CN101526955A - Method for automatically withdrawing draft-based network graphics primitives and system thereof - Google Patents
Method for automatically withdrawing draft-based network graphics primitives and system thereof Download PDFInfo
- Publication number
- CN101526955A CN101526955A CN200910081069A CN200910081069A CN101526955A CN 101526955 A CN101526955 A CN 101526955A CN 200910081069 A CN200910081069 A CN 200910081069A CN 200910081069 A CN200910081069 A CN 200910081069A CN 101526955 A CN101526955 A CN 101526955A
- Authority
- CN
- China
- Prior art keywords
- pel
- matching degree
- sketch
- picture
- key word
- 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
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method for automatically withdrawing draft-based network graphics primitives. The method consists of the following steps: firstly, receiving keywords and drafts of target-describing graphics primitives input by subscribers and then acquiring a set of corresponding pictures according to keywords; utilizing visual attention calculation to analyze the visual attention area of each picture; taking the visual attention area as the initial value and acquiring a set of initial graphics primitives by an image withdrawing method; finally ordering the acquired initial graphics primitives according to the matching degree of keywords and drafts and selecting a plurality of graphics primitives with high matching degree as the target graphics primitives. The invention also discloses a system for automatically withdrawing draft-based network graphics primitives. The invention can automatically acquire the graphics primitives expected by subscribers according to keywords and drafts input by subscribers and utilize a picture search engine to withdraw graphics primitives in huge volumes of network pictures. In addition, the invention with simple input and less artificial interference reduces the cost and improves the efficiency.
Description
Technical field
The present invention relates to technical field of image processing, particularly relate to a kind of network pel extraction method and system based on sketch.
Background technology
Make up at virtual scene, in many important use fields such as image is synthetic, material database structure, obtaining of pel material is one of stubborn problem the most always.Though the method for artificial mark can access the very high pictorial element of quality.But this method workload is too big, often need a large amount of special mark personnel to carry out, thereby the procurement cost of material is very high.In above-mentioned application, the quality of net result and obtain efficient and depend critically upon quality and the efficient of setting up above-mentioned pel storehouse.Be difficult to provide kind to enrich, satisfy the pel storehouse that design requirement also can make up rapidly by artificial mode.And the cost of artificial mark also can cause the cost of these application sharply to rise.
Summary of the invention
The problem to be solved in the present invention provides a kind of network pel extraction method and system based on sketch, sets up the poor efficiency of material database, expensive, defective that material is incomplete to overcome in the prior art artificial mark mode.
For achieving the above object, technical scheme of the present invention provides a kind of network pel extraction method based on sketch, said method comprising the steps of: the key word and the sketch that receive the description target pel of user's input; Obtain the set of diagrams sheet corresponding according to described key word with described key word; Utilize the vision attention zone in the every width of cloth picture of vision attention Algorithm Analysis, and utilize described vision attention zone, adopt image extraction method to obtain one group of initial pel as initial value; The one group of initial pel that is obtained is sorted according to the matching degree with input key word and input sketch, and choosing several higher pels of coupling is the target pel.
Wherein, in the described step of obtaining the set of diagrams sheet corresponding according to key word, be specially and utilize image search engine to obtain described picture with key word.
Wherein, before the described step of obtaining the set of diagrams sheet corresponding according to key word, also comprise: set the quantity that comprises picture in the set of diagrams sheet with key word.
Wherein, described to one group of initial pel being obtained according to the step that the matching degree with input key word and input sketch sorts, be specially: the matching degree according to matching degree, shape information and the user expectation of bottom visual signature and user expectation sorts to described initial pel.
Wherein, according to a maximum class distances of clustering centers in the bottom visual signature of each picture and all picture bottom visual signatures, determine the matching degree of described bottom visual signature and user expectation, distance is more little, and matching degree is high more, otherwise matching degree is low more.A work (Arnold W.M.Smeulders who delivers at Arnold W.M.Smeulders in 2000, Marcel Worring, Simone Santini and Amarnath Gupta and Ramesh Jain.Content-BasedImage Retrieval at the End of the Early Years.IEEE Trans.Pattern Anal.Mach.Intell, 22 (12), pp.1349-1380,2000.) have many about obtaining the implementation method of the bottom layer image feature that is used to retrieve in.
