CN101872468A - Image scaling method for keeping visual quality of sensitive target - Google Patents
Image scaling method for keeping visual quality of sensitive target Download PDFInfo
- Publication number
- CN101872468A CN101872468A CN201010185241A CN201010185241A CN101872468A CN 101872468 A CN101872468 A CN 101872468A CN 201010185241 A CN201010185241 A CN 201010185241A CN 201010185241 A CN201010185241 A CN 201010185241A CN 101872468 A CN101872468 A CN 101872468A
- Authority
- CN
- China
- Prior art keywords
- image
- sensitivity
- target
- triangle
- triangle gridding
- 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
Abstract
The invention discloses an image scaling method for keeping visual quality of a sensitive target, which comprises the following steps of: calculating the sensitivity of a pixel point in an image by using four characteristics comprising color, gradient, brightness and center distance of the pixel point in the image to obtain an image sensitivity graph in accordance with the primary image, then generating a triangle mesh covering the primary image by adopting Delaunay triangulation according to the image sensitivity graph, establishing a visual loss function of the image on the basis, obtaining a target mesh with the same size as a target image by solving the optimization problem of the minimum value of the visual loss function under integrity constraint of the image, and finally generating a final result image in a texture mapping mode.
Description
Technical field
The present invention relates to a kind of image-scaling method that keeps visual quality of sensitive target, the susceptibility that relates to pixel in the image calculates and based on the image distortion method of triangle gridding, high-quality zoomed image keeping obtaining under the indeformable substantially prerequisite of sensitive target the best visual effect belongs to technical field of computer vision.
Background technology
At present all be to adopt the mode of direct linear scale that original image is carried out convergent-divergent in actual applications, the two class problems that the image-scaling method of direct linear scale exists: 1) directly may cause important goal undersized after the linear scale, be not easy to observe.2) if when the length breadth ratio of the length breadth ratio of original image and mobile terminal screen differs big, can cause the gross distortion distortion of target in the image.The image-scaling method that keeps visual quality of sensitive target is exactly the image zoom technology that is widely studied for the problem that solves direct Zoom method existence.The hot university of University of Wisconsin-Madison, Microsoft Research, Israel Tel Aviv university, Mitsubishi electrical equipment research institute, Israel Ci Man Wei Science Institute and domestic Tsing-Hua University, Institute of Automation, CAS, Hong Kong Chinese University, Taiwan success university etc. have all carried out number of research projects, and have obtained the correlative study achievement.According to the principle difference of these achievements in research, they can be divided into following four classes: based on the image-scaling method of window clipping, based on the image-scaling method of distortion of the mesh, based on the image-scaling method of resampling and the image-scaling method that merges based on several different methods.
Basic ideas based on the image-scaling method of window clipping are: search out a window identical with the target size size by certain strategy of choosing in original image, then the picture material in the window is cut out and be used as last scaled results and export.Because the content of just having chosen in original image in the window identical with target size is as a result of exported, as Fig. 1 (a) is original image, Fig. 1 (b) is an image behind the convergent-divergent, only kept original image center position image information behind the image zoom, all the other original image informations are all lost.So, Zoom method based on window clipping is not suitable for the situation that sensitive target distributes and do not concentrate, for example sensitive target be distributed in image about two edges, a crop window can't comprise all sensitive targets wherein, at this moment must cause sensitive information to lose.
Mainly adopted the thought of distortion of the mesh based on the image-scaling method of distortion of the mesh.At first image is out of shape with the grid covering of an equal size and to grid, wherein similarity transformation is adopted in the sensitive target in grid zone or remained unchanged, the distortion of big coefficient is then carried out in other zone, obtain with the measure-alike distortion of target image after grid; According to the coordinate of each pixel of target gridding in original mesh, the interpolation method that the pixel value employing of the pixel in the original image is certain generates the pixel value in the target gridding then, thereby generates the target image behind the convergent-divergent.Based on the method for distortion of the mesh,, may introduce the deformation distortion owing to adopted morphing.
Based on the image-scaling method that resamples mainly is according to certain strategy the pixel of original image to be carried out resampling, generates class methods of final objective image.Cause the distortion of structural stronger sensitive target in the image easily based on the method that resamples.As Fig. 2 (a) is original image, and Fig. 2 (b) is a result images behind the convergent-divergent.Circular sensitive target among Fig. 2 (a) is deformed into the ellipse that serious deformation takes place in Fig. 2 (b).
The image-scaling method that merges based on several different methods mainly is that the Zoom method that multiple sensitive target keeps is combined into a kind of new Zoom method according to certain strategy.The technique computes amount that merges based on several different methods is bigger, is unfavorable for using on the limited mobile device of processing power.
Summary of the invention
The purpose of this invention is to provide a kind of image-scaling method that keeps visual quality of sensitive target.This method is utilized the sensitivity of a plurality of low-level image feature computed image pixels of image pixel, sensitivity according to pixel adopts the triangle gridding of different densities that original image is covered then, by the optimization procedure that satisfies the vision loss minimum under the image integrity constraint original image triangle gridding is deformed to the target image grid again, passes through the mode productive target image of texture at last.This method has reduced sensitive target and has produced the probability of deformation owing to convergent-divergent.Realized high-quality sensitive target zooming effect.
For achieving the above object, the present invention adopts following technical scheme.It is characterized in that may further comprise the steps:
Step 1: calculate the sensitivity of original image pixels point by the low-level image feature that extracts the image slices vegetarian refreshments, obtain comprising the sensitivity figure of all pixel sensitivitys in the original image;
Step 2: according to the sensitivity figure that obtains in the step 1, in original image, choose the summit of the coordinate points of some as triangle gridding, original image is carried out triangulation, obtain the triangle gridding identical with original image size, wherein the triangle gridding number of vertex chosen of the high more zone of sensitivity is many more;
Step 3: according to the triangle gridding that step 2 obtains, the sensitivity sum of pixel that calculates all delta-shaped regions in the triangle gridding is as this regional sensitivity;
Step 4: utilize the triangle gridding area sensitive degree that obtains in the step 3,, calculate the target triangle gridding identical with target image size by the loss of the visual effect in the formula (1) S is minimized,
The wherein visual effect of S presentation video loss, i, j represent the summit on any limit in the triangle gridding;
Represent summit i respectively, the position behind the j convergent-divergent; λ
IjBe the vision loss energy coefficient,
J-1, j+1 are respectively and summit i, and j constitutes vertex of a triangle, Saliency (i, j, j-1) expression triangle (i, j, sensitivity j-1), Saliency (i, j, j+1) expression triangle (i, j, sensitivity j+1);
Expression limit (i, length j);＜P
i, P
j, P
J-1Expression triangle (i, j, area j-1);＜P
i, P
j, P
J+1Expression triangle (i, j, area j+1);
Step 5: according to the target triangle gridding that obtains in the step 4, the method by texture obtains target image.
More excellent, the image low-level image feature in the described step 1 comprises: color of pixel, gradient, brightness and with the distance of central pixel point.
More excellent, the calculating of image slices vegetarian refreshments sensitivity is by being weighted average acquisition to each low-level image feature in the described step 1.
More excellent, choosing of described step 2 intermediate cam grid vertex is by original image being divided into the identical square area of size, count the sensitivity sum of all pixels in each square area, above-mentioned sensitivity sum and predetermined threshold value are compared the quantity of determining each square area intermediate cam grid vertex.
The image-scaling method of maintenance visual quality of sensitive target provided by the present invention can reduce image effectively because the distortion of the sensitive target that convergent-divergent causes and the information dropout of sensitive target.Relevant test result shows that this method all has zooming effect preferably for all kinds of images, and especially for the large percentage of sensitive target in image, and the situation effect of image zoom large percentage is more obvious.
Description of drawings
Fig. 1 window clipping class keeps visual quality of sensitive target method design sketch.
Fig. 2 class that resamples keeps visual quality of sensitive target method design sketch.
The image-scaling method process flow diagram that Fig. 3 sensitive target of the present invention keeps.
Embodiment
Before address, the present invention obtains sensitivity figure with the original image correspondence by the sensitivity of analyzing original image, adopt the triangulation method to generate the triangle gridding identical on this basis, make the optimization procedure of image vision energy loss minimum obtain the target gridding consistent by finding the solution then with target image size with original image size.Mode by texture generates result images at last.
Below in conjunction with description of drawings implementation of the present invention, clearly represented process of the present invention among Fig. 3.At first, the sensitivity of computed image; Secondly, generate the original image triangle gridding according to sensitivity; Then, triangle gridding is carried out the optimization distortion; At last, adopt the mode of texture to generate the final objective image.
It should be noted that following only is the exemplary one embodiment of the present invention of having enumerated:
Step 1: the calculating of image sensitivity
Existing image sensitivity computing method can be divided into two classes: the sensitivity that adopts level image feature calculation image; The mode of employing pattern-recognition identifies the sensitive target in the image.Adopt at present the method major part of level image feature to concentrate on to adopt single low-level image feature such as gradient, be used as the sensitivity of image.The characteristics of these class methods are simple, quick.Shortcoming is that the robustness of these class methods is bad, and is not accurate enough.And the advantage of the method for the sensitive target in the method recognition image of employing pattern-recognition is: accurately.Shortcoming is: can only discern extensive poor-performing to a few type objects.Among the present invention, adopt average weighted mode to calculate the sensitivity of image by four kinds of features of color, gradient, brightness, centre distance to the image slices vegetarian refreshments, have simply, characteristics have improved the accuracy that image susceptibility is judged simultaneously fast, and have robustness and extensive performance preferably.
A kind of exemplary implementation step of step 1 is as follows:
(1) calculating of color of image susceptibility
The calculating of color of image susceptibility has showed the effect of color aspect the image sensitivity.It is generally acknowledged that the less color of distribution has higher susceptibility in the image, and general color distributed more widely has lower susceptibility.The computing method of color of image susceptibility can be by adding up this row pixel at R for the pixel of every row, G, color histogram in the B triple channel (size of bin is 9) adopts formula (2) to calculate the color sensitivity Sc of this pixel for each pixel then.
Sc＝0.34S
Cr+0.33S
Cg+0.33S
Cb
S in the formula (2)
CiRemarked pixel point is at the sensitivity of i Color Channel, f
iBe the frequency of occurrences of pixel in color value correspondence in color histogram of i Color Channel,
The minimum value of color histogram medium frequency under the expression i Color Channel, similarly
The maximal value of color histogram medium frequency under the expression i Color Channel, the color sensitivity of Sc remarked pixel point.
Because it is less that the color of noise pixel point often distributes in image, in order to reduce the influence that noise spot calculates the image sensitivity in the image, if the frequency of occurrences of pixel color under Color Channel i is less than threshold value T in the process of computed image sensitivity, can be directly with the sensitivity value S of pixel under this passage
CiBe made as 0.
(2) calculating of image gradient susceptibility
Because gradient has reflected the information such as structure of image, for structural stronger sensitive target, such as: straight line, circle etc.The Grad of this class sensitive target is higher.The canny operator has noise resisting ability preferably, and the gradient detectability, therefore can adopt the gradient of canny operator computed image pixel, at last with the gradient sensing value of the Grad after the normalization as image.
(3) calculating of image irradiation susceptibility
Generally speaking, in the image the higher zone of brightness often the sensitizing range in the image since in the LAB space L component represented the illumination of image, therefore, can be the LAB color space by the RGB color space conversion with image, directly the value of L component is carried out normalized, and with the photoperiod sensitivity value of end value as image.
(4) calculating of picture centre distance sensitive
It is bigger that a large amount of experiments shows that the sensitive target in the image appears at the probability of center of image.Therefore this method adopts formula (3) computed image centre distance susceptibility.
S in the formula (3)
pThe centre distance susceptibility of presentation video, l is that (x is y) with picture centre pixel (x for pixel
0, y
0) between distance.
(5) calculating of image susceptibility
Finally can be by average weighted mode, the method shown in the formula (4) for example, the susceptibility of computed image.
S
I＝α
1×S
C+α
2×S
L+α
3×S
G+α
4×S
P (4)
α
1+α
2+α
3+α
4＝1
S in the formula (4)
IBe the susceptibility of image, S
CBe the color sensitivity of image, S
LBe the photoperiod sensitivity of image, S
GBe the gradient sensing of image, S
PBe the centre distance susceptibility of image, α
1, α
2, α
3, α
4Be respectively color sensitivity, photoperiod sensitivity, the weighting factor of gradient sensing and centre distance susceptibility.
Step 2: the generation of triangle gridding
Triangulation is meant a several picture is divided into mutually disjoint one by one leg-of-mutton process with it, and the set that the triangle of these divisions is formed is called triangle gridding.In image-scaling method in the past, all be the rectangular node that adopts the rectangle identical to be divided into original image size based on the maintenance visual quality of sensitive target of grid.The problem of rectangular node is accurately to express whole sensitive target.Among the present invention, can not accurately express the problem of sensitive target, propose a kind of triangle gridding based on the Delaunay triangulation at rectangular node.This triangle gridding can determine the distribution density of grid intermediate cam shape according to the sensitivity of image, thereby adopt more triangle to be described to sensitive target, and adopt less triangle to be described for non-sensitive target, with respect to rectangular node sensitive target has been described more accurately.
An exemplary implementation step of step 2 is as follows:
(1) the Delaunay triangulation vertex determines
In order to guarantee the existence of Delaunay subdivision, must guarantee that the convex closure of the set of all triangulation points exists.Therefore, it being sampled according to the density of 5 pixels for the edge of image pixel, and sampled point is added the set of triangulation point, in image inside, is the set of the patch of 25*25 pixel size with image division.Calculate for each patch pixel wherein susceptibility and, the patch that surpasses certain threshold value for sensitivity, 8 triangle gridding summits of picked at random in this grid elements, otherwise for the patch that does not surpass threshold value 1 triangle gridding summit of picked at random quantity then.
(2) Delaunay triangulation
For the Delaunay triangulation, if triangulation vertex determines that the result of Delaunay triangulation is unique so.Method will go up the triangulation vertex of step generation to be gathered as the triangle that input obtains the Delaunay triangulation, i.e. the triangle gridding of original image.
Step 3: the optimization deformation of triangle gridding
The triangle gridding that generates in the step 2 is out of shape obtains the triangle gridding identical, at the zoomed image that keeps obtaining under the indeformable substantially situation of sensitive target visual effect the best with target zoomed image size.Among the present invention, at first defined the visual effect loss function of image in the convergent-divergent process, thus by find the solution that optimization problem at the following vision loss function minimum of image integrity constraint obtains having minimum vision loss with the identical triangle gridding of target zoomed image size.
An exemplary implementation step of step 3 is as follows:
(1) visual effect loss function
In order to realize that sensitive target does not produce deformation in the convergent-divergent process of image, the convergent-divergent that requires sensitive target is a linear scale, and promptly the angle at the leg-of-mutton mapping angle in triangle in the triangle gridding after distortion and the original triangle gridding remains unchanged.Also just require original triangle gridding is carried out conformal transformation.Formula among the present invention (1) has defined the visual effect loss function of image in the convergent-divergent process.
(2) generation of optimum vision loss triangle gridding
In order to obtain the target zoomed image of optimum visual effect, just require image in the convergent-divergent process, to have minimum vision loss, therefore the present invention is converted into the image lattice problem on deformation optimization problem of finding the solution at the following vision loss function minimum of image integrity constraint, in order to guarantee that final result images is a rectangle, the present invention adopts the constraint condition of formula (5) as optimization vision loss problem simultaneously.Because this problem is a quadratic programming problem, by finding the solution this optimization problem, has obtained having the target triangle gridding of minimum vision loss.
(5)
In the formula (5)
Be respectively pixel P
iCoordinate behind the convergent-divergent.M ', n ' are the width and the height of image behind the convergent-divergent.
Step 4: texture
Generated the target triangle gridding with minimum vision loss in step 3, the present invention has generated the triangle gridding of original image in step 2.Because the target triangle gridding is to have original triangle gridding distortion to obtain, so the summit in the summit of target triangle gridding and the original triangle gridding is for concerning one to one, therefore to the target triangle gridding with corresponding vertex position in the original triangle gridding as texture coordinate, in conjunction with original image by the final target zoomed image that generates of the mode of texture with optimum vision loss.
More than disclosed only be instantiation of the present invention, according to thought provided by the invention, those skilled in the art can think and variation, all should fall within the scope of protection of the present invention.
Claims (4)
1. image-scaling method that keeps visual quality of sensitive target is characterized in that may further comprise the steps:
Step 1: calculate the sensitivity of original image pixels point by the low-level image feature that extracts the image slices vegetarian refreshments, obtain comprising the sensitivity figure of all pixel sensitivitys in the original image;
Step 2: according to the sensitivity figure that obtains in the step 1, in original image, choose the summit of the coordinate points of some as triangle gridding, original image is carried out triangulation, obtain the triangle gridding identical with original image size, wherein the triangle gridding number of vertex chosen of the high more zone of sensitivity is many more;
Step 3: according to the triangle gridding that step 2 obtains, the sensitivity sum of pixel that calculates all delta-shaped regions in the triangle gridding is as this regional sensitivity;
Step 4: utilize the triangle gridding area sensitive degree that obtains in the step 3,, calculate the target triangle gridding identical with target image size by the loss of the visual effect in the formula (1) S is minimized,
The wherein visual effect of S presentation video loss, i, j represent the summit on any limit in the triangle gridding;
Represent summit i respectively, the position behind the j convergent-divergent; λ
IjBe the vision loss energy coefficient,
J-1, j+1 are respectively and summit i, and j constitutes vertex of a triangle, Saliency (i, j, j-1) expression triangle (i, j, sensitivity j-1), Saliency (i, j, j+1) expression triangle (i, j, sensitivity j+1);
Expression limit (i, length j);＜P
i, P
j, P
J-1Expression triangle (i, j, area j-1);＜P
i, P
j, P
J+1Expression triangle (i, j, area j+1);
Step 5: according to the target triangle gridding that obtains in the step 4, the method by texture obtains target image.
2. the method for claim 1, it is characterized in that: the image low-level image feature in the described step 1 comprises: color of pixel, gradient, brightness and with the distance of central pixel point.
3. the method for claim 1, it is characterized in that: the calculating of image slices vegetarian refreshments sensitivity is by being weighted average acquisition to each low-level image feature in the described step 1.
4. the method for claim 1, it is characterized in that: choosing of described step 2 intermediate cam grid vertex is by original image being divided into the identical square area of size, count the sensitivity sum of all pixels in each square area, above-mentioned sensitivity sum and predetermined threshold value are compared the quantity of determining each square area intermediate cam grid vertex.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010185241.XA CN101872468B (en) | 2010-05-27 | 2010-05-27 | Image scaling method for keeping visual quality of sensitive target |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010185241.XA CN101872468B (en) | 2010-05-27 | 2010-05-27 | Image scaling method for keeping visual quality of sensitive target |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101872468A true CN101872468A (en) | 2010-10-27 |
| CN101872468B CN101872468B (en) | 2014-12-10 |
Family
ID=42997314
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010185241.XA Expired - Fee Related CN101872468B (en) | 2010-05-27 | 2010-05-27 | Image scaling method for keeping visual quality of sensitive target |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101872468B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102354388A (en) * | 2011-09-22 | 2012-02-15 | 北京航空航天大学 | Method for carrying out adaptive computing on importance weights of low-level features of image |
| CN106779042A (en) * | 2016-12-27 | 2017-05-31 | 北京农业信息技术研究中心 | A kind of aquaculture shoal of fish aggregate index computing device and computational methods |
| CN113311433A (en) * | 2021-05-28 | 2021-08-27 | 北京航空航天大学 | InSAR interferometric phase two-step unwrapping method combining quality map and minimum cost flow |
-
2010
- 2010-05-27 CN CN201010185241.XA patent/CN101872468B/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| YANWEN GUO ET.AL: "Image Retargeting Using Mesh Parametrization", 《MULTIMEDIA,IEEE TRANSACTIONS ON》 * |
| 时健等: "一种基于网格参数化的图像适应方法", 《软件学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102354388A (en) * | 2011-09-22 | 2012-02-15 | 北京航空航天大学 | Method for carrying out adaptive computing on importance weights of low-level features of image |
| CN102354388B (en) * | 2011-09-22 | 2013-03-20 | 北京航空航天大学 | Method for carrying out adaptive computing on importance weights of low-level features of image |
| CN106779042A (en) * | 2016-12-27 | 2017-05-31 | 北京农业信息技术研究中心 | A kind of aquaculture shoal of fish aggregate index computing device and computational methods |
| CN106779042B (en) * | 2016-12-27 | 2023-08-04 | 北京农业信息技术研究中心 | Device and method for calculating aggregation index of aquaculture fish shoal |
| CN113311433A (en) * | 2021-05-28 | 2021-08-27 | 北京航空航天大学 | InSAR interferometric phase two-step unwrapping method combining quality map and minimum cost flow |
| CN113311433B (en) * | 2021-05-28 | 2022-08-02 | 北京航空航天大学 | InSAR interferometric phase two-step unwrapping method combining quality map and minimum cost flow |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101872468B (en) | 2014-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111462225B (en) | Centroid identification and positioning method of infrared light spot image | |
| Zhang et al. | Multiple-level feature-based measure for retargeted image quality | |
| CN104751166B (en) | Remote Image Classification based on spectral modeling and Euclidean distance | |
| CN109859226B (en) | Detection method of checkerboard corner sub-pixels for graph segmentation | |
| CN109712153A (en) | A kind of remote sensing images city superpixel segmentation method | |
| CN108428220B (en) | Automatic geometric correction method for ocean island reef area of remote sensing image of geostationary orbit satellite sequence | |
| CN106447679B (en) | A kind of conspicuousness detection method cut based on iteration diagram with self-adaption cluster | |
| CN108121949B (en) | A kind of harbour Ore stockpile recognition methods based on remote sensing scene classification | |
| CN109726649B (en) | Remote sensing image cloud detection method and system and electronic equipment | |
| CN107330875A (en) | Based on the forward and reverse heterogeneous water body surrounding enviroment change detecting method of remote sensing images | |
| CN106778680A (en) | A kind of hyperspectral image band selection method and device extracted based on critical bands | |
| CN116012364B (en) | SAR image change detection method and device | |
| CN105549009A (en) | SAR image CFAR target detection method based on super pixels | |
| CN107256571A (en) | A kind of Fractal Dimension Estimation based on deep learning Yu adaptive differential box | |
| CN103162818A (en) | Laser beam width evaluation method based on moment invariance | |
| CN107992856A (en) | High score remote sensing building effects detection method under City scenarios | |
| CN101872468B (en) | Image scaling method for keeping visual quality of sensitive target | |
| CN109389593A (en) | A kind of detection method, device, medium and the equipment of infrared image Small object | |
| CN109063717B (en) | Method for acquiring center point of instrument | |
| CN109190452B (en) | Crop row identification method and device | |
| CN114998365A (en) | Ground feature classification method based on polarimetric interference SAR | |
| CN102938148A (en) | High-spectrum image texture analysis method based on V-GLCM (Gray Level Co-occurrence Matrix) | |
| CN116129191B (en) | Multi-target intelligent identification and fine classification method based on remote sensing AI | |
| CN112435249A (en) | Dynamic small target detection method based on periodic scanning infrared search system | |
| CN116958962A (en) | Method for detecting pre-fruit-thinning pomegranate fruits based on improved YOLOv8s |
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: 20141210 Termination date: 20150527 |
|
| EXPY | Termination of patent right or utility model |