CN103077384A - Method and system for positioning and recognizing vehicle logo - Google Patents
Method and system for positioning and recognizing vehicle logo Download PDFInfo
- Publication number
- CN103077384A CN103077384A CN201310009960XA CN201310009960A CN103077384A CN 103077384 A CN103077384 A CN 103077384A CN 201310009960X A CN201310009960X A CN 201310009960XA CN 201310009960 A CN201310009960 A CN 201310009960A CN 103077384 A CN103077384 A CN 103077384A
- Authority
- CN
- China
- Prior art keywords
- car
- car mark
- mark
- vehicle
- texture
- 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 discloses a method for positioning and recognizing a vehicle logo. The method comprises the following steps of: collecting the vehicle head image of a vehicle by a high-definition camera; positioning the vehicle and positioning and recognizing a vehicle registration plate according to the vehicle head image, and generating the estimated position region of the vehicle logo according to the prior information of the vehicle logo and the vehicle registration plate; in the estimated position region of the vehicle logo, filtering the background region disturbance of the vehicle logo according to extracted texture information, carrying out binaryzation on the texture grey-scale map of the vehicle logo, on which the background region disturbance of the vehicle logo is filtered, accurately positioning the vehicle logo on the generated binary image through a method of finding connected domains in the estimated position region of the vehicle logo and generating the accurate position information of the vehicle logo; according to the accurate position information of the vehicle logo, extracting the accurate grey-scale map of the vehicle logo, carrying out binaryzation on the accurate grey-scale map of the vehicle logo, and generating a binary map corresponding to the vehicle logo; and matching the binary map corresponding to the vehicle logo with a prestored vehicle logo format, carrying out weighing calculation according to the position information, and viewing the vehicle logo type corresponding to the highest score obtained by calculation as the recognizing result of the vehicle logo.
Description
Technical field
The present invention relates to vehicular traffic fixation and recognition field, relate in particular to a kind of car and demarcate position recognition methods and system.
Background technology
The car mark is the significant image of vehicle, has not only comprised vehicle information, has also comprised the information of manufacturer, and more important is that it is difficult to change.If car mark information is combined with license board information, can improve greatly the reliability of vehicle identification.In some criminal activity, although car plate is easily changed, but the car mark is unique, car identifies automatic safe management that other technology can be widely used in the vehicles such as traffic hazard processing, vehicles peccancy automatic monitoring, automatic charging, airport, harbour, parking lot, residential quarter, hit the field such as vehicle crime, thus the car sign other to vehicle recongnition technique in the intelligent transportation improve and development has important theory significance and using value.
Can effectively detect the car mark at present, the Main Means of location and identification is based on the method for video, but utilize video method to carry out a car demarcation problem that identification exists mainly is:
1, the impact of self background of vehicle image and vehicle body background;
2, have a plurality of car marks in the same vehicle, and the car mark is subjected to weather and colour of sky variable effect serious;
3, the car target is different and the size difference plot is large, image blurring, and the position is not quite similar;
4, the radiator-grid texture at car mark place is varied, and the radiator-grid height is uneven;
5, there be a large amount of the interference around the car mark, such as car light, bumper etc.The car mark on car body and some and body color similar.
The existence of the problems referred to above so that it is not high to use video method to carry out the degree of accuracy of car target fixation and recognition, and does not also have a kind of checkout equipment accurately to locate and effectively identification the car mark at present.
Summary of the invention
The invention provides a kind of car and demarcate the method and system of position identification, to solve the not high problem of precision of method of existing video location.
In order to achieve the above object, the embodiment of the invention discloses the method that a kind of car is demarcated position identification, comprising: the headstock image that utilizes the high-definition camera collection vehicle; Utilize described headstock image, carry out vehicle location and car plate fixation and recognition, and according to the prior imformation of car mark and car plate, generate the car mark and estimate the band of position; Estimate in the band of position at described car mark, disturb according to the texture information filtering car mark background area of extracting, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark and carry out the car mark in the band of position and accurately locate by seeking the connected domain method, generate car mark precise position information; According to described car mark precise position information, extract car mark accurate greyscale figure, and described car mark accurate greyscale figure is carried out binaryzation, generate binary map corresponding to described car mark; The binary map that described car mark is corresponding is mated with the car mark template that prestores, and is weighted calculating according to positional information, and car mark type corresponding to the mark soprano who calculates is that car identifies other result.
In order to achieve the above object, the embodiment of the invention also discloses the system that a kind of car is demarcated position identification, comprising: image collecting device, for the headstock image of collection vehicle; The car mark is estimated band of position generating apparatus, is used for utilizing described headstock image, carries out vehicle location and car plate fixation and recognition, and according to the prior imformation of car mark and car plate, generation car mark is estimated the band of position; Car mark precise position information generating apparatus, be used for estimating the band of position at described car mark, disturb according to the texture information filtering car mark background area of extracting, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark and carry out the car mark in the band of position and accurately locate by seeking the connected domain method, generate car mark precise position information; Car mark binary map generating apparatus is used for according to described car mark precise position information, extracts car mark accurate greyscale figure, and described car mark accurate greyscale figure is carried out binaryzation, generates binary map corresponding to described car mark; Car identifies other as a result generating apparatus, is used for the binary map that described car mark is corresponding and mates with the car mark template that prestores, and is weighted calculating according to positional information, and car mark type corresponding to the mark soprano who calculates is that car identifies other result.
Demarcation position recognition methods of the present invention and system can accurately locate and effectively identification the car mark.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, the below will do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art, apparently, accompanying drawing in the following describes only is some embodiments of the present invention, for those skilled in the art, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is that the car of the embodiment of the invention is demarcated the process flow diagram of position recognition methods;
Fig. 2 is the method flow diagram of the substep of the step S102 in embodiment illustrated in fig. 1;
Fig. 3 generates the result schematic diagram that the car mark is estimated a specific embodiment in zone for certain two field picture that gathers is carried out vehicle location, car plate fixation and recognition;
Fig. 4 is the method flow diagram of the substep of the step S103 in embodiment illustrated in fig. 1;
Fig. 5 is the schematic diagram that the car mark in embodiment illustrated in fig. 4 is estimated the cross grain in the zone;
Fig. 6 is the schematic diagram that the car mark in embodiment illustrated in fig. 4 is estimated the vertical texture in the zone;
Fig. 7 is for to the binary map that generates after the pre-service of carrying out embodiment illustrated in fig. 6;
Fig. 8 is on the binary map basis of Fig. 7, carries out the figure of longitudinal projection of longitudinal projection;
Fig. 9 is the final pinpoint result schematic diagram of car mark embodiment illustrated in fig. 8;
Figure 10 is the method flow diagram of the substep of the step S104 in embodiment illustrated in fig. 1;
Figure 11 is the method flow diagram of the substep of the step S105 in embodiment illustrated in fig. 1;
Figure 12 is that middle location embodiment illustrated in fig. 9 car out identifies other result;
Figure 13 be car of the present invention demarcate the position recognition methods the method flow diagram of another embodiment;
Figure 14 is that the car of the embodiment of the invention is demarcated the structural representation of position recognition system;
Figure 15 is the structural representation that the car mark in embodiment illustrated in fig. 14 is estimated band of position generating apparatus 101;
Figure 16 is the structural representation of the car mark precise position information generating apparatus 102 in embodiment illustrated in fig. 14;
Figure 17 is the structural representation of the car mark binary map generating apparatus 103 in embodiment illustrated in fig. 14;
Figure 18 is the structural representation that the car in embodiment illustrated in fig. 14 identifies other as a result generating apparatus 104;
Figure 19 is the structural representation that car of the present invention is demarcated another embodiment of position recognition system;
Figure 20 (a) and (b), (c), (d) are for utilizing car of the present invention to demarcate position recognition methods and system, the treatment scheme schematic diagram that the car mark of several vehicles is identified.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that obtains under the creative work prerequisite.
