CN102779268B - Hand swing motion direction judging method based on direction motion historigram and competition mechanism - Google Patents
Hand swing motion direction judging method based on direction motion historigram and competition mechanism Download PDFInfo
- Publication number
- CN102779268B CN102779268B CN201210190678.1A CN201210190678A CN102779268B CN 102779268 B CN102779268 B CN 102779268B CN 201210190678 A CN201210190678 A CN 201210190678A CN 102779268 B CN102779268 B CN 102779268B
- Authority
- CN
- China
- Prior art keywords
- motion
- tau
- skin
- gesture
- greaterequal
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a hand swing motion direction judging method based on direction motion historigram and competition mechanism. The method comprises the following steps of: A1. obtaining the current motion image from video stream, establishing a skin color probability model, and obtaining four light stream directional diagrams; A3. respectively calculating four direction motion historigrams according to the four light stream directional diagrams in step A2; A4. obtaining four direction motion energy diagrams according to a gesture motion historigram; A5. calculating and encoding the motion direction of each frame of image; A6. judging the motion direction of the whole motion sequence; and an interested region is extracted by frame difference and skin color cooperatively to reduce the interference of background and the skin color of a human face. The integral motion direction is calculated by the four motion historigrams and the energy diagrams in a statistic mode to effectively overcome the motion coverage problem. The motion direction is determined by the action competition mechanism to greatly reduce the noise effect and improve the judging accuracy and stability of the motion direction.
Description
Technical field
The present invention relates to computer vision technical field of hand gesture recognition, in particular a kind of hand based on direction motion history figure and competition mechanism waves movement direction decision method.
Background technology
Gesture interaction is a kind of nature, intuitively man-machine interaction mode, and using gesture directly as the input of computing machine, the communication of between humans and machines no longer will need medium.Gesture identification is the technical foundation of gesture interaction, and visual gesture identification is simple due to it, is the main flow of Gesture Recognition at present.
Although visual gesture identification is the main development direction of gesture identification, but also there is a lot of problem, such as be subject to the impact of change of background, illumination variation, motion artifacts, also easily there is non-rigid distortion in hand itself, has the information etc. that a lot of redundancy is unnecessary in gesture.How to overcome these technical difficulties in gesture identification, can fully apply to change following interactive mode, be the key issue of gesture interaction technology.
Visual gesture identification is generally by camera collection gesture video stream data.At present, video frequency pick-up head is divided into 2D camera and 3D camera.What 3D camera obtained is the three-dimensional information of gesture, and the kinect that such as Microsoft releases just can gather 3D gesture information, but its cost expensive being unfavorable for is popularized.Therefore research and develop the Gesture Recognition of the USB 2D camera based on cheapness, there is positive effect
But because hand is non-Rigid Bodies, and gesture tracking generally carries out based on the colour of skin, causing this method based on following the tracks of unstable, being subject to the interference of ambient lighting and background colour, correct pursuit path cannot be formed.
At present, also occurred utilizing motion history figure to carry out the method for gesture motion direction determining.First this method utilizes frame difference method to generate hand and waves motion history figure, then calculates its gradient vector to obtain direction of motion.The subject matter of this method is: cannot overcome motion covering problem when hand is brandished; Easily by the impact of motion artifacts, cause walking direction and instability thereof.
Brandishing direct Control-Menu up and down by hand or regulate some function, is the quick application model widely of one of gesture interaction.Now, the correct direction judging that hand moves up and down will be the technical foundation of this kind of gesture interaction.
Existing gesture motion direction determining is mainly followed the tracks of based on gesture motion, by following the tracks of gesture, calculating hand shift length thus judging the method for its direction of motion.
1, direction determining depends on the effect of gesture tracking, and follows the tracks of the impact being subject to the distortion of illumination variation, face complexion and hand, weak effect, unstable;
2, track algorithm is complicated, and speed is slow, can not judge that hand waves direction fast.
Utilize motion history figure to carry out the method for gesture motion direction determining.First this method utilizes background subtraction or frame difference method to generate motion history figure, and then calculate its gradient vector to obtain direction of motion, advantage is that speed is fast.
1, motion covering problem when hand is brandished cannot be overcome;
2, easily by the impact of motion artifacts, cause walking direction inaccurate.
Summary of the invention
Technical matters to be solved by this invention provides a kind of hand based on direction motion history figure and competition mechanism to wave movement direction decision method for the deficiencies in the prior art.
Technical scheme of the present invention is as follows:
Hand based on direction motion history figure and competition mechanism waves a movement direction decision method, comprises the following steps:
A1, from video flowing, obtain current kinetic image, set up skin similarity model, the foundation of hand skin cluster in process is waved as hand, in hsv color space, add up a large amount of colour of skin and non-Skin Color Information, set up normalized H-S colour of skin histogram, for the skin color segmentation of subsequent video two field picture, the threshold probability of skin cluster is set as 0.85, the skin color probability map of present frame
X, y represent location of pixels, and t represents the time; Obtain three width frame difference images of current video moving image, the data after being multiplied with frame difference result by skin color probability map carry out filtering through median filter, obtain current gesture area grayscale figure
A2, calculating gesture area grayscale figure
light stream vector field F (x, y, t) after, be first broken down into horizontal and vertical two scalar field F
x(x, y, t), F
y(x, y, t), then carries out half-wave rectification, resolves into four non-negative passages further and carries out Gaussian smoothing and normalization, finally obtains four light stream direction area grayscale figure
wherein
A3, according to the light stream directional diagram of four in steps A 2, calculate four direction motion history figure respectively,
A4, according to gesture motion historigram, obtain four direction kinergety figure:
A5, calculate every two field picture direction of motion and encode:
wherein Sx
+=size [E
x+(x, y, t) > 0], Sx
-=size [E
x-(x, y, t) > 0], Sy
+=size [E
y+(x, y, t) > 0], Sy
-=size [E
y-(x, y, t) > 0], wherein, size [E
x+(x, y, t), t, size [E
x-(x, y, t) > 0], size [E
y+(x, y, t) > 0] and size [E
y-(x, y, t) > 0] kinergety figure size when representing the parameter τ of the time range of all directions known motions respectively; If Sx
+-Sx
-=0,
or
occurrence is according to Sy
+-Sy
-be greater than or less than 0 to judge, if valy
+-valy
-also be 0, then judge current not moving, be encoded to-1; By according to the angle calculated during other situation, its value is carried out as required divide and encode, carry out " upper and lower, left and right " four direction and judge; (-50 °, 50 °) are encoded to " 0 ", and (50 °, 140 °) are encoded to " 1 ", and (140 °, 230 °) are " 2 ", and (230 °, 310 °) are " 3 ", and not in the value in these intervals, assignment is "-1 "; Analyze one section of gesture motion video, all above-mentioned calculating is carried out to every frame and travel direction coding, thus obtain direction of motion coded sequence O (o
1, o
2... o
n), wherein o
i∈-1,0,1,2,3}, i=1 ... N, N are efficient coding sequence lengths;
A6, judge the direction of motion of whole motion sequence; For overcoming motion artifacts, each coding in this direction of motion coded sequence is sent into successively competition device and to be at war with ballot, final direction of overflowing the coding correspondence of (triumph) is exactly the direction of motion that this gesture brandishes correspondence; Specifically: first set up 4 little lattice, and mark 0,1,2, corresponding each direction of motion coding of 3 difference, arranges poll extreme value V simultaneously
maxand poll overflows threshold value V
th; Each direction encoding entered adds 1 by making the little lattice of its correspondence, and makes other little lattice subtract 1; Every little lattice poll increases until V always
max, and any instant, its poll reaches V
th, then label corresponding to these little lattice will be overflowed.
Described method, in described steps A 6, arranges V
th=6, V
max=8.
Described method, in described steps A 1, the computing method of frame difference are:
wherein f (x, y, t) represents current frame image, and t represents the time, and x, y represent location of pixels (lower with), symbol || represent and take absolute value, function min represents and gets minimum value; For taking noise and light change, an adaptive threshold can be set
here
with
expression is got
minimum and maximum value, then finally calculating frame difference is
finally according to skin color probability map Skin (x, y, t) and frame difference figure
gesture area grayscale figure can be obtained
wherein symbol × expression is multiplied.
Described method, calculates every two field picture direction of motion in described steps A 5 and encodes, and also can adopt to compare size and the method for direct coding is carried out, i.e. o
i=code (max (Sx
+, Sx
-, Sy
+, Sy
-)), wherein code (Sx
+) be " 0 ", code (Sx
-) be " 1 ", code (Sy
+) be " 2 ", code (Sy
+) be " 3 ".
Described method, in described steps A 6, after obtaining direction of motion coded sequence, is judged the final direction of motion of whole sequence, can be estimated by Hidden Markov Model (HMM) travel direction.
The present invention has following beneficial effect:
1, owing to adopting frame difference and colour of skin combined extracting area-of-interest, the interference of background and face complexion will be reduced.;
2, owing to adopting four direction motion history figure and energygram to calculate mass motion direction in the mode of directional statistics, motion covering problem can effectively be overcome.
3。Owing to have employed action competition mechanism determination direction of motion, greatly can reduce the impact of action noise, improve the Stability and veracity of movement direction decision.
Accompanying drawing explanation
Fig. 1 is the inventive method process flow diagram;
Fig. 2 is that in embodiment 1, competition ballot is won the schematic diagram of the method determining mass motion direction.
Embodiment
Below in conjunction with specific embodiment, the present invention is described in detail.
Embodiment 1
With reference to figure 1, the hand based on direction motion history figure and competition mechanism waves movement direction decision method, comprises the following steps:
A1, from video flowing, obtain current kinetic image, set up skin similarity model, the foundation of hand skin cluster in process is waved as hand, in hsv color space, add up a large amount of colour of skin and non-Skin Color Information, set up normalized H-S colour of skin histogram, for the skin color segmentation of subsequent video two field picture, the threshold probability of skin cluster is set as 0.85, present frame skin color probability map
X, y represent location of pixels, and t represents the time.Obtain three subframe difference images of current video moving image, the data after being multiplied with frame difference result by skin color probability map carry out filtering through median filter, obtain current gesture area grayscale figure
The computing method of frame difference are:
wherein f (x, y, t) represents current frame image, and t represents the time, and x, y represent location of pixels (lower with), symbol || represent and take absolute value, function min represents and gets minimum value.For taking noise and light change, an adaptive threshold can be set
here
with
expression is got
minimum and maximum value, then finally calculating frame difference is
finally according to skin color probability map Skin (x, y, t) and frame difference figure
gesture area grayscale figure can be obtained
wherein symbol × expression is multiplied.
A2, calculating gesture area grayscale figure
light stream vector field F (x, y, t) after, be first broken down into horizontal and vertical two scalar field F
x(x, y, t), F
y(x, y, t), then carries out half-wave rectification, resolves into four non-negative passages further and carries out Gaussian smoothing and normalization, finally obtains four light stream direction area grayscale figure
wherein
A3, basis four light stream directional diagrams above, calculate four direction motion history figure respectively,
A4, according to gesture motion historigram, obtain four direction kinergety figure (MEI):
A5, calculate every two field picture direction of motion and encode:
wherein Sx
+=size [E
x+(x, y, t) > 0], Sx
-=size [E
x-(x, y, t) > 0], Sy
+=size [E
y+(x, y, t) > 0], Sy
-=size [E
y-(x, y, t) > 0], wherein, size [E
x+(x, y, t), size [E
x-(x, y, t) > 0], size [E
y+(x, y, t) > 0] and size [E
y-(x, y, t) > 0] kinergety figure size when representing the parameter τ of the time range of all directions known motions respectively; If Sx
+-Sx
-=0,
or
occurrence is according to Sy
+-Sy
-be greater than or less than 0 to judge, if valy
+-valy
-also be 0, then judge current not moving, be encoded to-1; By according to the angle calculated during other situation, its value is carried out as required divide and encode, carry out " upper and lower, left and right " four direction and judge.(-50 °, 50 °) are encoded to " 0 ", and (50 °, 140 °) are encoded to " 1 ", and (140 °, 230 °) are " 2 ", and (230 °, 310 °) are " 3 ", and not in the value in these intervals, assignment is "-1 ".Analyze one section of gesture motion video, all above-mentioned calculating is carried out to every frame and travel direction coding, thus obtain direction of motion coded sequence O (o
1, o
2... o
n), wherein o
i∈-1,0,1,2,3}, i=1 ... N, N are efficient coding sequence lengths;
A6, judge the direction of motion of whole motion sequence.As shown in Figure 2, for overcoming motion artifacts, each coding in this direction of motion coded sequence is sent into successively competition device and to be at war with ballot, final direction of overflowing the coding correspondence of (triumph) is exactly the direction of motion that this gesture brandishes correspondence.Specifically: first set up 4 little lattice, and mark 0,1,2, corresponding each direction of motion coding of 3 difference, arranges poll extreme value V simultaneously
maxand poll overflows threshold value V
th; Each direction encoding entered adds 1 by making the little lattice of its correspondence, and makes other little lattice subtract 1.Every little lattice poll increases until V always
max, and any instant, its poll reaches V
th, then label corresponding to these little lattice will be overflowed.By this competition mechanism, effectively can overcome motion artifacts, improve the Stability and veracity of direction of motion identification.Rule of thumb can V be set
th=6, V
max=8.
Embodiment 2
As different from Example 1:
A5, calculates every two field picture direction of motion and encodes, and also can adopt to compare size and the method for direct coding is carried out, i.e. o
i=code (max (Sx
+, Sx
-, Sy
+, Sy
-)), wherein code (Sx
+) be " 0 ", code (Sx
-) be " 1 ", code (Sy
+) be " 2 ", code (Sy
+) be " 3 ".
Embodiment 3
As different from Example 1:
A6: after obtaining direction of motion coded sequence, judges the final direction of motion of whole sequence, estimates by Hidden Markov Model (HMM) travel direction.
Should be understood that, for those of ordinary skills, can be improved according to the above description or convert, and all these improve and convert the protection domain that all should belong to claims of the present invention.
Claims (5)
1. the hand based on direction motion history figure and competition mechanism waves a movement direction decision method, it is characterized in that, comprises the following steps:
A1, from video flowing, obtain current kinetic image, set up skin similarity model, the foundation of hand skin cluster in process is waved as hand, in hsv color space, add up a large amount of colour of skin and non-Skin Color Information, set up normalized H-S colour of skin histogram, for the skin color segmentation of subsequent video two field picture, the threshold probability of skin cluster is set as 0.85, the skin color probability map of present frame
X, y represent location of pixels, and t represents the time; Obtain three width frame difference images of current video moving image, the data after being multiplied with frame difference result by skin color probability map carry out filtering through median filter, obtain current gesture area grayscale figure
A2, calculating gesture area grayscale figure
light stream vector field F (x, y, t) after, be first broken down into horizontal and vertical two scalar field F
x(x, y, t), F
y(x, y, t), then carries out half-wave rectification, resolves into four non-negative passages further and carries out Gaussian smoothing and normalization, finally obtains four light stream direction area grayscale figure
Wherein
A3, according to the light stream directional diagram of four in steps A 2, calculate four direction motion history figure respectively,
A4, according to gesture motion historigram, obtain four direction kinergety figure:
A5, calculate every two field picture direction of motion and encode:
wherein Sx
+=size [E
x+(x, y, t) > 0], Sx
-=size [E
x-(x, y, t) > 0], Sy
+=size [E
y+(x, y, t) > 0], Sy
-=size [E
y-(x, y, t) > 0], wherein, size [E
x+(x, y, t), size [E
x-(x, y, t) > 0], size [E
y+(x, y, t) > 0] and size [E
y-(x, y, t) > 0] kinergety figure size when representing the parameter τ of the time range of all directions known motions respectively; If Sx
+-Sx
-=0,
or
occurrence is according to Sy
+-Sy
-be greater than or less than 0 to judge, if valy
+-valy
-also be 0, then judge current not moving, be encoded to-1; By according to the angle calculated during other situation, its value is carried out as required divide and encode, carry out " upper and lower, left and right " four direction and judge; (-50 °, 50 °) are encoded to " 0 ", and (50 °, 140 °) are encoded to " 1 ", and (140 °, 230 °) are " 2 ", and (230 °, 310 °) are " 3 ", and not in the value in these intervals, assignment is "-1 "; Analyze one section of gesture motion video, all above-mentioned calculating is carried out to every frame and travel direction coding, thus obtain direction of motion coded sequence O (o
1, o
2... o
n), wherein o
i∈-1,0,1,2,3}, i=1 ... N, N are efficient coding sequence lengths;
A6, judge the direction of motion of whole motion sequence; For overcoming motion artifacts, each coding in this direction of motion coded sequence being sent into successively competition device and to be at war with ballot, the direction of final coding correspondence of overflowing is exactly the direction of motion that this gesture brandishes correspondence; Specifically: first set up 4 little lattice, and mark 0,1,2, corresponding each direction of motion coding of 3 difference, arranges poll extreme value V simultaneously
maxand poll overflows threshold value V
th; Each direction encoding entered adds 1 by making the little lattice of its correspondence, and makes other little lattice subtract 1; Every little lattice poll increases until V always
max, and any instant, its poll reaches V
th, then label corresponding to these little lattice will be overflowed.
2. method according to claim 1, is characterized in that, in described steps A 6, arranges V
th=6, V
max=8.
3. method according to claim 1, is characterized in that, in described steps A 1, the computing method of frame difference are:
wherein f (x, y, t) represents current frame image, and t represents the time, and x, y represent location of pixels, symbol || represent and take absolute value, function min represents and gets minimum value; For taking noise and light change, an adaptive threshold can be set
here
with
expression is got
minimum and maximum value, then finally calculating frame difference is
finally according to skin color probability map Skin (x, y, t) and frame difference figure
gesture area grayscale figure can be obtained
wherein symbol × expression is multiplied.
4. method according to claim 1, is characterized in that, calculates every two field picture direction of motion and encode in described steps A 5, also can adopt to compare size and the method for direct coding is carried out, i.e. o
i=code (max (Sx
+, Sx
-, Sy
+, Sy
-)), wherein code (Sx
+) be " 0 ", code (Sx
-) be " 1 ", code (Sy
+) be " 2 ", code (Sy
+) be " 3 ".
5. method according to claim 1, is characterized in that, in described steps A 6, after obtaining direction of motion coded sequence, judges the final direction of motion of whole sequence, can be estimated by Hidden Markov Model (HMM) travel direction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210190678.1A CN102779268B (en) | 2012-02-06 | 2012-06-04 | Hand swing motion direction judging method based on direction motion historigram and competition mechanism |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2012100247170 | 2012-02-06 | ||
CN201210024717.0 | 2012-02-06 | ||
CN201210024717 | 2012-02-06 | ||
CN201210190678.1A CN102779268B (en) | 2012-02-06 | 2012-06-04 | Hand swing motion direction judging method based on direction motion historigram and competition mechanism |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102779268A CN102779268A (en) | 2012-11-14 |
CN102779268B true CN102779268B (en) | 2015-04-22 |
Family
ID=47124178
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210190678.1A Expired - Fee Related CN102779268B (en) | 2012-02-06 | 2012-06-04 | Hand swing motion direction judging method based on direction motion historigram and competition mechanism |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102779268B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104331151B (en) * | 2014-10-11 | 2018-02-23 | 中国传媒大学 | Gesture motion direction recognizing method based on optical flow method |
US10254845B2 (en) * | 2016-01-05 | 2019-04-09 | Intel Corporation | Hand gesture recognition for cursor control |
CN107633252B (en) | 2017-09-19 | 2020-04-21 | 广州市百果园信息技术有限公司 | Skin color detection method, device and storage medium |
CN111277780B (en) * | 2018-12-04 | 2021-07-20 | 阿里巴巴集团控股有限公司 | Method and device for improving frame interpolation effect |
CN113642493B (en) * | 2021-08-20 | 2024-02-09 | 北京有竹居网络技术有限公司 | Gesture recognition method, device, equipment and medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102236502A (en) * | 2010-04-21 | 2011-11-09 | 上海三旗通信科技有限公司 | Pressure touch gesture recognition human-computer interaction way for mobile terminal |
CN102236412A (en) * | 2010-04-30 | 2011-11-09 | 宏碁股份有限公司 | Three-dimensional gesture recognition system and vision-based gesture recognition method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5601045B2 (en) * | 2010-06-24 | 2014-10-08 | ソニー株式会社 | Gesture recognition device, gesture recognition method and program |
-
2012
- 2012-06-04 CN CN201210190678.1A patent/CN102779268B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102236502A (en) * | 2010-04-21 | 2011-11-09 | 上海三旗通信科技有限公司 | Pressure touch gesture recognition human-computer interaction way for mobile terminal |
CN102236412A (en) * | 2010-04-30 | 2011-11-09 | 宏碁股份有限公司 | Three-dimensional gesture recognition system and vision-based gesture recognition method |
Non-Patent Citations (2)
Title |
---|
Action Recognition by Employing Combined Directional Motion History and Energy Images;Md. Atiqur Rahman Ahad;《2010IEEE》;20101231;73-78 * |
一种快速鲁棒的动态人手跟踪方法;吕金刚等;《微型机与应用》;20120110;第31卷(第1期);33-36 * |
Also Published As
Publication number | Publication date |
---|---|
CN102779268A (en) | 2012-11-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107679491B (en) | 3D convolutional neural network sign language recognition method fusing multimodal data | |
CN110147743B (en) | Real-time online pedestrian analysis and counting system and method under complex scene | |
CN104050488B (en) | A kind of gesture identification method of the Kalman filter model based on switching | |
CN107967695B (en) | A kind of moving target detecting method based on depth light stream and morphological method | |
CN102592112B (en) | Method for determining gesture moving direction based on hidden Markov model | |
CN102779268B (en) | Hand swing motion direction judging method based on direction motion historigram and competition mechanism | |
CN103020992B (en) | A kind of video image conspicuousness detection method based on motion color-associations | |
CN103279791B (en) | Based on pedestrian's computing method of multiple features | |
CN105528794A (en) | Moving object detection method based on Gaussian mixture model and superpixel segmentation | |
CN107273835A (en) | Act of violence intelligent detecting method based on video analysis | |
CN111523378B (en) | Human behavior prediction method based on deep learning | |
CN104318263A (en) | Real-time high-precision people stream counting method | |
CN102063727B (en) | Covariance matching-based active contour tracking method | |
CN105243376A (en) | Living body detection method and device | |
Heo et al. | Appearance and motion based deep learning architecture for moving object detection in moving camera | |
CN110543917B (en) | Indoor map matching method by utilizing pedestrian inertial navigation track and video information | |
CN104102904A (en) | Static gesture identification method | |
CN103902989A (en) | Human body motion video recognition method based on non-negative matrix factorization | |
CN113312973A (en) | Method and system for extracting features of gesture recognition key points | |
US20140328516A1 (en) | Gesture Recognition Method, An Apparatus and a Computer Program for the Same | |
CN104778670A (en) | Fractal-wavelet self-adaption image denoising method based on multivariate statistical model | |
CN114241379A (en) | Passenger abnormal behavior identification method, device and equipment and passenger monitoring system | |
CN102938153B (en) | Video image splitting method based on restrain spectral clustering and markov random field | |
CN103077383B (en) | Based on the human motion identification method of the Divisional of spatio-temporal gradient feature | |
CN105118073A (en) | Human body head target identification method based on Xtion camera |
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: 20150422 Termination date: 20180604 |
|
CF01 | Termination of patent right due to non-payment of annual fee |