CN114545959A - Remote sensing satellite platform control based on flutter information and image correction method thereof - Google Patents
Remote sensing satellite platform control based on flutter information and image correction method thereof Download PDFInfo
- Publication number
- CN114545959A CN114545959A CN202210172160.9A CN202210172160A CN114545959A CN 114545959 A CN114545959 A CN 114545959A CN 202210172160 A CN202210172160 A CN 202210172160A CN 114545959 A CN114545959 A CN 114545959A
- Authority
- CN
- China
- Prior art keywords
- flutter
- image
- information
- satellite
- flutter information
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000003702 image correction Methods 0.000 title claims abstract description 18
- 230000033001 locomotion Effects 0.000 claims abstract description 20
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012937 correction Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 6
- 238000013136 deep learning model Methods 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 4
- 230000002708 enhancing effect Effects 0.000 abstract description 3
- 238000004088 simulation Methods 0.000 description 7
- 238000003384 imaging method Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/08—Control of attitude, i.e. control of roll, pitch, or yaw
- G05D1/0808—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft
-
- G06T5/80—
-
- 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]
Abstract
A remote sensing satellite platform control and image correction method based on flutter information. The method comprises the following processing flows: simulating satellite motion and constructing a flutter image data set, enhancing the data volume through data, preprocessing the flutter image data, extracting image flutter information by using a ResNet algorithm, feeding the flutter information back to a satellite attitude control system for control, and guiding a Transformer model to carry out image correction by using the flutter information. Aiming at the technical problem that space tasks such as high-resolution reconnaissance cannot be imaged clearly due to the restriction of high-frequency low-amplitude micro-vibration of a spacecraft, satellite flutter information is resolved by utilizing flutter images, and the flutter information is used for controlling a satellite platform and guiding image correction network learning, so that a satellite flutter estimation, attitude control feedback and image correction closed-loop system is set up. The invention provides an intelligent method for estimating satellite flutter information, feeding back attitude control information and correcting flutter images based on a Transformer model based on ResNet algorithm, which effectively improves the intelligent processing capability of spatial information in China.
Description
Technical Field
The invention relates to the field of target identification, in particular to a remote sensing satellite platform control based on flutter information and an image correction method thereof.
Background
In a space mission, the remote sensing satellite has extremely high military and commercial values for earth observation, however, the influence of high-frequency low-amplitude micro vibration of a spacecraft can cause the remote sensing satellite to be incapable of imaging clearly, so that the use effect of a remote sensing image is severely restricted. The satellite flutter is high-frequency low-amplitude micro-vibration which is difficult to detect and correct, satellite flutter information is effectively estimated, a satellite platform is controlled according to the flutter information, and finally a distorted remote sensing image is corrected by using a related algorithm, so that the influence of the satellite flutter on the remote sensing image can be effectively solved. The remote sensing satellite platform control and image correction method based on flutter information improves the intelligent processing capacity of the spatial information in China and provides theoretical and technical basis for promoting the independent intelligent development of space equipment.
The traditional remote sensing satellite image processing directly corrects images, cannot effectively combine causes causing image distortion, combines satellite flutter information, and can deeply analyze distorted remote sensing images so as to better correct the images.
The patent number of the invention publication CN201711173663.3 discloses a satellite platform flutter detection method based on TDICCD splicing area images, which estimates satellite platform flutter information by calculating the relative imaging position difference of the same target in two overlapped images, but fails to integrate a satellite platform control and image correction method. The patent number CN202110264265.2 discloses a method for detecting and modeling flutter of a high-resolution remote sensing satellite cooperating with multi-load data, adopts the multi-load data and uses an adaptive Bayes algorithm to model the flutter of a satellite platform, and the method has higher difficulty in implementation and operation.
Disclosure of Invention
Objects of the invention
The invention aims to provide a remote sensing satellite platform control method based on flutter information and an image correction method thereof. The method forms closed-loop feedback by estimating the flutter information of the satellite platform and controlling the satellite platform, and further corrects the distorted image shot by the satellite platform which reaches the stable index.
(II) technical scheme
The technical scheme of the invention is that a remote sensing satellite platform control and image correction method based on flutter information is adopted, and the method comprises the following steps: simulating satellite motion and constructing a flutter image data set, enhancing data volume through data, preprocessing the flutter image data, extracting image flutter information by using a ResNet algorithm, feeding the flutter information back to a satellite attitude control system for control, and guiding a Transformer model to perform image correction by using the flutter information.
Simulating satellite motion and constructing a flutter image data set, and using a spacecraft attitude control full-physical simulation system to realize full-physical simulation of the spacecraft attitude control system, wherein the system simulates pitching and deflecting motions of a satellite platform to complete three-degree-of-freedom motion. The method includes the steps that a vibration generator is used for simulating the flutter phenomenon of a satellite, an optical camera is used for shooting fixed images in the motion process of the satellite, and the attitude information of the triaxial air floating platform during shooting is recorded. And matching the image with the attitude information of the triaxial air-floating platform to construct a data set.
The data enhancement module is used for expanding the data volume; and the flutter image data preprocessing module is used for preprocessing image information and facilitating input into the deep learning model for training.
And the image flutter information estimation module extracts image flutter information by using a deep learning algorithm, inputs the preprocessed flutter picture into a ResNet network, extracts the image flutter information by using a deep ResNet network, and outputs motion information in each direction through a full connection layer.
And the satellite attitude control module is used for inputting the flutter information solved by the image flutter information estimation module into a satellite attitude control system to form a feedback link and carry out satellite attitude correction.
And the image correction module is used for guiding the deep learning model to correct the flutter image, inputting the flutter information finally solved in the image flutter information estimation module into the Transformer model as priori knowledge, firstly vectorizing and coding the flutter information solved by the image flutter information estimation module, splicing the vectorized and coded flutter information on a position encoder of the Transformer, and training through a gradient descent algorithm to finish the final correction of the flutter image.
The simulated satellite motion, image flutter information estimation module and the satellite attitude control module form a closed-loop system, the image flutter information estimation module carries out flutter information estimation on a satellite image, and feeds the result back to the satellite attitude control module to carry out simulated satellite attitude correction, shooting is carried out again after the attitude correction is finished, the image flutter information estimation module carries out flutter information estimation on the image shot again, and the simulated satellite motion, the image flutter information estimation and the satellite attitude control are repeated until an image evaluation index reaches a set threshold value.
(III) major advantages of the invention
The technical scheme of the invention has the following beneficial technical effects: the method is used for solving the problems of flutter and clear imaging of the satellite platform, constructing closed loop feedback consisting of an image flutter information estimation module and a satellite attitude control module, and further correcting the distorted image shot by the satellite platform which reaches a stable index.
Drawings
FIG. 1 is a flow diagram of the present invention;
FIG. 2 is a flow chart of a ResNet network in the present invention;
FIG. 3 is a flow chart of the Transformer model of the present invention;
Detailed Description
In order to make the technical scheme, advantages and purposes of the invention clearer, the technical scheme of the invention is further explained by combining a method flow described by a specific example and referring to the attached drawings.
The embodiment 1 of the invention relates to a remote sensing satellite platform control based on flutter information and an image correction method thereof, which are shown in figure 1 and are carried out according to the following steps:
the simulation system simulates the motion of a satellite and constructs a flutter image data set, uses a spacecraft attitude control full-physical simulation system to realize full-physical simulation of the spacecraft attitude control system, realizes the real simulation of the attitude dynamics of the spacecraft in a weightless state based on a three-axis air floatation platform, and completes the simulation of the motion of the spacecraft by the complete integration of real hardware components of the attitude control system. The system is used for simulating the motion of a satellite platform, a vibration generator is installed on a three-axis air floatation platform, and an optical camera is fixedly connected to the vibration generator. The three-axis air floating platform simulates the pitching and the deflecting motion of a satellite to complete the three-degree-of-freedom motion. The vibration generator simulates the flutter phenomenon of a satellite, an optical camera is used for photographing a fixed image in the motion process of the simulated satellite, and the attitude information of the triaxial air floating platform during photographing is recorded. And matching the image with the attitude information of the triaxial air-floating platform, constructing a data set, and dividing the data set into a training set and a test set according to a ratio of 4: 1.
And the data enhancement module is used for enhancing the data of the image in the data set, and can adopt methods of adding white Gaussian noise, average blurring, color interference and the like for expanding the data volume.
And the flutter image data preprocessing module is used for preprocessing image information and facilitating input into the deep learning model for training.
The image flutter information estimation module extracts image flutter information by using a deep learning algorithm, a preprocessed flutter picture is input into a ResNet network, the image flutter information is extracted by using the deep ResNet network, the ResNet network is connected by using residual errors, and the residual error connection function is as follows:
xl+1=xl+F(xl,Wl) (1)
wherein x isl+1And xlRepresenting two connected layers of a feature matrix, F (x)l,Wl) And (3) building a 36-layer ResNet module through a connected two-layer characteristic convolution operation process, and finally outputting motion information in each direction by using a full-connection layer.
