CN111553405B - Group fog recognition algorithm based on pixel density K-means clustering - Google Patents
Group fog recognition algorithm based on pixel density K-means clustering Download PDFInfo
- Publication number
- CN111553405B CN111553405B CN202010330521.9A CN202010330521A CN111553405B CN 111553405 B CN111553405 B CN 111553405B CN 202010330521 A CN202010330521 A CN 202010330521A CN 111553405 B CN111553405 B CN 111553405B
- Authority
- CN
- China
- Prior art keywords
- fog
- cluster
- pixel density
- mass
- image
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a cluster fog recognition algorithm based on pixel density K-means clustering, which comprises the following steps: s1, capturing an area image through a high-speed camera, inputting an image set, dividing an image grid, calculating pixel density of the mass fog in each grid, and judging a mass fog area; s2, extracting the density of the group fog pixels in the group fog area; s3, calculating the mean value and variance of the density of the group fog pixels, forming the mean value and the variance into two-dimensional data points representing the group fog images, and forming all the image data points into a data set; s4, randomly selecting five two-dimensional data points in a data set as an initial clustering center according to the cluster fog grade division index; s5, traversing all data, calculating the distance between the data and a clustering center one by one, and classifying all data into five types; s6, iterating the calculation criterion function until the threshold requirement is met, converging the algorithm, and outputting a group fog recognition level result. The method has the advantages of high detection and identification speed, high accuracy, high convergence speed and great reduction of false detection rate.
Description
Technical Field
The application relates to the technical field of cluster fog image detection and identification processing, in particular to a cluster fog identification algorithm based on pixel density K-means clustering.
Background
In recent years, traffic safety accidents frequently occur on highways and marine vessels due to the influence of 'mist clusters', and suspension drops of mist clusters are very small, usually in air, belonging to microscopic drops, and generally have a horizontal visibility of less than 1 km on the ground surface according to the definition of WMO weather organization. It is shown by meteorological studies that most of the mist is radiation mist, caused by cooling of the air mass near the underlying surface. Therefore, the mist has the characteristics of burst property, locality, small scale and large concentration.
In the technical field of mass fog detection and recognition, mass fog detection and recognition are mostly applied to highways, mass fog is subjected to fuzzy recognition by utilizing a video image analysis processing technology through automatic weather station acquisition, but the mass fog detection is only dependent on real-time mass fog detection, and the early warning effect cannot be achieved, so that mass fog image data information is acquired, mass fog element characteristics are analyzed, a typical mass fog element characteristic set is constructed, rapid targeting recognition of weather conditions is realized, and the method has important significance for improving weather early warning.
The existing method for monitoring the mist in real time judges the mist grade according to the intensity of a reflected signal by penetrating the mist layer through an optical beam, is greatly influenced by the precision of an instrument device and the installation manufacturability, has large error between an analysis result and the actual situation, and has insufficient detection precision and reliability, and most importantly, the method cannot fully combine all image information to extract the mist element characteristics and cannot quickly and accurately identify the mist.
Disclosure of Invention
In order to solve the problems, the application provides a cluster fog recognition algorithm based on pixel density K-means, which is based on massive cluster fog image data, and performs targeted matching recognition through the image features of an actually measured area and sample features by extracting massive cluster fog image features as a sample set and taking cluster fog grades as discrimination boundaries. The technical proposal is that,
a cluster fog recognition algorithm based on pixel density K-means clustering comprises the following steps:
s1, capturing an area image through a high-speed camera, inputting an image set, dividing an image grid, calculating pixel density of the mass fog in each grid, and judging a mass fog area;
s2, calculating the density of the mass fog pixels in the mass fog region;
s3, calculating the mean value and variance of the density of the group fog pixels, forming the mean value and the variance into two-dimensional data points representing the group fog images, and forming all the image data points into a data set;
s4, randomly selecting five two-dimensional data points in a data set as an initial clustering center according to the cluster fog grade division index;
s5, traversing all the data, calculating the distance between the data and the clustering center one by one, and classifying all the data into five types according to the distance;
s6, iterating the calculation criterion function until the threshold requirement is met, converging the algorithm, and outputting a group fog recognition level result.
