CN106022268A - Identification method and device of speed limiting sign - Google Patents
Identification method and device of speed limiting sign Download PDFInfo
- Publication number
- CN106022268A CN106022268A CN201610343756.5A CN201610343756A CN106022268A CN 106022268 A CN106022268 A CN 106022268A CN 201610343756 A CN201610343756 A CN 201610343756A CN 106022268 A CN106022268 A CN 106022268A
- Authority
- CN
- China
- Prior art keywords
- image
- speed limit
- mark
- edge
- edge pattern
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides an identification method and device of a speed limiting sign. The method comprises that one frame of image, which is shot by a camera, of the road condition is obtained; an effective part is captured from the image, the captured image serves as an image to be identified, edge extraction is carried out an object in the image to be identified, an edge image including an edge figure of the object is obtained, and whether the object is a traffic sign is determined according to the edge image; if the object is determined to be the traffic sign and traffic sign is determined to be the speed limiting sign, an edge figure of the speed limiting sign is segmented from the edge image to for a speed-limiting sign image, and according to the edge figure of the speed limiting sign, interference is removed from the speed-limiting sign image; and numbers in the edge figure in the speed-limiting sign image from which interference is removed are identified. According to the invention, the identification rate of the speed limiting sign is improved.
Description
Technical field
The present invention relates to field of image recognition, particularly relate to recognition methods and the device of a kind of speed limit mark.
Background technology
Driver is in driving vehicle processes, sometimes due to attention compares concentration, can ignore the speed limit mark in roadside, thus
Hypervelocity or car speed is caused not to mate with road；Or owing to speed is very fast, the speed limit mark in unsighted roadside, cause surpassing
Speed.
Prior art provides a lot of method pointing out speed, as by vehicle location or image recognition speed limit mark
Knowing, comparatively speaking, the accuracy rate that the speed that the method identified by image recognition speed limit is provided limits prompting is some higher, but
At present in the method for image recognition speed limit mark, there is also and cause the highest the asking of discrimination because of the problem such as light, shelter
Topic.
Therefore, it is necessary to improvement drawbacks mentioned above.
Summary of the invention
Based on problem above, the present invention proposes the recognition methods of a kind of speed limit mark, by obtaining camera head shooting
The image of road conditions, intercepts the live part of image, and making the image after intercepting is image to be identified, in image to be identified
Object carry out edge extracting, it is thus achieved that comprise the edge image of the edge pattern of object, according to edge pattern judgment object whether
For traffic mark, if judgment object is traffic mark, and if judge that traffic mark is speed limit mark, then will limit from edge image
The edge pattern of speed mark splits, and forms speed limit identification image, according to the edge pattern of speed limit mark, speed limit is identified figure
Processing as carrying out interference, the numeral within the edge pattern in speed limit identification image after processing going interference is identified,
The numeral identified is the mode of the speed limit numeral of speed limit mark correspondence, makes to improve the discrimination of speed limit mark.
On the one hand, the present invention proposes the recognition methods of a kind of speed limit mark, including:
Obtain image step, including: obtaining a two field picture, described image is the image of the road conditions shot by camera head；
Judge traffic mark step, including: intercepting the live part of described image, the image after intercepting is figure to be identified
Picture, carries out edge extracting to the object in described image to be identified, it is thus achieved that comprise the edge image of the edge pattern of described object,
Judge whether described object is traffic mark according to described edge pattern, including at least a described thing in described image to be identified
Body；
Judge speed limit identification of steps, including: if judging, described object is described traffic mark, and if judging that described traffic mark is
Described speed limit identifies, then the edge pattern that described speed limit identifies split from described edge image, forms speed limit mark
Image, goes interference to process described speed limit identification image according to the edge pattern that described speed limit identifies；
Identification step, including: the numeral within the edge pattern in described speed limit identification image after processing going interference is carried out
Identifying, the numeral identified is the speed limit numeral that described speed limit mark is corresponding.
Additionally, live part to described image described in described judgement traffic mark step carries out intercepting and specifically includes:
The more than vanishing line live part that image is described image in described image, intercepts image more than vanishing line.
Judge to judge whether described object is traffic according to described edge pattern described in traffic mark step additionally, described
Mark specifically includes: judge according to the ratio of width to height in the size in the surrounded region of described edge pattern, the shape in region and/or region
Whether described object is described traffic mark.
If additionally, judging described in described judgement speed limit identification of steps that described traffic mark is the concrete bag of described speed limit mark
Include: the edge pattern outward expansion pixel to described traffic mark, after the method for employing support vector machine is to outward expansion pixel
Described edge pattern be identified, if identify described edge pattern be described speed limit mark, the most described traffic mark is institute
State speed limit mark.
Additionally, the described edge pattern outward expansion pixel to described traffic mark specifically includes: to described traffic mark
1 to 5 pixel of edge pattern outward expansion.