The embodiment of the invention provides a kind of car to demarcate position recognition methods and system, under the prerequisite of automobile storage at effective board, can orient fast the car cursor position, and carry out single frames identification, multiframe is confirmed, determine the uniqueness of vehicle and vehicle by car plate and car mark, report background system, vehicle is registered affirmation.
Fig. 1 is that the car of the embodiment of the invention is demarcated the process flow diagram of position recognition methods.As shown in the figure, the car of a present embodiment demarcation position recognition methods comprises:
Step S101 utilizes the headstock image of high-definition camera collection vehicle; Step S102 utilizes described headstock image, carries out vehicle location and car plate fixation and recognition, and according to the prior imformation of car mark and car plate, generates the car mark and estimate the band of position; Step S103, estimate in the band of position at described car mark, disturb according to the texture information filtering car mark background area of extracting, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark and carry out the car mark in the band of position and accurately locate by seeking the connected domain method, generate car mark precise position information; Step S104 according to described car mark precise position information, extracts car mark accurate greyscale figure, and described car mark accurate greyscale figure is carried out binaryzation, generates binary map corresponding to described car mark; Step S105, the binary map that described car mark is corresponding is mated with the car mark template that prestores, and is weighted calculating according to positional information, and car mark type corresponding to the mark soprano who calculates is that car identifies other result.
In the step S101 of the present embodiment, roll end away from the expressway, high-definition camera is installed, catch the headstock image that comes into view, 200W scene frame lower p.s.s 15, the image size is 1600*1200, is frames p.s.s 7 under the 500W scene, size is 2432*2048, and image resolution ratio can be adjusted according to the difference of high-definition camera.In the following step of the present embodiment, be that each two field picture of taking among the step S101 is processed, the car that generates corresponding each frame identifies other result.
In the step S102 of the present embodiment, as shown in Figure 2, utilize described headstock image, carry out vehicle location and car plate fixation and recognition, and according to the prior imformation of car mark and car plate, generate the car mark and estimate the step of the band of position and comprise following substep: step S1021, the headstock image of described collection is compressed in proportion, generate gray level image; Step S1022 sets up background according to Gauss model on described gray level image, according to background subtraction and poor vehicle location, the generation vehicle location zone of carrying out of frame; Step S1023 in described vehicle location zone, carries out the car plate fixation and recognition according to texture information and the colouring information of vehicle, generates the car plate band of position; Step S1024 in the described car plate band of position, carries out car according to the prior imformation of described car mark and car plate and demarcates the position, generates the car mark and estimates the zone.
In step S1021, receive the headstock image that high-definition camera gathers after, in order to reduce complexity, coloured image is compressed in proportion, carry out simultaneously dimensionality reduction and extract gray level image as an input of next carrying out car and demarcate position identification.
In step S1022, on gray level image, set up background according to Gauss model, seek connected region according to background subtraction and carry out vehicle location, do poor " get with " by adjacent three frames the vehicle that carries out is out confirmed, thereby it is regional to generate vehicle location.
In the present embodiment, it is poor to be done by present frame gray-scale map and background gray-scale map, to background subtraction gray-scale map binaryzation, seeks the white point connected domain, and area is just tentatively thought vehicle region greater than certain threshold value.Because being the poor poor figure of frame that obtains, adjacent two frames can cause the vehicle motion blur phenomenon, the vehicle region is elongated, introduce and disturb, so adopt continuous three frames, adjacent two frames are done the poor then binaryzation of frame in twos, to two poor binary map of frame that obtain get with, namely same position has the available point of thinking of the poor information of frame on the poor figure of two width of cloth frames, obtain the poor binary map of final frame after " get with ", seek connected domain and be the vehicle region, confirm by the poor location of frame vehicle out and the vehicle of locating out by background subtraction, obtain final vehicle location.
In step S1023, in described vehicle location zone, carry out the car plate fixation and recognition of corresponding vehicle according to texture information and colouring information, generate the car plate band of position.
In the present embodiment, in the vehicle location zone, cromogram carries out gray processing, then ask cross grain, common car plate has 7 characters, and cross grain has 14 saltus steps, so seek in texture maps and to satisfy saltus step in the car plate width range greater than 10 UNICOM zone, be license plate candidate area.Carrying out license plate locating method by colouring information is, car plate mainly is blue board and yellow card, in the coloured picture scope in vehicle location zone, extract blue dot and yellow dots will information, that satisfies that blue dot connected domain or yellow dots will connected domain size meet the car plate rule is candidate's car plate, candidate's car plate gray-scale map is carried out binaryzation, Character segmentation, the identification of monocase stencil matching, an average matching degree asked in seven characters of car plate, average matching degree is the highest and reach the real car plate that is of setting recognition threshold, real car plate gray-scale map is carried out binaryzation, carry out the precise cutting of character according to binary map projection and connected domain method, obtain accurately marginal position up and down of seven characters, according to up and down edge fitting straight line and the horizontal line contrast of character, thereby draw license plate sloped angle.
In the present embodiment, also comprise by Character segmentation and thick identification, then optimum car plate is accurately identified in the searching car plate location, if being lower than default recognition threshold, recognition confidence thinks that this vehicle does not contain effective board, the vehicle location result might have deviation, do not contain effective car plate and car mark zone in the vehicle region, otherwise think that this vehicle location is effective.
In step S1024, in the described car plate band of position, carry out car according to the prior imformation of described car mark and car plate and demarcate the position, generate the car mark and estimate the zone.In the present embodiment, on the basis of car plate fixation and recognition, can estimate the zone at car mark place according to prioris such as the positional information of car mark and car plate, car mark shape, vehicle symmetry, namely the car mark is estimated the zone.
In the present embodiment, step S1021-1024 is the method flow that car mark zone is estimated.With reference to figure 3, be that certain two field picture that gathers is carried out vehicle location, car plate fixation and recognition, generate the specific embodiment that the car mark is estimated the zone.Wherein, 1. empty frame is that step S1022 is according to background subtraction and the poor location of frame vehicle region out, 2. empty frame is that the car plate of step S1023 fixation and recognition on empty frame basis 1., vehicle location zone is estimated the band of position, and 3. empty frame is the car plate band of position of seeking the last location of effective bridge queen among the step S1023, then according to prioris such as the positional information of car mark and car plate, car mark shape, vehicle symmetry, searched out the empty frame of empty frame above 3. and 4. estimated the zone for the car mark, carried out car in 4. at this empty frame and demarcate.
In the step S103 of the present embodiment, as shown in Figure 4, estimate in the band of position at described car mark, disturb according to the texture information filtering car mark background area of extracting, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark and carry out the car mark in the band of position and accurately locate by seeking the connected domain method, generate car mark precise position information, realize by following substep: step S1031, estimate extraction sobel cross grain and vertical texture in the band of position at described car mark; Step S1032, the average gray of the gray-scale map after relatively extracting the gray-scale map behind the cross grain and extracting vertical texture is determined the grain direction that disturb described car mark background area by both differences; Step S1033 extracts the mutually anti-phase texture image of grain direction that disturbs with described car mark background area, and generation car mark is estimated the car mark texture gray-scale map in the band of position; Step S1034 carries out binary conversion treatment to described car mark veining gray-scale map, and carries out morphologic filtering and corrosion expansion, generates binary map corresponding to described car mark texture gray-scale map; Step S1035 carries out projection on binary map basis corresponding to described car mark veining gray-scale map, seeks border, the left and right sides corresponding to car mark, generates car mark precise position information.