Further, in the step S1, the specific step of judging the mist area comprises,
s11, unifying the sizes of the images, dividing the grids of the images, calculating the density of the mass fog pixels in each grid one by one,
in the formula (1), A i Representing a cluster fog pixel density within an ith grid; n is n i Represents the number of the fog pixels in the ith grid, N i Representing the number of all pixels in the ith grid;
s12, judging the fog pixel density in each image segmentation grid, and reserving A i The effective grids more than or equal to 0.3 are removed A i Invalid grid < 0.3; and connecting the effective grids to form a group fog area in the image.
Further, summarizing the density of the group fog pixels in the effective grid, and calculating the density of the group fog pixels in the group fog area;
in the formula (2), S represents the density of the mass fog pixels in the mass fog region, D represents the number of effective grids, A j The cluster fog pixel density of the jth active grid is represented.
Further, in step S3, a mean value and a variance of the density of the mass fog pixels in the mass fog areas of all the images in the image set are calculated;
in the formula (3), M represents the average value of the density of the mass fog pixels in the mass fog region, Q represents the number of images and S k Representing a mass fog pixel density in a mass fog region of the kth image; p represents the variance of the mass fog pixel density in the mass fog region.
Further, in step S3, the mean value M and the variance P of the pixel density of the cluster fog in the cluster fog region are formed into two-dimensional data points (M, P) representing the cluster fog image, and all the two-dimensional data points (M, P) representing the cluster fog image in the image set are formed into a data set I.
Further, in step S4, 5 two-dimensional data points b are randomly selected from the data set I 1 、b 2 、b 3 、b 4 、b 5 As an initial cluster center.
Further, in step S5, all the two-dimensional data points in the dataset I are traversed, and the initial clustering center b is calculated one by one 1 、b 2 、b 3 、b 4 、b 5 Is a distance of the mahalanobis of (a) in the drawing,
d(x,b r )=[(x-b r ) T H -1 (x-b r )] 1/2 (r=1,2,3,4,5) (5)
in the formula (5), d (x, b) r ) Represents the mahalanobis distance of the data-by-data from the initial cluster center, x represents any data point in the data set I, (x-b) r ) T Representing the transpose matrix, H representing the covariance matrix of all data points in the dataset I, and r representing five classes.
Further, in step S5, the data points are classified into the distance from the initial cluster center b according to the March distance calculation value r In the nearest class, calculating the average value in the classes;
in the formula (6), m r Represents the average value of data in class r, F r Representing the number of data points in class r, x rq Representing the q-th data point in class r;
and taking the calculated various average values as new clustering centers in the class to replace the initial clustering centers.
Further, in step S6, a criterion function is established for calculating the distance between the new cluster center and the initial cluster center in each class,
Δ=|m r -b r | (r=1,2,...,5) (7)
in equation (7), Δ represents the distance between the new cluster center and the initial cluster center.
Further, in step S6, given a threshold δ, if Δ is less than or equal to δ, the algorithm converges, and a cluster fog recognition result is output; if delta is larger than delta, returning to S5, carrying out algorithm iteration until delta is smaller than or equal to delta, converging the algorithm, and outputting a cluster fog recognition result.
Advantageous effects
The application relates to a cluster fog feature targeted matching recognition algorithm for improving K-means clustering based on cluster fog region pixel density, which effectively improves a traditional visual image analysis algorithm for cluster fog feature recognition, extracts mass cluster fog image pixel density features to form a sample set, and compares and recognizes and matches measured region image pixel density features with the sample features.
Drawings
FIG. 1 is a flow chart of a cluster fog feature targeted matching recognition algorithm of the application;
FIG. 2 is a diagram of a cluster fog area grid determination of the present application;
FIG. 3 is a diagram of experimental verification of a cluster fog recognition algorithm of the present application;
Detailed Description
The techniques are further described below in conjunction with figures 1-3 and the specific embodiments to aid in understanding the present application.
A cluster fog recognition algorithm based on pixel density K-means clustering comprises the following steps:
s1, capturing an area image through a high-speed camera, inputting an image set, dividing an image grid, calculating pixel density of the mass fog in each grid, and judging a mass fog area;
s11, dividing the image into grids with uniform sizes of 20cm long and 20cm wide, dividing the image into grids of not less than 1000 x 1000, dividing the image into grids of 10000 x 10000 in order to improve accuracy, calculating the density of the mass fog pixels in each grid one by one,
in the formula (1), A i Representing a cluster fog pixel density within an ith grid; n is n i Represents the number of the fog pixels in the ith grid, N i Representing the number of all pixels in the ith grid;
s12, judging the fog pixel density in each image segmentation grid, and reserving A i The effective grids more than or equal to 0.3 are removed A i Invalid grid < 0.3; and connecting the effective grids to form a group fog area in the image.
S2, calculating the density of the mass fog pixels in the mass fog region;
summarizing the density of the mass fog pixels in the effective grid in the step S1, and calculating the density of the mass fog pixels in the mass fog region;
in the formula (2), S represents the density of the mass fog pixels in the mass fog region, D represents the number of effective grids, A j The cluster fog pixel density of the jth active grid is represented.
S3, calculating the mean value and variance of the density of the group fog pixels, forming two-dimensional data points (M, P) representing the group fog images by the mean value and the variance, and forming a data set I by all the two-dimensional data points (M, P) representing the group fog images in the image set;
in the formula (3), M represents the average value of the density of the mass fog pixels in the mass fog region, Q represents the number of images and S k Representing a mass fog pixel density in a mass fog region of the kth image; p represents the variance of the mass fog pixel density in the mass fog region.
S4, randomly selecting five two-dimensional data points b from the data set I according to the group fog grade division index (clear, light fog, medium fog, large fog and thick fog) 1 、b 2 、b 3 、b 4 、b 5 As an initial cluster center;
s5, traversing all the data, calculating the distance between the data and the clustering center one by one, and classifying all the data into five types according to the distance;
s51, traversing all two-dimensional data points in the data set I, and calculating b in initial clustering one by one 1 、b 2 、b 3 、b 4 、b 5 Is a distance of the mahalanobis of (a) in the drawing,
d(x,b r )=[(x-b r ) T H -1 (x-b r )] 1/2 (r=1,2,3,4,5) (5)
in the formula (5), d (x, b) r ) Represents the mahalanobis distance of the data-by-data from the initial cluster center, x represents any data point in the data set I, (x-b) r ) T Representing the transpose matrix, H representing the covariance matrix of all data points in the dataset I, and r representing five classes.
S52, classifying the data points to be away from an initial clustering center b according to the Marsh distance calculation value r In the nearest class, calculating the average value in the classes;
in the formula (6), m r Represents the average value of data in class r, F r Representing the number of data points in class r, x rq Representing the q-th data point in class r;
s53, taking the calculated various average values as new clustering centers in the category, replacing the initial clustering centers,
finally determining 5 clusters according to the cluster center mean value m r The small to large represents clear, light fog, medium fog, large fog and thick fog.
S6, iterating to calculate a criterion function until the threshold requirement is met, converging an algorithm, and outputting a group fog recognition level result;
s61, establishing a criterion function for calculating the distance between the new cluster center and the initial cluster center in various types,
Δ=|m r -b r | (r=1,2,...,5) (7)
in equation (7), Δ represents the distance between the new cluster center and the initial cluster center.
S62, setting a threshold delta, if delta is less than or equal to delta, converging an algorithm, and outputting a group fog recognition result; if delta is larger than delta, returning to S5, carrying out algorithm iteration until delta is smaller than or equal to delta, converging the algorithm, and outputting a cluster fog recognition result.
The method is mainly applied to prediction of marine mist.
Of course, the above description is not intended to limit the present technology, and the present technology is not limited to the above examples, but rather, changes, modifications, additions or substitutions made by those skilled in the art within the spirit and scope of the present application are also within the scope of the present technology.
Claims (9)
1. The cluster fog recognition algorithm based on the pixel density K-means clustering is characterized by comprising the following steps of:
s1, capturing an area image through a high-speed camera, inputting an image set, dividing an image grid, calculating pixel density of the mass fog in each grid, and judging a mass fog area;
the specific step of determining the haze region includes,
s11, unifying the sizes of the images, dividing the grids of the images, calculating the density of the mass fog pixels in each grid one by one,
in the formula (1), A i Representing a cluster fog pixel density within an ith grid; n is n i Represents the number of the fog pixels in the ith grid, N i Representing the number of all pixels in the ith grid;
s12, judging the fog pixel density in each image segmentation grid, and reserving A i The effective grids more than or equal to 0.3 are removed A i Invalid grid < 0.3; connecting the effective grids to form a group fog area in the image;
s2, calculating the density of the mass fog pixels in the mass fog region;
s3, calculating the mean value and variance of the density of the group fog pixels, forming the mean value and the variance into two-dimensional data points representing the group fog images, and forming all the image data points into a data set;
s4, randomly selecting five two-dimensional data points in a data set as an initial clustering center according to the cluster fog grade division index;
s5, traversing all the data, calculating the distance between the data and the clustering center one by one, and classifying all the data into five types according to the distance;
s6, iterating the calculation criterion function until the threshold requirement is met, converging the algorithm, and outputting a group fog recognition level result.
2. The cluster fog recognition algorithm based on the pixel density K-means clustering is characterized in that the cluster fog pixel densities in the effective grids are summarized, and the cluster fog pixel densities in a cluster fog area are calculated;
in the formula (2), S represents the density of the mass fog pixels in the mass fog region, D represents the number of effective grids, A j The cluster fog pixel density of the jth active grid is represented.
3. The cluster fog recognition algorithm based on the pixel density K-means clustering of claim 1, wherein in step S3, the cluster fog pixel density mean and variance in the cluster fog areas of all images in the image set are calculated;
in the formula (3), M represents the average value of the density of the mass fog pixels in the mass fog region, Q represents the number of images and S k Representing a mass fog pixel density in a mass fog region of the kth image; p represents the variance of the mass fog pixel density in the mass fog region.
4. The cluster fog recognition algorithm based on the pixel density K-means clustering as claimed in claim 1, wherein in step S3, the cluster fog pixel density mean M and the variance P in the cluster fog region are combined to form two-dimensional data points (M, P) representing the cluster fog image, and all the two-dimensional data points (M, P) representing the cluster fog image in the image set are combined to form a data set I.
5. The pixel density K-means cluster fog recognition algorithm according to claim 1, wherein in step S4, 5 two-dimensional data points b are randomly selected from the data set I 1 、b 2 、b 3 、b 4 、b 5 As an initial cluster center.
6. The method for recognizing cluster fog based on pixel density K-means clustering according to claim 5, wherein in step S5, all two-dimensional data points in the data set I are traversed, and the initial clustering center b is calculated one by one 1 、b 2 、b 3 、b 4 、b 5 Is a distance of the mahalanobis of (a) in the drawing,
d(x,b r )=[(x-b r ) T H -1 (x-b r )] 1/2 (r=1,2,3,4,5) (5)
in the formula (5), d (x, b) r ) Represents the mahalanobis distance of the data-by-data from the initial cluster center, x represents any data point in the data set I, (x-b) r ) T Representing the transpose matrix, H representing the covariance matrix of all data points in the dataset I, and r representing five classes.
7. The method as claimed in claim 1, wherein in step S5, the data points are classified as being away from the initial cluster center b according to the Mahalanobis distance calculation value r In the nearest class, calculating the average value in the classes;
in the formula (6), m r Represents the average value of data in class r, F r Representing the number of data points in class r, x rq Representing the q-th data point in class r;
and taking the calculated various average values as new clustering centers in the class to replace the initial clustering centers.
8. The method for recognizing cluster fog based on pixel density K-means according to claim 1, wherein in step S6, a criterion function is established for calculating the distance between the new cluster center and the initial cluster center in each class,
Δ=|m r -b r | (r=1,2,...,5) (7)
in formula (7), Δ represents m r And b r Is a distance of (3).
9. The cluster fog recognition algorithm based on the pixel density K-means clustering according to claim 8, wherein in the step S6, given a threshold delta, if delta is less than or equal to delta, the algorithm converges, and a cluster fog recognition result is output; if delta is larger than delta, returning to S5, carrying out algorithm iteration until delta is smaller than or equal to delta, converging the algorithm, and outputting a cluster fog recognition result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010330521.9A CN111553405B (en) | 2020-04-24 | 2020-04-24 | Group fog recognition algorithm based on pixel density K-means clustering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010330521.9A CN111553405B (en) | 2020-04-24 | 2020-04-24 | Group fog recognition algorithm based on pixel density K-means clustering |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111553405A CN111553405A (en) | 2020-08-18 |
CN111553405B true CN111553405B (en) | 2023-08-18 |
Family
ID=72002470
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010330521.9A Active CN111553405B (en) | 2020-04-24 | 2020-04-24 | Group fog recognition algorithm based on pixel density K-means clustering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111553405B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116699730B (en) * | 2023-08-04 | 2023-10-27 | 江西师范大学 | Road group fog prediction method based on edge calculation |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104200445A (en) * | 2014-09-26 | 2014-12-10 | 常熟理工学院 | Image defogging method with optimal contrast ratio and minimal information loss |
CN105844295A (en) * | 2016-03-21 | 2016-08-10 | 北京航空航天大学 | Video smog fine classification method based on color model and motion characteristics |
WO2016207875A1 (en) * | 2015-06-22 | 2016-12-29 | Photomyne Ltd. | System and method for detecting objects in an image |
US10255670B1 (en) * | 2017-01-08 | 2019-04-09 | Dolly Y. Wu PLLC | Image sensor and module for agricultural crop improvement |
JP2020057236A (en) * | 2018-10-03 | 2020-04-09 | ホーチキ株式会社 | Smoke detection device and smoke identification method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
IL300998A (en) * | 2016-04-07 | 2023-04-01 | Carmel Haifa Univ Economic Corporation Ltd | Image dehazing and restoration |
-
2020
- 2020-04-24 CN CN202010330521.9A patent/CN111553405B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104200445A (en) * | 2014-09-26 | 2014-12-10 | 常熟理工学院 | Image defogging method with optimal contrast ratio and minimal information loss |
WO2016207875A1 (en) * | 2015-06-22 | 2016-12-29 | Photomyne Ltd. | System and method for detecting objects in an image |
CN105844295A (en) * | 2016-03-21 | 2016-08-10 | 北京航空航天大学 | Video smog fine classification method based on color model and motion characteristics |
US10255670B1 (en) * | 2017-01-08 | 2019-04-09 | Dolly Y. Wu PLLC | Image sensor and module for agricultural crop improvement |
JP2020057236A (en) * | 2018-10-03 | 2020-04-09 | ホーチキ株式会社 | Smoke detection device and smoke identification method |
Non-Patent Citations (1)
Title |
---|
山西省高速公路团雾影响路段分级评价研究;牛彦峰;刘艳强;戎浩;;山西交通科技(第06期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111553405A (en) | 2020-08-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110992683B (en) | Dynamic image perception-based intersection blind area early warning method and system | |
CN111079640B (en) | Vehicle type identification method and system based on automatic amplification sample | |
Liu et al. | Visibility classification and influencing-factors analysis of airport: A deep learning approach | |
CN110120218A (en) | Expressway oversize vehicle recognition methods based on GMM-HMM | |
CN111582380B (en) | Ship track density clustering method and device based on space-time characteristics | |
CN109164450B (en) | Downburst prediction method based on Doppler radar data | |
CN115205796B (en) | Rail line foreign matter intrusion monitoring and risk early warning method and system | |
CN113409252B (en) | Obstacle detection method for overhead transmission line inspection robot | |
CN113609895A (en) | Road traffic information acquisition method based on improved Yolov3 | |
CN112580575A (en) | Electric power inspection insulator image identification method | |
CN111553405B (en) | Group fog recognition algorithm based on pixel density K-means clustering | |
CN113378905B (en) | Small target detection method based on distribution distance | |
Sun et al. | Objects detection with 3-d roadside lidar under snowy weather | |
CN111832463A (en) | Deep learning-based traffic sign detection method | |
Wang et al. | Forewarning method of downburst based on feature recognition and extrapolation | |
CN110515081A (en) | A kind of radar return zero_dynamics system intelligent recognition method for early warning | |
CN116110230A (en) | Vehicle lane crossing line identification method and system based on vehicle-mounted camera | |
CN116415163A (en) | Unmanned aerial vehicle identification method based on radar data | |
CN114019503A (en) | FOD detection system-based airport runway foreign matter detection method, device and storage medium | |
CN110309802B (en) | Convection monomer detection method based on extended maximum value transformation | |
CN113591714A (en) | Flood detection method based on satellite remote sensing image | |
JP4723771B2 (en) | Lightning determination system and lightning determination method | |
CN110533034A (en) | A kind of automobile front face brand classification method | |
CN108986542A (en) | A kind of automatic distinguishing method of city intersection accident potential stain | |
CN112799156B (en) | Meteorological integrated emergency early warning issuing method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |