CN104361333A - Traffic speed limit sign recognition method and device - Google Patents
Traffic speed limit sign recognition method and device Download PDFInfo
- Publication number
- CN104361333A CN104361333A CN201410754286.2A CN201410754286A CN104361333A CN 104361333 A CN104361333 A CN 104361333A CN 201410754286 A CN201410754286 A CN 201410754286A CN 104361333 A CN104361333 A CN 104361333A
- Authority
- CN
- China
- Prior art keywords
- sign
- pixel point
- candidate
- image
- monocase
- 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
Abstract
The invention discloses a traffic speed limit sign recognition method and a traffic speed limit sign recognition device. The method comprises the following steps: carrying out binarization on a sign image to obtain a binarized sign image composed of foreground and background pixel points with corresponding numbers, continuing to conduct segmentation on various connected domains in the binarized images composed of the foreground pixel points to segment connected domains meeting a preset single-character rule (i.e., single-character characteristic of speed limit sign) and extracting the connected domains to obtain candidate single characters; then combining the extracted candidate single characters to obtain a candidate sequence, and recognizing various single characters in the candidate sequence depending on the sequence to obtain required speed limit value. Obviously, the method disclosed by the invention realizes recognition of the traffic speed limit sign by virtue of a single-character recognition way; and compared with an overall recognition way, the recognition method based on character segmentation is not high in location precision on the traffic sign and is little affected by overall brightness and background of licenses plate as well as shape of the traffic sign; therefore, traffic speed limit sign recognition accuracy is finally improved.
Description
Technical field
The invention belongs to image procossing, recognition technology field, particularly relate to the recognition methods of a kind of traffic speed(-)limit sign and device.
Background technology
According to statistics, the traffic hazard of 70% is all relevant with overspeed of vehicle, according to the speed limit on road indicated by traffic speed(-)limit sign, carries out safe driving prompting, can effectively reduce traffic hazard incidence to driver.Based on this, provide a kind of and can identify that the recognition methods of traffic speed(-)limit sign on road is very necessary.
Traditional recognition methods, generally carry out entirety based on pattern-recognition or template matches to locating segmentation traffic sign out to identify, overall identification requirement traffic sign locating segmentation is accurate, and being subject to the impact of number plate overall brightness, background and traffic sign shape (as distortion), recognition accuracy is lower.
Summary of the invention
In view of this, the object of the present invention is to provide the recognition methods of a kind of traffic speed(-)limit sign and device, effectively to identify the traffic speed(-)limit sign on road, improve recognition accuracy.
For this reason, the present invention's openly following technical scheme:
The recognition methods of a kind of traffic speed(-)limit sign, comprising:
Obtain the sign image of object speed(-)limit sign on traffic indication board;
Binary conversion treatment is carried out to described sign image, obtains binaryzation sign image, wherein, the image of described binaryzation sign image for being made up of foreground pixel point and the background pixel point of respective number;
From each connected domain that described binaryzation sign image comprises, be partitioned into the connected domain that geometric properties meets default speed(-)limit sign monocase feature, and extract each connected domain split, by each connected domain alternatively monocase extracted; Wherein, the connected region of described connected domain for being made up of foreground pixel point;
According to the assemblage characteristic between each connected domain of being partitioned into, and each candidate's monocase and the geometric similarity of each connected domain be partitioned into, each candidate's monocase is combined, obtains candidate sequence;
Utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence, obtain speed limit numerical value.
Said method, preferably, describedly carries out binary conversion treatment to described sign image and comprises following foreground pixel point leaching process:
Utilize formula
In calculation flag image current pixel point p contiguous range in average brightness ave (p) of each pixel, wherein, w is the half size of neighborhood window, and x, y represent horizontal ordinate, the ordinate of pixel p respectively, and f is luminance function;
If brightness value f (p) of described current pixel point p is greater than predetermined threshold value, then using the foreground pixel point of described current pixel point p as described sign image with the difference of described average brightness ave (p).
Said method, preferably, also comprised following preprocessing process before carrying out binary conversion treatment to described sign image:
When the overall brightness of described sign image is lower than preset standard, brightness is carried out to described sign image and strengthens process;
Sign image after brightness strengthens is normalized.
Said method, preferably, utilizes before respective symbols sorter identifies according to the order of sequence to each candidate's monocase in described candidate sequence described, also comprises:
Described candidate sequence and the speed(-)limit sign template preset are compared, and judges whether described candidate sequence conforms to described speed(-)limit sign template;
If the determination result is YES, then following steps are triggered: utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence;
If judged result is no, then terminate.
Said method, preferably, described geometric similarity comprises high similarity, width similarity, distance similarity.
Said method, preferably, described speed(-)limit sign template comprises two Character mother plates ' x0 ' and three-character doctrine template ' xx0 '.
A kind of traffic speed(-)limit sign recognition device, is characterized in that, comprising:
Acquisition module, for obtaining the sign image of object speed(-)limit sign on traffic indication board;
Binary conversion treatment module, for carrying out binary conversion treatment to described sign image, obtains binaryzation sign image, wherein, and the image of described binaryzation sign image for being made up of foreground pixel point and the background pixel point of respective number;
Segmentation and extraction module, for being partitioned into the connected domain that geometric properties meets default speed(-)limit sign character feature in each connected domain of comprising from described binaryzation sign image, and extract each connected domain split, by each connected domain alternatively monocase extracted; Wherein, the connected region of described connected domain for being made up of foreground pixel point;
Composite module, for according to the assemblage characteristic between each connected domain of being partitioned into, and each candidate's monocase and the geometric similarity of each connected domain be partitioned into, each candidate's monocase is combined, obtains candidate sequence;
Identification module, for utilizing respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence, obtains speed limit numerical value.
Said apparatus, preferably, described binary conversion treatment module comprises foreground pixel point extraction module, and described foreground pixel point extraction module comprises:
Computing unit, for utilizing formula
In calculation flag image current pixel point p contiguous range in average brightness ave (p) of each pixel, wherein, w is the half size of neighborhood window, and x, y are respectively horizontal ordinate, the ordinate of pixel p, and f is luminance function;
Extraction unit, for being greater than predetermined threshold value, using the foreground pixel point of described current pixel point p as described sign image in brightness value f (p) of current pixel point p and the difference of described average brightness ave (p).
Said apparatus, preferably, also comprises pretreatment module, and described pretreatment module comprises:
Brightness enhancement unit, for when the overall brightness of described sign image is lower than preset standard, carries out brightness to described sign image and strengthens process;
Normalization unit, is normalized for the sign image after strengthening brightness.
Said apparatus, preferably, also comprises template matches module, and described module matching module comprises:
Judging unit, for described candidate sequence and the speed(-)limit sign template preset being compared, and judges whether described candidate sequence conforms to described speed(-)limit sign template;
Trigger element, for when judged result is for being, triggers following steps: utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence;
End unit, during for being no in judged result, terminates identification process.
From above scheme, the present invention is carrying out binary conversion treatment to sign image, after obtaining the binaryzation sign image be made up of the foreground pixel point of respective number and background pixel point, continue to carry out dividing processing to each connected domain be made up of foreground pixel point in binary image, be partitioned into the connected domain meeting default monocase rule (i.e. speed(-)limit sign monocase feature), and extracted alternatively monocase; Follow-up candidate's monocase to extracting carries out combination and obtains candidate sequence, and by identifying according to the order of sequence each monocase in candidate sequence, obtains required speed limit numerical value.Visible, present invention achieves and by monocase recognition method, traffic speed(-)limit sign is identified, compared to overall recognition method, the recognition methods that the present invention is based on Character segmentation is not high to the positioning accuracy request of traffic sign, and less by the impact of number plate overall brightness, background and traffic sign shape, thus finally improve the recognition accuracy of traffic speed(-)limit sign.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only embodiments of the invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to the accompanying drawing provided.
Fig. 1 is a kind of process flow diagram of traffic speed(-)limit sign recognition methods disclosed in the embodiment of the present invention one;
Fig. 2 is connected area segmentation exemplary plot disclosed in the embodiment of the present invention one;
Fig. 3 is the another kind of process flow diagram of traffic speed(-)limit sign recognition methods disclosed in the embodiment of the present invention two;
Fig. 4 is another process flow diagram of traffic speed(-)limit sign recognition methods disclosed in the embodiment of the present invention three;
Fig. 5 is a kind of structural representation of traffic speed(-)limit sign recognition device disclosed in the embodiment of the present invention four;
Fig. 6 is the another kind of structural representation of traffic speed(-)limit sign recognition device disclosed in the embodiment of the present invention four;
Fig. 7 is another structural representation of traffic speed(-)limit sign recognition device disclosed in the embodiment of the present invention four.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Embodiment one
The present embodiment one discloses the recognition methods of a kind of traffic speed(-)limit sign, and with reference to figure 1, described method can comprise the following steps:
S101: the sign image obtaining object speed(-)limit sign on traffic indication board.
S102: carry out binary conversion treatment to described sign image, obtains binaryzation sign image, wherein, and the image of described binaryzation sign image for being made up of foreground pixel point and the background pixel point of respective number.
In order to solve the binaryzation problem of negative and positive board, illumination brightness irregularities, the present invention proposes following local binarization method: the distinguishing characteristics investigating each pixel in current pixel point and its field, and when described distinguishing characteristics meets default distinguishing characteristics standard, extract described current pixel point as foreground pixel point, provide support to be embodied as follow-up speed(-)limit sign identification, on the contrary, when not meeting described distinguishing characteristics standard, then by described current pixel point pixel as a setting.
Based on described local binarization method, the present embodiment extracts foreground pixel point especially by following process implementation from sign image:
A, utilize current pixel point p in formula (1) calculation flag image contiguous range in average brightness ave (p) of each pixel:
Wherein, w is the half size of neighborhood window, and x, y represent horizontal ordinate, the ordinate of pixel p respectively, and f is luminance function;
If brightness value f (p) of b current pixel point p is greater than predetermined threshold value with the difference of described average brightness ave (p), i.e. ave (p)-f (p) >T, then extract the foreground pixel point of current pixel point p as sign image, wherein, T is the size of set threshold value.
The local binarization method that the present invention proposes, relative to the overall binarization method of traditional images, is not subject to noise, uneven illumination is even, contrast is low impact, can reaches satisfied binaryzation effect.
S103: be partitioned into the connected domain that geometric properties meets default speed(-)limit sign monocase feature from each connected domain that described binaryzation sign image comprises, and extract each connected domain split, by each connected domain alternatively monocase extracted; Wherein, the connected region of described connected domain for being made up of foreground pixel point.
When carrying out binary conversion treatment to sign image, after obtaining the binaryzation sign image be made up of a series of foreground pixel point and background pixel point, this step identifies the connected region be made up of foreground pixel point in described binary image, identifies, obtain each connected domain, and the monocase rule continued according to presetting, height, width, the ratio of width to height etc. in each connected domain of mark are met the connected area segmentation of speed(-)limit sign monocase feature out, and the connected domain be partitioned into can with reference to the content in the white rectangle frame shown in figure 2.
On this basis, extract each connected domain of segmentation gained, and using each connected domain of extraction as the follow-up candidate's monocase carrying out the identification of traffic speed(-)limit sign.
S104: according to the assemblage characteristic between each connected domain of being partitioned into, and each candidate's monocase and the geometric similarity of each connected domain be partitioned into, combine each candidate's monocase, obtain candidate sequence.
Particularly, each candidate's monocase extracted carries out being combined to form candidate sequence according to the high similarity of each connected domain split in itself and binary image, width similarity, distance by this step.
S105: utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence, obtain speed limit numerical value.
Each connected domain (i.e. candidate's monocase) that this step comprises candidate sequence identifies according to the order of sequence.
In practical application scene, for the situation having two connected regions, first of speed(-)limit sign can not be 0, second character is generally chosen " 0 " or " 5 ", therefore, candidate sequence is comprised to the situation of two connected domains, first identifies with adopting the sorter of " 1-9 ", and the second character sorter that " 0 " and " 5 " is formed identifies.
Correspondingly, according to the span of speed(-)limit sign character in practical application scene, candidate sequence is comprised to the situation of three connected domains, the sorter of " 1-9 " is adopted to identify it to first character of candidate sequence, the second sorter of " 0-2 " identifies, the 3rd is all that " 0 " need not identify.
The present embodiment specifically adopts the mode training classifier of BP (Back Propagation, backpropagation) feature+neural network, and through checking, its monocase discrimination accuracy rate of sorter that training generates is more than 98.5%.
By identifying according to the order of sequence each connected domain of candidate sequence, finally obtaining required speed limit numerical value, on this basis, safe driving prompting can be carried out according to the speed limit identified to driver further, effectively reducing traffic hazard incidence.
From above scheme, the present invention is carrying out binary conversion treatment to sign image, after obtaining the binaryzation sign image be made up of the foreground pixel point of respective number and background pixel point, continue to carry out dividing processing to each connected domain be made up of foreground pixel point in binary image, be partitioned into the connected domain meeting default monocase rule (i.e. speed(-)limit sign monocase feature), and extracted alternatively monocase; Follow-up candidate's monocase to extracting carries out combination and obtains candidate sequence, and by identifying according to the order of sequence each monocase in candidate sequence, obtains required speed limit numerical value.Visible, present invention achieves and by monocase recognition method, traffic speed(-)limit sign is identified, compared to overall recognition method, the locating segmentation of recognition methods to traffic sign that the present invention is based on Character segmentation is less demanding, and less by the impact of number plate overall brightness, background and traffic sign shape, thus finally improve the recognition accuracy of traffic speed(-)limit sign.
Embodiment two
In the present embodiment two, with reference to figure 3, described method can also comprise following preprocessing process:
S106: when the overall brightness of described sign image is lower than preset standard, carries out brightness to described sign image and strengthens process;
S107: the sign image after brightness strengthens is normalized.
Particularly, before binary conversion treatment is carried out to sign image, the overall brightness of the present embodiment first analysis mark image, brightness is carried out to the lower sign image not reaching established standards of brightness and strengthens process, afterwards, continue to be normalized the sign image after brightness strengthens, to facilitate choosing of Character segmentation threshold value in subsequent step.
Wherein, described normalization specifically refers to and sign image is zoomed to fixed size.
The present embodiment, by carrying out brightness processed to sign image, reduces the influence degree of luminance factor to traffic speed(-)limit sign identification adverse effect.
Embodiment three
In the present embodiment three, with reference to figure 4, described method can also comprise the following steps before S105:
S108: described candidate sequence and the speed(-)limit sign template preset are compared, and judges whether described candidate sequence conforms to described speed(-)limit sign template, if the determination result is YES, then triggered step S105; If judged result is no, then terminate.
Before each connected domain in candidate sequence is identified according to the order of sequence, first candidate sequence and the speed(-)limit sign template preset compare by the present embodiment, according to actual service condition, the present invention has formulated two kinds of speed(-)limit sign templates: the template ' xx0 ' of two Character mother plates ' x0 ' of less than 100 and the three-character doctrine of more than 100, thus, when the number of connected domain in candidate sequence is more than 3 or when being less than 2, show that it is non-speed(-)limit sign, need not participate in identifying, only when the number of connected domain in candidate sequence is 2 or 3, just it is identified.
The present embodiment is by comparing candidate sequence and the speed(-)limit sign template preset, and the candidate sequence achieved not meeting pattern rule filters, and improves the recognition accuracy of traffic speed(-)limit sign.
Embodiment four
The present embodiment four discloses a kind of traffic speed(-)limit sign recognition device, and this device is corresponding with traffic speed(-)limit sign recognition methods disclosed in above each embodiment.
First, corresponding to embodiment one, with reference to figure 5, described device comprises acquisition module 100, binary conversion treatment module 200, segmentation and extraction module 300, composite module 400 and identification module 500.
Acquisition module 100, for obtaining the sign image of object speed(-)limit sign on traffic indication board.
Binary conversion treatment module 200, for carrying out binary conversion treatment to described sign image, obtains binaryzation sign image, wherein, and the image of described binaryzation sign image for being made up of foreground pixel point and the background pixel point of respective number.
Wherein, binary conversion treatment module 200 comprises foreground pixel point extraction module, and described foreground pixel point extraction module comprises computing unit and extraction unit.
Computing unit, for utilizing formula
In calculation flag image current pixel point p contiguous range in average brightness ave (p) of each pixel, wherein, w is the half size of neighborhood window, and x, y are respectively horizontal ordinate, the ordinate of pixel p, and f is luminance function;
Extraction unit, for being greater than predetermined threshold value, using the foreground pixel point of described current pixel point p as described sign image in brightness value f (p) of current pixel point p and the difference of described average brightness ave (p).
Segmentation and extraction module 300, for being partitioned into the connected domain that geometric properties meets default speed(-)limit sign character feature in each connected domain of comprising from described binaryzation sign image, and extract each connected domain split, by each connected domain alternatively monocase extracted; Wherein, the connected region of described connected domain for being made up of foreground pixel point.
Composite module 400, for according to the assemblage characteristic between each connected domain of being partitioned into, and each candidate's monocase and the geometric similarity of each connected domain be partitioned into, each candidate's monocase is combined, obtains candidate sequence.
Identification module 500, for utilizing respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence, obtains speed limit numerical value.
Corresponding to embodiment two, with reference to figure 6, described device also comprises pretreatment module 600, and this module comprises brightness enhancement unit and normalization unit.
Brightness enhancement unit, for when the overall brightness of described sign image is lower than preset standard, carries out brightness to described sign image and strengthens process;
Normalization unit, is normalized for the sign image after strengthening brightness.
Corresponding to embodiment three, with reference to figure 7, described device also comprises template matches module 700, and this module comprises judging unit, trigger element and end unit.
Judging unit, for described candidate sequence and the speed(-)limit sign template preset being compared, and judges whether described candidate sequence conforms to described speed(-)limit sign template;
Trigger element, for when judged result is for being, triggers following steps: utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence;
End unit, during for being no in judged result, terminates identification process.
For traffic speed(-)limit sign recognition device disclosed in the embodiment of the present invention four, because it is corresponding with traffic speed(-)limit sign recognition methods disclosed in above each embodiment, so description is fairly simple, relevant similarity refers to the explanation of traffic speed(-)limit sign recognition methods part in above each embodiment, no longer describes in detail herein.
In sum, the present invention can solve the impalpable problem of speed(-)limit sign in negative and positive number plate, brightness irregularities number plate effectively by proposing local binarization method, effectively can carry out prospect to it, background pixel is separated; And the present invention is based on monocase rule and entirety is regular filters it, effectively can remove non-traffic sign candidate region, obtain the candidate sequence that reference value is higher, and realize identifying candidate sequence by monocase recognition method, improve the recognition accuracy of traffic speed(-)limit sign.
It should be noted that, each embodiment in this instructions all adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar part mutually see.
As seen through the above description of the embodiments, those skilled in the art can be well understood to the mode that the application can add required general hardware platform by software and realizes.Based on such understanding, the technical scheme of the application can embody with the form of software product the part that prior art contributes in essence in other words, this computer software product can be stored in storage medium, as ROM/RAM, magnetic disc, CD etc., comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform the method described in some part of each embodiment of the application or embodiment.
The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (10)
1. the recognition methods of traffic speed(-)limit sign, is characterized in that, comprising:
Obtain the sign image of object speed(-)limit sign on traffic indication board;
Binary conversion treatment is carried out to described sign image, obtains binaryzation sign image, wherein, the image of described binaryzation sign image for being made up of foreground pixel point and the background pixel point of respective number;
From each connected domain that described binaryzation sign image comprises, be partitioned into the connected domain that geometric properties meets default speed(-)limit sign monocase feature, and extract each connected domain split, by each connected domain alternatively monocase extracted; Wherein, the connected region of described connected domain for being made up of foreground pixel point;
According to the assemblage characteristic between each connected domain of being partitioned into, and each candidate's monocase and the geometric similarity of each connected domain be partitioned into, each candidate's monocase is combined, obtains candidate sequence;
Utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence, obtain speed limit numerical value.
2. method according to claim 1, is characterized in that, describedly carries out binary conversion treatment to described sign image and comprises following foreground pixel point leaching process:
Utilize formula
In calculation flag image current pixel point p contiguous range in average brightness ave (p) of each pixel, wherein, w is the half size of neighborhood window, and x, y represent horizontal ordinate, the ordinate of pixel p respectively, and f is luminance function;
If brightness value f (p) of described current pixel point p is greater than predetermined threshold value, then using the foreground pixel point of described current pixel point p as described sign image with the difference of described average brightness ave (p).
3. method according to claim 1, is characterized in that, before carrying out binary conversion treatment to described sign image, also comprise following preprocessing process:
When the overall brightness of described sign image is lower than preset standard, brightness is carried out to described sign image and strengthens process;
Sign image after brightness strengthens is normalized.
4. method according to claim 1, is characterized in that, utilizes before respective symbols sorter identifies according to the order of sequence to each candidate's monocase in described candidate sequence, also comprise described:
Described candidate sequence and the speed(-)limit sign template preset are compared, and judges whether described candidate sequence conforms to described speed(-)limit sign template;
If the determination result is YES, then following steps are triggered: utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence;
If judged result is no, then terminate.
5. method according to claim 1, is characterized in that, described geometric similarity comprises high similarity, width similarity, distance similarity.
6. method according to claim 4, is characterized in that, described speed(-)limit sign template comprises two Character mother plates ' x0 ' and three-character doctrine template ' xx0 '.
7. a traffic speed(-)limit sign recognition device, is characterized in that, comprising:
Acquisition module, for obtaining the sign image of object speed(-)limit sign on traffic indication board;
Binary conversion treatment module, for carrying out binary conversion treatment to described sign image, obtains binaryzation sign image, wherein, and the image of described binaryzation sign image for being made up of foreground pixel point and the background pixel point of respective number;
Segmentation and extraction module, for being partitioned into the connected domain that geometric properties meets default speed(-)limit sign character feature in each connected domain of comprising from described binaryzation sign image, and extract each connected domain split, by each connected domain alternatively monocase extracted; Wherein, the connected region of described connected domain for being made up of foreground pixel point;
Composite module, for according to the assemblage characteristic between each connected domain of being partitioned into, and each candidate's monocase and the geometric similarity of each connected domain be partitioned into, each candidate's monocase is combined, obtains candidate sequence;
Identification module, for utilizing respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence, obtains speed limit numerical value.
8. device according to claim 7, is characterized in that, described binary conversion treatment module comprises foreground pixel point extraction module, and described foreground pixel point extraction module comprises:
Computing unit, for utilizing formula
In calculation flag image current pixel point p contiguous range in average brightness ave (p) of each pixel, wherein, w is the half size of neighborhood window, and x, y are respectively horizontal ordinate, the ordinate of pixel p, and f is luminance function;
Extraction unit, for being greater than predetermined threshold value, using the foreground pixel point of described current pixel point p as described sign image in brightness value f (p) of current pixel point p and the difference of described average brightness ave (p).
9. device according to claim 7, is characterized in that, also comprises pretreatment module, and described pretreatment module comprises:
Brightness enhancement unit, for when the overall brightness of described sign image is lower than preset standard, carries out brightness to described sign image and strengthens process;
Normalization unit, is normalized for the sign image after strengthening brightness.
10. device according to claim 7, is characterized in that, also comprises template matches module, and described module matching module comprises:
Judging unit, for described candidate sequence and the speed(-)limit sign template preset being compared, and judges whether described candidate sequence conforms to described speed(-)limit sign template;
Trigger element, for when judged result is for being, triggers following steps: utilize respective symbols sorter to identify according to the order of sequence each candidate's monocase in described candidate sequence;
End unit, during for being no in judged result, terminates identification process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410754286.2A CN104361333A (en) | 2014-12-10 | 2014-12-10 | Traffic speed limit sign recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410754286.2A CN104361333A (en) | 2014-12-10 | 2014-12-10 | Traffic speed limit sign recognition method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104361333A true CN104361333A (en) | 2015-02-18 |
Family
ID=52528591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410754286.2A Pending CN104361333A (en) | 2014-12-10 | 2014-12-10 | Traffic speed limit sign recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104361333A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106022268A (en) * | 2016-05-23 | 2016-10-12 | 广州鹰瞰信息科技有限公司 | Identification method and device of speed limiting sign |
| CN109558767A (en) * | 2017-09-25 | 2019-04-02 | 比亚迪股份有限公司 | The recognition methods of automobile and road speed(-)limit sign, device |
| CN110390224A (en) * | 2018-04-16 | 2019-10-29 | 高德软件有限公司 | A kind of recognition methods of traffic sign and device |
| CN111783781A (en) * | 2020-05-22 | 2020-10-16 | 平安国际智慧城市科技股份有限公司 | Malicious clause identification method, device and equipment based on product agreement character identification |
| CN113487656A (en) * | 2021-07-26 | 2021-10-08 | 推想医疗科技股份有限公司 | Image registration method and device, training method and device, control method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706873A (en) * | 2009-11-27 | 2010-05-12 | 东软集团股份有限公司 | Identification method and device of digital-class limitation marking |
| CN101751785A (en) * | 2010-01-12 | 2010-06-23 | 杭州电子科技大学 | Automatic license plate recognition method based on image processing |
| CN101761038A (en) * | 2009-12-31 | 2010-06-30 | 华南理工大学 | System and method for recognizing railway temporary speed-limiting sign |
| CN102542260A (en) * | 2011-12-30 | 2012-07-04 | 中南大学 | Method for recognizing road traffic sign for unmanned vehicle |
| CN103984949A (en) * | 2014-06-11 | 2014-08-13 | 四川九洲电器集团有限责任公司 | License plate positioning method and system based on high and low cap transformation and connected domain |
| CN104156706A (en) * | 2014-08-12 | 2014-11-19 | 华北电力大学句容研究中心 | Chinese character recognition method based on optical character recognition technology |
-
2014
- 2014-12-10 CN CN201410754286.2A patent/CN104361333A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706873A (en) * | 2009-11-27 | 2010-05-12 | 东软集团股份有限公司 | Identification method and device of digital-class limitation marking |
| CN101761038A (en) * | 2009-12-31 | 2010-06-30 | 华南理工大学 | System and method for recognizing railway temporary speed-limiting sign |
| CN101751785A (en) * | 2010-01-12 | 2010-06-23 | 杭州电子科技大学 | Automatic license plate recognition method based on image processing |
| CN102542260A (en) * | 2011-12-30 | 2012-07-04 | 中南大学 | Method for recognizing road traffic sign for unmanned vehicle |
| CN103984949A (en) * | 2014-06-11 | 2014-08-13 | 四川九洲电器集团有限责任公司 | License plate positioning method and system based on high and low cap transformation and connected domain |
| CN104156706A (en) * | 2014-08-12 | 2014-11-19 | 华北电力大学句容研究中心 | Chinese character recognition method based on optical character recognition technology |
Non-Patent Citations (1)
| Title |
|---|
| 刘玉洁: "限速标志识别算法的研究与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106022268A (en) * | 2016-05-23 | 2016-10-12 | 广州鹰瞰信息科技有限公司 | Identification method and device of speed limiting sign |
| CN109558767A (en) * | 2017-09-25 | 2019-04-02 | 比亚迪股份有限公司 | The recognition methods of automobile and road speed(-)limit sign, device |
| CN110390224A (en) * | 2018-04-16 | 2019-10-29 | 高德软件有限公司 | A kind of recognition methods of traffic sign and device |
| CN110390224B (en) * | 2018-04-16 | 2021-06-25 | 阿里巴巴（中国）有限公司 | Traffic sign recognition method and device |
| CN111783781A (en) * | 2020-05-22 | 2020-10-16 | 平安国际智慧城市科技股份有限公司 | Malicious clause identification method, device and equipment based on product agreement character identification |
| CN111783781B (en) * | 2020-05-22 | 2024-04-05 | 深圳赛安特技术服务有限公司 | Malicious term recognition method, device and equipment based on product agreement character recognition |
| CN113487656A (en) * | 2021-07-26 | 2021-10-08 | 推想医疗科技股份有限公司 | Image registration method and device, training method and device, control method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Raghunandan et al. | Riesz fractional based model for enhancing license plate detection and recognition | |
| CN105447851B (en) | The sound hole defect inspection method and system of a kind of glass panel | |
| Babu et al. | Vehicle number plate detection and recognition using bounding box method | |
| Ashtari et al. | An Iranian license plate recognition system based on color features | |
| CN102043950B (en) | Vehicle outline recognition method based on canny operator and marginal point statistic | |
| CN108268867B (en) | License plate positioning method and device | |
| CN103824091B (en) | A kind of licence plate recognition method for intelligent transportation system | |
| CN104361333A (en) | Traffic speed limit sign recognition method and device | |
| CN103268489B (en) | Automotive number plate recognition methods based on sliding window search | |
| Jusoh et al. | Application of freeman chain codes: An alternative recognition technique for Malaysian car plates | |
| CN103324930A (en) | License plate character segmentation method based on grey level histogram binaryzation | |
| Bhat et al. | Recognition of vehicle number plate using matlab | |
| CN104751142A (en) | Natural scene text detection algorithm based on stroke features | |
| CN108520212A (en) | Method for traffic sign detection based on improved convolutional neural networks | |
| Azad et al. | New method for optimization of license plate recognition system with use of edge detection and connected component | |
| Prabhakar et al. | A novel design for vehicle license plate detection and recognition | |
| Ingole et al. | Characters feature based Indian vehicle license plate detection and recognition | |
| CN107578048A (en) | A kind of long sight scene vehicle checking method based on vehicle rough sort | |
| CN104168462B (en) | Camera scene change detection method based on image angle point set feature | |
| Yilmaz | A smart hybrid license plate recognition system based on image processing using neural network and image correlation | |
| Rani et al. | Automatic Vehicle Tracking System Based on Fixed Thresholding and Histogram Based Edge Processing. | |
| CN107392115B (en) | Traffic sign identification method based on hierarchical feature extraction | |
| Goel et al. | Vehicle registration plate recognition system using template matching | |
| Dev | A novel approach for car license plate detection based on vertical edges | |
| CN108009480A (en) | A kind of image human body behavioral value method of feature based identification |
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: 20150218 |