In step S1031, preceding step 102 has found and has comprised car target rough position zone, namely such as the empty frame among Fig. 3 4. shown in the inner region, on the basis of this area grayscale figure to empty frame 4. inner region extract sobel cross grain and vertical texture, following Fig. 5 and shown in Figure 6.Wherein, Fig. 5 is that the car mark is estimated the cross grain in the zone, and Fig. 6 is that the car mark is estimated the vertical texture in the zone.
In step S1032 and step S1033, extract respectively the average gray of the gray-scale map behind the texture by comparison diagram 5 and Fig. 6, (the texture threshold value generally gets 15 if both differences are greater than the texture threshold value of setting, certainly, can be other numerical value according to actual conditions), think that then car mark background texture is cross grain, estimate the interior vertical texture in zone so should extract the car mark, filtering heatsink transverse window interference etc., only remaining car mark texture information.After if the car mark is estimated extracted region transverse and longitudinal background texture, if vertically texture mean value is much larger than horizontal mean value, think that then background is vertical texture, just need to extract cross grain, vertical interference of texture of wiping out background and remaining car mark information only, if but cross grain is close with vertical texture average, then background is thought reticular texture, and will extract Texture this moment by the sobel filter operator.That is: extract the mutually anti-phase texture image of grain direction that disturbs with described car mark background area, estimate car mark texture gray-scale map in the band of position with generation car mark.Texture herein extracts, relatively reaches the computing method of sobel filter operator for well known to a person skilled in the art technology, so locate to repeat no more.
In step S1034, gray level image in the rough filtering interfering of Fig. 6 is sought car mark exact position through pre-service, ask the binary-state threshold of gray-scale map shown in Figure 6 with the method for OTSU, the phenomenon of the light and shade inequality that causes for fear of light, shade etc., and car mark background area reflector lamp impact, mean value correction OTSU with gray-scale map obtains next binary-state threshold, Fig. 6 is carried out binary conversion treatment, and then carry out morphologic filtering, corrosion is expanded, eliminate the processing such as horizontal long line, the pretreated effect of Fig. 6 as shown in Figure 7.Can see, gray level image is turned to binary map by two-value.
In step S1035, carry out projection on binary map basis corresponding to described car mark veining gray-scale map, seek border, the left and right sides corresponding to car mark, generate car mark precise position information.In the embodiment of Fig. 2-shown in Figure 7, on the basis of the binary map of Fig. 7, carry out longitudinal projection, and ask for the mean value nAvg of longitudinal projection, the certain proportion k that averages is as the threshold value nTh=k*nAvg that seeks border, the left and right sides, take the car plate axis as benchmark, about on the transverse projection figure, expand searching border, the car mark left and right sides, when a continuous fixed point ratio when seeking the threshold value on border, think that then the border stops, otherwise the node zero clearing breaks from the new connection are sought about always again, and Fig. 7 longitudinal projection schemes as shown in Figure 8, after determining border, the car target left and right sides, carry out transverse projection according to same method, seeking car target up-and-down boundary, the final pinpoint result of car mark as empty frame in the middle of Fig. 9 5. shown in.
In step S104, as shown in figure 10, according to described car mark precise position information, extract car mark accurate greyscale figure, and described car mark accurate greyscale figure carried out binaryzation, and generate the step of binary map corresponding to described car mark, comprise following substep: step S1041, according to the license plate sloped angle that generates in the described car plate location, described car mark accurate greyscale figure is carried out slant correction; Step S1042 carries out binary conversion treatment to the car mark accurate greyscale figure after the described correction according to the method for bilinearity OTSU, generates binary map corresponding to described car mark.
In the present embodiment, demarcate the car mark precise position information that the position provides according to car, extract car mark gray-scale map, the license plate sloped angle calculated in the car plate fixation and recognition of carrying out by the front, slant correction is carried out in car mark zone, then be normalized to the figure of 40*40, uneven for the light and shade that prevents from causing because of disturbing effects such as light inequality, the shadow of the trees, car mark gray-scale map is carried out binaryzation according to the method for bilinearity OTSU.
In step S105, as shown in figure 11, the step that the binary map that described car mark is corresponding and the car mark template that prestores are mated comprises: step S1051, carry out the outline match by the car mark on binary map corresponding to the described car mark of hough transfer pair; Step S1052 classifies to described car mark according to fitting result; Step S1053 selects to mate with the car mark template of described car mark corresponding types.
In the present embodiment, on the basis of binary map, carry out the outline match by hough transfer pair car mark, simultaneously according to prioris such as car mark depth-width ratio shapes, be a class in ellipse, circle and the irregular figure with this car mark rough segmentation, then select the car mark masterplate of the type to mate.The hough conversion is the method for fitting circle on binary map, that is: according to car plate and car mark relation, can draw car mark least radius and maximum radius, namely want match radius of a circle minimum value and maximal value, according to symmetry, on the binary map of car mark zone take central point as central coordinate of circle, according to radius scanning circumference, calculate white point number on the circumference, when the white point number is higher than certain threshold value with the girth ratio, think that this circle effectively, this car is designated as circular outline car mark, otherwise take car mark regional center point as basic point, take least radius and maximum radius two times as the long scope of transverse or square side size range run-down, square is satisfied in judgement or oval-shaped point accounts for the ratio of girth, just think ellipse or square outline car mark when being higher than certain threshold value, if circular, oval, the words that all lost efficacy of square match are then thought and are belonged to irregular type car mark.
Car mark masterplate is the representative car mark binary map that extracts according to poor principle maximum, class interpolation maximum between class, size is 40*40, in order to save space complexity and time complexity, with eight synthetic bytes of binary map, the masterplate boil down to 40*5 of 40*40 size, the car of each type indicates three templates at present, and number can be adjusted according to actual conditions.
When carrying out car mark template matches, consider that the location is difficult to be accurate to the location of pixels on car mark border, and certain boundary interference information may be arranged also when choosing masterplate, so adopt the method for shiding matching, car mark binary map is upper, lower, left, can carry out the slip of 1 to 2 pixel on the right four direction, blank parts is mended 0 alignment, then eight synthetic bytes of binary map and template are mated, consider the car mark of same type, such as Ford and BYD, Kia, popular, Buick and benz etc., outline is identical all to be circular or oval, only have inside center information different, so when stencil matching, be weighted according to positional information, on the car mark binary map of 40*40,8 layers of weight of outermost are 1, and middle 16 layers of weight are 2, and the 16 layers of weight in center are 3, if not having coupling upper is zero, be set to respectively 1 minute according to the position weight difference after the match is successful, 2 minutes, 3 minutes is 0-100 minutes with score normalization, with all sort results behind the shiding matching at last, choose mark the highest identify other result for final single frames car, the single frames car is set identifies other degree of confidence, think identification effectively if high confidence level is higher than this value, otherwise refuse identification.
Location car out identifies other result as shown in figure 12 among Fig. 9, wherein, the first from left figure is original location gray-scale map among Figure 12, the second from left figure is through pretreated figures such as slant correction, normalization, right two figure are the binary map after the bilinearity OTSU binaryzation, a right figure be mate in right two figure and the template base after, the template that similarity is the highest, the coupling mark is Audi's template of 89 minutes, so this car mark single frames recognition result is Audi.
In embodiments of the present invention, as shown in figure 13, except the method step that comprises Fig. 1-embodiment illustrated in fig. 11, also comprise step S106, set up the vehicle tracking chain, the a plurality of cars that obtain described vehicle identify other result, and car mark type corresponding to mark soprano of selecting described a plurality of car to identify among the other result is that final car identifies other result.
Its concrete grammar is: the vehicle in coming into view is set up followed the tracks of chain, each node of following the tracks of on the chain has the car of this vehicle to identify other result, when vehicle satisfies the condition of publishing picture or rolls the visual field away from, add up single frames car on this tracking chain and identify other highest score and corresponding car mark type, the car mark that occurrence number is maximum and corresponding degree of confidence, if two types identical then identify other result for final car, otherwise in the maximum car mark of occurrence number, ask for an average degree of confidence, if the high confidence level of the first is higher than the certain mark of this average degree of confidence, by the highest final recognition result of car mark Sort positioning of mark occurring, be final recognition result otherwise select the maximum car mark type of occurrence number then.
Set up the tracking chain and carry out Optimum Matching, can improve matching precision, use the raising car and identify other precision.
Figure 14 is that the car of the embodiment of the invention is demarcated the structural representation of position recognition system.As shown in the figure, the car of a present embodiment demarcation position recognition system comprises:
Image collecting device 101 is for the headstock image of collection vehicle; The car mark is estimated band of position generating apparatus 102, is used for utilizing described headstock image, carries out vehicle location and car plate fixation and recognition, and according to the prior imformation of car mark and car plate, generation car mark is estimated the band of position; Car mark precise position information generating apparatus 103, be used for estimating the band of position at described car mark, disturb according to texture information filtering car mark background area, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark and carry out the car mark in the band of position and accurately locate by seeking the connected domain method, generate car mark precise position information; Car mark binary map generating apparatus 104 is used for according to described car mark precise position information, extracts car mark accurate greyscale figure, and described car mark accurate greyscale figure is carried out binaryzation, generates binary map corresponding to described car mark; Car identifies other as a result generating apparatus 105, is used for the binary map that described car mark is corresponding and mates with the car mark template that prestores, and is weighted calculating according to positional information, and car mark type corresponding to the mark soprano who calculates is that car identifies other result.
In the present embodiment, image collecting device can be high-definition camera, catch the headstock image that comes into view, 200W scene frame lower p.s.s 15, the image size is 1600*1200, be frames p.s.s 7 under the 500W scene, size is 2432*2048, and image resolution ratio can be adjusted according to the difference of high-definition camera.In the following step of the present embodiment, be that each two field picture of taking among the step S101 is processed, the car that generates corresponding each frame identifies other result.
In the present embodiment, as shown in figure 15, the car mark is estimated band of position generating apparatus 101 and is comprised:
Vehicle region generation unit 1012, be used on described gray level image, setting up background according to Gauss model, carry out vehicle location according to background subtraction and frame are poor, namely by adjacent three frames " do poor get with " vehicle that is undertaken is out confirmed, thereby generated the vehicle location zone.
License plate area generation unit 1013 is used in described vehicle location zone, carries out the car plate location according to texture information and the colouring information of vehicle, generates the car plate band of position.In the present embodiment, also comprise by Character segmentation and thick identification, then optimum car plate is accurately identified in the searching car plate location, if being lower than default recognition threshold, recognition confidence thinks that this vehicle does not contain effective board, the vehicle location result might have deviation, do not contain effective car plate and car mark zone in the vehicle region, otherwise think that this vehicle location is effective.
The car mark is estimated regional generation unit 1014, be used in the described car plate band of position, carry out car according to the prior imformation of described car mark and car plate and demarcate the position, generate the car mark and estimate the zone, wherein, the prior imformation of described car mark and car plate comprises positional information, car mark shape and the vehicle symmetry of car mark and car plate.
In the present embodiment, as shown in figure 16, car mark precise position information generating apparatus 102 comprises:
Texture fetch unit 1021 is used for estimating extraction sobel cross grain and vertical texture in the band of position at described car mark.
Background interference texture determining unit 1022, the average gray of the gray-scale map after being used for relatively extracting the gray-scale map behind the cross grain and extracting vertical texture is determined the grain direction that disturb described car mark background area by both differences.
Car mark texture gray-scale map generation unit 1023 is used for extracting the mutually anti-phase texture image of grain direction that disturbs with described car mark background area, and generation car mark is estimated the car mark texture gray-scale map in the band of position.In the present embodiment, (the texture threshold value generally gets 15 if both differences are greater than the texture threshold value of setting, certainly, can be other numerical value according to actual conditions), think that then car mark background texture is cross grain, estimate the interior vertical texture in zone, filtering heatsink transverse window interference etc., only remaining car mark texture information so should extract the car mark.After if the car mark is estimated extracted region transverse and longitudinal background texture, if vertically texture mean value is much larger than horizontal mean value, think that then background is vertical texture, just need to extract cross grain, vertical interference of texture of wiping out background and remaining car mark information only, if but cross grain is close with vertical texture average, then background is thought reticular texture, and will extract Texture this moment by the sobel filter operator.That is: extract the mutually anti-phase texture image of grain direction that disturbs with described car mark background area, estimate car mark texture gray-scale map in the band of position with generation car mark.
Car mark texture binary map generation unit 1024 is used for described car mark veining gray-scale map is carried out binary conversion treatment, and carries out morphologic filtering and corrosion expansion, generates binary map corresponding to described car mark texture gray-scale map.
Car mark boundary alignment unit 1025 is used for carrying out projection on binary map basis corresponding to described car mark veining gray-scale map, seeks border, the left and right sides corresponding to car mark, generates car mark precise position information.
In the present embodiment, as shown in figure 17, car mark binary map generating apparatus 103 comprises:
In the present embodiment, as shown in figure 18, car identifies other as a result generating apparatus 104 and comprises: match unit 1041 is used for carrying out the outline match by the car mark on binary map corresponding to the described car mark of hough transfer pair; Taxon 1042 is used for according to fitting result described car mark being classified; Matching unit 1043 is used for the car mark template of described car mark and described car mark corresponding types is mated.
In the present embodiment, on the basis of binary map, carry out the outline match by hough transfer pair car mark, simultaneously according to prioris such as car mark depth-width ratio shapes, be a class in ellipse, circle and the irregular figure with this car mark rough segmentation, then select the car mark masterplate of the type to mate.Car mark masterplate is the representative car mark binary map that extracts according to poor principle maximum, class interpolation maximum between class, size is 40*40, in order to save space complexity and time complexity, with eight synthetic bytes of binary map, the masterplate boil down to 40*5 of 40*40 size, the car of each type indicates three templates at present, and number can be adjusted according to actual conditions.
When carrying out car mark template matches, consider that the location is difficult to be accurate to the location of pixels on car mark border, and certain boundary interference information may be arranged also when choosing masterplate, so adopt the method for shiding matching, car mark binary map is upper, lower, left, can carry out the slip of 1 to 2 pixel on the right four direction, blank parts is mended 0 alignment, then eight synthetic bytes of binary map and template are mated, consider the car mark of same type, such as Ford and BYD, Kia, popular, Buick and benz etc., outline is identical all to be circular or oval, only have inside center information different, so when stencil matching, be weighted according to positional information, on the car mark binary map of 40*40,8 layers of weight of outermost are 1, and middle 16 layers of weight are 2, and the 16 layers of weight in center are 3, if not having coupling upper is zero, be set to respectively 1 minute according to the position weight difference after the match is successful, 2 minutes, 3 minutes is 0-100 minutes with score normalization, with all sort results behind the shiding matching at last, choose mark the highest identify other result for final single frames car, the single frames car is set identifies other degree of confidence, think identification effectively if high confidence level is higher than this value, otherwise refuse identification.