Additionally, the edge pattern identified according to described speed limit described in described judgement speed limit identification of steps is to described speed limit mark
Knowledge image carries out interference process and specifically includes: the edge pattern of described speed limit mark is circle, calculates the radius of described circle,
Determining the center of circle of described circle, by described speed limit identification image, the image beyond circle being made up of the center of circle and radius removes.
Additionally, described identification step specifically includes: the image after described speed limit identification image goes interference process is for treating threshold
Value process image, treat that threshold process image carries out histogrammic numerical statistic to described, according to the maximum in numerical statistic with
Minima calculates threshold value, utilizes described threshold value to treat that threshold process image carries out threshold process to described, obtains described speed limit mark
Binary image, described binary image is carried out numeral identification, obtains the described speed limit numeral that described speed limit mark is corresponding.
Additionally, also include pointing out step, including: to described by the way of voice broadcast and/or by the way of display
Speed limit numeral is pointed out.
Additionally, camera head is arranged on the front windshield of its place vehicle described in described acquisition image step.
On the other hand, the present invention proposes the identification device of a kind of speed limit mark, including:
Obtaining image module, be used for: obtain a two field picture, described image is the image of the road conditions shot by camera head；
Judging traffic mark module, be used for: intercept the live part of described image, the image after intercepting is figure to be identified
Picture, carries out edge extracting to the object in described image to be identified, it is thus achieved that comprise the edge image of the edge pattern of described object,
Judge whether described object is traffic mark according to described edge pattern, including at least a described thing in described image to be identified
Body；
Judging that speed limit identifies module, being used for: described object is described traffic mark if judging, and if judging that described traffic mark is
Described speed limit identifies, then the edge pattern that described speed limit identifies split from described edge image, form described speed limit
Identification image, goes interference to process described speed limit identification image according to the edge pattern that described speed limit identifies；
Identification module, is used for: the numeral within the edge pattern in described speed limit identification image after processing going interference is carried out
Identifying, the numeral identified is the speed limit numeral that speed limit mark is corresponding.
Use technique scheme, have the advantages that
By obtaining the image of the road conditions of camera head shooting, the live part of image is intercepted, makes the image after intercepting
For image to be identified, the object in image to be identified is carried out edge extracting, it is thus achieved that comprise the edge graph of the edge pattern of object
Whether picture, be traffic mark according to edge pattern judgment object, if judgment object is traffic mark, and if judging that traffic mark is
Speed limit identifies, then the edge pattern that speed limit identifies split from edge image, forms speed limit identification image, according to speed limit
Speed limit identification image is gone interference to process by the edge pattern of mark, the limit in speed limit identification image after processing going interference
Numeral within edge figure is identified, and the numeral identified is the mode of the speed limit numeral of speed limit mark correspondence, makes speed limit
The discrimination of mark improves.
Accompanying drawing explanation
Fig. 1 is the flow chart of the recognition methods of speed limit mark according to an embodiment of the invention；
Fig. 2 is the schematic diagram of the image of middle according to a further embodiment of the invention camera head shooting；
Fig. 3 be according to a further embodiment of the invention in the image more than vanishing line of image is intercepted after the showing of image
It is intended to；
Fig. 4 be according to a further embodiment of the invention in the schematic diagram removing the speed limit identification image behind corner；
Fig. 5 be according to a further embodiment of the invention in schematic diagram to binary image；
Fig. 6 be according to a further embodiment of the invention in be syncopated as numeral schematic diagram；
Fig. 7 is the schematic diagram that incomplete situation occurs in middle according to a further embodiment of the invention numeral top；
Fig. 8 is the schematic diagram that incomplete situation occurs in middle according to a further embodiment of the invention numeral middle part；
Fig. 9 is the schematic diagram occurring adhesion between middle according to a further embodiment of the invention numeral；
Figure 10 is the flow chart of the recognition methods of speed limit mark in accordance with another embodiment of the present invention；
Figure 11 is the block diagram identifying device of speed limit mark in accordance with another embodiment of the present invention.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
The a part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under not making creative work premise, broadly falls into the scope of protection of the invention.
With reference to Fig. 1, the present invention proposes the recognition methods of a kind of speed limit mark, including:
Obtain image step S001, including: obtaining a two field picture, image is the image of the road conditions shot by camera head；
Judge traffic mark step S002, including: intercepting the live part of image, the image after intercepting is figure to be identified
Picture, carries out edge extracting to the object in image to be identified, it is thus achieved that comprise the edge image of the edge pattern of object, according to edge
Whether figure judgment object is traffic mark, including at least an object in image to be identified；
Judge speed limit identification of steps S003, including: if judgment object is traffic mark, and if judging that traffic mark is speed limit mark
Know, then from edge image, the edge pattern that speed limit identifies is split, form speed limit identification image, according to speed limit mark
Speed limit identification image is gone interference to process by edge pattern；
Identification step S004, including: the numeral within the edge pattern in speed limit identification image after processing going interference is carried out
Identifying, the numeral identified is the speed limit numeral that speed limit mark is corresponding.
Obtaining in image step S001, obtain a two field picture, a two field picture is to obtain from the continuously recording of shooting road conditions
A two field picture, or be from shooting road conditions discontinuous image obtain an image.Filming apparatus is photographic head,
In one of them embodiment, photographic head is arranged on the anterior windshield of vehicle, in one of which embodiment, takes the photograph
The position of the centre of the anterior windshield of vehicle it is arranged on as head.
Judge traffic mark step S002, when the road conditions before vehicle heading are shot by photographic head, if
Photographic head install for wide-angle lens, then can photograph the image of below vanishing line, vanishing line refers to: modified line in perspectivity of focus
Extend to the line segment between vanishing point.Fig. 2 is the image that camera head photographs, and the image below vanishing line does not comprise the friendship in roadside
Logical mark, so need not this parts of images is carried out image procossing, needing to remove this parts of images, only taking more than vanishing line
The original image that processes as successive image of image.Fig. 3 is the image after intercepting the image more than vanishing line of image,
Image after intercepting is image to be identified.
And during because image is carried out edge extracting, if image is the biggest, its required time consumed is the longest.At it
In an embodiment in, the two field picture obtained from filming apparatus is intercepted, the image after intercepting is vanishing line and disappear
Lose the image in region between the above 50mm of line.Thus save the process time to image.
Image to be identified is first carried out noise reduction process, as used Gauss noise reduction process.
Object in image to be identified after noise reduction is carried out edge extracting, and the method for edge extracting includes: Roberts
Cross operator, Prewitt operator, Sobel operator, Kirsch operator, compass operator；Marr-Hildreth, Canny calculate
Son, Laplacian operator etc..
Canny operator is conventional edge detection method, is characterized in that the candidate pixel attempted independent limit is assembled into wheel
Wide.
John Canny proposed Canny operator, it and Marr(LoG in 1986) edge detection method is similar, falls within
It is first to smooth the method differentiated afterwards.
John Canny have studied the characteristic needed for optimal edge detection method, gives evaluation edge detection performance good and bad
Three indexs:
First, good signal to noise ratio, will be judged to that the probability of marginal point is low by non-edge point, marginal point is judged to non-edge point
Probability is low；
Second, high positioning performance, the marginal point i.e. detected will be as far as possible at the center of actual edge；
3rd, single edge is only had unique response, the probability that the most single edge produces multiple responses is low, and false response
Edge should obtain maximum suppression.
Canny operator wishes, while improving the sensitivity to scenery edge, the method for noise to be suppressed to be only
Edge extracting method.
In one of which embodiment, Canny operator is utilized to ask marginal point specific algorithm step as follows:
Step S201, uses Gaussian filter smoothed image.
Step S202, with single order local derviation finite difference formulations gradient magnitude and direction.
Step S203, carries out non-maxima suppression to gradient magnitude.
Step S204, by the detection of dual threshold algorithm and adjoining edge.
In the function library of some image procossing, the method to various edge extractings has carried out realization and the envelope of function
Dress, as a example by Canny operator, in OpenCV function library, the function of its correspondence is cvCanny.
During because using camera head shooting road conditions, the image of shooting would generally comprise multiple object: as other vehicle,
The direction board in roadside, pedestrian or trees etc., so including at least an object in image to be identified, generally comprising plural
Object.
In judging traffic mark step S002, it is thus achieved that after the edge image of the edge pattern comprising object, according to edge
Whether figure judgment object is traffic mark.
As gone to judge, go to sentence according to the shape in region that edge pattern surrounds according to the size in region that edge pattern surrounds
Break or go to judge according to the ratio of width to height in the surrounded region of edge pattern.As: preset the wide high proportion in region 0.9-
Numerical value between 1.2, meeting edge pattern preliminary judgement corresponding to wide high proportion region between 0.9-1.2 is traffic mark
Know.In one of which embodiment, make above two or three kinds of methods combine and go to judge, it is judged that result can be more accurate.
After preliminary judgement is traffic mark, next edge image is identified, in one of which embodiment, adopts
By support vector machine, edge pattern is identified, real traffic mark is identified.
In judging speed limit identification of steps S003, when judgment object is traffic mark, the most next judge this traffic mark
Whether know is speed limit mark.In one of which embodiment, the edge pattern outward expansion pixel to traffic mark, use and prop up
Edge pattern after outward expansion pixel is identified by the method holding vector machine, if identifying edge pattern is speed limit mark,
Then traffic mark is speed limit mark.Because the contour pattern of traffic mark is outline figure in the template used in support vector machine
Picture, and the edge pattern in edge image is the Internal periphery of object, so needing the profile of the edge pattern of object to extending out
Several pixels so that edge pattern is the outline of object, thus with the traffic mark in the template in support vector machine
Contour pattern mates, to improve discrimination.In one of which embodiment, by the profile of the edge pattern of object to extending out 1-
5 pixels.In one of which embodiment, by the profile of the edge pattern of object to extending out 3 pixels.
Next judge whether traffic mark is speed limit mark, first speed limit mark is split from edge image, shape
Become speed limit identification image.In one of which embodiment, it is 50*50 pixel size that speed limit identification image is unified size.
Although speed limit mark is the most circular and two numerals, but in the image of actual photographed, speed limit identifies around
There may be leaves or some other shelter, need from speed limit identification image, remove these shelters, simultaneously as
Speed limit is designated circle, so being also required to remove the corner of speed limit identification image.In one of which embodiment, calculate circle
The radius of shape, determines the center of circle of circle, and by speed limit identification image, the image beyond circle being made up of the center of circle and radius removes.
Thus being eliminated the corner of original speed limit identification image, complete interference and process, Fig. 4 is the speed limit mark after removing corner
Know image.
In identification step S004, the numeral within the edge pattern in speed limit identification image after processing going interference is carried out
Identifying, the numeral identified is the speed limit numeral that speed limit mark is corresponding.
Image after speed limit identification image goes interference process, for treating threshold process image, will be treated to limit in threshold process image
The circle of speed mark removes, and then treats threshold process image when carrying out histogrammic numerical statistic, is divided into by threshold process image
Four regions are carried out histogrammic numerical statistic by four regions: upper left, upper right, lower-left and bottom right respectively.Find statistics Nogata
Numerical value maximum inside figure, finds numerical value minimum inside statistic histogram, with the numerical value * 0.5 that maximum numerical value * 0.5+ is minimum
Calculate threshold value.Utilize threshold value to treat threshold process image and carry out threshold process, obtain the binary image of speed limit mark, Fig. 5
For binary image.To binary image, do upright projection and floor projection finds the corresponding position of numeral, be syncopated as number
Word, Fig. 6 is the numeral being syncopated as, to the numeral being syncopated as, utilize afterwards 10 class models that support vector machine trains (0,1,2,3,
4,5,6,7,8,9) to numeral identification, the speed limit numeral that speed limit mark is corresponding is obtained.
When numeral is identified, it sometimes appear that the numeral of incompleteness, for numeral, incomplete situation occurs, for not
Same situation does different corresponding process, the situation as shown in Fig. 7, needs incomplete digital height and the highest numeral
Height alignment supplement.Situation as shown in Fig. 8, there is fracture in number glyphs, now with the peak of numeral and
Low spot is the height of numeral.
In the case of threshold value is chosen improperly by prior art, it may appear that the problem of Characters Stuck, as shown in Fig. 9.And adopt
By the method in the present embodiment, after intercepting out by numeral from artwork, re-use threshold value and speed limit mark is carried out at binaryzation
Reason, then the problem not havinging adhesion.
By obtaining the image of the road conditions of camera head shooting, the live part of image is intercepted, after making intercepting
Image is image to be identified, and the object in image to be identified is carried out edge extracting, it is thus achieved that comprise the limit of the edge pattern of object
Whether edge image, be traffic mark according to edge pattern judgment object, if judgment object is traffic mark, and if judging traffic mark
Know and identify for speed limit, then from edge image, the edge pattern that speed limit identifies is split, form speed limit identification image, according to
Speed limit identification image is gone interference to process by the edge pattern of speed limit mark, in the speed limit identification image after processing going interference
The numeral within edge pattern be identified, the numeral identified is the mode of speed limit numeral corresponding to speed limit mark, and it is right to make
The discrimination of speed limit mark improves.
In one of which embodiment, it is judged that in traffic mark step, the live part to image intercepts concrete bag
Include: the more than vanishing line live part that image is image in image, image more than vanishing line is intercepted.Vanishing line with
Under image do not comprise the traffic mark in roadside, so need not this parts of images is carried out image procossing, need this part
Image removes, and only takes the original image that the image of more than vanishing line processes as successive image.And because image is carried out edge
During extraction, if image is the biggest, its required time consumed is the longest.In one of which embodiment, will be from filming apparatus
The two field picture obtained intercepts, and the image after intercepting is the image in region between vanishing line and the above 50mm of vanishing line.Logical
Cross and image more than vanishing line is intercepted, save the process time to image.
In one of which embodiment, it is judged that whether traffic mark step is traffic according to edge pattern judgment object
Mark specifically includes: according to the size in the surrounded region of edge pattern, the shape in region and/or the ratio of width to height judgment object in region
Whether it is traffic mark.By the method in the present embodiment, object is tentatively judged, thus save follow-up identification time
Between, improve the efficiency of identification.
In one of which embodiment, it is judged that if judging in speed limit identification of steps, traffic mark is the concrete bag of speed limit mark
Include: the edge pattern outward expansion pixel to traffic mark, use the method for support vector machine to the limit after outward expansion pixel
Edge figure is identified, if identifying edge pattern is speed limit mark, then traffic mark is speed limit mark.
In one of which embodiment, the edge pattern outward expansion pixel of traffic mark is specifically included: to traffic
1 to 5 pixel of edge pattern outward expansion of mark.
Because the contour pattern of traffic mark is outer profile image in the template used in support vector machine, and edge image
In the Internal periphery that edge pattern is object, so needing the profile of the edge pattern of object to extending out several pixel, with
The outline making edge pattern be object, thus mate with the contour pattern of the traffic mark in the template in support vector machine,
To improve discrimination.In one of which embodiment, by the profile of the edge pattern of object to extending out 1-5 pixel.?
In one of them embodiment, by the profile of the edge pattern of object to extending out 3 pixels.
In one of which embodiment, it is judged that in speed limit identification of steps, the edge pattern according to speed limit mark is to speed limit mark
Knowledge image carries out interference process and specifically includes: the edge pattern of speed limit mark is circle, and the radius of calculating circle determines circle
The center of circle, by speed limit identification image, the image beyond circle being made up of the center of circle and radius removes.It is determined by the center of circle and radius
Method, speed limit identification image has carried out going the process of corner and interference.
In one of which embodiment, identification step specifically includes: the figure after speed limit identification image goes interference process
As for treating threshold process image, treating threshold process image and carry out histogrammic numerical statistic, according to the maximum in numerical statistic
Value calculates threshold value with minima, utilizes threshold value to treat threshold process image and carries out threshold process, obtains the binaryzation of speed limit mark
Image, carries out numeral identification to binary image, obtains the speed limit numeral that speed limit mark is corresponding.Find inside statistic histogram
Big numerical value, finds numerical value minimum inside statistic histogram, and the numerical value * 0.5 minimum with maximum numerical value * 0.5+ calculates threshold
Value.Utilize this threshold value to treat threshold process image and carry out binaryzation, make to treat that the numeral in threshold process image does not haves adhesion
Situation.
In one of which embodiment, also include pointing out step, including: by the way of voice broadcast and/or pass through
Speed limit numeral is pointed out by the mode of display.Driver is carried by the way of voice broadcast by the numeral that will identify that
Show, make driver it is known that the speed limit of current road segment.After utilizing the method that speed limit mark is identified, driver is entered
The lower error rate of row speed limit prompting.
In one of which embodiment, obtain camera head in image step and be arranged on the front windshield glass of its place vehicle
On glass.Camera head is arranged on the centre position of front windshield, makes camera head the carrying out at maximum visual angle to shoot, remove
Beyond shooting traffic mark, it is also possible to as other purposes.
With reference to Figure 10, the flow process of one embodiment of the invention is described.
Step S110, obtains a two field picture from the image of the road conditions of camera head shooting；
Step S111, saves as pending image by the image in region between above to vanishing line in image and vanishing line 50mm；
Step S112, carries out Gauss noise reduction process to pending image, and the object in the pending image after noise reduction is carried out limit
Edge detects, and after rim detection, obtains edge image, the corresponding respective edge pattern of each object in edge image；
Step S113, the ratio of width to height surrounding region according to the edge pattern that object is corresponding primarily determines that object is traffic mark；
By support vector machine, step S114, judges that object is as traffic identification；
Step S115, it is determined that this traffic mark is speed limit mark；
Step S116, splits speed limit mark from edge image, forms speed limit identification image, united by speed limit identification image
One size is 50*50 pixel size；
Step S117, removes the corner of speed limit identification image, removes the circle of speed limit mark, obtains treating threshold process image；
Step S118, treats threshold process image when carrying out histogrammic numerical statistic, threshold process image is divided into four districts
Four regions are carried out histogrammic numerical statistic and find inside statistic histogram by territory: upper left, upper right, lower-left and bottom right respectively
Maximum numerical value, finds numerical value minimum inside statistic histogram, and the numerical value * 0.5 minimum with maximum numerical value * 0.5+ calculates
Threshold value.Utilize this threshold value to treat threshold process image and carry out binaryzation；
Step S119, to binary image, does upright projection and floor projection finds the corresponding position of numeral, be syncopated as number
Word, 10 class models (0,1,2,3,4,5,6,7,8,9) utilizing support vector machine to train afterwards carry out numeral to binary image
Identify, obtain the speed limit numeral that speed limit mark is corresponding；
Step S120, points out driver by the way of speech play.
With reference to Figure 11, the present invention also proposes the identification device of a kind of speed limit mark, including:
Obtaining image module 10, be used for: obtain a two field picture, image is the image of the road conditions shot by camera head；
Judging traffic mark module 20, be used for: intercept the live part of image, the image after intercepting is figure to be identified
Picture, carries out edge extracting to the object in image to be identified, it is thus achieved that comprise the edge image of the edge pattern of object, according to edge
Whether figure judgment object is traffic mark, including at least an object in image to be identified；
Judge speed limit identify module 30, be used for: if judgment object is traffic mark, and if judge traffic mark be speed limit identify,
Then from edge image, the edge pattern that speed limit identifies is split, form speed limit identification image, according to the limit of speed limit mark
Speed limit identification image is gone interference to process by edge figure；
Identification module 40, is used for: the numeral within the edge pattern in speed limit identification image after processing going interference is known
, the numeral not identified is the speed limit numeral that speed limit mark is corresponding.
In one of which embodiment, it is judged that in traffic mark module 20, the live part to image carries out intercepting specifically
Including: the more than vanishing line live part that image is image in image, image more than vanishing line is intercepted.
In one of which embodiment, it is judged that whether traffic mark module 20 is friendship according to edge pattern judgment object
Logical mark specifically includes: judge thing according to the length-width ratio of the size in the surrounded region of edge pattern, the shape in region and/or region
Whether body is traffic mark.
In one of which embodiment, it is judged that if judging in speed limit mark module 30, traffic mark is that speed limit mark is concrete
Including: the edge pattern outward expansion pixel to traffic mark, after the method for employing support vector machine is to outward expansion pixel
Edge pattern is identified, if identifying edge pattern is speed limit mark, then traffic mark is speed limit mark.
In one of which embodiment, the edge pattern outward expansion pixel of traffic mark is specifically included: to traffic
1 to 5 pixel of edge pattern outward expansion of mark.
In one of which embodiment, it is judged that in speed limit mark module 30, the edge pattern according to speed limit mark is to speed limit
Identification image carries out interference process and specifically includes: the edge pattern of speed limit mark is circle, calculates the radius of circle, determines circle
The center of circle of shape, by speed limit identification image, the image beyond circle being made up of the center of circle and radius removes.
In one of which embodiment, identification module 40 specifically includes: after speed limit identification image goes interference process
Image, for treating threshold process image, is treated threshold process image and is carried out histogrammic numerical statistic, according in numerical statistic
Big value calculates threshold value with minima, utilizes threshold value to treat threshold process image and carries out threshold process, obtains the two-value of speed limit mark
Change image, binary image is carried out numeral identification, obtain the speed limit numeral that speed limit mark is corresponding.
In one of which embodiment, also include reminding module, be used for: by the way of voice broadcast and/or pass through
Speed limit numeral is pointed out by the mode of display.
In one of which embodiment, obtain camera head in image module and be arranged on the front windshield glass of its place vehicle
On glass.
Device embodiment described above is only schematically, and the wherein said unit illustrated as separating component can
To be or to may not be physically separate, the parts shown as unit can be or may not be physics list
Unit, i.e. may be located at a place, or can also be distributed on multiple NE.Can be selected it according to the actual needs
In some or all of module realize the purpose of the present embodiment scheme.Those of ordinary skill in the art are not paying creativeness
Work in the case of, be i.e. appreciated that and implement.
Through the above description of the embodiments, those skilled in the art it can be understood that to each embodiment can
The mode adding required general hardware platform by software realizes, naturally it is also possible to pass through hardware.Based on such understanding, on
State the part that prior art contributes by technical scheme the most in other words to embody with the form of software product, should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD etc., including some fingers
Make with so that a computer equipment (can be personal computer, server, or the network equipment etc.) performs each and implements
The method described in some part of example or embodiment.
Last it is noted that above example is only in order to illustrate technical scheme, it is not intended to limit；Although
With reference to previous embodiment, the present invention is described in detail, it will be understood by those within the art that: it still may be used
So that the technical scheme described in foregoing embodiments to be modified, or wherein portion of techniques feature is carried out equivalent；
And these amendment or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (10)
1. the recognition methods of a speed limit mark, it is characterised in that including:
Obtain image step, including: obtaining a two field picture, described image is the image of the road conditions shot by camera head；
Judge traffic mark step, including: intercepting the live part of described image, the image after intercepting is figure to be identified
Picture, carries out edge extracting to the object in described image to be identified, it is thus achieved that comprise the edge image of the edge pattern of described object,
Judge whether described object is traffic mark according to described edge pattern, including at least a described thing in described image to be identified
Body；
Judge speed limit identification of steps, including: if judging, described object is described traffic mark, and if judging that described traffic mark is
Described speed limit identifies, then the edge pattern that described speed limit identifies split from described edge image, forms speed limit mark
Image, goes interference to process described speed limit identification image according to the edge pattern that described speed limit identifies；
Identification step, including: the numeral within the edge pattern in described speed limit identification image after processing going interference is carried out
Identifying, the numeral identified is the speed limit numeral that described speed limit mark is corresponding.
The recognition methods of speed limit the most according to claim 1 mark, it is characterised in that:
Live part to described image described in described judgement traffic mark step carries out intercepting and specifically includes: in described image
Image more than vanishing line is the live part of described image, intercepts image more than vanishing line.
The recognition methods of speed limit the most according to claim 1 mark, it is characterised in that:
Described judge to judge whether described object is that traffic mark is concrete according to described edge pattern described in traffic mark step
Including: judge described object according to the ratio of width to height of the size in the surrounded region of described edge pattern, the shape in region and/or region
Whether it is described traffic mark.
The recognition methods of speed limit the most according to claim 1 mark, it is characterised in that:
If judging described in described judgement speed limit identification of steps, described traffic mark is that described speed limit mark specifically includes: to described
The edge pattern outward expansion pixel of traffic mark, uses the method for support vector machine to the described edge after outward expansion pixel
Figure is identified, if identifying described edge pattern is described speed limit mark, the most described traffic mark is described speed limit mark.
The recognition methods of speed limit the most according to claim 4 mark, it is characterised in that:
The described edge pattern outward expansion pixel to described traffic mark specifically includes: the edge pattern to described traffic mark
1 to 5 pixel of outward expansion.
The recognition methods of speed limit the most according to claim 1 mark, it is characterised in that:
Described speed limit identification image is entered by the edge pattern identified according to described speed limit described in described judgement speed limit identification of steps
Row goes interference process to specifically include: the edge pattern of described speed limit mark is circle, calculates the radius of described circle, determines described
The circular center of circle, by described speed limit identification image, the image beyond circle being made up of the center of circle and radius removes.
The recognition methods of speed limit the most according to claim 1 mark, it is characterised in that:
Described identification step specifically includes: the image after described speed limit identification image goes interference process is for treating threshold process figure
To described, picture, treats that threshold process image carries out histogrammic numerical statistic, according to the maxima and minima meter in numerical statistic
Calculate threshold value, utilize described threshold value to treat that threshold process image carries out threshold process to described, obtain the binaryzation of described speed limit mark
Image, carries out numeral identification to described binary image, obtains the described speed limit numeral that described speed limit mark is corresponding.
8. the recognition methods identified according to the speed limit described in any one of claim 1 to 7, it is characterised in that:
Also include pointing out step, including: by the way of voice broadcast and/or by the way of display, described speed limit numeral is entered
Row prompting.
9. the recognition methods identified according to the speed limit described in any one of claim 1 to 7, it is characterised in that:
Described in described acquisition image step, camera head is arranged on the front windshield of its place vehicle.
10. the identification device of a speed limit mark, it is characterised in that including:
Obtaining image module, be used for: obtain a two field picture, described image is the image of the road conditions shot by camera head；
Judging traffic mark module, be used for: intercept the live part of described image, the image after intercepting is figure to be identified
Picture, carries out edge extracting to the object in described image to be identified, it is thus achieved that comprise the edge image of the edge pattern of described object,
Judge whether described object is traffic mark according to described edge pattern, including at least a described thing in described image to be identified
Body；
Judging that speed limit identifies module, being used for: described object is described traffic mark if judging, and if judging that described traffic mark is
Described speed limit identifies, then the edge pattern that described speed limit identifies split from described edge image, form described speed limit
Identification image, goes interference to process described speed limit identification image according to the edge pattern that described speed limit identifies；
Identification module, is used for: the numeral within the edge pattern in described speed limit identification image after processing going interference is carried out
Identifying, the numeral identified is the speed limit numeral that speed limit mark is corresponding.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610343756.5A CN106022268A (en) | 2016-05-23 | 2016-05-23 | Identification method and device of speed limiting sign |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610343756.5A CN106022268A (en) | 2016-05-23 | 2016-05-23 | Identification method and device of speed limiting sign |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106022268A true CN106022268A (en) | 2016-10-12 |
Family
ID=57096778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610343756.5A Pending CN106022268A (en) | 2016-05-23 | 2016-05-23 | Identification method and device of speed limiting sign |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106022268A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107944425A (en) * | 2017-12-12 | 2018-04-20 | 北京小米移动软件有限公司 | The recognition methods of road sign and device |
| CN109300334A (en) * | 2018-11-28 | 2019-02-01 | 深圳市元征科技股份有限公司 | Car speed control method, device, equipment and computer readable storage medium |
| CN109579867A (en) * | 2018-11-30 | 2019-04-05 | 惠州市德赛西威汽车电子股份有限公司 | A kind of air navigation aid and its system with marker prompt |
| CN110222599A (en) * | 2019-05-21 | 2019-09-10 | 西安理工大学 | A kind of gait recognition method based on Gauss Map |
| WO2020173056A1 (en) * | 2019-02-25 | 2020-09-03 | 百度在线网络技术（北京）有限公司 | Traffic image recognition method and apparatus, and computer device and medium |
| US11293762B2 (en) | 2019-06-18 | 2022-04-05 | Here Global B.V. | System and methods for generating updated map data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101761038A (en) * | 2009-12-31 | 2010-06-30 | 华南理工大学 | System and method for recognizing railway temporary speed-limiting sign |
| CN102999756A (en) * | 2012-11-09 | 2013-03-27 | 重庆邮电大学 | Method for recognizing road signs by PSO-SVM (particle swarm optimization-support vector machine) based on GPU (graphics processing unit) |
| CN103971128A (en) * | 2014-05-23 | 2014-08-06 | 北京理工大学 | Traffic sign recognition method for driverless car |
| CN104361333A (en) * | 2014-12-10 | 2015-02-18 | 东方网力科技股份有限公司 | Traffic speed limit sign recognition method and device |
| US9460355B2 (en) * | 2014-10-14 | 2016-10-04 | Here Global B.V. | Lateral sign placement determination |
-
2016
- 2016-05-23 CN CN201610343756.5A patent/CN106022268A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101761038A (en) * | 2009-12-31 | 2010-06-30 | 华南理工大学 | System and method for recognizing railway temporary speed-limiting sign |
| CN102999756A (en) * | 2012-11-09 | 2013-03-27 | 重庆邮电大学 | Method for recognizing road signs by PSO-SVM (particle swarm optimization-support vector machine) based on GPU (graphics processing unit) |
| CN103971128A (en) * | 2014-05-23 | 2014-08-06 | 北京理工大学 | Traffic sign recognition method for driverless car |
| US9460355B2 (en) * | 2014-10-14 | 2016-10-04 | Here Global B.V. | Lateral sign placement determination |
| CN104361333A (en) * | 2014-12-10 | 2015-02-18 | 东方网力科技股份有限公司 | Traffic speed limit sign recognition method and device |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107944425A (en) * | 2017-12-12 | 2018-04-20 | 北京小米移动软件有限公司 | The recognition methods of road sign and device |
| CN109300334A (en) * | 2018-11-28 | 2019-02-01 | 深圳市元征科技股份有限公司 | Car speed control method, device, equipment and computer readable storage medium |
| CN109579867A (en) * | 2018-11-30 | 2019-04-05 | 惠州市德赛西威汽车电子股份有限公司 | A kind of air navigation aid and its system with marker prompt |
| WO2020173056A1 (en) * | 2019-02-25 | 2020-09-03 | 百度在线网络技术（北京）有限公司 | Traffic image recognition method and apparatus, and computer device and medium |
| CN110222599A (en) * | 2019-05-21 | 2019-09-10 | 西安理工大学 | A kind of gait recognition method based on Gauss Map |
| CN110222599B (en) * | 2019-05-21 | 2021-09-10 | 西安理工大学 | Gait recognition method based on Gaussian mapping |
| US11293762B2 (en) | 2019-06-18 | 2022-04-05 | Here Global B.V. | System and methods for generating updated map data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bilal et al. | Real-time lane detection and tracking for advanced driver assistance systems | |
| CN106022268A (en) | Identification method and device of speed limiting sign | |
| Wu et al. | Lane-mark extraction for automobiles under complex conditions | |
| Gomez et al. | Traffic lights detection and state estimation using hidden markov models | |
| CN110660254A (en) | Traffic signal lamp detection and intelligent driving method and device, vehicle and electronic equipment | |
| CN109635656A (en) | Vehicle attribute recognition methods, device, equipment and medium neural network based | |
| CN106326822B (en) | Method and device for detecting lane line | |
| US20160148063A1 (en) | Traffic light detection | |
| CN108052904B (en) | Method and device for acquiring lane line | |
| CN103902985B (en) | High-robustness real-time lane detection algorithm based on ROI | |
| CN101470806A (en) | Vehicle lamp detection method and apparatus, interested region splitting method and apparatus | |
| Chang et al. | An efficient method for lane-mark extraction in complex conditions | |
| Qin et al. | A general framework for road marking detection and analysis | |
| CN101369312B (en) | Method and equipment for detecting intersection in image | |
| CN112580447B (en) | Edge second-order statistics and fusion-based power line detection method | |
| Devane et al. | Lane detection techniques using image processing | |
| KR20130000023A (en) | Method for dectecting front vehicle using scene information of image | |
| Premachandra et al. | Image based automatic road surface crack detection for achieving smooth driving on deformed roads | |
| CN103093229A (en) | Positioning method and device of vehicle logo | |
| KR101236223B1 (en) | Method for detecting traffic lane | |
| FAN et al. | Robust lane detection and tracking based on machine vision | |
| JP7264428B2 (en) | Road sign recognition device and its program | |
| CN111680580A (en) | Red light running identification method and device, electronic equipment and storage medium | |
| CN109145797B (en) | Square induction method for road rescue equipment based on vehicle bottom shadow positioning license plate | |
| CN103971347A (en) | Method and device for treating shadow in video image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161012 |