CN102982335A - Intelligent safe multi-license-plate positioning identification method based on cellular neural network - Google Patents
Intelligent safe multi-license-plate positioning identification method based on cellular neural network Download PDFInfo
- Publication number
- CN102982335A CN102982335A CN2012105597433A CN201210559743A CN102982335A CN 102982335 A CN102982335 A CN 102982335A CN 2012105597433 A CN2012105597433 A CN 2012105597433A CN 201210559743 A CN201210559743 A CN 201210559743A CN 102982335 A CN102982335 A CN 102982335A
- Authority
- CN
- China
- Prior art keywords
- color
- license plate
- neural network
- license
- plate area
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses an intelligent safe multi-license-plate positioning identification method based on a cellular neural network, which comprises the steps that after front-end camera-shooting equipment collects color monitoring images of a plurality of lanes, license plate areas are positioned based on cellular neural network edge detection, fusion color quantification division color code matrices and definite length-width ratio characteristics; the license plate areas are magnified by a bilinear interpolation method, and are binarized by a gradation adaptive method; characters are divided by a binary projection method; and the characters and color types of license plates are identified based on a template matching fusion color code matrix. The license plate areas that are as small as possible is positioned and identified from the images containing the license plates; the positioning speed is high; and the identification accuracy is high.
Description
Technical field
The invention belongs to technical field of image processing, particularly the method for the car plate fixation and recognition in the intelligent and safe traffic system.
Background technology
Develop rapidly along with socioeconomic, vehicle is popularized becomes inexorable trend, and traditional traffic administration mode can not satisfy people's demand, and modern traffic administration mode has progressively trended towards intellectuality, safe, robotization.Mostly car plate fixation and recognition in intelligent and safe traffic system (Intelligent Transportion System, ITS) is need to expend more manpower, device resource for single lane design at present.Therefore research and development are imperative for real time monitoring and the supervisory system on multilane highway crossing and thoroughfare, city.In the ITS for multilane, all may contain a plurality of dissimilar license plate areas in each the frame ITS image that collects, it is significant that accurate complete fixation and recognition goes out each license plate area.
Existing many car plates positioning identifying method mainly is based on various traditional images Processing Algorithm at present, in the car plate positioning identification system, play an important role, but ubiquity algorithm practicality shortcoming, operand and arithmetic speed can't requirement of real time, are unfavorable for the shortcomings such as hardware realization.In addition, existing many car plates positioning identifying method is not considered the identification to car plate background color and character color.
In sum, existing many car plates positioning identification system lacks car plate color recognition function, or is unfavorable for the hardware realization, and fixation and recognition efficient is low, can't satisfy the performance that real-time fixation and recognition goes out extensive car plate.
Summary of the invention
A kind of many car plates of intelligent and safe positioning identifying method based on cell neural network that the embodiment of the invention provides, in order to solve many car plates fixation and recognition length consuming time, efficient low, can't Real time identification go out the problems such as car plate color.
After the front end picture pick-up device collects multilane monitoring coloured image, based on cell neural network rim detection, Fusion of Color quantize to cut apart the colour coding matrix, length and width are decided ratio characteristic and are oriented each license plate area; Bilinear interpolation amplifies each license plate area, gray scale adaptive method license plate area binaryzation, and the two-value sciagraphy is partitioned into each character; Merge the colour coding matrix based on template matches and identify each character and each car plate color type.
The embodiment of the invention can realize that fixation and recognition goes out as far as possible little license plate area from contain many license plate images, and locating speed is fast, and recognition accuracy is high, but and Real time identification go out the car plate color type.
Description of drawings
The method flow diagram that Fig. 1 provides for the embodiment of the invention;
Fig. 2 for the embodiment of the invention provide based on cell neural network edge detection algorithm process flow diagram;
Many threshold values color quantizing partitioning algorithm process flow diagram that Fig. 3 provides for the embodiment of the invention.
Embodiment
Low for existing many car plates positioning identification system ubiquity computing length consuming time, efficient, can't Real time identification go out the problems such as car plate color, but the embodiment of the invention is utilized the cell neural network parallel computation, processing speed and image size are irrelevant, be beneficial to the advantage raising arithmetic speeds such as large scale integrated circuit hardware realization, and Fusion of Color quantizes color and type that partitioning algorithm identifies car plate fast.
What as shown in Figure 1, the embodiment of the invention provided comprises the following steps: based on many car plates of cell neural network positioning identifying method
The result of step 107, fusion (4)～(6) locates, gathers out each candidate's license plate area;
Gray scale adaptive method binaryzation is adopted in step 109, the zone that will amplify;
What as shown in Figure 2, the embodiment of the invention provided comprises the following steps: based on the cell neural network edge detection algorithm
Many license plate images behind step 201, the input gray level;
The template parameter value that step 202, input are set by performance analysis;
If step 205 network is convergence not, then repeatedly (4)-(5) are until the cycle index of setting is restrained or executed to complete network;
As shown in Figure 3, many threshold values color quantizing partitioning algorithm of providing of the embodiment of the invention comprises the following steps:
The rgb format that step 301, input collect contains the coloured image of many license plate areas;
The threshold range that step 304, basis are determined carries out multimedia threshold quantization to be cut apart, and generates the colour coding matrix that only contains the color of object code value:
If f (x, y) ∈ is [f
i(x, y) (1-θ
i), f
i(x, y) (1+ θ
i)], g
i(x, y)=50i.
Wherein, f
i(x, y) is H corresponding to candidate region color, S, I dimension color standard value, θ
iBe color permission, g
i(x, y) (i ∈ [0,5], i ∈ R) is the quantification code value of candidate region color.
Can find out from above-described embodiment: the embodiment of the invention is for monitoring that multilane, monitor area contain the situation of many vehicles, have proposed the new system of many car plates fixation and recognition based on cell neural network, Fusion Feature.This system can orient as far as possible little license plate area from contain many car plates collection images, locating speed is fast, and recognition accuracy is high, and the energy Real time identification goes out color and the type of car plate.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (6)
1. many car plates of intelligent and safe positioning identifying method based on cell neural network is characterized in that the method comprises:
After the front end picture pick-up device collects multilane monitoring coloured image, based on cell neural network rim detection, Fusion of Color quantize to cut apart the colour coding matrix, length and width are decided ratio characteristic and are oriented each license plate area; Bilinear interpolation amplifies each license plate area, gray scale adaptive method license plate area binaryzation, and the two-value sciagraphy is partitioned into each character; Merge the colour coding matrix based on template matches and identify each character and each car plate color type.To realize that fixation and recognition goes out as far as possible little license plate area from contain many license plate images, locating speed is fast, and recognition accuracy is high.
2. the method for claim 1 is characterized in that, the described method of orienting each license plate area specifically comprises:
After the front end picture pick-up device collects multilane monitoring coloured image, carry out gray processing, cell neural network rim detection, twice expansion form processing, then the Fusion of Color space is changed the colour coding that quantizes to obtain and is covered plate acquisition candidate region, and utilize length and width surely than carrying out the candidate region authenticity verification, locate at last, gather out each candidate's license plate area.
3. the method for claim 1 is characterized in that, the method for described Character segmentation specifically comprises:
Adopt bilinear interpolation that each license plate area is amplified, gray scale adaptive method license plate area binaryzation, the two-value sciagraphy is partitioned into each character.
4. the method for claim 1 is characterized in that, the method for described character and color identification specifically comprises:
Identify each character based on template matches, the colour coding matrix that Fusion of Color is cut apart generation obtains each car plate color type.
5. method as claimed in claim 2 is characterized in that, described cell neural network edge detection method specifically comprises:
Adopt analysis of cells neural network dynamic iterative process to obtain the template parameter value, bring the parameter value that obtains into state equation, input-output equation, iteration output edge detection results.
6. method as claimed in claim 2 is characterized in that, described color quantizing is cut apart generation colour coding matrix method and specifically comprised:
The rgb format coloured image that collects is converted into the HSI pattern, according to the size of car plate color probability of occurrence: blueness, yellow, white, black, redness, carry out multimedia threshold quantization and cut apart:
If f (x, y) ∈ is [f
i(x, y) (1-θ
i), f
i(x, y) (1+ θ
i)], g
i(x, y)=50i.
Wherein, f
i(x, y) is H corresponding to candidate region color, S, I dimension color standard value, θ
iBe color permission, g
i(x, y (i ∈ [0,5], i ∈ R) is the quantification code value of candidate region color.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012105597433A CN102982335A (en) | 2012-12-21 | 2012-12-21 | Intelligent safe multi-license-plate positioning identification method based on cellular neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012105597433A CN102982335A (en) | 2012-12-21 | 2012-12-21 | Intelligent safe multi-license-plate positioning identification method based on cellular neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102982335A true CN102982335A (en) | 2013-03-20 |
Family
ID=47856318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012105597433A Pending CN102982335A (en) | 2012-12-21 | 2012-12-21 | Intelligent safe multi-license-plate positioning identification method based on cellular neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102982335A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110476864A (en) * | 2019-09-17 | 2019-11-22 | 广东工业大学 | A kind of Intelligent fish tank system |
| CN110717865A (en) * | 2019-09-02 | 2020-01-21 | 苏宁云计算有限公司 | Picture detection method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101408942A (en) * | 2008-04-17 | 2009-04-15 | 浙江师范大学 | Method for locating license plate under a complicated background |
| CN101630360A (en) * | 2008-07-14 | 2010-01-20 | 上海分维智能科技有限公司 | Method for identifying license plate in high-definition image |
-
2012
- 2012-12-21 CN CN2012105597433A patent/CN102982335A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101408942A (en) * | 2008-04-17 | 2009-04-15 | 浙江师范大学 | Method for locating license plate under a complicated background |
| CN101630360A (en) * | 2008-07-14 | 2010-01-20 | 上海分维智能科技有限公司 | Method for identifying license plate in high-definition image |
Non-Patent Citations (2)
| Title |
|---|
| 曹被军: ""关于车牌字符识别技术的研究"", 《中国优秀硕士学位论文全文数据库（电子期刊） 信息科技辑》, 15 March 2012 (2012-03-15), pages 138 - 2010 * |
| 谢康: ""基于CNN的智能交通***多车牌定位方法的研究"", 《中国优秀硕士学位论文全文数据库（电子期刊）信息科技辑》, 15 June 2012 (2012-06-15), pages 138 - 1856 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717865A (en) * | 2019-09-02 | 2020-01-21 | 苏宁云计算有限公司 | Picture detection method and device |
| CN110717865B (en) * | 2019-09-02 | 2022-07-29 | 苏宁云计算有限公司 | Picture detection method and device |
| CN110476864A (en) * | 2019-09-17 | 2019-11-22 | 广东工业大学 | A kind of Intelligent fish tank system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109977812B (en) | Vehicle-mounted video target detection method based on deep learning | |
| CN103996041B (en) | Vehicle color identification method and system based on matching | |
| CN102096821B (en) | Number plate identification method under strong interference environment on basis of complex network theory | |
| CN102509291B (en) | Pavement disease detecting and recognizing method based on wireless online video sensor | |
| CN108268867B (en) | License plate positioning method and device | |
| CN107016362B (en) | Vehicle weight recognition method and system based on vehicle front windshield pasted mark | |
| CN110728283A (en) | License plate type identification method and device | |
| CN106934374B (en) | Method and system for identifying traffic signboard in haze scene | |
| CN106919910B (en) | Traffic sign identification method based on HOG-CTH combined features | |
| CN102867183B (en) | Method and device for detecting littered objects of vehicle and intelligent traffic monitoring system | |
| CN103810505A (en) | Vehicle identification method and system based on multilayer descriptors | |
| CN104715239A (en) | Vehicle color identification method based on defogging processing and weight blocking | |
| CN114998852A (en) | Intelligent detection method for road pavement diseases based on deep learning | |
| CN102902957A (en) | Video-stream-based automatic license plate recognition method | |
| CN103268489A (en) | Motor vehicle plate identification method based on sliding window searching | |
| CN103310199B (en) | Based on the vehicle model recognition methods of high-definition remote sensing data | |
| CN102915544A (en) | Video image motion target extracting method based on pattern detection and color segmentation | |
| CN103886760A (en) | Real-time vehicle type detection system based on traffic video | |
| CN110610153A (en) | Lane recognition method and system for automatic driving | |
| CN104200228A (en) | Recognizing method and system for safety belt | |
| CN105426863A (en) | Method and device for detecting lane line | |
| CN116168356B (en) | Vehicle damage judging method based on computer vision | |
| CN111144301A (en) | Road pavement defect quick early warning device based on degree of depth learning | |
| CN111462140A (en) | Real-time image instance segmentation method based on block splicing | |
| CN102306276A (en) | Method for identifying color of vehicle body in video vehicle image based on block clustering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130320 |