CN110321826B - Unmanned aerial vehicle side slope vegetation classification method based on plant height - Google Patents
Unmanned aerial vehicle side slope vegetation classification method based on plant height Download PDFInfo
- Publication number
- CN110321826B CN110321826B CN201910563029.3A CN201910563029A CN110321826B CN 110321826 B CN110321826 B CN 110321826B CN 201910563029 A CN201910563029 A CN 201910563029A CN 110321826 B CN110321826 B CN 110321826B
- Authority
- CN
- China
- Prior art keywords
- image
- aerial vehicle
- unmanned aerial
- dsm
- classification
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an unmanned aerial vehicle side slope vegetation classification method based on plant height, which comprises the following steps: (1) Enabling the unmanned aerial vehicle to fly along a plurality of fixed-height equal-sharing interval routes in a snake-shaped mode and hover on sampling points on the fixed-height equal-sharing interval routes to acquire images; (2) Importing the image into a Pix4DMapper, synthesizing high-density point cloud data, obtaining DSM and an orthoimage from the high-density point cloud data, importing the DSM and the orthoimage into ArcGIS, manually selecting ground points in the ArcGIS according to the orthoimage, extracting the elevation of the ground points from the DSM, and generating a DTM (digital time frame model) by using an inverse distance weight interpolation method; nsm = DSM-DTM where nsm is plant height; DSM is a digital surface model; DTM is a digital terrain model; (3) constructing a sample training manager; (4) Acting the ecd file generated by the sample training manager based on the scheme management category and the sample library in the ortho-image by combining the classification method to obtain the classification result of the slope plant image; the method enables the classification precision to reach 90%.
Description
Technical Field
The invention relates to an unmanned aerial vehicle side slope vegetation classification method based on plant height.
Background
The slope ecological engineering comprises six main links such as planning design, engineering construction, engineering supervision, engineering acceptance, engineering management and engineering research, wherein the development of the four links such as the engineering supervision, the engineering acceptance, the engineering management and the engineering research needs to focus on plant space distribution, no matter the process and the quality in the engineering supervision, the stability evaluation in the engineering acceptance, the maintenance, the replanting, the thinning and the growth management in the engineering management, or the spatial pattern in a group in the engineering research, the biodiversity succession, the ecological evaluation and the like, and the dynamic high-frequency plant space distribution data are used as the basis. And the most important data dimension in plant spatial distribution monitoring is plant classification.
At present, the classification of slope plants is mainly based on a ground investigation method, and has the disadvantages of time and labor consumption and higher cost. The existing unmanned aerial vehicle plant species classification research is basically carried out in areas with flat terrain and no obvious fluctuation, and classification objects are simpler. The slope, whether viewed from the terrain or community structure, is very different from existing research objects. The geological environment of China is rich, but natural disasters are frequent, earthquake disasters or natural landslides are frequently encountered, and activities such as human mining or construction and the like can also damage natural plants, so that a plurality of exposed side slopes are formed, and the side slope restoration is required to be normal high frequency. The effect of slope restoration needs to be evaluated through slope investigation and evaluation. However, the flight mode of the unmanned aerial vehicle, the data quality and the information mining are affected by the complex slope terrain, the slope plants are rich in levels and are matched in height, so that the plant classification is difficult to a certain degree, no research report for searching plant species under the slope condition by the unmanned aerial vehicle exists at present, and no existing research method can be used for reference, so that the research provided by the invention is an innovative attempt.
Disclosure of Invention
In view of this, the invention aims to provide an unmanned aerial vehicle side slope vegetation classification method based on plant height, which can improve the classification precision of side slope plants.
The purpose of the invention is realized by the following technical scheme:
an unmanned aerial vehicle side slope vegetation classification method based on plant height comprises the following steps:
(1) Enabling the unmanned aerial vehicle to fly along a plurality of fixed-height uniform spacing route lines in a snake-shaped mode and hover at sampling points on the fixed-height uniform spacing route lines to acquire images, wherein the fixed-height uniform spacing route lines are located in the same plane above a side slope and are arranged in parallel at even intervals along the slope direction of the side slope, and the route lines are provided with the sampling points at even intervals;
(2) Importing the image into a Pix4DMapper, synthesizing high-density point cloud data, obtaining a DSM (digital surface model) and an orthographic image from the high-density point cloud data, importing the DSM and the orthographic image into an ArcGIS (geographic information system), manually selecting a ground point in the ArcGIS according to the orthographic image, extracting the elevation of the ground point from the DSM, and generating a DTM (digital time frame model) by using an inverse distance weight interpolation method;
wherein nDSM is the plant height image; DSM is a digital surface model; DTM is a digital terrain model;
(3) Constructing a sample training manager, specifically, adding the manually distinguished plant species into ArcGIS to generate a scheme management class; then managing the categories according to the scheme, and circling the object categories to be classified in the near-earth orthographic images to obtain a sample library; the method comprises the steps that low-altitude flight is conducted through an unmanned aerial vehicle, collected images are guided into a Pix4DMapper to synthesize high-density point cloud data, and a near-earth orthographic image is obtained from the high-density point cloud data;
(4) And (4) acting the ecd file generated by the sample training manager based on the scheme management category and the sample library in combination with a classification method on the nDSM to obtain a classification result of the plant image of the side slope.
Further, sampling points on the air route corresponding to the slope toe to the air route corresponding to the slope top are gradually increased.
Further, a certain number of precision evaluation points are randomly generated in the slope plant image, the precision evaluation points are compared with manual classification, the accuracy of precision evaluation point classification is obtained, and finally a precision inspection result set is generated.
Further, a confusion matrix is generated by using the accuracy test result set, and the classification accuracy is analyzed according to the counting, the user accuracy, the producer accuracy and the FScore, and the applicability of the classification method is evaluated according to the Kappa score.
Still further, the classification method is a random forest algorithm.
The beneficial effects of the invention are:
according to the unmanned aerial vehicle side slope vegetation classification method based on the plant height, the classification method combining nDSM and the orthographic image is added under the side slope environment, the plant types with similar textures in the remote sensing image can be well distinguished, the classification precision of plant species is obviously improved, the classification precision reaches 90%, DTM is obtained by adopting a mode of manually selecting ground points, the root mean square error of DTM is 0.28m, then relatively accurate nDSM is obtained, and the classification precision can be effectively improved.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof.
Drawings
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings, in which:
FIG. 1 is a schematic flow chart of the present invention.
FIG. 2 is a schematic view of a constant-height and uniform-spacing route.
Fig. 3 is a schematic distribution diagram of the height-setting and average-dividing spacing routes above the side slope.
Fig. 4 is an image acquired by the unmanned aerial vehicle along the fixed-height equal-division interval air route.
Fig. 5 is the DSM image obtained.
Fig. 6 is an orthophotograph obtained.
Fig. 7 is a schematic diagram of selected ground points during DTM generation.
Fig. 8 is a DTM image.
Fig. 9 is the resulting nsmd image.
Fig. 10 is a diagram illustrating the classification result.
Fig. 11 is a schematic diagram of training sample distribution.
Fig. 12 is a classification accuracy check point.
Detailed Description
Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be understood that the preferred embodiments are illustrative of the invention only and are not limiting upon the scope of the invention.
The method is implemented on an artificial side slope and a natural mountain land in the northern area respectively, the slope height of the side slope is 75m, the slope length is 150m, and the slope ratio is 1/2.
As shown in fig. 1, an unmanned aerial vehicle side slope vegetation classification method based on plant height includes the following steps:
(1) As shown in fig. 2, the unmanned aerial vehicle flies along a plurality of constant-height evenly-divided spaced routes in a snake-shaped manner and hovers over the constant-height evenly-divided spaced routes to acquire images (as shown in fig. 4), the constant-height evenly-divided spaced routes are located in the same plane above the side slope and are arranged in parallel at even intervals along the slope direction of the side slope, and the routes are provided with a plurality of sampling points at even intervals; the unmanned aerial vehicle comprises a four-rotor carbon fiber frame, a three-axis camera holder and a flying firefly 8S motion camera, wherein a hardware system of the AutoPilot adopts Pix-hawk 2.4.8, and a software system is AutoPilot. As shown in fig. 3, the toe of the sloping field in the figure is α, h1, h2, and h3 are the vertical distances from the unmanned aerial vehicle to the ground, and b1, b2, and b3 respectively correspond to the actual ground frames of the images of the three routes at the sampling points P1, P2, and P3.
(2) Importing the image into a Pix4 DMaper and synthesizing high-density point cloud data, obtaining DSM (shown in figure 5) and an orthoimage (shown in figure 6) from the high-density point cloud data, importing the DSM and the orthoimage into ArcGIS, manually selecting ground points in the ArcGIS according to the orthoimage, extracting the elevation of the ground points from the DSM, and generating a DTM (digital time frame) by an inverse distance weight interpolation method;
nDSM=DSM-DTM
wherein nDSM is the plant height image; DSMs are digital surface models representing the true elevation of the plant with a resolution of 1 x GSD (5.59 cm/pixel); DTM is a digital terrain model; the orthoimage is an image corrected by a vertical visual angle and corrected by adopting adjacent pixels, so that the projection difference of the plant image is corrected, the plant on the ground can be more accurately shown, and the purpose of generating the DTM by using an inverse distance weight interpolation method is as follows: considering that the high-density point cloud data is obtained by three-dimensional reconstruction, compared with LiDAR point cloud, in a forest land area with dense trees, ground points are sparser, and also considering that an orthoimage obtained by digital differential correction may have problems of partial stretching or missing and the like, 1200 ground points are selected in total, as shown in fig. 7, DTM (as shown in fig. 8) is generated by interpolation fitting, and the generation result of an nmsm image (as shown in fig. 9) can clearly see the distribution of middle lanes and higher plants at the periphery.
(3) Constructing a sample training manager, specifically, adding the manually distinguished plant species into ArcGIS to generate a scheme management category; then managing the categories according to the scheme, and enclosing the object categories to be classified in the near-earth orthographic images to obtain a sample library, wherein the number of the near-earth orthographic images is 13, and as shown in fig. 11, 187 training samples are finally selected on the slope of the northern and Sichuan sample plot, and 78 training samples are selected in the natural sample plot; the method for acquiring the near-earth ortho-image comprises the following steps: the method comprises the steps that an unmanned aerial vehicle flies in low-altitude mode and guides an image into a Pix4 DMaper to synthesize high-density point cloud data, a ground-near orthographic image is obtained from the high-density point cloud data, and the distance between a sampling point of the unmanned aerial vehicle and the ground is 30m; the classification method in the invention is a random forest algorithm, namely, the random forest algorithm is used for feature selection and arbor species diversity estimation.
(4) The method is characterized in that the ecd file generated by the sample training manager based on the scheme management category and the sample library is combined with a classification method to act on the nDSM, and a classification result of the plant image of the side slope is obtained (as shown in figure 10).
As shown in figure 2, sampling points on the route corresponding to the toe of the slope and the route corresponding to the top of the slope are gradually increased, so that the density of the sampling points on the top of the slope is increased, the heading coverage rates of the top of the slope and the toe of the slope are kept consistent, and cavities possibly caused by the images on the top of the slope in three-dimensional reconstruction can be effectively avoided.
The method also comprises precision inspection, namely randomly generating a certain number of precision evaluation points in the slope plant image, comparing the precision evaluation points with artificial classification to obtain the accuracy of precision evaluation point classification, and finally generating a precision inspection result set, wherein 595 random sample points and 352 random sample points are selected as precision verification points on the identified plum, coriaria sinica and other ground objects respectively for the artificial slope of the northern sample plot and the natural mountain of the northern sample plot to determine the effectiveness of the classification method, and the distribution condition of the precision inspection points of the northern sample plot is shown in figure 12.
And generating a confusion matrix by using the precision test result set, analyzing the classification precision according to the counting, the user precision, the producer precision and the FScore, evaluating the applicability of the classification method according to the Kappa score, comparing the classification result of each sample with the real result by using the confusion matrix, and forming a cross table about the attribute set, wherein the content is the frequency of the occurrence of the comparison result.
The precision test shows that the classification precision of 27 plants is improved, and the improvement range is obvious particularly for cogongrass, miscanthus sinensis, quercus kwangsieboldii, poplar, ranunculus japonicus, gossypium hirsutum, cypress, osmanthus fragrans, walnut, bamboo grass, eucalyptus and small carex.
The method can play a great role in a complex northern slope environment, and the classification precision can reach 90% in the dimension of the seeds.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (5)
1. An unmanned aerial vehicle side slope vegetation classification method based on plant height is characterized in that: the method comprises the following steps:
(1) Enabling the unmanned aerial vehicle to fly along a plurality of fixed-height uniform spacing route lines in a snake-shaped mode and hover at sampling points on the fixed-height uniform spacing route lines to acquire images, wherein the fixed-height uniform spacing route lines are located in the same plane above a side slope and are arranged in parallel at even intervals along the slope direction of the side slope, and the route lines are provided with the sampling points at even intervals;
(2) Importing the image into a Pix4DMapper, synthesizing high-density point cloud data, obtaining a DSM (digital surface model) and an orthographic image from the high-density point cloud data, importing the DSM and the orthographic image into an ArcGIS (geographic information system), manually selecting a ground point in the ArcGIS according to the orthographic image, extracting the elevation of the ground point from the DSM, and generating a DTM (digital time frame model) by using an inverse distance weight interpolation method;
nDSM=DSM-DTM
wherein nDSM is the plant height image; DSM is a digital surface model; DTM is a digital terrain model;
constructing a sample training manager, specifically, adding the manually distinguished plant species into ArcGIS to generate a scheme management class; then managing the categories according to a scheme, and enclosing the categories of the objects to be classified in the near-earth orthographic images to obtain a sample library; the method for acquiring the near-earth ortho-image comprises the following steps: the method comprises the steps that an unmanned aerial vehicle flies in a low-altitude mode and guides an acquired image into a Pix4DMapper to synthesize high-density point cloud data, and a near-earth orthographic image is obtained from the high-density point cloud data;
and acting the ecd file generated by the sample training manager based on the scheme management category and the sample library in combination with a classification method on the nDSM to obtain a classification result of the slope plant image.
2. The unmanned aerial vehicle side slope vegetation classification method based on plant height as claimed in claim 1, wherein: and gradually increasing sampling points from the route corresponding to the base of the slope to the route corresponding to the top of the slope.
3. The unmanned aerial vehicle side slope vegetation classification method based on plant height as claimed in claim 1 or 2, characterized in that: randomly generating a certain number of precision evaluation points in the slope plant image, comparing the precision evaluation points with manual classification to obtain the accuracy of precision evaluation point classification, and finally generating a precision test result set.
4. The unmanned aerial vehicle side slope vegetation classification method based on plant height as claimed in claim 3, wherein: and generating a confusion matrix by using the precision test result set, analyzing and classifying precision according to the counting, the user precision, the producer precision and the FScore, and evaluating the applicability of the classification method according to the Kappa score.
5. The method for classifying vegetation on unmanned aerial vehicle slopes based on plant height as claimed in claim 1, wherein: the classification method is a random forest algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910563029.3A CN110321826B (en) | 2019-06-26 | 2019-06-26 | Unmanned aerial vehicle side slope vegetation classification method based on plant height |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910563029.3A CN110321826B (en) | 2019-06-26 | 2019-06-26 | Unmanned aerial vehicle side slope vegetation classification method based on plant height |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110321826A CN110321826A (en) | 2019-10-11 |
| CN110321826B true CN110321826B (en) | 2023-02-24 |
Family
ID=68120385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910563029.3A Active CN110321826B (en) | 2019-06-26 | 2019-06-26 | Unmanned aerial vehicle side slope vegetation classification method based on plant height |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110321826B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144213B (en) * | 2019-11-26 | 2023-08-18 | 北京华捷艾米科技有限公司 | Object detection method and related equipment |
| CN112683171B (en) * | 2020-12-31 | 2023-04-14 | 数字港湾科技(湖北)有限公司 | Method and device for measuring and calculating earthwork amount |
| CN112668534B (en) * | 2021-01-05 | 2022-08-23 | 长沙理工大学 | Forest zone vegetation height inversion method based on digital orthographic images and digital surface models |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009131108A1 (en) * | 2008-04-23 | 2009-10-29 | 株式会社パスコ | Building roof outline recognizing device, building roof outline recognizing method, and building roof outline recognizing program |
| WO2016064045A1 (en) * | 2014-10-20 | 2016-04-28 | 대한민국(산림청 국립산림과학원장) | Stand height estimation system and method based on digital aerial photos |
| CN107330279A (en) * | 2017-07-05 | 2017-11-07 | 贵州省草业研究所 | A kind of high mountain permafrost area vegetation pattern Forecasting Methodology |
| CN108846352A (en) * | 2018-06-08 | 2018-11-20 | 广东电网有限责任公司 | A kind of vegetation classification and recognition methods |
| CN109684929A (en) * | 2018-11-23 | 2019-04-26 | 中国电建集团成都勘测设计研究院有限公司 | Terrestrial plant ECOLOGICAL ENVIRONMENTAL MONITORING method based on multi-sources RS data fusion |
-
2019
- 2019-06-26 CN CN201910563029.3A patent/CN110321826B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009131108A1 (en) * | 2008-04-23 | 2009-10-29 | 株式会社パスコ | Building roof outline recognizing device, building roof outline recognizing method, and building roof outline recognizing program |
| WO2016064045A1 (en) * | 2014-10-20 | 2016-04-28 | 대한민국(산림청 국립산림과학원장) | Stand height estimation system and method based on digital aerial photos |
| CN107330279A (en) * | 2017-07-05 | 2017-11-07 | 贵州省草业研究所 | A kind of high mountain permafrost area vegetation pattern Forecasting Methodology |
| CN108846352A (en) * | 2018-06-08 | 2018-11-20 | 广东电网有限责任公司 | A kind of vegetation classification and recognition methods |
| CN109684929A (en) * | 2018-11-23 | 2019-04-26 | 中国电建集团成都勘测设计研究院有限公司 | Terrestrial plant ECOLOGICAL ENVIRONMENTAL MONITORING method based on multi-sources RS data fusion |
Non-Patent Citations (2)
| Title |
|---|
| Classification of slopes and prediction of factor of safety using differential evolution neural networks;Sarat Kumar Das,et al;《Environmental Earth Sciences》;20110122;201-210 * |
| 四川盆地路域边坡植被的数量分类与排序;庞亮;《植物研究》;20130515;360-366 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110321826A (en) | 2019-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guo et al. | Lidar boosts 3D ecological observations and modelings: A review and perspective | |
| CN110321826B (en) | Unmanned aerial vehicle side slope vegetation classification method based on plant height | |
| Bongers | Methods to assess tropical rain forest canopy structure: an overview | |
| Kelly et al. | Mapping forests with Lidar provides flexible, accurate data with many uses | |
| MX2012006717A (en) | Method and system for estimating vegetation growth relative to an object of interest. | |
| CN108195767A (en) | Estuarine wetland denizen monitoring method | |
| CN110298299B (en) | Unmanned aerial vehicle side slope vegetation classification method based on contour course | |
| Arnau et al. | Detecting and understanding historic landscapes | |
| Millikan et al. | Automated individual tree detection in Amazon tropical forest from airborne laser scanning data | |
| Malinowski et al. | Mapping rill soil erosion in agricultural fields with UAV‐borne remote sensing data | |
| Torma | Estimation of tree species proportions of forest st ands using laser scanning | |
| Robertson et al. | Determination of the compositional change (1999–2006) in the pine forests of British Columbia due to mountain pine beetle infestation | |
| van Leeuwen et al. | 3.08 vegetation structure (lidar) | |
| JP2012103155A (en) | Apparatus, method and program for estimating laser density distribution | |
| CN116403125B (en) | Grassland caterpillar suitable living area dividing method, system and terminal based on unmanned aerial vehicle aerial photography | |
| Nordiana et al. | Land Evaluation for Oil Palm Cultivation using Geospatial Information Technologies | |
| Viana et al. | Training samples from open data for satellite imagery classification: using K-means clustering algorithm | |
| Hatami | Carbon estimation of individual trees using high laser density data of airborne lidar: A case study in Bois Noir, France | |
| Zhang | Single tree detection from airborne laser scanning data: a stochastic approach | |
| Nakajima et al. | A growth prediction system for local stand volume derived from lidar data | |
| Hnila et al. | Quality Assessment of Digital Elevation Models in a Treeless High-Mountainous Landscape: A Case Study from Mount Aragats, Armenia | |
| Stals et al. | Identification of wet areas in agricultural lands using remote sensing data | |
| Reddy | Automated Estimation of Forest Inventories using Terrestrial LiDAR | |
| Barlow | A comparative study of raster and vector based approaches in vegetation mapping on Five Islands off the coast of Port Kembla. | |
| Sims | Remote Sensing Data and Methods in NFI |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Han Chao Inventor after: Liu Pin Inventor after: Li Linxia Inventor after: Yang Yin Inventor after: Guo Ke Inventor before: Han Chao Inventor before: Yang Yin Inventor before: Guo Ke |
|
| CB03 | Change of inventor or designer information |