CN111091087B - Land coverage extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing image - Google Patents
Land coverage extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing image Download PDFInfo
- Publication number
- CN111091087B CN111091087B CN201911270329.9A CN201911270329A CN111091087B CN 111091087 B CN111091087 B CN 111091087B CN 201911270329 A CN201911270329 A CN 201911270329A CN 111091087 B CN111091087 B CN 111091087B
- Authority
- CN
- China
- Prior art keywords
- view
- typical
- forest
- remote sensing
- space
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/245—Classification techniques relating to the decision surface
- G06F18/2451—Classification techniques relating to the decision surface linear, e.g. hyperplane
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/259—Fusion by voting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a land cover extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing images, which comprises the steps of constructing and generating multi-views by using spectral-texture-morphological section characteristics of high-resolution remote sensing images; performing characteristic space transformation between the high-resolution remote sensing multi-view characteristics by using typical correlation analysis; constructing multi-view linear combination classification for selecting non-leaf node characteristics of a tilting tree by utilizing a multivariate tilting decision tree; projecting the original multi-view space into a canonical correlation space; in consideration of the relationship between the input features and the output categories, projecting the original input features to a typical component space by adopting typical correlation analysis on an input feature subspace and an output category; performing exhaustive search on the integrated multi-view data set in a projection feature space to complete hyperplane splitting, and generating a typical relevant forest by integrating trees; and performing majority voting on the prediction result of each tree in the typical relevant forest so as to obtain high-precision land cover category information.
Description
Technical Field
The invention belongs to the technical field of remote sensing mode recognition, and particularly relates to a multi-view cooperation typical relevant forest remote sensing image land cover extraction algorithm, which can improve the final classification precision and extraction efficiency of high-resolution remote sensing land cover information aiming at a complex land surface environment and is suitable for various relevant applications of extraction, dynamic expansion monitoring, area planning and the like of the land cover detailed information of a high-resolution remote sensing image in complex environments such as an urban area, a mining area, a wetland and the like.
Background
For the data processing and image information presentation of high-resolution images, there are several difficulties as follows: (1) The intra-class spectral difference of the same class of ground objects in the image is increased, and the inter-class spectral difference of different class of ground objects is reduced, so that the phenomena of same object, different spectrum and same spectrum foreign matter commonly exist; (2) The traditional method is difficult to effectively describe the complex natural environment and various artificial ground objects existing in the scene; (3) The image spatial resolution and the number of features are increased sharply, so that the calculated amount is increased exponentially; these problems make the conventional classification method suitable for the medium-low resolution remote sensing image difficult to satisfy the requirement of high resolution remote sensing image classification, and a new data processing method is urgently needed to be explored to fully utilize all the potentials of the high resolution image.
Under the background, the invention provides a multi-view collaborative typical forest related remote sensing image land cover classification algorithm aiming at the characteristics of a high spatial resolution remote sensing image, and a multi-view data set is constructed by extracting spectral, texture and morphological characteristics; calculating the relation between the input features and the output categories by utilizing typical correlation analysis to perform feature space projection; integrating a multi-view data set and a Gini index to perform exhaustive search in a projection characteristic space so as to complete hyperplane splitting and construct a typical correlation tree; and combining the generated plurality of typical correlation trees to generate a typical correlation forest, and obtaining a final classification result by adopting a majority voting method.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a soil coverage extraction algorithm based on multi-view collaborative typical relevant forest remote sensing images.
In order to achieve the purpose, the invention adopts the following technical scheme:
a land cover extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing images comprises the following steps:
(1) And constructing and generating a multi-view by using the spectrum-texture-morphology section characteristics of the high-resolution remote sensing image.
(2) And performing feature space transformation between the high-resolution remote sensing multi-view features by using typical correlation analysis.
(3) And constructing multi-view linear combination classification for selecting non-leaf node features of the tilt tree by utilizing the multivariate tilt decision tree.
(4) And projecting the original multi-view space to a typical correlation space, sequencing the multi-view correlation on the feature set and the attribute set of the non-leaf node, and growing the multi-feature with the maximum value.
(5) Taking into account the relationship between the input features and the output classes, a canonical correlation analysis is applied to the input feature subspace and the output classes to project the original input features into a canonical component space.
(6) And performing exhaustive search on the integrated multi-view data set in a projection feature space to complete hyperplane splitting, further completing tree model training to generate a typical relevant tree, and generating a typical relevant forest by integrating the tree.
(7) And performing majority voting on the prediction result of each tree in the typical relevant forest so as to obtain high-precision land cover category information.
By adopting the technical scheme, the invention has the beneficial effects that:
the invention aims at the high-spatial resolution remote sensing image and solves the problem of effectively utilizing the spectrum-spatial characteristics of the high-resolution remote sensing data to extract the land cover information with high precision; firstly, performing band operation on a remote sensing image to obtain spectral characteristics, performing gray level co-occurrence matrix to obtain texture characteristics, and performing morphological operation and reconstruction to obtain morphological characteristics; then, a multi-view data set is constructed by utilizing the spectral-spatial characteristics obtained by calculation; considering the relation between input features and output categories, and projecting a feature space by adopting typical correlation analysis at a node trained by the model; integrating a multi-view data set and a tilt splitting rule to perform exhaustive search in a projection feature space to complete hyperplane splitting so as to construct a typical relevant tilt tree; and combining the prediction results of a plurality of typical correlation trees by adopting a majority voting method to form a typical correlation forest, and finally achieving the purpose of high-precision land coverage classification of high-resolution remote sensing images.
Drawings
FIG. 1 is a flow chart of steps (1) - (7) of a land cover extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing images.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further explained by combining the specific embodiments as follows:
therefore, the following detailed description of the embodiments of the present invention, provided in the accompanying drawings, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention, and all other embodiments, which can be derived by a person of ordinary skill in the art based on the embodiments of the present invention without inventive faculty, are within the scope of the invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined or explained in subsequent figures.
The terms "first," "second," "third," and the like are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should also be noted that, unless otherwise explicitly specified or limited, the terms "disposed," "mounted," "connected," and "connected" are to be construed broadly and may, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; the two elements can be directly connected or indirectly connected through an intermediate medium, and the two elements are communicated with each other, so that the specific meaning of the terms in the invention can be understood by those skilled in the art.
In conjunction with FIG. 1:
a land cover extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing images comprises the following steps:
step 1: inputting high spatial resolution remote sensing image data;
the algorithm can process the high-spatial-resolution remote sensing images acquired by different remote sensing platform (aerospace, aviation and ground) sensors.
Step 2: extracting and generating multi-view features;
such as spectral-texture-morphological-spatial structure, etc.; calculating and extracting texture information of a gray level co-occurrence matrix (GLCM) of a research area by utilizing a moving window; the spatial index is segmented, and space analysis is carried out by using morphology, and a positive autocorrelation region and a negative autocorrelation region which are high relative to the earth surface type to be classified are respectively calculated and extracted; removing isolated small-area objects, and performing growth calculation on meaningful objects by using mathematical morphology filling; and performing intersection operation on the optimized object graph, and identifying and extracting an area which has positive autocorrelation and negative autocorrelation simultaneously.
And step 3: converting the orthogonal space into an inclined projection characteristic space;
and adopting typical correlation analysis on input features and output categories in the multi-view data set at nodes of the tree model, and increasing the correlation of the features and the categories in the common projection space to the maximum extent so as to construct a projection feature space.
And 4, step 4: generating a typical correlation tree;
and projecting the original multi-view space to a typical correlation space, and selecting multi-view features segmented at each time by utilizing the established linear classifier on the feature set and the attribute set of the non-leaf nodes so as to construct a typical correlation tree.
And 5: generating typical relevant forests in a combined mode;
and (4) constructing a plurality of typical relevant trees according to the method from the step 3 to the step 4, performing exhaustive search on the integrated multi-view data set in a projection feature space to finish hyperplane splitting, further generating typical relevant trees, and integrating the trees to generate typical relevant forests.
Step 6: obtaining accurate land coverage information;
and performing majority voting on the prediction result of each tree in the typical relevant forest so as to obtain high-precision land cover type information.
A land cover extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing images is described in detail for the specific implementation steps of the invention:
(1) Constructing a high-score multi-view collaborative model; and constructing and generating a multi-view by using the spectrum-texture-morphology section characteristics of the high-resolution remote sensing image.
(2) Multi-dimensional orthogonal space to tilted space conversion; and performing feature space transformation between the high-resolution remote sensing multi-view features by using typical correlation analysis.
(3) Constructing a single inclined subtree model element; for a tree model in a typical relevant forest model, not an orthogonal decision tree in a traditional random forest, but a multivariable inclined decision tree is adopted, a linear classifier is formed by a linear combination test on multi-view attributes in the decision tree,
a i is the view attribute x i The weight of (c).
(4) Growing an inclined subtree model; and projecting the original multi-view space to a typical correlation space, and sequencing the multi-view correlation and solving the multi-feature of the maximum value to grow on the feature set and the attribute set of the non-leaf node by using the established linear classifier.
(5) Constructing a typical component feature space; taking into account the relationship between the input features and the output classes, a canonical correlation analysis is applied to the input feature subspace and the output classes,
where λ represents the correlation of multiple feature views for projecting the original input features into the canonical component space.
(6) Generating a typical relevant forest; and performing exhaustive search on the integrated multi-view data set in a projection feature space to complete hyperplane splitting, further generating a typical correlation tree, and generating a typical correlation forest for tree integration.
(7) Obtaining accurate land coverage information; the prediction results for each tree in a typical related forest are majority voted, which for the two categories of questions can be simply expressed as,
where ε is the probability that a member tree is in error, for the general case >>
Wherein +>
Represents h i In class label c j And then high-precision land cover type information is obtained.
In this embodiment, as described in step (1), the algorithm supports inputting different characteristic views of the high-resolution remote sensing image to construct a multi-view.
In this embodiment, in the step (2), the orthogonal to oblique space conversion is performed on the multidimensional feature of the high-resolution remote sensing multi-view through a typical correlation analysis.
In this embodiment, in the generation process of the single tree model in step (3), the traditional orthogonal decision tree is not used, but a tilted decision tree with enhanced diversity is used.
In this embodiment, the oblique subtree model in step (4) grows; through multi-view typical correlation analysis and maximum correlation value sequencing, an inclined subtree model is generated, and the operation efficiency and the calculation accuracy of the model can be effectively improved.
In this embodiment, in the step (5), a characteristic component feature space is constructed; and performing feature selection and multivariate cooperative calculation based on input feature and output category typical correlation analysis to construct an optimal typical correlation space based on feature optimization.
In this embodiment, a typical forest is generated in the step (6); and (3) finding a final splitting plane by carrying out novel exhaustive search on the space after the multi-view transformation, iteratively circulating to finish the growth of a single inclined tree, and generating a typical relevant forest through multiple iterations.
In this embodiment, the step (7) adopts a majority voting method for the predictive flag result of each typical related tree in the forest to obtain the final classification result.
Aiming at the high-spatial resolution remote sensing image, the invention solves the problem of effectively utilizing the spectrum-space characteristics of the high-resolution remote sensing data to extract the land cover information with high precision; firstly, obtaining spectral characteristics by adopting band operation on a remote sensing image, obtaining texture characteristics by adopting a gray level co-occurrence matrix, and obtaining morphological characteristics by adopting morphological operation and reconstruction; then, a multi-view data set is constructed by utilizing the spectral-spatial characteristics obtained by calculation; considering the relation between input features and output categories, and projecting a feature space by adopting typical correlation analysis at a node trained by the model; integrating a multi-view data set and a tilt splitting rule to perform exhaustive search in a projection feature space to complete hyperplane splitting so as to construct a typical relevant tilt tree; and combining the prediction results of a plurality of typical correlation trees by adopting a majority voting method to form a typical correlation forest, and finally achieving the purpose of high-precision land coverage classification of high-resolution remote sensing images.
The present embodiment is not intended to limit the shape, material, structure, etc. of the present invention in any way, and any simple modification, equivalent change and modification made to the above embodiments according to the technical spirit of the present invention are within the protection scope of the technical solution of the present invention.
Claims (8)
1. A land cover extraction method based on multi-view collaborative canonical correlation forest remote sensing images comprises the following steps:
(1) Constructing a high-score multi-view collaborative model; constructing and generating a multi-view by using the spectrum-texture-morphology profile characteristics of the high-resolution remote sensing image;
(2) Converting the multidimensional orthogonal space into an inclined space; performing characteristic space transformation between the high-resolution remote sensing multi-view characteristics by using typical correlation analysis;
(3) Constructing a single inclined subtree model element; for a tree model in a typical relevant forest model, not an orthogonal decision tree in a traditional random forest but a multivariable slant decision tree is adopted, a linear classifier is formed in the decision tree through a linear combination test on multi-view attributes,
a i is the view attribute x i The weight of (c);
(4) Growing an inclined subtree model; projecting an original multi-view space to a typical correlation space, sorting the multi-view correlation and solving the multi-feature of the maximum value to grow on the feature set and the attribute set of the non-leaf node by using the established linear classifier;
(5) Constructing a typical component feature space; taking into account the relationship between the input features and the output classes, a canonical correlation analysis is applied to the input feature subspace and the output classes,
wherein λ represents the correlation of multiple feature views for projecting the original input features into a typical component space;
(6) Generating a typical relevant forest; performing exhaustive search on the integrated multi-view data set in a projection feature space to complete hyperplane splitting, further generating a typical relevant tree, and generating a typical relevant forest for tree integration;
(7) Obtaining accurate land coverage information; the prediction results for each tree in a typical related forest are majority voted, which for the two categories of questions can be simply expressed as,
where ε is the probability of a member tree error, for the general case
Wherein->
Represents h i Class mark c j And then high-precision land cover type information is obtained.
2. The method for extracting the land cover based on the multi-view collaborative canonical correlation forest remote sensing image according to claim 1, characterized by comprising the following steps: in the step (1), the method supports the input of different characteristic views of the high-resolution remote sensing image so as to construct multiple views.
3. The method for extracting the land cover based on the multi-view collaborative canonical correlation forest remote sensing image as claimed in claim 1, characterized in that: and (3) performing orthogonal to inclined space conversion on the multi-dimensional features of the high-resolution remote sensing multi-view through typical correlation analysis in the step (2).
4. The method for extracting the land cover based on the multi-view collaborative canonical correlation forest remote sensing image according to claim 1, characterized by comprising the following steps: in the generation process of the single tree model in the step (3), the traditional orthogonal decision tree is not used, but the inclined decision tree with the difference enhanced.
5. The method for extracting the land cover based on the multi-view collaborative canonical correlation forest remote sensing image according to claim 1, characterized by comprising the following steps: growing an inclined subtree model in the step (4); the inclined sub-tree model is generated through multi-view typical correlation analysis and correlation maximum value sequencing, and the operation efficiency and the calculation precision of the model can be effectively improved.
6. The method for extracting the land cover based on the multi-view collaborative canonical correlation forest remote sensing image according to claim 1, characterized by comprising the following steps: constructing a typical component feature space in the step (5); and performing feature selection and multivariate cooperative calculation based on input feature and output category typical correlation analysis to construct an optimal typical correlation space based on feature optimization.
7. The method for extracting the land cover based on the multi-view collaborative canonical correlation forest remote sensing image according to claim 1, characterized by comprising the following steps: generating a typical forest in the step (6); and (3) finding a final splitting plane by carrying out novel exhaustive search on the space after the multi-view transformation, iteratively circulating to finish the growth of a single inclined tree, and generating a typical relevant forest through multiple iterations.
8. The method for extracting the land cover based on the multi-view collaborative canonical correlation forest remote sensing image as claimed in claim 1, characterized in that: and (7) adopting a majority voting method for the predictive marking result of each typical relevant tree in the forest to obtain a final classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911270329.9A CN111091087B (en) | 2019-12-12 | 2019-12-12 | Land coverage extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911270329.9A CN111091087B (en) | 2019-12-12 | 2019-12-12 | Land coverage extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111091087A CN111091087A (en) | 2020-05-01 |
| CN111091087B true CN111091087B (en) | 2023-04-07 |
Family
ID=70394904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911270329.9A Active CN111091087B (en) | 2019-12-12 | 2019-12-12 | Land coverage extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111091087B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114022782B (en) * | 2021-11-01 | 2024-05-10 | 中国石油大学(华东) | Sea fog detection method based on MODIS satellite data |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3848749B2 (en) * | 1997-08-13 | 2006-11-22 | 松下電器産業株式会社 | Image search device |
| CN203964904U (en) * | 2014-07-26 | 2014-11-26 | 刘浙 | Trees detector perpendicularity detector |
| CN108038448A (en) * | 2017-12-13 | 2018-05-15 | 河南理工大学 | Semi-supervised random forest Hyperspectral Remote Sensing Imagery Classification method based on weighted entropy |
-
2019
- 2019-12-12 CN CN201911270329.9A patent/CN111091087B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111091087A (en) | 2020-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111062282B (en) | Substation pointer instrument identification method based on improved YOLOV3 model | |
| CN110245709B (en) | 3D point cloud data semantic segmentation method based on deep learning and self-attention | |
| Guan et al. | Integration of orthoimagery and lidar data for object-based urban thematic mapping using random forests | |
| Lv et al. | Iterative training sample expansion to increase and balance the accuracy of land classification from VHR imagery | |
| CN108428220B (en) | Automatic geometric correction method for ocean island reef area of remote sensing image of geostationary orbit satellite sequence | |
| Cavallaro et al. | Automatic attribute profiles | |
| Ko et al. | Tree genera classification with geometric features from high-density airborne LiDAR | |
| Zeybek | Classification of UAV point clouds by random forest machine learning algorithm | |
| Chen | Mapping mangrove in Dongzhaigang, China using Sentinel-2 imagery | |
| CN111598045A (en) | Remote sensing farmland change detection method based on object spectrum and mixed spectrum | |
| Su et al. | Machine learning-assisted region merging for remote sensing image segmentation | |
| Kozoderov et al. | Retrieval of forest stand attributes using optical airborne remote sensing data | |
| CN112508936A (en) | Remote sensing image change detection method based on deep learning | |
| CN102521624A (en) | Classification method for land use types and system | |
| Kozoderov et al. | A system for processing hyperspectral imagery: application to detecting forest species | |
| Zafari et al. | A multiscale random forest kernel for land cover classification | |
| Dmitriev et al. | Spectral texture classification of high-resolution satellite images for the state forest inventory in Russia | |
| Quan et al. | Tree species classification in a typical natural secondary forest using UAV-borne LiDAR and hyperspectral data | |
| CN111091087B (en) | Land coverage extraction algorithm based on multi-view collaborative canonical correlation forest remote sensing image | |
| Sjahputera et al. | Clustering of detected changes in high-resolution satellite imagery using a stabilized competitive agglomeration algorithm | |
| Kuzera et al. | Importance of matrix construction for multiple-resolution categorical map comparison | |
| CN116740474A (en) | Remote sensing image classification method based on anchoring stripe attention mechanism | |
| CN117392533A (en) | Construction method of high-spectrum remote sensing image lithology sample set | |
| CN110427956B (en) | LiDAR point cloud data single tree extraction method based on spectral clustering algorithm | |
| Jusoff et al. | Mapping of individual oil palm trees using airborne hyperspectral sensing: an overview |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220712 Address after: 570100 No.12 Baiju Avenue, Meilan District, Haikou City, Hainan Province Applicant after: HAINAN ACADEMY OF OCEAN AND FISHERIES SCIENCES Address before: 454000 No. 2001 Century Avenue, hi tech Zone, Henan, Jiaozuo Applicant before: HENAN POLYTECHNIC University |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |