CN103279944A - Image division method based on biogeography optimization - Google Patents
Image division method based on biogeography optimization Download PDFInfo
- Publication number
- CN103279944A CN103279944A CN2013101412945A CN201310141294A CN103279944A CN 103279944 A CN103279944 A CN 103279944A CN 2013101412945 A CN2013101412945 A CN 2013101412945A CN 201310141294 A CN201310141294 A CN 201310141294A CN 103279944 A CN103279944 A CN 103279944A
- Authority
- CN
- China
- Prior art keywords
- rate
- cluster centre
- image
- clustering
- individuality
- 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 belongs to the technical field of image processing, in particular to an image division method based on biogeography optimization capable of being used for image enhancement, mode identification, target tracking and the like. The method comprises the following steps that parameters are initialized; pictures to be divided are input, and clustering centers are initialized; a fuzzy matrix is calculated; the clustering centers are calculated again; the fitness value of each clustering center is extracted; the immigration rate and the emigration rate of each emigration center are extracted; the emigration centers are updated according to mutation operators; and division results are output. Because the biogeography migration strategy is adopted for optimizing the clustering centers, the calculation quantity is reduced, the selected clustering centers after the optimization have the overall situation characteristics, the problem of initialization sensitivity of the traditional clustering algorithm is solved, and the stability and the clustering performance of the clustering algorithm are improved. Because the immigration and emigration strategies are introduced for optimizing the clustering centers, the data center distribution characteristics can be more accurately reflected, in addition, the corresponding clustering center updating rules are designed, and the calculation quantity is reduced.
Description
Technical field
The invention belongs to technical field of image processing, be specifically related to the image partition method in the technical fields such as a kind of figure of can be used for image intensifying, pattern-recognition, target following.
Background technology
Image Segmentation Technology is one of problem that the difficulty biggest obstacle is maximum in machine vision and the multimedia application technology always.This is because the researchist does not also work out a kind of automatic partitioning algorithm general, good reliability up to now.So-called image Segmentation Technology, existing analytical approach all only are at the image in a certain field or the image of a certain kind.Therefore, in several years of future, the image technique field presses for the problem of solution, remains the technological breakthrough that image is cut apart.
The fuzzy clustering algorithm is cut apart the application in field at image, has become a kind of requirement the most widely.1979, people such as Coleman were cut apart with regard to proposing to carry out image with clustering algorithm.Through people's practice and study successively in decades, the researchist has proposed multiple gray level image based on fuzzy clustering one after another and has cut apart new method.These algorithmic methods have obtained very big achievement at aspects such as cutting apart texture image, remote sensing images, sequence image.
Inborn advantages such as fuzzy K mean algorithm is fast with its computing velocity, local search ability strong, simplicity of design, the simple easy to understand of algorithm have obtained the favor that image is cut apart the field.Yet traditional fuzzy K mean algorithm is also because its born shortcoming
[4], make it when the application that image is cut apart, demonstrated very big problem.
On the one hand, the function of handling noise image owing to traditional fuzzy K mean algorithm is very poor, just poor robustness.This is because this kind algorithm has only utilized the half-tone information of image, and does not utilize other features such as image space information, thereby this parted pattern is incomplete, thereby causes it can only be used in the middle of the very low image of noise content.
On the other hand, the selection of the initial cluster center of traditional fuzzy K mean algorithm and determine has very big influence to the cluster result of back, and is wrong if initial cluster center is selected, and then can cause result far from each other.If it is improper that initial cluster center is selected, objective function might can not get global optimum, and is absorbed in local minimum; The ability of searching optimum of algorithm own is also poor, and for the data of multi-peak, simple fuzzy K mean algorithm is easy to be absorbed in local minimum.Unsettled like this performance makes traditional fuzzy K mean algorithm in the application that image is cut apart, and the mistake of cutting apart often occurs.
Cut apart the trouble that runs in the field in order to overcome traditional fuzzy K mean algorithm at image, the researchist has done a large amount of improvements.These improvements, two problems at above-mentioned also can be divided into two aspects: at the improvement of fuzzy K mean algorithm itself, with improved algorithm application in image is cut apart, in the hope of obtaining better segmentation effect; Enlarge the feature extraction scope of fuzzy K mean algorithm, utilize other space characteristics except pixel grey scale information, reaching more objective and cluster effect widely, thus the quality that the raising image is cut apart.
By the selection introduction of above-mentioned cluster centre as can be known, cause the main reason of these defectives, being exactly the final cluster effect of traditional fuzzy K mean algorithm relies on too strong to the selection of initial cluster center.In case it is improper to cause initial cluster center to be selected, then cluster will be failed.Therefore, when improving initial cluster center, find out and can make algorithm no longer rely on initial cluster center, also being that algorithm is improved has a force direction.Biological intelligence is optimized algorithm, has unsupervised learning and ability of searching optimum, is combining with traditional fuzzy K mean algorithm, and there is very big conduct the ability of searching optimum aspect of improving algorithm.At present, there has been the scholar that ant group algorithm and immune genetic are optimized algorithm in the middle of the improvement of bluring the K mean algorithm, referring to: Yang Licai, Zhao Lina, Wu Xiaoqing is based on the fuzzy mean cluster medical image segmentation of ant group algorithm, journal of Shandong university (engineering version), 37(3), 51-56; Sun Xiujuan is based on the K means clustering algorithm analysis and research of genetic algorithm, Shandong Normal University's Master's thesis, 2009.Liu Yuying is with the look for food clustering criteria function of the fuzzy K average of algorithm optimization of bacterium, referring to: bacterium clustering algorithm and the application of cutting apart at image thereof, Harbin Engineering University's Master's thesis, 2011.
Concrete improvement algorithm based on these types is varied, in application category separately, is greatly improved for the applicability of algorithm.But, under different application directions, but do not have a general image partitioning algorithm.Therefore, the task of top priority that to study a kind of effective image partition method be the art scientific and technical personnel.
Summary of the invention
The objective of the invention is to propose a kind ofly reduce traditional cluster segmentation method to the susceptibility of initial cluster center, make that the image clustering segmentation result is more stable, the edge is more level and smooth, the fuzzy K mean cluster center of regional consistance better optimize to be to carry out the method that image is cut apart.
The object of the present invention is achieved like this:
The present invention includes following steps:
(1) classification is counted C, fuzzy exponent m, population size n, iterations N, the rate of moving into function maximal value I, the rate of moving out function maximal value E, mobility p
Modify, individual number k and the maximum variation probability m of retaining of elite
MaxCarry out initialization;
(2) input picture to be split, cromogram is converted to gray-scale map, and image information is converted to the gradation data of image, initialization cluster centre;
(3) calculate the fuzzy matrix that k cluster centre degree of membership constitutes;
(4) recomputate k cluster centre;
(5) extract the fitness value of each cluster centre, judge whether to satisfy maximum iteration time T stop condition, just stop if satisfying, the output optimum solution; If do not satisfy then execution in step (6);
(6) extract the rate of moving into of each cluster centre and the rate of moving out, revise cluster centre according to migration strategy, extract suitability degree again;
(7) upgrade cluster centre according to mutation operator, extract suitability degree again;
(8) if do not reach iterations T, execution in step (3) is carried out next iteration; If reach iterations, the output segmentation result.
The step that migration strategy is revised cluster centre is:
(1) initialization: the iteration algebraically counter t=0 of migration strategy is set, migration strategy greatest iteration algebraically T is set, with the cluster centre individuality that generates as initial population P (0);
(2) individual evaluation: calculate fitness and population number individual among the P of colony (t), extract each individuality move into rate λ and the rate μ that moves out, and each individuality in the population is deposited into pbest (t), the rate of will moving into is minimum, the highest individuality of the rate of moving out is deposited among the gbest (t), wherein, pbest (t) is the matrix identical with P (t) size structure, and gbest (t) only stores a solution;
(3) operator of moving into of moving out: all individualities are carried out the entry/leave operation according to move into rate λ and the rate μ that moves out of each individuality;
(4) mutation operator: according to the variation probability of being obtained by population number half minimum individuality is made a variation, replace old solution with obtaining new solution;
(5) judge end condition: if t=T then exports as optimum cluster centre solution with the individuality with maximum adaptation degree.
Beneficial effect of the present invention is:
The present invention has reduced calculated amount owing to adopt the biogeography migration strategy to optimize cluster centre; Optimize thus and select cluster centre and have global property, thereby overcome the initialization tender subject of traditional clustering algorithm, promoted stability and the cluster performance of clustering algorithm; The present invention reflects data center's characteristic distributions more accurately, and has designed corresponding cluster centre update rule owing to introduce the policy optimization cluster centre of entry/leave, reduced calculated amount.
Description of drawings
Fig. 1 is the FB(flow block) of performing step of the present invention;
Fig. 2 is standard picture Lena to be split;
Fig. 3 carries out image with the inventive method to Fig. 2 to cut apart the The simulation experiment result figure that obtains;
Fig. 4 carries out image with existing fuzzy K mean algorithm to Fig. 2 to cut apart the The simulation experiment result figure that obtains.
Fig. 5 is standard picture Peppers to be split;
Fig. 6 carries out image with the inventive method to Fig. 5 to cut apart the The simulation experiment result figure that obtains;
Fig. 7 carries out image with existing fuzzy K mean algorithm to Fig. 5 to cut apart the The simulation experiment result figure that obtains.
Embodiment
Technical scheme of the present invention is that the biogeography migration strategy is introduced in the clustering algorithm to reach the better cluster performance, cluster centre optimization method based on the biogeography migration strategy has been proposed, solve the computation complexity problem that fuzzy clustering center uncertainty is brought, obtain new image partition method.
With reference to Fig. 1, specific implementation step of the present invention is as follows:
Step 1: a plurality of parameters of initialization.
It is artificial fixed that classification is counted C, and according to the difference of picture, the classification number of cutting apart is also different.
Fuzzy exponent m according to present stage applying frequency the highest, the optimal value of generally acknowledging: 2.
Population size n is artificial fixed, and this paper is decided to be 30; Iterations N is 500; Elite's encumbrance k is 2.The rate of moving into function maximal value I and the rate function maximal value E that moves out are made as 1, maximum variation probability m
MaxBe 0.005, mobility p
ModifyBe 0.8.
Step 2:
If colour picture will change it into the grey picture, according to the variation of gray scale pictorial information is processed into a pair of data.Picked at random n cluster centre.
Step 3:
If V=[v
1, v
2..., v
c], v
i∈ R
nBe the cluster centre vector, R is real number space, and n is dimension, i=1, and 2 ..., c, c are the cluster centre number,
Represent given sample, S is the sample space dimension.
Data set x
k(k=1,2 ... n, n are the data number) and cluster centre v
iEuclidean distance be calculated as follows:
Calculate the degree of membership μ that i sample belongs to k cluster centre
Ik, formula is as follows:
U=μ
IjIt is the fuzzy partition matrix of a c * n.
Step 4:
According to formula
Upgrade each cluster centre, then new cluster centre is replaced original cluster centre.
Step 5:
Criterion function with the Fuzzy C mean algorithm
As the fitness function of BBO-FCM algorithm, fitness value is more little, and corresponding solution is just more good.Bring the cluster centre that last step draws into criterion function, if satisfy end condition ‖ v
(k)-v
(k+1)‖＜e then stops, the output optimum solution, otherwise continue next step.
Step 6:
Utilize P
ModifyThe habitat i operation of whether moving into is judged in circulation (island quantity is as cycle index).Be determined the operation of moving into as if island i, then the recycling rate λ that moves into
iJudge the characteristic component X of island i
IjThe operation (problem dimension D is as cycle index) of whether moving into.If the characteristic component X of island i
i(j) be determined, then utilize the rate of the moving out μ on other island
iCarrying out wheel disc selects.The corresponding position of island i is replaced in the corresponding position of selecting island k.Recomputate the suitability degree of island i.
Step 7:
Upgrade the population quantity probability P on each island according to formula
iCalculate the mutation rate on each island then according to formula, carry out mutation operation, m is used on each non-elite island that make a variation
iWhether certain characteristic component of judging island i suddenlys change.Recomputate the fitness of island i;
Step 8:
Judge whether iterations reaches, and has reached then to stop, otherwise forward step 3 to.
This algorithm advantage:
Biogeography is optimized algorithm, itself has good ability of searching optimum.And the shortcoming of traditional fuzzy K means clustering algorithm maximum is exactly to be absorbed in local minimum easily, and ability of searching optimum is poor.So the ability of searching optimum of utilizing biogeography to optimize algorithm is helped fuzzy K average and found global minimum, just makes the value of criterion function minimum, the advantage of the two kinds of algorithms that so just neutralized also is the core place of new algorithm.Effect of the present invention can further specify by following emulation:
1. simulated conditions and emulation content:
This example is under Intel (R) Core (TM) 2Duo CPU2.4GHz Windows XP system, and on the Matlab7.0 operation platform, the image of finishing the present invention and fuzzy K mean cluster is cut apart emulation experiment.
2. emulation experiment content
A. the emulation of image partition method of the present invention
The present invention is applied in as shown in Figure 2 on the standard testing image Lena, and this image can roughly be divided into six zones such as cap, face, background (forward and backward), shoulder, hair, so C is made as 6.Fig. 3 is cut apart the The simulation experiment result figure that obtains for the inventive method Fig. 2 being carried out image.
The picture that Fig. 5 peppers is made up of pimiento and green chili is so C is made as 8.Fig. 6 is cut apart the The simulation experiment result figure that obtains for the inventive method Fig. 5 being carried out image.
B. the emulation of existing fuzzy K average image clustering dividing method, the cut zone complexity is established C=8;
Existing fuzzy K mean cluster method is applied on as shown in Figure 2 the standard picture, and The simulation experiment result as shown in Figure 4.Fig. 7 is cut apart the The simulation experiment result figure that obtains for existing fuzzy K mean cluster method is applied in the image that obtains as Fig. 5.
3. The simulation experiment result
As can be seen from Figure 3, the The simulation experiment result that the present invention obtains has subjective vision effect preferably, and by the segmentation result of lena figure as can be seen, the effect of cutting apart based on the clustering algorithm image of biogeography optimization is best.As a whole, the fully aware of of the boundary segmentation between lena and the background understood, do not have fuzzy situation.
As can be seen from Figure 4, the The simulation experiment result subjective vision effect that existing clustering method obtains is relatively poor, and erroneous segmentation is serious, and edge fog is unclear, and regional consistance is low, accurately six class zones in the component-bar chart 2.
Can be illustrated by above emulation experiment, at cutting apart of this image, there is certain advantage in the present invention, overcome existing fuzzy K average cutting techniques and be applied in deficiency on the Lena image, no matter be visual effect or sliced time, the present invention all is better than existing fuzzy K mean cluster cutting techniques.
As can be seen from Figure 6, can tell red and green based on the clustering algorithm of biogeography optimization, be black after pimiento is cut apart, be white after green chili is cut apart, red and green boundary segmentation fully aware of, it is very coherent that the shape lines of capsicum are also cut apart, and can successfully be partitioned into red and algorithm green chili.Existing fuzzy K mean algorithm is distinguished the clustering algorithm of optimizing not as based on biogeography to red, green chili.
In sum, the segmentation effect that the present invention is directed to Lena image and Peppers image obviously is better than existing fuzzy K average cutting techniques to the segmentation effect of Lena image.
Claims (2)
1. an image partition method of optimizing based on biogeography is characterized in that, comprises the steps:
(1) classification is counted C, fuzzy exponent m, population size n, iterations N, the rate of moving into function maximal value I, the rate of moving out function maximal value E, mobility p
Modify, individual number k and the maximum variation probability m of retaining of elite
MaxCarry out initialization;
(2) input picture to be split, cromogram is converted to gray-scale map, and image information is converted to the gradation data of image, initialization cluster centre;
(3) calculate the fuzzy matrix that k cluster centre degree of membership constitutes;
(4) recomputate k cluster centre;
(5) extract the fitness value of each cluster centre, judge whether to satisfy maximum iteration time T stop condition, just stop if satisfying, the output optimum solution; If do not satisfy then execution in step (6);
(6) extract the rate of moving into of each cluster centre and the rate of moving out, revise cluster centre according to migration strategy, extract suitability degree again;
(7) upgrade cluster centre according to mutation operator, extract suitability degree again;
(8) if do not reach iterations T, execution in step (3) is carried out next iteration; If reach iterations, the output segmentation result.
2. a kind of image partition method of optimizing based on biogeography according to claim 1 is characterized in that, the step of described migration strategy modification cluster centre is:
(1) initialization: the iteration algebraically counter t=0 of migration strategy is set, migration strategy greatest iteration algebraically T is set, with the cluster centre individuality that generates as initial population P (0);
(2) individual evaluation: calculate fitness and population number individual among the P of colony (t), extract each individuality move into rate λ and the rate μ that moves out, and each individuality in the population is deposited into pbest (t), the rate of will moving into is minimum, the highest individuality of the rate of moving out is deposited among the gbest (t), wherein, pbest (t) is the matrix identical with P (t) size structure, and gbest (t) only stores a solution;
(3) operator of moving into of moving out: all individualities are carried out the entry/leave operation according to move into rate λ and the rate μ that moves out of each individuality;
(4) mutation operator: according to the variation probability of being obtained by population number half minimum individuality is made a variation, replace old solution with obtaining new solution;
(5) judge end condition: if t=T then exports as optimum cluster centre solution with the individuality with maximum adaptation degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101412945A CN103279944A (en) | 2013-04-22 | 2013-04-22 | Image division method based on biogeography optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101412945A CN103279944A (en) | 2013-04-22 | 2013-04-22 | Image division method based on biogeography optimization |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103279944A true CN103279944A (en) | 2013-09-04 |
Family
ID=49062451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013101412945A Pending CN103279944A (en) | 2013-04-22 | 2013-04-22 | Image division method based on biogeography optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103279944A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105976379A (en) * | 2016-05-11 | 2016-09-28 | 南京邮电大学 | Fuzzy clustering color image segmentation method based on cuckoo optimization |
| CN106708901A (en) * | 2015-11-17 | 2017-05-24 | 北京国双科技有限公司 | Clustering method and device of search terms in website |
| CN107832885A (en) * | 2017-11-02 | 2018-03-23 | 南京航空航天大学 | A kind of fleet Algorithm of Firepower Allocation based on adaptive-migration strategy BBO algorithms |
| CN108241911A (en) * | 2018-01-30 | 2018-07-03 | 合肥工业大学 | A kind of Site Selection Method of Distribution Center for optimization algorithm of being looked for food based on bacterium |
| CN109509196A (en) * | 2018-12-24 | 2019-03-22 | 广东工业大学 | A kind of lingual diagnosis image partition method of the fuzzy clustering based on improved ant group algorithm |
| CN114240807A (en) * | 2022-02-28 | 2022-03-25 | 山东慧丰花生食品股份有限公司 | Peanut aflatoxin detection method and system based on machine vision |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655911A (en) * | 2009-05-11 | 2010-02-24 | 华北电力大学（保定） | Mode identification method based on immune antibody network |
-
2013
- 2013-04-22 CN CN2013101412945A patent/CN103279944A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655911A (en) * | 2009-05-11 | 2010-02-24 | 华北电力大学（保定） | Mode identification method based on immune antibody network |
Non-Patent Citations (1)
| Title |
|---|
| 朱丽莉，李真真: "基于生物地理学模糊C均值聚类的图像分割算法", 《应用科技》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106708901A (en) * | 2015-11-17 | 2017-05-24 | 北京国双科技有限公司 | Clustering method and device of search terms in website |
| CN105976379A (en) * | 2016-05-11 | 2016-09-28 | 南京邮电大学 | Fuzzy clustering color image segmentation method based on cuckoo optimization |
| CN107832885A (en) * | 2017-11-02 | 2018-03-23 | 南京航空航天大学 | A kind of fleet Algorithm of Firepower Allocation based on adaptive-migration strategy BBO algorithms |
| CN107832885B (en) * | 2017-11-02 | 2022-02-11 | 南京航空航天大学 | Ship formation fire power distribution method based on self-adaptive migration strategy BBO algorithm |
| CN108241911A (en) * | 2018-01-30 | 2018-07-03 | 合肥工业大学 | A kind of Site Selection Method of Distribution Center for optimization algorithm of being looked for food based on bacterium |
| CN108241911B (en) * | 2018-01-30 | 2021-06-29 | 合肥工业大学 | Distribution center site selection method based on bacterial foraging optimization algorithm |
| CN109509196A (en) * | 2018-12-24 | 2019-03-22 | 广东工业大学 | A kind of lingual diagnosis image partition method of the fuzzy clustering based on improved ant group algorithm |
| CN109509196B (en) * | 2018-12-24 | 2023-01-17 | 广东工业大学 | Tongue diagnosis image segmentation method based on fuzzy clustering of improved ant colony algorithm |
| CN114240807A (en) * | 2022-02-28 | 2022-03-25 | 山东慧丰花生食品股份有限公司 | Peanut aflatoxin detection method and system based on machine vision |
| CN114240807B (en) * | 2022-02-28 | 2022-05-17 | 山东慧丰花生食品股份有限公司 | Peanut aflatoxin detection method and system based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110287932B (en) | Road blocking information extraction method based on deep learning image semantic segmentation | |
| CN109801256B (en) | Image aesthetic quality assessment method based on region of interest and global features | |
| CN105741279B (en) | The quick image partition method for suppressing fuzzy clustering based on rough set | |
| Rahaman et al. | An efficient multilevel thresholding based satellite image segmentation approach using a new adaptive cuckoo search algorithm | |
| CN107977671A (en) | A kind of tongue picture sorting technique based on multitask convolutional neural networks | |
| CN103279944A (en) | Image division method based on biogeography optimization | |
| CN108537102A (en) | High Resolution SAR image classification method based on sparse features and condition random field | |
| CN105096259A (en) | Depth value restoration method and system for depth image | |
| CN106780485A (en) | SAR image change detection based on super-pixel segmentation and feature learning | |
| CN104408469A (en) | Firework identification method and firework identification system based on deep learning of image | |
| CN109740588A (en) | The X-ray picture contraband localization method reassigned based on the response of Weakly supervised and depth | |
| CN103136504A (en) | Face recognition method and device | |
| CN107808138B (en) | Communication signal identification method based on FasterR-CNN | |
| CN105574063A (en) | Image retrieval method based on visual saliency | |
| CN112837344B (en) | Target tracking method for generating twin network based on condition countermeasure | |
| CN100545865C (en) | A kind of automatic division method that image initial partitioning boundary is optimized | |
| CN104537647A (en) | Target detection method and device | |
| CN102496023A (en) | Region of interest extraction method of pixel level | |
| CN113569724B (en) | Road extraction method and system based on attention mechanism and dilation convolution | |
| CN103473786A (en) | Gray level image segmentation method based on multi-objective fuzzy clustering | |
| CN114399480A (en) | Method and device for detecting severity of vegetable leaf disease | |
| CN104156943B (en) | Multi objective fuzzy cluster image change detection method based on non-dominant neighborhood immune algorithm | |
| Xing et al. | Traffic sign recognition using guided image filtering | |
| CN109191434A (en) | Image detecting system and detection method in a kind of cell differentiation | |
| CN104408731B (en) | Region graph and statistic similarity coding-based SAR (synthetic aperture radar) image segmentation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130904 |