Wherein, according to profile information and the pel shape matching method institute calculated distance that described sketch provides, determine the matching degree of described shape information and user expectation, distance is more little, and matching degree is high more, otherwise matching degree is low more.Work (the BELONGIE that delivered in 2002 people such as Serge Belongie, S., MALIK, J., AND PUZICHA, J.2002.Shapematching and object recognition using shape contexts.IEEE Transactionson Pattern Analysis and Machine Intelligence 24,4,509-522.) in relevant for a good implementation method of form fit.
At present, the partitioning algorithm that obtains pel interested alternately by ease of user from picture emerges in an endless stream.Yuri Y.Boykov equals calendar year 2001 and has proposed a kind of method (Y.Y.Boykov andM.P.Jolly.Interactive Graph Cuts for Optimal Boundary and RegionSegmentation of Objects in N-D Images.ICCV, p.I:105-112,2001), this method utilizes max-flow/minimal cut technology classical in the graph theory to obtain accurate relatively foreground graphical elements from the simple preceding background parts markup information of user's input.Leo Grady has proposed a kind of method (L.Grady.Random Walks for Image Segmentation.IEEETrans.Pattern Analysis and Machine Intelligence in 04 year, 28 (11), pp.1768-1783, November 2006.), this method utilizes immediately that the migration model reaches and the similar purpose of said method.People such as Carsten Rother have proposed a kind of method (ROTHER in 04 year, C., KOLMOGOROV, V., AND BLAKE, A.2004. " grab-cut ": interactiveforeground extraction using iterated graph cuts.ACM Transactions onGraphics (SIGGRAPH2004) 23,3,309-314.), further simplified user's input, this method allows the user to choose the pel that will extract with a rectangle frame in the input picture, and algorithm is removed background irrelevant in the rectangle frame automatically and obtained accurate pel.
Now, there is serial of methods to carry out the detection in visual importance zone in the computer vision field.Itti L. and Koch C. have proposed a kind of method (L.Itti and C.Koch.Computational modelling of visual attention.Nature ReviewsNeuroscience in calendar year 2001, Vol.2, pp.194-203,2001.), this method to candidate region and neighboring area in early days the conspicuousness of the difference on the visual signature carry out modeling and determine the vision attention zone.A kind of method based on the image spectrum analysis that Xiaodi Hou proposed in 2007 detects visual importance zone (X.D.Hou and L.Q.Zhang.Saliency Detection:ASpectral Residual Approach.CVPR, pp.1-8,2007.).Chenlei Guo equals to propose in succession again to analyze with the spectrum phase techniques in 08 year method (the C.L.Guoand Q.Ma and L.M.Zhang.Spatio-temporal Saliency detection usingphase spectrum ofquaternion fourier transform.CVPR of visual importance, pp.1-8,2008.).
Technical scheme of the present invention also provides a kind of network pel automatic extracting system based on sketch, and described system comprises: input block is used to receive the key word and the sketch of the description target pel of user's input; The picture acquiring unit is used for obtaining the set of diagrams sheet corresponding with described key word according to described key word; Initial pel acquiring unit is used for utilizing the vision attention zone of the every width of cloth picture of vision attention Algorithm Analysis, and utilizes described vision attention zone as initial value, adopts image extraction method to obtain one group of initial pel; Target pel acquiring unit is used for the one group of initial pel that is obtained is sorted according to the matching degree with input key word and input sketch, and choosing several higher pels of coupling is the target pel.
Wherein, described system also comprises the picture number setup unit, is used for setting the quantity that the set of diagrams sheet comprises picture.
Wherein, described initial pel acquiring unit comprises: the vision attention zone obtains subelement, is used for utilizing the vision attention zone of the every width of cloth picture of vision attention Algorithm Analysis; Picture is cut apart subelement, utilizes described vision attention zone as initial value, adopts image extraction method to obtain one group of initial pel.
Wherein, described target pel acquiring unit comprises: bottom visual signature matching degree is obtained subelement, be used for bottom visual signature and the maximum class distances of clustering centers of all picture bottom visual signatures according to each picture, determine the matching degree of bottom visual signature and user expectation, distance is more little, matching degree is high more, otherwise matching degree is low more; The form fit degree obtains subelement, is used for the profile information that provides according to described sketch and pel with shape matching method institute calculated distance, determines the matching degree of shape information and user expectation, and distance is more little, and matching degree is high more, otherwise matching degree is low more; The ordering subelement is used for by the weighting to the matching degree of the matching degree of described bottom visual signature and user expectation and shape information and user expectation the initial pel that is obtained being sorted.
Compared with prior art, technical scheme of the present invention has following advantage:
The pel that the present invention can obtain user expectation automatically according to the key word and the sketch of user's input utilizes photographic search engine to carry out the pel extraction at the network picture of magnanimity, and material is more complete.In addition, the present invention imports simply, and manual intervention is few, thereby has reduced cost, has improved efficient.
Description of drawings
Fig. 1 is the process flow diagram of a kind of network pel extraction method based on sketch of the embodiment of the invention;
Fig. 2 is the structural representation of a kind of network pel automatic extracting system based on sketch of the embodiment of the invention.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used to illustrate the present invention, but are not used for limiting the scope of the invention.
A kind of network pel extraction method based on sketch of the embodiment of the invention may further comprise the steps as shown in Figure 1:
Step s101, the key word and the sketch of the description target pel of reception user input.Making up during virtual scene that a dog holds flying disc in the month uses in the present embodiment, the user simply imports the posture that key word " dog jumping " and simple profile are described dog.
Step s102 sets the quantity that comprises picture in the set of diagrams sheet.The picture number of a key word correspondence of system default is 3000, can adopt the picture number of system default in the present embodiment, also can be set by the user a picture number.
Step s103 obtains the set of diagrams sheet corresponding with described key word according to key word.The key word of user's input can be used for reducing searches the space, by this key word, present embodiment can automatically utilize existing photographic search engine, as Google (http://images.***.com/), Baidu (http://image.***.com/), Flickr (http://www.flickr.com/) etc., obtain one group of picture relevant with this key word.
Step s104 utilizes the vision attention zone in the every width of cloth picture of vision attention Algorithm Analysis, and utilizes described vision attention zone as initial value, adopts image extraction method to obtain one group of initial pel.The vision attention algorithm can provide some information about visual importance, is undertaken just obtaining the vision attention zone after the region growing by these information.We can carry out figure according to this vision attention zone and cut apart then, thereby obtain foreground graphical elements.For example, the foreground graphical elements that a given rectangular area will obtain wherein to comprise can adopt the Grabcut method, this method allows the user to choose the pel that will extract with a rectangle frame in the input picture, and algorithm is removed background irrelevant in the rectangle frame automatically and obtained accurate pel.
Step s105 sorts according to the matching degree with input key word and input sketch to the one group of initial pel that is obtained, and choosing several higher pels of coupling is the target pel.Matching degree according to matching degree, shape information and the user expectation of bottom visual signature and user expectation in the present embodiment sorts to described initial pel, can be in earlier position in the hope of the result of user's expectation.The matching degree of bottom visual signature and user expectation can be weighed with a maximum class distances of clustering centers in the bottom visual signature of each picture and all picture bottom visual signatures, a because maximum class pel that class that is exactly user expectation in most cases.The sketch of user's input also can be used to instruct the later stage ordering in addition, and profile information that sketch provides and pel are the matching degree of shape information and user expectation with shape matching method institute calculated distance.Present embodiment passes through the weighting to the matching degree of the matching degree of described bottom visual signature and user expectation and shape information and user expectation, and the initial pel that is obtained is sorted.A large amount of test findings show that the method for employing present embodiment can be used for the pel extraction work based on key word and sketch effectively.
A kind of network pel automatic extracting system based on sketch of the embodiment of the invention comprises input block 21, picture number setup unit 22, picture acquiring unit 23, initial pel acquiring unit 24 and target pel acquiring unit 25 as shown in Figure 2.Wherein, picture acquiring unit 23 is connected with input block 21, picture number setup unit 22 and initial pel acquiring unit 24 respectively, and initial pel acquiring unit 24 is connected with target pel acquiring unit 25.
Initial pel acquiring unit 24 comprises that the vision attention zone obtains subelement 241 and picture is cut apart subelement 242, and wherein the vision attention zone obtains subelement 241 and cuts apart subelement 242 with picture and be connected.The vision attention zone obtains the vision attention zone that subelement 241 is used for utilizing the every width of cloth picture of vision attention Algorithm Analysis; Picture is cut apart subelement 242 and is utilized described vision attention zone as initial value, adopts image extraction method to obtain one group of initial pel.
Target pel acquiring unit 25 comprises that bottom visual signature matching degree obtains subelement 251, form fit degree and obtain subelement 252 and ordering subelement 253, and the subelement 253 that wherein sorts obtains subelement 251 with bottom visual signature matching degree respectively and obtains subelement 252 with the form fit degree and be connected.Bottom visual signature matching degree is obtained subelement 251 and is used for according to the bottom visual signature of each picture and the maximum class distances of clustering centers of all picture bottom visual signatures, determine the matching degree of bottom visual signature and user expectation, distance is more little, matching degree is high more, otherwise matching degree is low more; The form fit degree obtains subelement 252 and is used for the profile information that provides according to described sketch and pel with shape matching method institute calculated distance, determines the matching degree of shape information and user expectation, and distance is more little, and matching degree is high more, otherwise matching degree is low more; Ordering subelement 253 is used for by the weighting to the matching degree of the matching degree of described bottom visual signature and user expectation and shape information and user expectation the initial pel that is obtained being sorted.
The pel that the present invention can obtain user expectation automatically according to the key word and the sketch of user's input.There is the Ku Genggao of the possibility of the pel of meeting consumers' demand more than any artificial mark in the network picture that the present invention utilizes photographic search engine to provide in the network picture of magnanimity.And the present invention imports simply, and manual intervention is few, thereby cost also is significantly less than artificial mark mode.After the pel process cluster and form fit ordering that the present invention obtains automatically, the preceding preferably several pels of matching degree often can allow the user satisfied.Generally speaking, the present invention obtains high-quality pel a brand-new solution is provided for low-cost, high efficiency.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the technology of the present invention principle; 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, a kind of network pel extraction method based on sketch is characterized in that, said method comprising the steps of:
Receive the key word and the sketch of the description target pel of user's input;
Obtain the set of diagrams sheet corresponding according to described key word with described key word;
Utilize the vision attention zone in the every width of cloth picture of vision attention Algorithm Analysis, and utilize described vision attention zone, adopt image extraction method to obtain one group of initial pel as initial value;
The one group of initial pel that is obtained is sorted according to the matching degree with input key word and input sketch, and choosing several higher pels of coupling is the target pel.
2, the network pel extraction method based on sketch as claimed in claim 1 is characterized in that, in the described step of obtaining the set of diagrams sheet corresponding with key word according to key word, is specially and utilizes image search engine to obtain described picture.
3, the network pel extraction method based on sketch as claimed in claim 1 or 2 is characterized in that, before the described step of obtaining the set of diagrams sheet corresponding with key word according to key word, also comprises: set the quantity that comprises picture in the set of diagrams sheet.
4, the network pel extraction method based on sketch as claimed in claim 1, it is characterized in that, described to one group of initial pel being obtained according to the step that the matching degree with input key word and input sketch sorts, be specially: the matching degree according to matching degree, shape information and the user expectation of bottom visual signature and user expectation sorts to described initial pel.
5, the network pel extraction method based on sketch as claimed in claim 4, it is characterized in that, according to a maximum class distances of clustering centers in the bottom visual signature of each picture and all picture bottom visual signatures, determine the matching degree of described bottom visual signature and user expectation, distance is more little, matching degree is high more, otherwise matching degree is low more.
6, the network pel extraction method based on sketch as claimed in claim 4, it is characterized in that, profile information that provides according to described sketch and pel are with shape matching method institute calculated distance, determine the matching degree of described shape information and user expectation, distance is more little, matching degree is high more, otherwise matching degree is low more.
7, a kind of network pel automatic extracting system based on sketch is characterized in that described system comprises:
Input block is used to receive the key word and the sketch of the description target pel of user's input;
The picture acquiring unit is used for obtaining the set of diagrams sheet corresponding with described key word according to described key word;
Initial pel acquiring unit is used for utilizing the vision attention zone of the every width of cloth picture of vision attention Algorithm Analysis, and utilizes described vision attention zone as initial value, adopts image extraction method to obtain one group of initial pel;
Target pel acquiring unit is used for the one group of initial pel that is obtained is sorted according to the matching degree with input key word and input sketch, and choosing several higher pels of coupling is the target pel.
8, the network pel automatic extracting system based on sketch as claimed in claim 7 is characterized in that described system also comprises the picture number setup unit, is used for setting the quantity that the set of diagrams sheet comprises picture.
9, the network pel automatic extracting system based on sketch as claimed in claim 7 is characterized in that, described initial pel acquiring unit comprises:
The vision attention zone obtains subelement, is used for utilizing the vision attention zone of the every width of cloth picture of vision attention Algorithm Analysis;
Picture is cut apart subelement, utilizes described vision attention zone as initial value, adopts image extraction method to obtain one group of initial pel.
10, the network pel automatic extracting system based on sketch as claimed in claim 7 is characterized in that, described target pel acquiring unit comprises:
Bottom visual signature matching degree is obtained subelement, is used for bottom visual signature and the maximum class distances of clustering centers of all picture bottom visual signatures according to each picture, determines the matching degree of bottom visual signature and user expectation, distance is more little, matching degree is high more, otherwise matching degree is low more;
The form fit degree obtains subelement, is used for the profile information that provides according to described sketch and pel with shape matching method institute calculated distance, determines the matching degree of shape information and user expectation, and distance is more little, and matching degree is high more, otherwise matching degree is low more;
The ordering subelement is used for by the weighting to the matching degree of the matching degree of described bottom visual signature and user expectation and shape information and user expectation the initial pel that is obtained being sorted.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009100810690A CN101526955B (en) | 2009-04-01 | 2009-04-01 | Method for automatically withdrawing draft-based network graphics primitives and system thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009100810690A CN101526955B (en) | 2009-04-01 | 2009-04-01 | Method for automatically withdrawing draft-based network graphics primitives and system thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101526955A true CN101526955A (en) | 2009-09-09 |
CN101526955B CN101526955B (en) | 2010-09-01 |
Family
ID=41094821
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009100810690A Active CN101526955B (en) | 2009-04-01 | 2009-04-01 | Method for automatically withdrawing draft-based network graphics primitives and system thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101526955B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102663794A (en) * | 2012-03-29 | 2012-09-12 | 清华大学 | Method and apparatus for image synthesis |
CN103034849A (en) * | 2012-12-19 | 2013-04-10 | 香港应用科技研究院有限公司 | Perception deviation level estimate for hand-drawn sketches in matching of sketches and photos |
CN103164539A (en) * | 2013-04-15 | 2013-06-19 | 中国传媒大学 | Interactive type image retrieval method of combining user evaluation and labels |
US9014467B2 (en) | 2011-05-13 | 2015-04-21 | Omron Corporation | Image processing method and image processing device |
CN107918942A (en) * | 2011-03-09 | 2018-04-17 | 想象技术有限公司 | The compression of the pel index list by surface subdivision in segment rendering system |
CN112613395A (en) * | 2020-12-18 | 2021-04-06 | 湖南特能博世科技有限公司 | Primitive position relation matching method and device and computer equipment |
CN113129447A (en) * | 2021-04-12 | 2021-07-16 | 清华大学 | Three-dimensional model generation method and device based on single hand-drawn sketch and electronic equipment |
US11430205B2 (en) | 2017-06-23 | 2022-08-30 | Huawei Technologies Co., Ltd. | Method and apparatus for detecting salient object in image |
-
2009
- 2009-04-01 CN CN2009100810690A patent/CN101526955B/en active Active
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107918942A (en) * | 2011-03-09 | 2018-04-17 | 想象技术有限公司 | The compression of the pel index list by surface subdivision in segment rendering system |
CN107918942B (en) * | 2011-03-09 | 2021-06-01 | 想象技术有限公司 | Method and device for compressing primitive list and graphics rendering system |
US9014467B2 (en) | 2011-05-13 | 2015-04-21 | Omron Corporation | Image processing method and image processing device |
CN102663794A (en) * | 2012-03-29 | 2012-09-12 | 清华大学 | Method and apparatus for image synthesis |
CN103034849A (en) * | 2012-12-19 | 2013-04-10 | 香港应用科技研究院有限公司 | Perception deviation level estimate for hand-drawn sketches in matching of sketches and photos |
CN103034849B (en) * | 2012-12-19 | 2016-01-13 | 香港应用科技研究院有限公司 | Estimate for the perception variance level of cartographical sketching in sketch mates with photo |
CN103164539A (en) * | 2013-04-15 | 2013-06-19 | 中国传媒大学 | Interactive type image retrieval method of combining user evaluation and labels |
US11430205B2 (en) | 2017-06-23 | 2022-08-30 | Huawei Technologies Co., Ltd. | Method and apparatus for detecting salient object in image |
CN112613395A (en) * | 2020-12-18 | 2021-04-06 | 湖南特能博世科技有限公司 | Primitive position relation matching method and device and computer equipment |
CN113129447A (en) * | 2021-04-12 | 2021-07-16 | 清华大学 | Three-dimensional model generation method and device based on single hand-drawn sketch and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN101526955B (en) | 2010-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101526955B (en) | Method for automatically withdrawing draft-based network graphics primitives and system thereof | |
Kumar et al. | Leafsnap: A computer vision system for automatic plant species identification | |
CN104182765B (en) | Internet image driven automatic selection method of optimal view of three-dimensional model | |
US8385654B2 (en) | Salience estimation for object-based visual attention model | |
CN101770578B (en) | Image characteristic extraction method | |
CN102024152B (en) | Method for recognizing traffic sings based on sparse expression and dictionary study | |
Yu et al. | Automatic interesting object extraction from images using complementary saliency maps | |
WO2017181892A1 (en) | Foreground segmentation method and device | |
CN101706780A (en) | Image semantic retrieving method based on visual attention model | |
CN106250431B (en) | A kind of Color Feature Extraction Method and costume retrieval system based on classification clothes | |
CN103853724A (en) | Multimedia data sorting method and device | |
CN106874421A (en) | Image search method based on self adaptation rectangular window | |
CN107223242A (en) | Efficient local feature description's symbol filtering | |
Nedovic et al. | Depth information by stage classification | |
Rangkuti et al. | Batik image retrieval based on similarity of shape and texture characteristics | |
Song et al. | Analyzing scenery images by monotonic tree | |
CN102306179B (en) | Image content retrieval method based on hierarchical color distribution descriptor | |
Putzu et al. | A mobile application for leaf detection in complex background using saliency maps | |
CN106066887A (en) | A kind of sequence of advertisements image quick-searching and the method for analysis | |
Kong et al. | SimLocator: robust locator of similar objects in images | |
Yu et al. | Automatic image captioning system using integration of N-cut and color-based segmentation method | |
Sima et al. | Texture superpixels merging by color-texture histograms for color image segmentation | |
CN104484676B (en) | A kind of interactive mode ancient wall disease identification method | |
Rubino et al. | Semantic multi-body motion segmentation | |
CN110807776A (en) | Crop hemiptera pest image automatic segmentation algorithm based on global region contrast |
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 |