In the present invention, as shown in figure 19, car of the present invention is demarcated the position recognition system except the structure that comprises Figure 14-embodiment illustrated in fig. 18, also comprise multiframe car mark preferred embodiment 106, be used for setting up the vehicle tracking chain, the a plurality of cars that obtain described vehicle identify other result, and car mark type corresponding to mark soprano of selecting described a plurality of car to identify among the other result is that final car identifies other result.
This device is set up the vehicle in coming into view and is followed the tracks of chain, each node of following the tracks of on the chain has the car of this vehicle to identify other result, when vehicle satisfies the condition of publishing picture or rolls the visual field away from, add up single frames car on this tracking chain and identify other highest score and corresponding car mark type, the car mark that occurrence number is maximum and corresponding degree of confidence, if two types identical then identify other result for final car, otherwise in the maximum car mark of occurrence number, ask for an average degree of confidence, if the high confidence level of the first is higher than the certain mark of this average degree of confidence, by the highest final recognition result of car mark Sort positioning of mark occurring, be final recognition result otherwise select the maximum car mark type of occurrence number then.Set up the tracking chain and carry out Optimum Matching, can improve matching precision, use the raising car and identify other precision.
Figure 20 (a) and (b), (c), (d) are for utilizing car of the present invention to demarcate position recognition methods and system, the processing flow chart that the car mark of several vehicles is identified.Its procedure for displaying is matching template binary map → template base after the image → bilinearity OTSU binaryzation after car is demarcated position gray-scale map → slant correction.Wherein, Figure 20 (a) is to Audi's car car target identifying, and its coupling mark is 85 minutes; Figure 20 (b) is to popular car car target recognition result, and its coupling mark is 82 minutes; Figure 20 (c) is to BMW car target recognition result, and its coupling mark is 77 minutes; Figure 20 (d) is to Suzuki car car target recognition result, and its coupling mark is 78 minutes.And on to certain expressway, several frequently seen car mark is carried out Recognition test, the index of test is as follows:
Audi: 99%; Popular: 99%; Suzuki: 90%; BMW: 86%; Lucky: 85%; Benz: 83%; Modern: 89%; Toyota: 97%; Honda: 98%.Therefore, demarcation position recognition methods of the present invention and system can accurately locate and effectively identification the car mark.
The car of the embodiment of the invention is demarcated position recognition methods and system, under the prerequisite of effective board, can orient fast the car cursor position in automobile storage, and carries out single frames identification, multiframe affirmation, determines the uniqueness of vehicle and vehicle by car plate and car mark.
Above-described specific embodiment; purpose of the present invention, technical scheme and beneficial effect are further described; institute is understood that; the above only is specific embodiments of the invention; the protection domain that is not intended to limit the present invention; within the spirit and principles in the present invention all, any modification of making, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (14)
1. a car is demarcated the method that the position is identified, and it is characterized in that, described method comprises:
Utilize the headstock image of high-definition camera collection vehicle;
Utilize described headstock image, carry out vehicle location and car plate fixation and recognition, and according to the prior imformation of car mark and car plate, generate the car mark and estimate the band of position;
Estimate in the band of position at described car mark, disturb according to the texture information filtering car mark background area of extracting, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark and carry out the car mark in the band of position and accurately locate by seeking the connected domain method, generate car mark precise position information;
According to described car mark precise position information, extract car mark accurate greyscale figure, and described car mark accurate greyscale figure is carried out binaryzation, generate binary map corresponding to described car mark;
The binary map that described car mark is corresponding is mated with the car mark template that prestores, and is weighted calculating according to positional information, and car mark type corresponding to the mark soprano who calculates is that car identifies other result.
2. car according to claim 1 is demarcated the method for position identification, it is characterized in that, describedly utilizes described headstock image, carries out vehicle location and car plate location, and according to the prior imformation of car mark and car plate, generation car mark is estimated the step of the band of position, and it comprises:
Headstock image to described collection compresses in proportion, generates gray level image;
On described gray level image, set up background according to Gauss model, according to background subtraction and poor vehicle location, the generation vehicle location zone of carrying out of frame;
In described vehicle location zone, carry out the car plate fixation and recognition according to texture information and the colouring information of vehicle, generate the car plate band of position;
In the described car plate band of position, carry out car according to the prior imformation of described car mark and car plate and demarcate the position, generate the car mark and estimate the zone.
3. each described car is demarcated the method for position identification according to claim 1-2, it is characterized in that, the prior imformation of described car mark and car plate comprises the positional information of car mark and car plate, car mark shape and vehicle symmetry.
4. car according to claim 1 is demarcated the method for position identification, it is characterized in that, describedly estimate in the band of position at described car mark, disturb according to the texture information filtering car mark background area of extracting, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark by the connected domain method and carry out the car mark in the band of position and accurately locate, generate the step of car mark precise position information, comprising:
Estimate extraction sobel cross grain and vertical texture in the band of position at described car mark;
The average gray of the gray-scale map after relatively extracting the gray-scale map behind the cross grain and extracting vertical texture is determined the grain direction that disturb described car mark background area by both differences;
Extract the mutually anti-phase texture image of grain direction that disturbs with described car mark background area, generation car mark is estimated the car mark texture gray-scale map in the band of position;
Described car mark veining gray-scale map is carried out binary conversion treatment, and carry out morphologic filtering and corrosion expansion, generate binary map corresponding to described car mark texture gray-scale map;
Carry out projection on binary map basis corresponding to described car mark veining gray-scale map, seek border, the left and right sides corresponding to car mark, generate car mark precise position information.
5. car according to claim 1 is demarcated the method for position identification, it is characterized in that, and is described according to described car mark precise position information, extract car mark accurate greyscale figure, and described car mark accurate greyscale figure carried out binaryzation, and generate the step of binary map corresponding to described car mark, comprising:
According to the license plate sloped angle that generates in the described car plate fixation and recognition, described car mark accurate greyscale figure is carried out slant correction;
Car mark accurate greyscale figure after the described correction is carried out binary conversion treatment according to the method for bilinearity OTSU, generate binary map corresponding to described car mark.
6. car according to claim 5 is demarcated the method for position identification, it is characterized in that, the step that described binary map that described car mark is corresponding and the car mark template that prestores are mated comprises:
Carry out the outline match by the car mark on binary map corresponding to the described car mark of hough transfer pair;
According to fitting result described car mark is classified;
Select to mate with the car mark template of described car mark corresponding types.
7. car according to claim 1 is demarcated the method for position identification, it is characterized in that, after obtaining car and identifying other result, also comprises:
Set up the vehicle tracking chain, a plurality of cars that obtain described vehicle identify other result, and car mark type corresponding to mark soprano of selecting described a plurality of car to identify among the other result is that final car identifies other result.
8. a car is demarcated the system that the position is identified, and it is characterized in that, described system comprises:
Image collecting device is for the headstock image of collection vehicle;
The car mark is estimated band of position generating apparatus, is used for utilizing described headstock image, carries out vehicle location and car plate fixation and recognition, and according to the prior imformation of car mark and car plate, generation car mark is estimated the band of position;
Car mark precise position information generating apparatus, be used for estimating the band of position at described car mark, disturb according to the texture information filtering car mark background area of extracting, car mark texture gray-scale map after disturb filtering car mark background area carries out binaryzation, on the binary image that generates, estimate at described car mark and carry out the car mark in the band of position and accurately locate by seeking the connected domain method, generate car mark precise position information;
Car mark binary map generating apparatus is used for according to described car mark precise position information, extracts car mark accurate greyscale figure, and described car mark accurate greyscale figure is carried out binaryzation, generates binary map corresponding to described car mark;
Car identifies other as a result generating apparatus, is used for the binary map that described car mark is corresponding and mates with the car mark template that prestores, and is weighted calculating according to positional information, and car mark type corresponding to the mark soprano who calculates is that car identifies other result.
9. car according to claim 8 is demarcated the system of position identification, it is characterized in that, described car mark is estimated band of position generating apparatus and comprised:
Image compression unit is used for the headstock image of described collection is compressed in proportion, generates gray level image;
The vehicle region generation unit is used for setting up background according to Gauss model on described gray level image, according to background subtraction and poor vehicle location, the generation vehicle location zone of carrying out of frame;
The license plate area generation unit is used in described vehicle location zone, carries out the car plate fixation and recognition according to texture information and the colouring information of vehicle, generates the car plate band of position;
The car mark is estimated regional generation unit, is used in the described car plate band of position, carries out car according to the prior imformation of described car mark and car plate and demarcates the position, generates the car mark and estimates the zone.
10. each described car is demarcated the system of position identification according to claim 8-9, it is characterized in that, the prior imformation of described car mark and car plate comprises the positional information of car mark and car plate, car mark shape and vehicle symmetry.
11. car according to claim 8 is demarcated the system of position identification, it is characterized in that, described car mark precise position information generating apparatus comprises:
Texture fetch unit is used for estimating extraction sobel cross grain and vertical texture in the band of position at described car mark;
Background interference texture determining unit, the average gray of the gray-scale map after being used for relatively extracting the gray-scale map behind the cross grain and extracting vertical texture is determined the grain direction that disturb described car mark background area by both differences;
Car mark texture gray-scale map generation unit is used for extracting the mutually anti-phase texture image of grain direction that disturbs with described car mark background area, and generation car mark is estimated the car mark texture gray-scale map in the band of position;
Car mark texture binary map generation unit is used for described car mark veining gray-scale map is carried out binary conversion treatment, and carries out morphologic filtering and corrosion expansion, generates binary map corresponding to described car mark texture gray-scale map;
Car mark boundary alignment unit is used for carrying out projection on binary map basis corresponding to described car mark veining gray-scale map, seeks border, the left and right sides corresponding to car mark, generates car mark precise position information.
12. car according to claim 8 is demarcated the system of position identification, it is characterized in that, described car mark binary map generating apparatus is according to described car mark precise position information, extract car mark accurate greyscale figure, and described car mark accurate greyscale figure carried out binaryzation, generate binary map corresponding to described car mark, comprising:
The slant correction unit for the license plate sloped angle that generates according to described car plate fixation and recognition, carries out slant correction to described car mark accurate greyscale figure;
Car mark binary map generation unit is used for the car mark accurate greyscale figure after the described correction is carried out binary conversion treatment according to the method for bilinearity OTSU, generates binary map corresponding to described car mark.
13. car according to claim 8 is demarcated the system of position identification, it is characterized in that, described car identifies other as a result generating apparatus and comprises:
The match unit is used for carrying out the outline match by the car mark on binary map corresponding to the described car mark of hough transfer pair;
Taxon is used for according to fitting result described car mark being classified;
Matching unit is used for the car mark template of described car mark and described car mark corresponding types is mated.
14. car according to claim 8 is demarcated the system of position identification, it is characterized in that, described car is demarcated the position recognition device and is also comprised:
Multiframe car mark preferred embodiment is used for setting up the vehicle tracking chain, and a plurality of cars that obtain described vehicle identify other result, and car mark type corresponding to mark soprano of selecting described a plurality of car to identify among the other result is that final car identifies other result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310009960.XA CN103077384B (en) | 2013-01-10 | 2013-01-10 | A kind of method and system of vehicle-logo location identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310009960.XA CN103077384B (en) | 2013-01-10 | 2013-01-10 | A kind of method and system of vehicle-logo location identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103077384A true CN103077384A (en) | 2013-05-01 |
| CN103077384B CN103077384B (en) | 2016-08-31 |
Family
ID=48153908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310009960.XA Active CN103077384B (en) | 2013-01-10 | 2013-01-10 | A kind of method and system of vehicle-logo location identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103077384B (en) |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473774A (en) * | 2013-09-09 | 2013-12-25 | 长安大学 | Vehicle locating method based on matching of road surface image characteristics |
| CN103530622A (en) * | 2013-11-07 | 2014-01-22 | 青岛海信网络科技股份有限公司 | Car logo identification method |
| CN103544489A (en) * | 2013-11-12 | 2014-01-29 | 公安部第三研究所 | Device and method for locating automobile logo |
| CN103559492A (en) * | 2013-11-12 | 2014-02-05 | 公安部第三研究所 | Car logo recognition device and method |
| CN103577829A (en) * | 2013-11-08 | 2014-02-12 | 中安消技术有限公司 | Car logo positioning method and device |
| CN104268501A (en) * | 2014-03-31 | 2015-01-07 | 张文强 | Community entrance and exit vehicle monitoring system |
| CN104331691A (en) * | 2014-11-28 | 2015-02-04 | 深圳市捷顺科技实业股份有限公司 | Vehicle logo classifier training method, vehicle logo recognition method and device |
| CN104391966A (en) * | 2014-12-03 | 2015-03-04 | 中国人民解放军国防科学技术大学 | Typical car logo searching method based on deep learning |
| CN104504398A (en) * | 2014-12-31 | 2015-04-08 | 深圳市捷顺科技实业股份有限公司 | Vehicle logo recognition method and vehicle logo recognition device |
| CN105205486A (en) * | 2015-09-15 | 2015-12-30 | 浙江宇视科技有限公司 | Vehicle logo recognition method and device |
| CN105262174A (en) * | 2015-11-02 | 2016-01-20 | 国网山东省电力公司电力科学研究院 | Ordered electromobile charging system and method of intelligent park based on intelligent identification |
| CN105426896A (en) * | 2015-11-16 | 2016-03-23 | 成都神州数码索贝科技有限公司 | Car logo automatic identification method and system |
| CN105550684A (en) * | 2015-12-14 | 2016-05-04 | 广东安居宝数码科技股份有限公司 | Image-based license plate positioning method and system thereof |
| CN105760876A (en) * | 2016-03-18 | 2016-07-13 | 中山大学 | Vehicle logo location method based on radiator grid background ablation |
| CN105868755A (en) * | 2015-11-24 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Number separation method and apparatus |
| CN105868683A (en) * | 2015-11-24 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Channel logo identification method and apparatus |
| CN106250555A (en) * | 2016-08-15 | 2016-12-21 | 东方网力科技股份有限公司 | Vehicle retrieval method and devices based on big data |
| CN106650550A (en) * | 2015-10-28 | 2017-05-10 | 中通服公众信息产业股份有限公司 | Vehicle model identification method and vehicle model identification system through image characteristics of vehicle mark and vehicle head |
| WO2017088478A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Number separating method and device |
| CN104268501B (en) * | 2014-03-31 | 2018-02-09 | 张文强 | Cell gateway vehicle monitoring system |
| CN108154127A (en) * | 2017-12-27 | 2018-06-12 | 天津智芯视界科技有限公司 | A kind of vehicle identification method based on video and radar |
| CN108154510A (en) * | 2018-01-17 | 2018-06-12 | 深圳市亿图视觉自动化技术有限公司 | Method for detecting surface defects of products, device and computer readable storage medium |
| CN108196729A (en) * | 2018-01-16 | 2018-06-22 | 安徽慧视金瞳科技有限公司 | A kind of finger tip point rapid detection method based on infrared video |
| CN108268811A (en) * | 2018-01-15 | 2018-07-10 | 深圳市易成自动驾驶技术有限公司 | Image processing method, device and computer readable storage medium |
| CN108614985A (en) * | 2016-12-12 | 2018-10-02 | 广西师范大学 | A kind of vehicle-logo location method |
| CN109141829A (en) * | 2018-07-23 | 2019-01-04 | 北京大恒图像视觉有限公司 | A kind of detection method of target line sharpness computation and industrial camera rear cut-off distance |
| CN109214386A (en) * | 2018-09-14 | 2019-01-15 | 北京京东金融科技控股有限公司 | Method and apparatus for generating image recognition model |
| CN109241973A (en) * | 2018-08-21 | 2019-01-18 | 南京工程学院 | A kind of full-automatic soft dividing method of character under grain background |
| CN109360236A (en) * | 2018-10-26 | 2019-02-19 | 郑州轻工业学院 | A kind of floral disc of sunflower dimension measurement method |
| CN109658584A (en) * | 2018-12-14 | 2019-04-19 | 泰康保险集团股份有限公司 | A kind of bill bank slip recognition method and device |
| CN110969135A (en) * | 2019-12-05 | 2020-04-07 | 中南大学 | Vehicle logo recognition method in natural scene |
| CN111695565A (en) * | 2020-06-14 | 2020-09-22 | 荆门汇易佳信息科技有限公司 | Automobile mark accurate positioning method based on road barrier fuzzy image |
| CN111815545A (en) * | 2020-07-14 | 2020-10-23 | 南京信息工程大学 | Intelligent terminal-based color image rapid processing method |
| CN112789208A (en) * | 2018-11-09 | 2021-05-11 | Sk电信有限公司 | Apparatus and method for estimating vehicle position |
| CN113111883A (en) * | 2021-03-23 | 2021-07-13 | 浙江大华技术股份有限公司 | License plate detection method, electronic equipment and storage medium |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154122A (en) * | 2017-12-26 | 2018-06-12 | 天津天地伟业投资管理有限公司 | A kind of automobile logo identification method based on image |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080075367A1 (en) * | 2006-09-21 | 2008-03-27 | Microsoft Corporation | Object Detection and Recognition System |
| CN101246551A (en) * | 2008-03-07 | 2008-08-20 | 北京航空航天大学 | Fast license plate locating method |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
| CN101719217A (en) * | 2009-12-15 | 2010-06-02 | 中国地质大学（武汉） | Elastic relaxation algorithm-based vehicle type identification system and elastic relaxation algorithm-based vehicle type identification method |
| CN101789080A (en) * | 2010-01-21 | 2010-07-28 | 上海交通大学 | Detection method for vehicle license plate real-time positioning character segmentation |
| CN102521891A (en) * | 2011-12-06 | 2012-06-27 | 北京万集科技股份有限公司 | Method and system for detecting vehicle congestion at toll station |
-
2013
- 2013-01-10 CN CN201310009960.XA patent/CN103077384B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080075367A1 (en) * | 2006-09-21 | 2008-03-27 | Microsoft Corporation | Object Detection and Recognition System |
| CN101246551A (en) * | 2008-03-07 | 2008-08-20 | 北京航空航天大学 | Fast license plate locating method |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
| CN101719217A (en) * | 2009-12-15 | 2010-06-02 | 中国地质大学（武汉） | Elastic relaxation algorithm-based vehicle type identification system and elastic relaxation algorithm-based vehicle type identification method |
| CN101789080A (en) * | 2010-01-21 | 2010-07-28 | 上海交通大学 | Detection method for vehicle license plate real-time positioning character segmentation |
| CN102521891A (en) * | 2011-12-06 | 2012-06-27 | 北京万集科技股份有限公司 | Method and system for detecting vehicle congestion at toll station |
Cited By (51)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473774A (en) * | 2013-09-09 | 2013-12-25 | 长安大学 | Vehicle locating method based on matching of road surface image characteristics |
| CN103473774B (en) * | 2013-09-09 | 2017-04-05 | 长安大学 | A kind of vehicle positioning method based on pavement image characteristic matching |
| CN103530622A (en) * | 2013-11-07 | 2014-01-22 | 青岛海信网络科技股份有限公司 | Car logo identification method |
| CN103530622B (en) * | 2013-11-07 | 2017-03-29 | 青岛海信网络科技股份有限公司 | A kind of car logo identification method |
| CN103577829B (en) * | 2013-11-08 | 2016-04-20 | 中安消技术有限公司 | A kind of vehicle-logo location method and apparatus |
| CN103577829A (en) * | 2013-11-08 | 2014-02-12 | 中安消技术有限公司 | Car logo positioning method and device |
| CN103544489A (en) * | 2013-11-12 | 2014-01-29 | 公安部第三研究所 | Device and method for locating automobile logo |
| CN103559492A (en) * | 2013-11-12 | 2014-02-05 | 公安部第三研究所 | Car logo recognition device and method |
| CN104268501B (en) * | 2014-03-31 | 2018-02-09 | 张文强 | Cell gateway vehicle monitoring system |
| CN104268501A (en) * | 2014-03-31 | 2015-01-07 | 张文强 | Community entrance and exit vehicle monitoring system |
| CN104331691A (en) * | 2014-11-28 | 2015-02-04 | 深圳市捷顺科技实业股份有限公司 | Vehicle logo classifier training method, vehicle logo recognition method and device |
| CN104391966A (en) * | 2014-12-03 | 2015-03-04 | 中国人民解放军国防科学技术大学 | Typical car logo searching method based on deep learning |
| CN104391966B (en) * | 2014-12-03 | 2017-09-29 | 中国人民解放军国防科学技术大学 | Typical logo searching method based on deep learning |
| CN104504398A (en) * | 2014-12-31 | 2015-04-08 | 深圳市捷顺科技实业股份有限公司 | Vehicle logo recognition method and vehicle logo recognition device |
| CN104504398B (en) * | 2014-12-31 | 2017-12-01 | 深圳市捷顺科技实业股份有限公司 | Automobile logo identification method and vehicle-logo recognition device |
| CN105205486A (en) * | 2015-09-15 | 2015-12-30 | 浙江宇视科技有限公司 | Vehicle logo recognition method and device |
| CN105205486B (en) * | 2015-09-15 | 2018-12-07 | 浙江宇视科技有限公司 | A kind of automobile logo identification method and device |
| CN106650550A (en) * | 2015-10-28 | 2017-05-10 | 中通服公众信息产业股份有限公司 | Vehicle model identification method and vehicle model identification system through image characteristics of vehicle mark and vehicle head |
| CN105262174A (en) * | 2015-11-02 | 2016-01-20 | 国网山东省电力公司电力科学研究院 | Ordered electromobile charging system and method of intelligent park based on intelligent identification |
| CN105426896A (en) * | 2015-11-16 | 2016-03-23 | 成都神州数码索贝科技有限公司 | Car logo automatic identification method and system |
| WO2017088478A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Number separating method and device |
| WO2017088479A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Method of identifying digital on-screen graphic and device |
| CN105868683A (en) * | 2015-11-24 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Channel logo identification method and apparatus |
| CN105868755A (en) * | 2015-11-24 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Number separation method and apparatus |
| CN105550684A (en) * | 2015-12-14 | 2016-05-04 | 广东安居宝数码科技股份有限公司 | Image-based license plate positioning method and system thereof |
| CN105760876A (en) * | 2016-03-18 | 2016-07-13 | 中山大学 | Vehicle logo location method based on radiator grid background ablation |
| CN105760876B (en) * | 2016-03-18 | 2019-03-08 | 中山大学 | A kind of vehicle-logo location method based on the ablation of radiator grid background |
| CN106250555B (en) * | 2016-08-15 | 2019-07-09 | 东方网力科技股份有限公司 | Vehicle retrieval method and device based on big data |
| CN106250555A (en) * | 2016-08-15 | 2016-12-21 | 东方网力科技股份有限公司 | Vehicle retrieval method and devices based on big data |
| CN108614985A (en) * | 2016-12-12 | 2018-10-02 | 广西师范大学 | A kind of vehicle-logo location method |
| CN108154127A (en) * | 2017-12-27 | 2018-06-12 | 天津智芯视界科技有限公司 | A kind of vehicle identification method based on video and radar |
| CN108268811A (en) * | 2018-01-15 | 2018-07-10 | 深圳市易成自动驾驶技术有限公司 | Image processing method, device and computer readable storage medium |
| CN108268811B (en) * | 2018-01-15 | 2021-02-02 | 深圳市易成自动驾驶技术有限公司 | Image processing method, image processing apparatus, and computer-readable storage medium |
| CN108196729A (en) * | 2018-01-16 | 2018-06-22 | 安徽慧视金瞳科技有限公司 | A kind of finger tip point rapid detection method based on infrared video |
| CN108154510A (en) * | 2018-01-17 | 2018-06-12 | 深圳市亿图视觉自动化技术有限公司 | Method for detecting surface defects of products, device and computer readable storage medium |
| CN109141829B (en) * | 2018-07-23 | 2020-10-09 | 北京大恒图像视觉有限公司 | Target line definition calculation and industrial camera rear intercept detection method |
| CN109141829A (en) * | 2018-07-23 | 2019-01-04 | 北京大恒图像视觉有限公司 | A kind of detection method of target line sharpness computation and industrial camera rear cut-off distance |
| CN109241973A (en) * | 2018-08-21 | 2019-01-18 | 南京工程学院 | A kind of full-automatic soft dividing method of character under grain background |
| CN109241973B (en) * | 2018-08-21 | 2022-02-08 | 南京工程学院 | Full-automatic soft segmentation method for characters under texture background |
| CN109214386A (en) * | 2018-09-14 | 2019-01-15 | 北京京东金融科技控股有限公司 | Method and apparatus for generating image recognition model |
| CN109360236A (en) * | 2018-10-26 | 2019-02-19 | 郑州轻工业学院 | A kind of floral disc of sunflower dimension measurement method |
| CN109360236B (en) * | 2018-10-26 | 2022-05-17 | 郑州轻工业学院 | Sunflower disc size measurement method |
| CN112789208A (en) * | 2018-11-09 | 2021-05-11 | Sk电信有限公司 | Apparatus and method for estimating vehicle position |
| CN109658584A (en) * | 2018-12-14 | 2019-04-19 | 泰康保险集团股份有限公司 | A kind of bill bank slip recognition method and device |
| CN109658584B (en) * | 2018-12-14 | 2021-01-12 | 泰康保险集团股份有限公司 | Bill information identification method and device |
| CN110969135A (en) * | 2019-12-05 | 2020-04-07 | 中南大学 | Vehicle logo recognition method in natural scene |
| CN110969135B (en) * | 2019-12-05 | 2023-04-18 | 中南大学 | Vehicle logo recognition method in natural scene |
| CN111695565A (en) * | 2020-06-14 | 2020-09-22 | 荆门汇易佳信息科技有限公司 | Automobile mark accurate positioning method based on road barrier fuzzy image |
| CN111815545A (en) * | 2020-07-14 | 2020-10-23 | 南京信息工程大学 | Intelligent terminal-based color image rapid processing method |
| CN113111883A (en) * | 2021-03-23 | 2021-07-13 | 浙江大华技术股份有限公司 | License plate detection method, electronic equipment and storage medium |
| CN113111883B (en) * | 2021-03-23 | 2023-06-06 | 浙江大华技术股份有限公司 | License plate detection method, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103077384B (en) | 2016-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103077384A (en) | Method and system for positioning and recognizing vehicle logo | |
| CN103116751B (en) | A kind of Method of Automatic Recognition for Character of Lcecse Plate | |
| CN103235938B (en) | The method and system of car plate detection and indentification | |
| CN104778444B (en) | The appearance features analysis method of vehicle image under road scene | |
| CN101030256B (en) | Method and apparatus for cutting vehicle image | |
| CN104036262B (en) | A kind of method and system of LPR car plates screening identification | |
| CN103116985B (en) | Detection method and device of parking against rules | |
| CN106650553A (en) | License plate recognition method and system | |
| Li et al. | Nighttime lane markings recognition based on Canny detection and Hough transform | |
| CN103136528B (en) | A kind of licence plate recognition method based on dual edge detection | |
| CN104050450A (en) | Vehicle license plate recognition method based on video | |
| CN105426864A (en) | Multiple lane line detecting method based on isometric peripheral point matching | |
| CN110414355A (en) | The right bit sky parking stall of view-based access control model and parking stall line detecting method during parking | |
| CN105488454A (en) | Monocular vision based front vehicle detection and ranging method | |
| CN107316486A (en) | Pilotless automobile visual identifying system based on dual camera | |
| CN104700072A (en) | Lane line historical frame recognition method | |
| CN105528609A (en) | Vehicle license plate location method based on character position | |
| CN107886034B (en) | Driving reminding method and device and vehicle | |
| CN102419820A (en) | Method for rapidly detecting car logo in videos and images | |
| CN104809433A (en) | Zebra stripe detection method based on maximum stable region and random sampling | |
| CN112287888B (en) | Track turning recognition method based on predictive weight | |
| CN112001216A (en) | Automobile driving lane detection system based on computer | |
| CN109886168B (en) | Ground traffic sign identification method based on hierarchy | |
| CN105718916A (en) | Lane line detection method based on Hough transform | |
| CN112949484B (en) | High-speed railway rockfall real-time detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Ge Leiming Inventor after: Fang Yanming Inventor after: Cheng Yujuan Inventor after: Wang Qin Inventor before: Ge Leiming Inventor before: Fang Yanming Inventor before: Cheng Yujuan |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: GE LEIMING FANG YANMING CHENG YUJUAN TO: GE LEIMING FANG YANMING CHENG YUJUAN WANG QIN |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221103 Address after: Room 601, 6th Floor, Building B, Building 9, Start Zone, Yangtze River Delta International R&D Community, No. 286, Qinglonggang Road, High-speed Rail New Town, Xiangcheng District, Suzhou City, Jiangsu Province, 215131 Patentee after: Suzhou Wanji IoV Technology Co.,Ltd. Address before: 100085 No. 5, building 1, East Road, Haidian District, Beijing, 601 Patentee before: BEIJING WANJI TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |