CN113902793B - Method, system and electronic equipment for predicting end-to-end building height based on single-vision remote sensing image - Google Patents
Method, system and electronic equipment for predicting end-to-end building height based on single-vision remote sensing image Download PDFInfo
- Publication number
- CN113902793B CN113902793B CN202111304222.9A CN202111304222A CN113902793B CN 113902793 B CN113902793 B CN 113902793B CN 202111304222 A CN202111304222 A CN 202111304222A CN 113902793 B CN113902793 B CN 113902793B
- Authority
- CN
- China
- Prior art keywords
- model
- remote sensing
- prediction
- building height
- training
- 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
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 42
- 230000003287 optical effect Effects 0.000 claims abstract description 8
- 238000004519 manufacturing process Methods 0.000 claims abstract description 7
- 238000005070 sampling Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 14
- 238000012795 verification Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 description 8
- 238000001514 detection method Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method, a system and electronic equipment for predicting the height of a building end to end based on single-vision remote sensing images, wherein the method comprises the following steps: step 1: manufacturing a building height prediction data set based on the high-resolution optical remote sensing image; step 2: improving RETINANET model, adding down sampling layer in FPN layer and using OHEM loss function and Fast-NMS optimizing model in prediction part; step 3: training the modified RETINANET model using the building height prediction dataset; step 4: and carrying out building height prediction on the remote sensing image containing the building by using the RETINANET model after training. The method realizes the function of directly predicting the building height with high precision based on the single-vision high-spatial-resolution remote sensing image.
Description
Technical Field
The invention belongs to the technical field of optical remote sensing image processing, and particularly relates to a method, a system and electronic equipment for predicting the end-to-end building height based on single-vision remote sensing images.
Background
Object detection is one of the core problems in the field of computer vision, whose task is to find all objects (objects) of interest in an image, determine their category and location. The target detection technology based on deep learning has wide application value in urban research. The target detection algorithm can integrate the positioning and the identification of the target based on the geometric and statistical characteristics of the target. Building height information is an important parameter for city planning and city information research, and has important significance for city planning and development. Along with the development of remote sensing technology, remote sensing images show the development trend of high spatial resolution, large breadth and short revisiting period, and at present, remote sensing has become one of the main means of building change investigation. In the field of building height prediction research at present, most methods adopt laser point cloud data and image stereo pairs to invert the building height, and although the methods can accurately calculate the building height, the acquisition of a data source and the prediction process of the building height require higher cost, so that the building height prediction based on single-vision high-resolution remote sensing images has higher engineering application value and scientific research value.
Based on a deep learning algorithm, a part of research results are currently available in the research of extracting building heights from high-resolution remote sensing images by adopting an end-to-end mode.
The Chao Ji et al propose a method for measuring the height of a building in a remote sensing image end to end, which takes the outline of the building as a detection target, takes the height of a floor corresponding to the building as a detection category, and trains based on a Mask-RCNN network model. Experiments show that the model obtained through training can identify building height end to end, and for buildings with 1-7 layers of heights, the prediction error of the buildings with 1.329,8-20 layers of heights is 3.546, and the prediction error of the buildings with higher than 20 layers is 8.317.
Xiang Li et al utilized convolutional neural networks and spatial pooling pyramids to achieve end-to-end building prediction. The method takes aerial images as training data and DSM data as labels to form a training set. The result shows that the root mean square error of the predicted building height by adopting the CNN model of the spatial pooling pyramid is 1.698.
Lichao Mou et al propose a complete convolution-deconvolution network architecture that enables prediction of building height in an end-to-end manner. The jump connection mode is introduced into the network model, and the low-layer image characteristics in the image are reserved. The method adopts 0.7m aerial image as training data and DSM data as sample labels to construct a training set. Experimental results show that the method has higher precision.
The research results show that the end-to-end prediction of the building height based on the convolutional neural network has a certain feasibility and has a large lifting space. The invention provides a deep learning model for predicting building height end to end based on an improved RETINANET network, which can improve prediction accuracy.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides a method, a system and electronic equipment for predicting the end-to-end building height based on single-vision remote sensing images. The method has important significance for urban planning and development, and can provide technology and data support for optimizing urban planning, evaluating urban economy, improving the utilization rate of domestic soil resources and formulating related policies.
The invention provides a single-vision remote sensing image-based end-to-end building height prediction method, which is realized according to the following scheme, and comprises the following steps of:
Step 1: manufacturing a building height prediction data set based on the high-resolution optical remote sensing image;
step 2: improving RETINANET model, adding down sampling layer in FPN layer and using OHEM loss function and Fast-NMS optimizing model in prediction part;
Step 3: training the modified RETINANET model using the building height prediction dataset;
Step 4: and carrying out building height prediction on the remote sensing image containing the building by using the RETINANET model after training.
Further, the step 1 specifically includes:
step 1.1: selecting a remote sensing image for training a building height prediction model, wherein the remote sensing image selects Jilin one-size broad 01-star shooting data;
step 1.2: cutting data, namely cutting the selected complete data to obtain 363 multiplied by 263 pixels, and filling the cut part with the size smaller than 363 multiplied by 263 with a value of 0;
step 1.3: dividing the training set and the verification set, obtaining 4751 images after cutting, and according to the training set and the verification set 3:1, 3565 sheets were selected as training set and 1186 sheets were selected as validation set.
Further, the step2 specifically includes:
Step 2.1: performing downsampling on an original FPN layer in a RETINANET model twice;
Step 2.2: replacing FOCAL LOSS loss functions with OHEM functions at the class prediction portion of the RETINANET model;
step 2.3: the bounding box prediction part of the RETINANET model is added to the Fast-NMS.
Further, in the training process, the super parameter is set as follows: initial learning rate learning_rate=0.001, batch size batch_size=4, training algebra epochs =100000, segmentation class n_ classes =41.
Further, the step 4 specifically includes:
Step 4.1: slicing the image for test into an image with the size of 363×263, and filling the cut part with a value of 0 for the part with the size of less than 363×263;
step 4.2: building height predictions are made for each cropped image using a trained RETINANET model.
The invention also provides an end-to-end building height prediction system based on the single-vision remote sensing image, which comprises:
and (3) a data set making module: the method is used for manufacturing a building height prediction data set based on the high-resolution optical remote sensing image;
model improvement module: for improving RETINANET model, adding downsampling layer in FPN layer and using OHEM loss function and Fast-NMS optimizing model in prediction part;
model training module: for training the improvement RETINANET model using the building height prediction dataset;
And a prediction module: the method is used for predicting the building height of the remote sensing image containing the building by using the RETINANET model after training.
The invention also proposes an electronic device comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the single vision remote sensing image based end-to-end building height prediction method.
The invention has the beneficial effects that:
(1) The original feature pyramid network in the RETINANET model is downsampled twice, so that the model can learn richer features, and targets with different sizes can be identified;
(2) Improving RETINANET a pre-measurement head part network, and improving the running speed by adopting a shared convolution network; screening the prior frame generated by the frame prediction part by using Fast-NMS, so as to shorten the screening time; using OHEM to predict categories, OHEM can distinguish between positive and negative samples in a training set, enhancing the impact on the model training of positive samples.
(3) When the solar altitude angle, the satellite altitude angle and the satellite side sway angle are kept consistent, the method provided by the invention can realize building altitude prediction based on a single remote sensing image, solves the problems of low accuracy, complex flow, low data updating frequency, difficult data acquisition and the like of the traditional algorithm, provides a new method support for urban planning research, and can also provide assistance for monitoring urban buildings in industries such as banks, insurance and the like.
Drawings
FIG. 1 is a flow chart of the end-to-end building height prediction method based on single vision remote sensing images according to the invention;
FIG. 2 is a graph showing the comparison of the FPN layer before and after improvement;
FIG. 3 is a diagram of a modified RETINANET model network architecture.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
1-3, The invention provides a method for predicting the height of an end-to-end building based on single-vision remote sensing images, which comprises the following steps:
Step 1: manufacturing a building height prediction data set based on the high-resolution optical remote sensing image;
the step 1 specifically comprises the following steps:
step 1.1: selecting a remote sensing image for training a building height prediction model, wherein the remote sensing image selects Jilin one-size broad 01-star shooting data;
step 1.2: cutting data, namely cutting the selected complete data to obtain 363 multiplied by 263 pixels, and filling the cut part with the size smaller than 363 multiplied by 263 with a value of 0;
step 1.3: dividing the training set and the verification set, obtaining 4751 images after cutting, and according to the training set and the verification set 3:1, 3565 sheets were selected as training set and 1186 sheets were selected as validation set.
Step 2: improving RETINANET model, adding down sampling layer in FPN layer and using OHEM loss function and Fast-NMS optimizing model in prediction part;
The step 2 specifically comprises the following steps:
Step 2.1: performing downsampling on an original FPN layer in a RETINANET model twice;
Step 2.2: replacing FOCAL LOSS loss functions with OHEM functions at the class prediction portion of the RETINANET model;
step 2.3: the bounding box prediction part of the RETINANET model is added to the Fast-NMS.
Step 3: training the modified RETINANET model using the building height prediction dataset;
The program will run on a machine with CPU Intel Core i7-9700, gpu NVIDIA GeForce RTX 2060 (Compute Capability =7.5, 1920 cudacores), memory 16GB, operating system Ubuntu 18.04 using Python version 3.6, pytorch version 1.8.0.
During training, the super parameters are set as follows: initial learning rate learning_rate=0.001, batch size batch_size=4, training algebra epochs =100000, segmentation class n_ classes =41.
Step 4: and carrying out building height prediction on the remote sensing image containing the building by using the RETINANET model after training.
The step 4 specifically comprises the following steps:
Step 4.1: slicing the image for test into an image with the size of 363×263, and filling the cut part with a value of 0 for the part with the size of less than 363×263;
step 4.2: building height predictions are made for each cropped image using a trained RETINANET model.
The invention also provides an end-to-end building height prediction system based on the single-vision remote sensing image, which comprises:
and (3) a data set making module: the method is used for manufacturing a building height prediction data set based on the high-resolution optical remote sensing image;
model improvement module: for improving RETINANET model, adding downsampling layer in FPN layer and using OHEM loss function and Fast-NMS optimizing model in prediction part;
model training module: for training the improvement RETINANET model using the building height prediction dataset;
And a prediction module: the method is used for predicting the building height of the remote sensing image containing the building by using the RETINANET model after training.
The invention also proposes an electronic device comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the single vision remote sensing image based end-to-end building height prediction method.
The method, the system and the electronic equipment for predicting the building height based on the single-vision remote sensing image end to end are described in detail, and specific examples are applied to the principle and the implementation mode of the method, the system and the electronic equipment for predicting the building height based on the single-vision remote sensing image end to end, and the description of the examples is only used for helping to understand the method and the core idea of the method; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Claims (6)
1. The end-to-end building height prediction method based on the single-vision remote sensing image is characterized by comprising the following steps of: the method comprises the following steps:
Step 1: manufacturing a building height prediction data set based on the high-resolution optical remote sensing image;
step 2: improving RETINANET model, adding down sampling layer in FPN layer and using OHEM loss function and Fast-NMS optimizing model in prediction part;
The step 2 specifically comprises the following steps:
Step 2.1: performing downsampling on an original FPN layer in a RETINANET model twice;
Step 2.2: replacing FOCAL LOSS loss functions with OHEM functions at the class prediction portion of the RETINANET model;
Step 2.3: adding Fast-NMS in boundary box prediction part of RETINANET model;
Step 3: training the modified RETINANET model using the building height prediction dataset;
Step 4: and carrying out building height prediction on the remote sensing image containing the building by using the RETINANET model after training.
2. The method according to claim 1, characterized in that: the step 1 specifically comprises the following steps:
step 1.1: selecting a remote sensing image for training a building height prediction model, wherein the remote sensing image selects Jilin one-size broad 01-star shooting data;
step 1.2: cutting data, namely cutting the selected complete data to obtain 363 multiplied by 263 pixels, and filling the cut part with the size smaller than 363 multiplied by 263 with a value of 0;
step 1.3: dividing the training set and the verification set, obtaining 4751 images after cutting, and according to the training set and the verification set 3:1, 3565 sheets were selected as training set and 1186 sheets were selected as validation set.
3. The method according to claim 2, characterized in that: during training, the super parameters are set as follows: initial learning rate learning_rate=0.001, batch size batch_size=4, training algebra epochs =100000, segmentation class n_ classes =41.
4. A method according to claim 3, characterized in that: the step 4 specifically comprises the following steps:
Step 4.1: slicing the image for test into an image with the size of 363×263, and filling the cut part with a value of 0 for the part with the size of less than 363×263;
step 4.2: building height predictions are made for each cropped image using a trained RETINANET model.
5. End-to-end building height prediction system based on single vision remote sensing image, its characterized in that: the system comprises:
and (3) a data set making module: the method is used for manufacturing a building height prediction data set based on the high-resolution optical remote sensing image;
model improvement module: for improving RETINANET model, adding downsampling layer in FPN layer and using OHEM loss function and Fast-NMS optimizing model in prediction part;
The model improvement module specifically comprises:
Performing downsampling on an original FPN layer in a RETINANET model twice;
replacing FOCAL LOSS loss functions with OHEM functions at the class prediction portion of the RETINANET model;
Adding Fast-NMS in boundary box prediction part of RETINANET model;
model training module: for training the improvement RETINANET model using the building height prediction dataset;
And a prediction module: the method is used for predicting the building height of the remote sensing image containing the building by using the RETINANET model after training.
6. An electronic device, the electronic device comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the single vision remote sensing image based end-to-end building height prediction method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111304222.9A CN113902793B (en) | 2021-11-05 | 2021-11-05 | Method, system and electronic equipment for predicting end-to-end building height based on single-vision remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111304222.9A CN113902793B (en) | 2021-11-05 | 2021-11-05 | Method, system and electronic equipment for predicting end-to-end building height based on single-vision remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113902793A CN113902793A (en) | 2022-01-07 |
| CN113902793B true CN113902793B (en) | 2024-05-14 |
Family
ID=79193429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111304222.9A Active CN113902793B (en) | 2021-11-05 | 2021-11-05 | Method, system and electronic equipment for predicting end-to-end building height based on single-vision remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113902793B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115859765B (en) * | 2022-09-29 | 2023-12-08 | 中山大学 | Urban expansion prediction method, device, equipment and storage medium |
| CN116503464B (en) * | 2023-06-25 | 2023-10-03 | 武汉理工大学三亚科教创新园 | Farmland building height prediction method based on remote sensing image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111709387A (en) * | 2020-06-22 | 2020-09-25 | 中国科学院空天信息创新研究院 | Building segmentation method and system for high-resolution remote sensing image |
| WO2021184891A1 (en) * | 2020-03-20 | 2021-09-23 | 中国科学院深圳先进技术研究院 | Remotely-sensed image-based terrain classification method, and system |
| CN113468993A (en) * | 2021-06-21 | 2021-10-01 | 天津大学 | Remote sensing image target detection method based on deep learning |
-
2021
- 2021-11-05 CN CN202111304222.9A patent/CN113902793B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021184891A1 (en) * | 2020-03-20 | 2021-09-23 | 中国科学院深圳先进技术研究院 | Remotely-sensed image-based terrain classification method, and system |
| CN111709387A (en) * | 2020-06-22 | 2020-09-25 | 中国科学院空天信息创新研究院 | Building segmentation method and system for high-resolution remote sensing image |
| CN113468993A (en) * | 2021-06-21 | 2021-10-01 | 天津大学 | Remote sensing image target detection method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 一种高分辨率遥感影像建筑物自动检测方法;张通;潘励;;测绘地理信息;20191122;第45卷(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113902793A (en) | 2022-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111553303B (en) | Remote sensing orthographic image dense building extraction method based on convolutional neural network | |
| Yang et al. | Deep learning‐based bolt loosening detection for wind turbine towers | |
| CN110929577A (en) | Improved target identification method based on YOLOv3 lightweight framework | |
| CN114049356B (en) | Method, device and system for detecting structure apparent crack | |
| CN113902793B (en) | Method, system and electronic equipment for predicting end-to-end building height based on single-vision remote sensing image | |
| CN111539355A (en) | Photovoltaic panel foreign matter detection system and detection method based on deep neural network | |
| CN111598942A (en) | Method and system for automatically positioning electric power facility instrument | |
| CN111932511A (en) | Electronic component quality detection method and system based on deep learning | |
| CN113902792B (en) | Building height detection method, system and electronic equipment based on improved RETINANET network | |
| Huang et al. | DEEP LEARNING-BASED DETECTION AND POSITIONING OF REBAR TYING NODES FOR AUTOMATED STRUCTURAL INTEGRITY | |
| CN110837839B (en) | High-precision unmanned aerial vehicle orthographic image manufacturing and data acquisition method | |
| CN115424237A (en) | Forward vehicle identification and distance detection method based on deep learning | |
| CN114463737A (en) | 3D target detection method and system based on implicit expression in 3D modeling | |
| CN117456198A (en) | Power transmission line fault detection method based on improved Yolov5 neural network | |
| CN114612803B (en) | Improved CENTERNET transmission line insulator defect detection method | |
| CN117237677B (en) | Precipitation prediction correction method for overall similarity of strong precipitation space based on deep learning | |
| CN116205918B (en) | Multi-mode fusion semiconductor detection method, device and medium based on graph convolution | |
| Jia et al. | Research on Industrial Production Defect Detection Method Based on Machine Vision Technology in Industrial Internet of Things. | |
| CN116912628A (en) | Method and device for training defect detection model and detecting defects | |
| Tan et al. | LiDAR-Based Automatic Pavement Distress Detection and Management Using Deep Learning and BIM | |
| CN116503765A (en) | Engineering scale wave identification method based on machine vision | |
| CN113947774B (en) | Lightweight vehicle target detection system | |
| CN118818222B (en) | Power grid space position analysis method combining GIS service and artificial intelligence technology | |
| Liu et al. | MPGI‐Terminal defect detection based on M‐FRCNN | |
| Chuang et al. | Bridge Crack Detection Based on Attention Mechanism. |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: No. 1299, Mingxi Road, Beihu science and Technology Development Zone, Changchun City, Jilin Province Applicant after: Changguang Satellite Technology Co.,Ltd. Address before: No. 1299, Mingxi Road, Beihu science and Technology Development Zone, Changchun City, Jilin Province Applicant before: CHANG GUANG SATELLITE TECHNOLOGY Co.,Ltd. Country or region before: China |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |