left corner coordinate on the right of left-hand lane, if represent that then car plate is in the color track, when the car plate central spot is half of in the frame of video right side, whether judge the car plate upper right corner on the left side of right-hand lane, if represent that then car plate is in the color track.
In the present embodiment, for ease of explanation, be yellow track with the color track, limit color and comprise the blue example that is, when car plate occurring in the frame of video, car plate is carried out color to be differentiated, when the color of car plate was blueness, the vehicle that expression hangs with this car plate was public vehicles, according to the positional information of car plate, detect car plate whether in yellow track, to judge whether these public vehicles are swarmed into, to take bus zone.
For ease of explanation, Fig. 6 shows the preferable sample figure that judges whether public vehicles are swarmed into, takies bus zone or other dedicated Lanes, wherein Fig. 6 A represents that public vehicles swarm into bus zone or other dedicated Lanes, and Fig. 6 B represents that public vehicles take bus zone or other dedicated Lanes.
In step S106, in the time of in described car plate is in described default color track, the described car plate of mark in described frame of video.
In the present embodiment, when car plate is in yellow track, represent that this car plate is car plate violating the regulations, at frame of video acceptance of the bid caravan board, the mode of mark car plate includes but not limited to the edge pixel of car plate is set at default color, adopts default marker graphic mark car plate etc.
As a preferred embodiment of the present invention, in described frame of video, after the described car plate of mark, also comprise:
Described car plate is carried out car plate identification, to obtain the described number-plate number.
Car plate identification can be adopted any recognition methods of prior art, and as adopting support vector machine identification, or neural network identifies, to obtain the number-plate number.Thereby can obtain to swarm into or take the number-plate number of the vehicle of bus zone, so that traffic police department can dispose the vehicle that swarm into or take bus zone according to the number-plate number.
Fig. 7 shows the present invention's preferable implementing procedure in actual applications, and details are as follows:
In step S701, import SVM and AdaBoost sorter;
In step S702, read in video frame images;
In step S703, extraction image effective coverage also dwindles, and RGB figure changes gray-scale map, smoothly reaches Canny operator extraction image border by Gauss;
In step S704, edge image is divided into left and right sides two parts, utilize the conversion of Hough straight line to find out the straight line of left and right sides two parts weight maximum;
In step S705, left and right sides straight line respectively extends certain pixel, utilizes SVM to judge that the pixel in the bandwidth meets the probability that yellow bus zone pixel rgb value distributes;
In step S706, judge whether probability whether greater than default threshold values, is carried out 707, otherwise carried out 702;
In step S707, the subregion of effective car plate can appear in the intercepting video frame images, and gray processing is handled;
In step S708, gray level image is carried out integrogram and integral square figure computing;
In step S709, the zone that chooses is divided into three zones by picture altitude, the zone in the middle of arranging is zone to be detected;
In step S710, zone line to be detected is sent into the AdaBoost cascade classifier that trains handle;
In step S711, judge whether to detect car plate, be then to carry out 712, otherwise carry out 720;
In step S712, merge algorithm merges all car plates that detect;
In step S713, the car plate after being combined carries out Sobel and detects the generation marginal density, and obtains the license plate area RGB three averages maximal value of absolute difference in twos;
In step S714, the maximal value of the difference of marginal density and RGB average is carried out threshold decision, be then to carry out 716, otherwise carry out 715
In step S715, do not handle;
In step S716, car plate is sent into SVM carry out color classification;
In step S717, judge that whether car plate is blue, is then to carry out 718, otherwise carries out 715;
In step S718, whether within lane line, be then to carry out 719, otherwise carry out 715;
In step S719, around car plate, carry out mark, and carry out S722;
In step S720, two zones are zone to be detected about arranging, and send into respectively in the Adaboost sorter and handle;
In step S721, detect car plate, carry out 712, otherwise carry out 722;
In step S722, judge that frame of video is the last frame image, be then to carry out 723, otherwise carry out 702.
In step S723, stop.
Fig. 8 shows the structured flowchart of the illegal road occupying pick-up unit of a kind of vehicle that the embodiment of the invention provides, this device can run on video acquisition device and various terminal, include but not limited to video camera, mobile phone, pocket computing machine (Pocket Personal Computer, PPC), palm PC, computing machine, notebook computer, personal digital assistant (Personal Digital Assistant, PDA) etc.For convenience of explanation, only show the part relevant with present embodiment.
With reference to Fig. 8, the illegal road occupying pick-up unit of this vehicle comprises:
Extraction unit 81 is used for extracting the image that frame of video comprises the track;
First judging unit 82 is used for judging whether described image track is default color track;
First detecting unit 83, be used for when described track when presetting the color track, in described frame of video, detect car plate;
Second judging unit 84 is used for when detecting car plate, described car plate is carried out color differentiate, and judges whether the color of described car plate is to limit color;
Second detecting unit 85, whether the color that is used for when described car plate is when limiting color, to detect described car plate to be in the described default color track;
Indexing unit 86, when being used in described car plate is in described default color track, the described car plate of mark in described frame of video.
Further, in this device, described first judging unit comprises:
First obtains subelement, is used for obtaining the central point of described image, generates zone, an image left side and the right zone of image according to divide described image by the straight line of described central point;
Second obtains subelement, is used for obtaining the straight line of described image zone, a left side and the right zone of described image weight maximum;
First generates subelement, is used for extending predetermined width left, to the right in the horizontal direction of the straight line of weight maximum, generates the quadrilateral track;
The 3rd obtains subelement, is used for the number of obtaining described quantity at quadrilateral track pixel and classifying as default color pixel point;
Second generates subelement, is used for generating the quadrilateral track and being the probability in default color track according to the described number of default color pixel point and the quantity of quadrilateral track pixel of being categorized as;
Judgment sub-unit is used for judging that described quadrilateral track is described default color track when described probability reaches pre-set threshold value.
Further, in this device, described first detecting unit comprises:
The 4th obtains subelement, is used for the effective coverage that obtains described frame of video;
Divide subelement, be used for described effective coverage is divided into a plurality of zones;
Detection sub-unit is used in described zone car plate being detected in regular turn by preset order.
Further, in this device, also comprise:
Reject the unit, be used for judging whether described car plate is effective car plate, keeps when described car plate is effective car plate, otherwise reject.
Further, in this device, described second detecting unit comprises:
The 5th obtains subelement, is used for obtaining the centre coordinate of described car plate;
Second detection sub-unit is used for representing that when car plate central spot during in the right in left-hand lane of frame of video left side one side of something and described car plate upper left corner coordinate detecting described car plate is in the described default color track; Or
When car plate central spot during in the left side in right-hand lane of the right one side of something of frame of video and described car plate upper right corner coordinate, expression detects described car plate and is in the described default color track.
In embodiments of the present invention, when detecting car plate, described car plate is carried out color differentiate, judge whether the color of described car plate is to limit color; When the color of described car plate is the restriction color, detect described car plate whether to be in the described default color track; In the time of in described car plate is in described default color track, the described car plate of mark in described frame of video.So that relevant department can be according to the car plate of mark, disposal takies the public vehicles of bus zone or other dedicated Lanes, thereby solved and to have detected the public vehicles that take bus zone or other dedicated Lanes, the operational efficiency and the low problem of security that cause public transit vehicle or other vehicle, the operational efficiency and the security that have improved public transit vehicle or other vehicle.
The above only is preferred embodiment of the present invention, not in order to limiting the present invention, all any modifications of doing within the spirit and principles in the present invention, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. the illegal road occupying detection method of vehicle is characterized in that, comprising:
Extract the image that comprises the track in the frame of video;
Judge whether the track is default color track in the described image;
When described track is when presetting the color track, to detect car plate in described frame of video;
When detecting car plate, described car plate is carried out color differentiate, judge whether the color of described car plate is to limit color;
When the color of described car plate is the restriction color, detect described car plate whether to be in the described default color track;
In the time of in described car plate is in described default color track, the described car plate of mark in described frame of video.
2. the method for claim 1 is characterized in that, describedly judges in the described image that whether the track is default color track, comprising:
Obtain the central point of described image, generate zone, an image left side and the right zone of image according to divide described image by the straight line of described central point;
Obtain the straight line of weight maximum in the right zone of described image zone, a left side and described image;
Horizontal direction at the straight line of weight maximum is extended predetermined width left, to the right, generates the quadrilateral track;
Obtain described in the quadrilateral track quantity of pixel and the number that classifies as default color pixel point;
Quantity according to pixel in the described number that is categorized as default color pixel point and the quadrilateral track generates the quadrilateral track for presetting the probability in color track;
When described probability reaches pre-set threshold value, judge that described quadrilateral track is described default color track.
3. the method for claim 1 is characterized in that, describedly detects car plate when described track during for default color track in described frame of video, comprising:
Obtain the effective coverage in the described frame of video;
Described effective coverage is divided into a plurality of zones;
In described zone, car plate is detected in regular turn by preset order.
4. as claim 1 or 3 described methods, it is characterized in that, also comprise:
Judge whether described car plate is effective car plate, keeps when described car plate is effective car plate, otherwise rejects.
5. the method for claim 1 is characterized in that, whether the described car plate of described detection is in the described default color track, comprising:
Obtain the centre coordinate of described car plate;
When car plate central spot during in the right in left-hand lane of frame of video left side one side of something and described car plate upper left corner coordinate, expression detects described car plate and is in the described default color track; Or
When car plate central spot during in the left side in right-hand lane of the right one side of something of frame of video and described car plate upper right corner coordinate, expression detects described car plate and is in the described default color track.
6. the illegal road occupying pick-up unit of vehicle is characterized in that, comprising:
Extraction unit is used for extracting the image that frame of video comprises the track;
First judging unit is used for judging whether described image track is default color track;
First detecting unit, be used for when described track when presetting the color track, in described frame of video, detect car plate;
Second judging unit is used for when detecting car plate, described car plate is carried out color differentiate, and judges whether the color of described car plate is to limit color;
Second detecting unit, whether the color that is used for when described car plate is when limiting color, to detect described car plate to be in the described default color track;
Indexing unit, when being used in described car plate is in described default color track, the described car plate of mark in described frame of video.
7. device as claimed in claim 6 is characterized in that, described first judging unit comprises:
First obtains subelement, is used for obtaining the central point of described image, generates zone, an image left side and the right zone of image according to divide described image by the straight line of described central point;
Second obtains subelement, is used for obtaining the straight line of described image zone, a left side and the right zone of described image weight maximum;
First generates subelement, is used for extending predetermined width left, to the right in the horizontal direction of the straight line of weight maximum, generates the quadrilateral track;
The 3rd obtains subelement, is used for the number of obtaining described quantity at quadrilateral track pixel and classifying as default color pixel point;
Second generates subelement, is used for generating the quadrilateral track and being the probability in default color track according to the described number of default color pixel point and the quantity of quadrilateral track pixel of being categorized as;
Judgment sub-unit is used for judging that described quadrilateral track is described default color track when described probability reaches pre-set threshold value.
8. device as claimed in claim 6 is characterized in that, described first detecting unit comprises:
The 4th obtains subelement, is used for the effective coverage that obtains described frame of video;
Divide subelement, be used for described effective coverage is divided into a plurality of zones;
Detection sub-unit is used in described zone car plate being detected in regular turn by preset order.
9. as claim 6 or 8 described devices, it is characterized in that, also comprise:
Reject the unit, be used for judging whether described car plate is effective car plate, keeps when described car plate is effective car plate, otherwise reject.
10. device as claimed in claim 6 is characterized in that, described second detecting unit comprises:
The 5th obtains subelement, is used for obtaining the centre coordinate of described car plate;
Second detection sub-unit is used for representing that when car plate central spot during in the right in left-hand lane of frame of video left side one side of something and described car plate upper left corner coordinate detecting described car plate is in the described default color track; Or
When car plate central spot during in the left side in right-hand lane of the right one side of something of frame of video and described car plate upper right corner coordinate, expression detects described car plate and is in the described default color track.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310133213.7A CN103268471B (en) | 2013-04-17 | 2013-04-17 | A kind of vehicle illegal road occupying detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310133213.7A CN103268471B (en) | 2013-04-17 | 2013-04-17 | A kind of vehicle illegal road occupying detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103268471A true CN103268471A (en) | 2013-08-28 |
| CN103268471B CN103268471B (en) | 2016-08-17 |
Family
ID=49012098
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310133213.7A Active CN103268471B (en) | 2013-04-17 | 2013-04-17 | A kind of vehicle illegal road occupying detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103268471B (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886745A (en) * | 2013-11-25 | 2014-06-25 | 天津思博科科技发展有限公司 | Automobile-dedicated lane early warning system |
| CN104424801A (en) * | 2013-08-29 | 2015-03-18 | 杭州海康威视系统技术有限公司 | System and method for snapping bus road occupation |
| CN104680794A (en) * | 2015-02-13 | 2015-06-03 | 深圳市图敏实业有限公司 | Vehicle-mounted monitoring terminal as well as bus lane monitoring method and system based on terminal |
| CN105206049A (en) * | 2015-08-18 | 2015-12-30 | 苏州玄禾物联网科技有限公司 | Bus-only lane occupation snapshot system based on ZigBee |
| CN105390014A (en) * | 2015-11-30 | 2016-03-09 | 四川诚品电子商务有限公司 | Public transport vehicle call system |
| CN105608906A (en) * | 2016-03-29 | 2016-05-25 | 成都理工大学 | System for monitoring illegal emergency lane occupancy of expressway motor vehicles and implementation method |
| CN105702049A (en) * | 2016-03-29 | 2016-06-22 | 成都理工大学 | DSP-based emergency lane monitoring system and realizing method thereof |
| CN105741559A (en) * | 2016-02-03 | 2016-07-06 | 安徽清新互联信息科技有限公司 | Emergency vehicle lane illegal occupation detection method based on lane line model |
| CN106340180A (en) * | 2015-07-06 | 2017-01-18 | 北京文安智能技术股份有限公司 | Vehicle-mounted illegal lane occupation behavior detecting method and device |
| CN107609555A (en) * | 2017-09-15 | 2018-01-19 | 北京文安智能技术股份有限公司 | Detection method of license plate, model recognizing method and relevant apparatus using it |
| WO2018129653A1 (en) * | 2017-01-10 | 2018-07-19 | 深圳市锐明技术股份有限公司 | Method for collecting evidence of not yielding to school bus, apparatus, and school bus |
| CN109509355A (en) * | 2019-01-07 | 2019-03-22 | 黄慧 | Wisdom traffic information taken terminal |
| CN110298837A (en) * | 2019-07-08 | 2019-10-01 | 上海天诚比集科技有限公司 | Fire-fighting road occupying exception object detecting method based on frame differential method |
| CN110659539A (en) * | 2018-06-28 | 2020-01-07 | 杭州海康威视数字技术股份有限公司 | Information processing method and device |
| CN111860219A (en) * | 2020-06-30 | 2020-10-30 | 杭州科度科技有限公司 | High-speed road occupation judging method and device and electronic equipment |
| CN112633062A (en) * | 2020-11-18 | 2021-04-09 | 合肥湛达智能科技有限公司 | Deep learning bus lane occupation detection method based on embedded terminal |
| CN112712708A (en) * | 2020-12-28 | 2021-04-27 | 上海眼控科技股份有限公司 | Information detection method, device, equipment and storage medium |
| CN113111884A (en) * | 2021-03-26 | 2021-07-13 | 沈阳天眼智云智能技术研究院有限公司 | Video detection method for special vehicle for hazardous chemical transportation |
| CN113642412A (en) * | 2021-07-16 | 2021-11-12 | 盛视科技股份有限公司 | Method, device and equipment for detecting vehicles occupying bus lane |
| CN114419885A (en) * | 2022-01-14 | 2022-04-29 | 重庆长安汽车股份有限公司 | Vehicle violation snapshot method and system based on high-precision positioning and vehicle-mounted sensing capability |
| CN115100619A (en) * | 2022-06-30 | 2022-09-23 | 重庆紫光华山智安科技有限公司 | Traffic violation state determination method, system, device, and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000268291A (en) * | 1999-03-18 | 2000-09-29 | Nec Corp | License plate recognition device |
| WO2002047373A2 (en) * | 2000-12-08 | 2002-06-13 | Koninklijke Philips Electronics N.V. | Apparatus and method for correcting signal and color degradation in a video image using known icons from video image |
| CN101826260A (en) * | 2010-05-04 | 2010-09-08 | 清华大学 | Method and system for automatically snapping traffic offence |
| CN202003508U (en) * | 2011-03-03 | 2011-10-05 | 苏州市慧视通讯科技有限公司 | Bus lane illegal vehicle intruding detection equipment |
| CN202018736U (en) * | 2011-04-22 | 2011-10-26 | 北京文安科技发展有限公司 | Vehicle-mounted monitoring system for illegal bus lane occupation |
| CN102622882A (en) * | 2012-03-20 | 2012-08-01 | 刘业兴 | System for snap-shooting illegal road occupation of large-scale vehicles on highway |
-
2013
- 2013-04-17 CN CN201310133213.7A patent/CN103268471B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000268291A (en) * | 1999-03-18 | 2000-09-29 | Nec Corp | License plate recognition device |
| WO2002047373A2 (en) * | 2000-12-08 | 2002-06-13 | Koninklijke Philips Electronics N.V. | Apparatus and method for correcting signal and color degradation in a video image using known icons from video image |
| CN101826260A (en) * | 2010-05-04 | 2010-09-08 | 清华大学 | Method and system for automatically snapping traffic offence |
| CN202003508U (en) * | 2011-03-03 | 2011-10-05 | 苏州市慧视通讯科技有限公司 | Bus lane illegal vehicle intruding detection equipment |
| CN202018736U (en) * | 2011-04-22 | 2011-10-26 | 北京文安科技发展有限公司 | Vehicle-mounted monitoring system for illegal bus lane occupation |
| CN102622882A (en) * | 2012-03-20 | 2012-08-01 | 刘业兴 | System for snap-shooting illegal road occupation of large-scale vehicles on highway |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104424801A (en) * | 2013-08-29 | 2015-03-18 | 杭州海康威视系统技术有限公司 | System and method for snapping bus road occupation |
| CN103886745A (en) * | 2013-11-25 | 2014-06-25 | 天津思博科科技发展有限公司 | Automobile-dedicated lane early warning system |
| CN104680794A (en) * | 2015-02-13 | 2015-06-03 | 深圳市图敏实业有限公司 | Vehicle-mounted monitoring terminal as well as bus lane monitoring method and system based on terminal |
| CN106340180A (en) * | 2015-07-06 | 2017-01-18 | 北京文安智能技术股份有限公司 | Vehicle-mounted illegal lane occupation behavior detecting method and device |
| CN105206049A (en) * | 2015-08-18 | 2015-12-30 | 苏州玄禾物联网科技有限公司 | Bus-only lane occupation snapshot system based on ZigBee |
| CN105390014A (en) * | 2015-11-30 | 2016-03-09 | 四川诚品电子商务有限公司 | Public transport vehicle call system |
| CN105741559B (en) * | 2016-02-03 | 2018-08-31 | 安徽清新互联信息科技有限公司 | A kind of illegal occupancy Emergency Vehicle Lane detection method based on track line model |
| CN105741559A (en) * | 2016-02-03 | 2016-07-06 | 安徽清新互联信息科技有限公司 | Emergency vehicle lane illegal occupation detection method based on lane line model |
| CN105702049A (en) * | 2016-03-29 | 2016-06-22 | 成都理工大学 | DSP-based emergency lane monitoring system and realizing method thereof |
| CN105608906A (en) * | 2016-03-29 | 2016-05-25 | 成都理工大学 | System for monitoring illegal emergency lane occupancy of expressway motor vehicles and implementation method |
| WO2018129653A1 (en) * | 2017-01-10 | 2018-07-19 | 深圳市锐明技术股份有限公司 | Method for collecting evidence of not yielding to school bus, apparatus, and school bus |
| CN107609555A (en) * | 2017-09-15 | 2018-01-19 | 北京文安智能技术股份有限公司 | Detection method of license plate, model recognizing method and relevant apparatus using it |
| CN107609555B (en) * | 2017-09-15 | 2020-10-27 | 北京文安智能技术股份有限公司 | License plate detection method, vehicle type identification method applying license plate detection method and related device |
| CN110659539A (en) * | 2018-06-28 | 2020-01-07 | 杭州海康威视数字技术股份有限公司 | Information processing method and device |
| CN109509355A (en) * | 2019-01-07 | 2019-03-22 | 黄慧 | Wisdom traffic information taken terminal |
| CN110298837A (en) * | 2019-07-08 | 2019-10-01 | 上海天诚比集科技有限公司 | Fire-fighting road occupying exception object detecting method based on frame differential method |
| CN111860219A (en) * | 2020-06-30 | 2020-10-30 | 杭州科度科技有限公司 | High-speed road occupation judging method and device and electronic equipment |
| CN111860219B (en) * | 2020-06-30 | 2024-01-05 | 杭州科度科技有限公司 | High-speed channel occupation judging method and device and electronic equipment |
| CN112633062A (en) * | 2020-11-18 | 2021-04-09 | 合肥湛达智能科技有限公司 | Deep learning bus lane occupation detection method based on embedded terminal |
| CN112712708A (en) * | 2020-12-28 | 2021-04-27 | 上海眼控科技股份有限公司 | Information detection method, device, equipment and storage medium |
| CN113111884A (en) * | 2021-03-26 | 2021-07-13 | 沈阳天眼智云智能技术研究院有限公司 | Video detection method for special vehicle for hazardous chemical transportation |
| CN113111884B (en) * | 2021-03-26 | 2024-05-24 | 沈阳天眼智云智能技术研究院有限公司 | Video detection method of special dangerous chemical transportation vehicle |
| CN113642412A (en) * | 2021-07-16 | 2021-11-12 | 盛视科技股份有限公司 | Method, device and equipment for detecting vehicles occupying bus lane |
| CN113642412B (en) * | 2021-07-16 | 2023-12-26 | 盛视科技股份有限公司 | Method, device and equipment for detecting vehicle occupying bus lane |
| CN114419885A (en) * | 2022-01-14 | 2022-04-29 | 重庆长安汽车股份有限公司 | Vehicle violation snapshot method and system based on high-precision positioning and vehicle-mounted sensing capability |
| CN115100619A (en) * | 2022-06-30 | 2022-09-23 | 重庆紫光华山智安科技有限公司 | Traffic violation state determination method, system, device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103268471B (en) | 2016-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103268471A (en) | Vehicle illegal land occupying detection method and device | |
| CN102722704B (en) | Method and system for recognizing vehicle license plate by integrating video dynamic tracking | |
| CN103235938B (en) | The method and system of car plate detection and indentification | |
| CN205992300U (en) | For providing the electronic installation of the guidance information using crossing recognition result | |
| CN102364496B (en) | Method and system for identifying automobile license plates automatically based on image analysis | |
| CN109711264B (en) | Method and device for detecting occupation of bus lane | |
| CN101872416B (en) | Vehicle license plate recognition method and system of road image | |
| CN105741559B (en) | A kind of illegal occupancy Emergency Vehicle Lane detection method based on track line model | |
| CN104537360B (en) | Vehicle does not give way peccancy detection method and its detecting system | |
| CN102867417B (en) | Taxi anti-forgery system and taxi anti-forgery method | |
| CN104050450A (en) | Vehicle license plate recognition method based on video | |
| CN105976392B (en) | Vehicle tyre detection method and device based on maximum output probability | |
| CN102902957A (en) | Video-stream-based automatic license plate recognition method | |
| CN108090459B (en) | Traffic sign detection and identification method suitable for vehicle-mounted vision system | |
| CN105426894B (en) | Railway plug pin image detecting method and device | |
| CN104616502A (en) | License plate identification and positioning system based on combined type vehicle-road video network | |
| CN105740836B (en) | A kind of illegal detection method for occupying Emergency Vehicle Lane | |
| CN109670449B (en) | Vehicle illegal judgment method based on vertical snapshot mode | |
| CN103164958B (en) | Method and system for vehicle monitoring | |
| CN104240515A (en) | Exclusive bus lane occupation snapshotting method based on image processing | |
| CN105184291A (en) | Method and system for detecting multiple types of license plates | |
| CN108304749A (en) | The recognition methods of road speed line, device and vehicle | |
| CN103021179A (en) | Real-time monitoring video based safety belt detection method | |
| Mammeri et al. | North-American speed limit sign detection and recognition for smart cars | |
| Janda et al. | Road boundary detection for run-off road prevention based on the fusion of video and radar |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Nanshan District Xueyuan Road in Shenzhen city of Guangdong province 518000 No. 1001 Nanshan Chi Park B1 building 22 floor Applicant after: STREAMAX TECHNOLOGY CO., LTD. Address before: 518057, Guangdong Shenzhen Nanshan District science and technology two road Shenzhen software park three, five floor Applicant before: Shenzhen Streaming Video Technology Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |