CN101964061A - Binary kernel function support vector machine-based vehicle type recognition method - Google Patents
Binary kernel function support vector machine-based vehicle type recognition method Download PDFInfo
- Publication number
- CN101964061A CN101964061A CN2010102714864A CN201010271486A CN101964061A CN 101964061 A CN101964061 A CN 101964061A CN 2010102714864 A CN2010102714864 A CN 2010102714864A CN 201010271486 A CN201010271486 A CN 201010271486A CN 101964061 A CN101964061 A CN 101964061A
- Authority
- CN
- China
- Prior art keywords
- support vector
- vehicle
- vector machine
- kernel function
- class
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a binary kernel function support vector machine-based vehicle type recognition method. The method is characterized by comprising the following steps of: (1) selecting effective vehicle characteristic information, standardizing the vehicle characteristic information for training and predication of support vector machines; (2) performing binary encoding on a vehicle type to be recognized, and constructing a support vector machine network according to the binary encoding; (3) training every binary support vector machine in the support vector machine network by using the selected vehicle characteristic information, simultaneously selecting an optimal parameter pair aiming at the network structure; and (4) inputting a sample characteristic vector to be classified by using the trained support vector machine network so as to acquire a predication result of the type and finish the vehicle type recognition. By performing the binary coding on the vehicle type, the support vector machine network used in the method of the invention has a simple construction and is easy to realize; the number of the support vector machines in the network is small, so that the recognition speed is improved, and the precision is higher, which contribute to practical application; simultaneously, by combining vehicle type characteristics with the experience data of the support vector machines in the method of selecting parameters of the binary support vector machines, the method has high fitness.
Description
Technical field
A kind of model recognizing method based on two class kernel function support vector machines of the present invention relates to the method that technology such as a kind of application mode identification is carried out type of vehicle identification, particularly utilizes the type of vehicle recognition methods of support vector machine.
Background technology
Vehicle detection and recognition technology are important subject in traffic monitoring, and it is specifically related to fields such as computer vision, Flame Image Process, data mining and pattern-recognition.Simultaneously, it also is basic information gathering technology in intelligent transportation and the wisdom city, is the basis of functions such as upper strata condition of road surface collection analysis, communications policy formulation.
Because information of vehicles mainly comes from the video of traffic monitoring, so the major function of vehicle detection is that information of vehicles is extracted from video; Vehicle identification then is the features of having extracted by analyzing such as contour of the vehicle, and vehicle is carried out Classification and Identification.Therefore, vehicle detection can be regarded the prerequisite of vehicle identification again as.At present, mainly contain background subtraction method, frame difference method and optical flow method etc. based on the vehicle checking method of video, these technology are comparative maturity all, and accuracy rate is higher.Can from image or video, extract the features such as frequency domain information of position, profile or the vehicle region of vehicle by vehicle inspection, utilize these features then, as the length breadth ratio of vehicle, the coefficient behind the frequency domain transform etc., carry out vehicle classification, behavior semanteme etc. by technology such as pattern-recognition and data minings and further analyze.And the present invention promptly is primarily aimed at the problem that vehicle characteristics that utilization extracted carries out vehicle identification, proposes a kind of model recognizing method based on support vector machine.
Support vector machine is the theory that proposes in late 1990s, and through development constantly, it has been widely used in pattern-recognition and data mining field.Support vector machine is mainly used in according to the proper vector of input target sample is classified the earliest from being used for the optimal planar theoretical developments of linear division.Suppose that sample characteristics is the vector x of a n dimension
i, and classification results y
i{ 0, the classification that sample is divided behind the 1} presentation class is 0 or 1 to ∈, that is:
Support vector machine is function sgn[f (x)], for the proper vector x of input, support vector machine is calculated corresponding output valve x ' by fundamental function f (x), then by discriminant function sgn[x '] calculate the classification (0 or 1) that sample belongs to.Early stage support vector machine can only be classified to two class problems of linear separability, but the proposition of kernel function method makes it can handle non-linear problem of dividing.Thereby so-called kernel function method promptly makes its linear separability by a kernel function with the inseparable sample characteristics of linearity DUAL PROBLEMS OF VECTOR MAPPING to a higher dimensional space; And, can solve the problem that multiclass is divided by support vector machine network by a plurality of two class support vector machines structures.But there are two shortcomings in existent method at present: the one, and support vector machine net structure complexity; The 2nd, the number of support vector machine is bigger in the support vector network, and speed is slow when directly causing classifying prediction, efficient is low, is difficult to practice.
Method described in the present invention is the two class support vector machines structure support vector machine network that utilizes the kernel function method by a plurality of, carries out the method for the type identification of multiclass vehicle according to selected contour of the vehicle feature.This method utilizes the binary coding mode to carry out the structure of support vector machine network, determines optimized parameter by cross validation, has reduced the number of vector machine in the vector machine network, simple structure, time efficiency height.
Summary of the invention
A kind of model recognizing method based on two class kernel function support vector machines of the present invention, the technical matters that solve is: utilize acquired vehicle characteristics information, the multiclass vehicle is carried out vehicle identification.The present invention proposes a kind of method of utilizing a plurality of support vector machine structure support vector machine networks to carry out vehicle identification, and this method is carried out binary coding to the type of vehicle that will discern, according to code construction support vector machine network; This network is input with selected vehicle characteristics vector, can discern type of vehicle quickly and efficiently; The support vector machine net structure is simple, is beneficial to realization, and efficient is higher.
The technical solution adopted for the present invention to solve the technical problems is: a kind of model recognizing method based on two class kernel function support vector machine networks, and it may further comprise the steps:
(1) chooses effective vehicle characteristics information, it is carried out training and the prediction of standardization to be used for support vector machine;
(2) type of vehicle that will discern is carried out binary coding, according to binary coding structure support vector machine network;
(3) utilize selected vehicle characteristics information, each two class support vector machines in the support vector machine network are trained, choose optimized parameter simultaneously at this network structure;
(4) the support vector machine network that utilizes training to finish is imported sample characteristics vector to be classified, and obtains its prediction result, finishes vehicle identification.
Effective characteristic information is meant acquired contour of the vehicle feature in the described step (1), mainly comprises following 6:
1) length to height ratio: the ratio of length of wagon and vehicle roof height;
2) schedule number: from the vehicle body side angle, the number of wheel;
3) the long ratio of wheel: wheel base from the ratio of length of wagon;
4) the long ratio in top: the ratio of vehicle roof length and vehicle body total length;
5) length ratio before and after: with the roof center line is the boundary, the length ratio of vehicle body front and rear part;
6) anterior and posterior height ratio: with the roof center line is the boundary, the average height ratio of vehicle body front and rear part.
The standardization of characteristic information is meant maximal value and the minimum value according to every eigenwert in the described step (1), and all data of every eigenwert are transformed into [0,1] interval linearly.
In the described step (2) type of vehicle is carried out binary coding and be meant,, then need to carry out if N kind type of vehicle is discerned to each type of vehicle distributes a binary numbering
The binary coding of position.As discerning to four kinds of type of vehicle, then only need carry out two binary coding, be respectively 00,01,10,11 totally four kinds.
The support vector machine net structure is carried out according to binary coding in the described step (2), according to digits of binary coding, and the number of two class vector machines in the decision network; Type of vehicle according to the corresponding confidential differentiation of support vector of binary coding 0/1 on each value decision.
Two all class support vector machines all adopt radially basic kernel function in the middle support vector machine network of described step (3), and the parameter that therefore needs to determine comprise punishment parameters C and kernel function parameter γ, and wherein radially basic kernel function formula is as follows:
All all shared identical parameters C of two class support vector machines and γ in the support vector machine network in the described step (3), both choose from multiple proportions is 2 Geometric Sequence respectively, and promptly C ∈ [2
12, 2
11, 2
10..., 2
-2] and γ ∈ [2
4, 2
3, 2
2..., 2
-10], then parameters C and γ have 225 kinds of combinations, finally choose the highest one group of its predictablity rate.
Choose the optimized parameter at this network structure in the described step (3), use cross validation method, be about to training and be divided into 5 equal portions with data, get wherein 4 parts of training respectively, its predictablity rate is write down in 1 part of prediction; So carry out 5 times, the mean value of getting 5 predictablity rates is as final predictablity rate.
The present invention proposes a kind of method of new structure two class support vector machines networks, and apply it in the vehicle identification.With respect to prior art, the method that proposes among the present invention has the following advantages:
(1) the support vector machine network utilisation in the method to the binary coding of vehicle, thereby make the number of support vector machine in the network much smaller than same class methods, thereby improved recognition speed, and have higher accuracy rate;
(2) construct the support vector machine network by binary coding in the method, simple structure is easy to realize, is beneficial to practical application;
(3) the selection of parameter method of two class support vector machines in the method has higher fitness in conjunction with the empirical data of the characteristics of vehicle feature and support vector machine.
Description of drawings
The model recognizing method process flow diagram of Fig. 1 for proposing;
Fig. 2 (a) is the binary-coded synoptic diagram of vehicle, and Fig. 2 (b) is for utilizing the synoptic diagram of vehicle binary coding structure support vector machine network.
Embodiment
Introduce the present invention in detail below in conjunction with the drawings and the specific embodiments.
The present invention relates to a kind of model recognizing method based on two class kernel function support vector machines, particular flow sheet as shown in Figure 1; Roughly can be divided into following four steps:
The first step, characteristic information pre-service.
Select the proper vector that is used for vehicle identification, and it is carried out standardization;
For improving the discrimination of vehicle, need choose suitable effective feature according to the profile characteristics of vehicle.According to the experience of vehicle identification and the main diacritical point of contour of the vehicle, mainly select following 6 features:
1) length to height ratio: the ratio of length of wagon and vehicle roof height;
2) schedule number: from the vehicle body side angle, the number of wheel;
3) the long ratio of wheel: wheel base from the ratio of length of wagon;
4) the long ratio in top: the ratio of vehicle roof length and vehicle body total length;
5) length ratio before and after: with the roof center line is the boundary, the length ratio of vehicle body front and rear part;
6) anterior and posterior height ratio: with the roof center line is the boundary, the average height ratio of vehicle body front and rear part.
Except that above-mentioned 6 features, also can choose other obtainable features, for specific vehicle, can choose its exclusive feature be used for identification, such as concavo-convex number feature of vehicle upper etc., also can utilize the coefficient that vehicle region is carried out behind the frequency domain transform to carry out the identification of vehicle as feature.Because the feature of choosing need be imported the support vector machine network as proper vector, so need carry out standardization to eigenwert.After the selected feature, all eigenwerts are carried out standardization, it is transformed into linearly [0,1] interval, promptly for each eigenwert x, get its maximal value Max (x) and minimum M in (x) respectively, the value x ' after its standardization calculates according to following formula:
After finishing the standardization of eigenwert, the eigenwert composition characteristic vector with all is used for the training and the prediction of support vector machine.Proper vector x is as shown in the formula expression:
x={x
1,x
2,x
3,x
4,x
5,x
6}
T
X wherein
1, x
2..., x
6Represent a concrete numerical value after 6 above-mentioned characteristic standardizations respectively.
Second step, structure support vector machine network.
At first the vehicle that will discern is carried out binary coding,, need so at most if the actual vehicle that will encode has the N kind
The position binary coding even will be discerned four kinds of type of vehicle, then only need carry out two binary coding, is respectively 00,01,10,11 totally four kinds.Shown in Fig. 2 (a), suppose and will discern 8 kinds of type of vehicle, then need to carry out 3 binary codings.
The structure of support vector machine is shown in Fig. 2 (b), and the value on each binary digit has determined the category division of one two class support vector machines accordingly.As shown in FIG., 0/1 value on binary-coded the 3rd has determined the category division of the support vector machine of ground floor, and wherein vehicle 1, the tri-bit encoding of 2,3,4 classes is 0, so actual 1 of vehicle that comprised of ' 0 ' class in two class support vector machines, 2,3,4 classes, and corresponding ' 1 ' class has comprised 5 of actual vehicle, 6,7,8 classes; And 0/1 value on binary-coded second has determined the category division of the support vector machine of the second layer, and wherein vehicle 1,2,5, the tri-bit encoding of 6 classes is 0, so actual 1,2 of the vehicle that comprised of ' 0 ' class in two class support vector machines, 5,6 classes, and corresponding ' 1 ' class has comprised 3,4 of actual vehicle, 7,8 classes.In such a manner, determine the category division of two all class support vector machines successively, shown in Fig. 2 (b).
The 3rd step, determine that optimized parameter is right, training support vector machine network.
Two all class support vector machines all use radially basic kernel function in the support vector machine network, and its formula is as follows:
Therefore the parameter that needs to determine comprises support vector machine punishment parameters C and kernel function parameter γ.For support vector machine different in the network, can use different parameters with its accuracy for predicting of further raising, but need carry out cross validation separately each support vector machine.So in reality was implemented, all two class support vector machines all used identical parameters C and γ in the setting consolidated network.And both choose from multiple proportions is 2 Geometric Sequence respectively, and promptly C ∈ [2
12, 2
11, 2
10..., 2
-2] and γ ∈ [2
4, 2
3, 2
2..., 2
-10], then parameter has 225 kinds of combinations to C and γ.
After structure is finished the support vector machine network structure, respectively each parameter is verified carrying out the M retransposing, be about to training data and be divided into M part, get M-1 part as predicted data, other 1 part as training data, its prediction accuracy of substitution network calculations, so iteration is M time, get at last the average accuracy predicted for M time as this parameter under final prediction accuracy.Cross validation can make each piece of data all once test as predicted data, thereby has improved the fitness of the utilization factor and the network structure of data.In actual the enforcement, need if data volume is less, can choose M=3 according to the concrete numerical value that how much determines M of data volume; If data volume is huge, can choose M=8 or M=10, and generally, according to data volume and in conjunction with the training experience data of SVM, can choose M=5 here and carry out cross validation.
After finishing cross validation, select wherein to predict the highest parameter of accuracy,, use all training datas that the support vector machine in the network is trained its substitution support vector machine network to C and γ.
The 4th goes on foot, imports proper vector to be measured, the output recognition result.
For vehicle characteristics vector to be measured, with its input support vector machine network.The support vector machine network only needs it is carried out at most
The two class support vector machines prediction of inferior (supposing has N kind vehicle) can obtain it is carried out the result that vehicle is discerned.According to encoded radio and the vehicle coding schedule of output, can obtain the vehicle classification number of reality.
In addition when vehicle is encoded, can investigate the similarity of vehicle between in twos and the ratio of possible data volume thereof in advance, by choosing different coded systems, further improve the recognition correct rate and the recognition speed of this method.
Claims (7)
1. model recognizing method based on two class kernel function support vector machines, described method is characterised in that and may further comprise the steps:
(1) chooses effective vehicle characteristics information, it is carried out training and the prediction of standardization to be used for support vector machine;
(2) type of vehicle that will discern is carried out binary coding, according to binary coding structure support vector machine network;
(3) utilize selected vehicle characteristics information, each two class support vector machines in the support vector machine network are trained, choose optimized parameter simultaneously at this network structure;
(4) the support vector machine network that utilizes training to finish is imported sample characteristics vector to be classified, and obtains its prediction result, finishes vehicle identification.
2. a kind of model recognizing method according to claim 1 based on two class kernel function support vector machines, it is characterized in that: effective vehicle characteristics information is meant acquired contour of the vehicle feature in the described step (1), mainly comprise length to height ratio, schedule number, take turns length ratio, the long ratio in top, front and back length ratio, anterior and posterior height ratio totally 6 vehicle body side characteristics.
3. a kind of model recognizing method according to claim 1 based on two class kernel function support vector machines, it is characterized in that: the standardization of vehicle characteristics information is meant maximal value and the minimum value according to every kind of eigenwert in the described step (1), all data of every eigenwert are transformed into [0,1] interval linearly.
4. a kind of model recognizing method according to claim 1 based on two class kernel function support vector machines, it is characterized in that: in the described step (2) type of vehicle is carried out binary coding and be meant and be that each type of vehicle distributes a binary numbering, if N kind type of vehicle is discerned, then need to carry out
The binary coding of position.
5. a kind of model recognizing method according to claim 1 based on two class kernel function support vector machines, it is characterized in that: the support vector machine net structure is carried out according to binary coding in the described step (2), according to digits of binary coding, the number of two class vector machines in the decision network; Type of vehicle according to the corresponding confidential differentiation of support vector of the value decision of binary coding on each.
6. a kind of model recognizing method according to claim 1 based on two class kernel function support vector machines, it is characterized in that: two all class support vector machines use identical punishment parameters C and radial kernel function parameter γ in the middle support vector machine network of described step (3), both choose from multiple proportions is 2 Geometric Sequence respectively, and promptly C ∈ [2
12, 2
11, 2
10..., 2
-2] and γ ∈ [2
4, 2
3, 2
2..., 2
-10], so parameters C and γ have 225 kinds of combinations.
7. a kind of model recognizing method according to claim 1 based on two class kernel function support vector machines, it is characterized in that: choose optimized parameter in the described step (3) at this network structure, use cross validation method, be about to training and be divided into the M equal portions with data, getting wherein respectively, M-1 part is used for training, other 1 part is used for prediction, writes down its predictablity rate; So carry out M time, each piece of data has all been carried out once test as predicted data, and the mean value of getting M predictablity rate is as final predictablity rate; Optimized parameter is to being chosen in one group of the highest parameters C of predictablity rate final in the cross validation and γ.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010271486.4A CN101964061B (en) | 2010-09-02 | 2010-09-02 | Binary kernel function support vector machine-based vehicle type recognition method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010271486.4A CN101964061B (en) | 2010-09-02 | 2010-09-02 | Binary kernel function support vector machine-based vehicle type recognition method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101964061A true CN101964061A (en) | 2011-02-02 |
CN101964061B CN101964061B (en) | 2014-06-18 |
Family
ID=43516926
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201010271486.4A Expired - Fee Related CN101964061B (en) | 2010-09-02 | 2010-09-02 | Binary kernel function support vector machine-based vehicle type recognition method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101964061B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102880881A (en) * | 2012-09-25 | 2013-01-16 | 常州大学 | Method for identifying car type on basis of binary support vector machines and genetic algorithm |
CN103136537A (en) * | 2012-12-12 | 2013-06-05 | 惠州学院 | Vehicle type identification method based on support vector machine |
CN105407364A (en) * | 2015-10-27 | 2016-03-16 | 四川长虹电器股份有限公司 | Channel comprehensive competitiveness realization method based on intelligent television rating system |
CN106127228A (en) * | 2016-06-16 | 2016-11-16 | 北方工业大学 | Remote sensing image ship detection candidate area identification method based on decision template classifier fusion |
CN106203464A (en) * | 2016-06-23 | 2016-12-07 | 长安大学 | A kind of based on the model recognizing method setting up membership function |
CN106971548A (en) * | 2017-05-18 | 2017-07-21 | 福州大学 | The Short-time Traffic Flow Forecasting Methods of optimizable multi-core adaptive SVMs |
CN108345794A (en) * | 2017-12-29 | 2018-07-31 | 北京物资学院 | The detection method and device of Malware |
-
2010
- 2010-09-02 CN CN201010271486.4A patent/CN101964061B/en not_active Expired - Fee Related
Non-Patent Citations (4)
Title |
---|
周磊 等: "最小二乘分解算法在车型识别中的应用", 《计算机仿真》 * |
尹玉梅 等: "基于支持向量机(SVM)的汽车车型识别", 《电子测量技术》 * |
范伊红 等: "基于SVM的车型识别系统的设计与实现", 《微计算机信息》 * |
黄洁; 张海: "利用支持向量机的飞机目标检测", 《电光与控制》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102880881A (en) * | 2012-09-25 | 2013-01-16 | 常州大学 | Method for identifying car type on basis of binary support vector machines and genetic algorithm |
CN103136537A (en) * | 2012-12-12 | 2013-06-05 | 惠州学院 | Vehicle type identification method based on support vector machine |
CN103136537B (en) * | 2012-12-12 | 2017-02-08 | 惠州学院 | Vehicle type identification method based on support vector machine |
CN105407364A (en) * | 2015-10-27 | 2016-03-16 | 四川长虹电器股份有限公司 | Channel comprehensive competitiveness realization method based on intelligent television rating system |
CN105407364B (en) * | 2015-10-27 | 2018-07-03 | 四川长虹电器股份有限公司 | Based on channel synthesized competitiveness implementation method under smart television audience ratings system |
CN106127228A (en) * | 2016-06-16 | 2016-11-16 | 北方工业大学 | Remote sensing image ship detection candidate area identification method based on decision template classifier fusion |
CN106203464A (en) * | 2016-06-23 | 2016-12-07 | 长安大学 | A kind of based on the model recognizing method setting up membership function |
CN106971548A (en) * | 2017-05-18 | 2017-07-21 | 福州大学 | The Short-time Traffic Flow Forecasting Methods of optimizable multi-core adaptive SVMs |
CN106971548B (en) * | 2017-05-18 | 2019-06-07 | 福州大学 | The Short-time Traffic Flow Forecasting Methods of optimizable multi-core adaptive support vector machines |
CN108345794A (en) * | 2017-12-29 | 2018-07-31 | 北京物资学院 | The detection method and device of Malware |
Also Published As
Publication number | Publication date |
---|---|
CN101964061B (en) | 2014-06-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101964061B (en) | Binary kernel function support vector machine-based vehicle type recognition method | |
CN102663100B (en) | Two-stage hybrid particle swarm optimization clustering method | |
CN101853392B (en) | Remote sensing hyperspectral image band selection method based on conditional mutual information | |
CN101729872B (en) | Video monitoring image based method for automatically distinguishing traffic states of roads | |
CN101551809B (en) | Search method of SAR images classified based on Gauss hybrid model | |
CN103366367B (en) | Based on the FCM gray-scale image segmentation method of pixel count cluster | |
CN107644426A (en) | Image, semantic dividing method based on pyramid pond encoding and decoding structure | |
CN113343563B (en) | Landslide susceptibility evaluation method based on automatic sample selection and surface deformation rate | |
CN103426004B (en) | Model recognizing method based on error correcting output codes | |
CN105913450A (en) | Tire rubber carbon black dispersity evaluation method and system based on neural network image processing | |
CN109558969A (en) | A kind of VANETs car accident risk forecast model based on AdaBoost-SO | |
CN104156943A (en) | Multi-target fuzzy cluster image variance detecting method based on non-control-neighborhood immune algorithm | |
CN105225227A (en) | The method and system that remote sensing image change detects | |
CN104794496A (en) | Remote sensing character optimization algorithm for improving mRMR (min-redundancy max-relevance) algorithm | |
CN113408561B (en) | Model generation method, target detection method, device, equipment and storage medium | |
CN104537353A (en) | Three-dimensional face age classifying device and method based on three-dimensional point cloud | |
CN103593855A (en) | Clustered image splitting method based on particle swarm optimization and spatial distance measurement | |
CN102945553A (en) | Remote sensing image partition method based on automatic difference clustering algorithm | |
CN116110234B (en) | Traffic flow prediction method and device based on artificial intelligence and application of traffic flow prediction method and device | |
Fang et al. | Multiscale CNNs ensemble based self-learning for hyperspectral image classification | |
CN102663684B (en) | SAR image segmentation method based on Gauss mixing model parameter block migration clustering | |
CN104268560B (en) | A kind of land use recognition methods based on remote Sensing Interpretation | |
Chen et al. | A mobile intelligent guide system for visually impaired pedestrian | |
CN106529391A (en) | Robust speed-limit traffic sign detection and recognition method | |
CN102880881A (en) | Method for identifying car type on basis of binary support vector machines and genetic algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140618 Termination date: 20160902 |
|
CF01 | Termination of patent right due to non-payment of annual fee |