And the satellite attitude control module is used for inputting the flutter information solved by the image flutter information estimation module into a satellite attitude control system, forming a feedback link by using a PID control method and correcting the satellite attitude.
And the image correction module is used for guiding the deep learning model to correct the flutter image, inputting the flutter information finally solved in the image flutter information estimation module into the Transformer model as priori knowledge, firstly vectorizing and coding the flutter information solved by the image flutter information estimation module, splicing the vectorized and coded flutter information on a position encoder of the Transformer, and training through a gradient descent algorithm to finish the final correction of the flutter image.
In the steps, a closed-loop system is formed by the simulated satellite motion and image flutter information estimation module and the satellite attitude control module, the image flutter information estimation module carries out flutter information estimation on a satellite image, the result is fed back to the satellite attitude control module to carry out simulated satellite attitude correction, shooting is carried out again after the attitude correction is finished, the image flutter information estimation module carries out flutter information estimation on the image shot again, and the simulated satellite motion, the image flutter information estimation and the satellite attitude control are repeated until an image evaluation index reaches a set threshold value.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.
Claims (4)
1. A remote sensing satellite platform control and image correction method based on flutter information is characterized by mainly comprising the following steps:
step 1, simulating satellite motion and constructing a flutter image data set;
step 2, a data enhancement module for expanding data volume;
step 3, a flutter image data preprocessing module;
step 4, an image flutter information estimation module extracts image flutter information by using a deep learning algorithm;
step 5, the satellite attitude control module feeds the flutter information back to the satellite attitude control system for control;
step 6, an image correction module guides the deep learning model to correct the flutter image;
the method comprises the following steps of 1, 4 and 5, wherein a closed-loop system is formed by the steps 1, 4, flutter information estimation is carried out on a satellite image, a result is fed back to a satellite attitude control module in the step 5 to carry out simulated satellite attitude correction of the step 1, shooting is carried out again after attitude correction is finished, an image flutter information estimation module in the step 4 carries out flutter information estimation on the image shot again, and the steps 1, 4 and 5 are repeated until an image evaluation index reaches a set threshold value.
2. The method according to claim 1, wherein in step 4, the method comprises:
and the image flutter information estimation module is used for inputting the flutter picture preprocessed in the step 3 into a ResNet network, extracting image flutter information by using a deep ResNet network, and outputting motion information in each direction through a full connection layer.
3. The method according to claim 1, wherein in step 5, the method comprises:
and the satellite attitude control module is used for inputting the flutter information obtained by the calculation in the step 4 into a satellite attitude control system to form a feedback link and carrying out satellite attitude correction.
4. The method according to claim 1, wherein in step 6, the method comprises:
and the image correction module is used for inputting the flutter information finally solved in the step 4 into the Transformer model as priori knowledge, firstly vectorizing and coding the flutter information solved in the step 4, splicing the vectorized and coded flutter information on a position encoder of the Transformer, and training through a gradient descent algorithm to finish the final correction of the flutter image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210172160.9A CN114545959B (en) | 2022-02-24 | 2022-02-24 | Remote sensing satellite platform control and image correction method based on flutter information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210172160.9A CN114545959B (en) | 2022-02-24 | 2022-02-24 | Remote sensing satellite platform control and image correction method based on flutter information |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114545959A true CN114545959A (en) | 2022-05-27 |
CN114545959B CN114545959B (en) | 2024-01-12 |
Family
ID=81678465
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210172160.9A Active CN114545959B (en) | 2022-02-24 | 2022-02-24 | Remote sensing satellite platform control and image correction method based on flutter information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114545959B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003284074A (en) * | 2002-03-25 | 2003-10-03 | Toshiba Corp | Apparatus and method for performing changeover processing of digital broadcast signal |
CN103093186A (en) * | 2011-07-28 | 2013-05-08 | 施乐公司 | Systems and methods for improving image recognition |
CN103438865A (en) * | 2013-08-15 | 2013-12-11 | 航天东方红卫星有限公司 | Obtaining method of optical remote sensing satellite image in-scene relative distortion indexes |
CN103791899A (en) * | 2014-02-14 | 2014-05-14 | 同济大学 | Satellite attitude fluttering detection method based on imaging sensor parallax error |
CN104864852A (en) * | 2015-05-08 | 2015-08-26 | 同济大学 | High resolution satellite attitude fluttering detection method based on intensive control points |
CN105136164A (en) * | 2015-08-13 | 2015-12-09 | 航天恒星科技有限公司 | Staring imaging simulation and quality evaluation method and device taking regard of satellite comprehensive motion |
CN109901166A (en) * | 2019-03-28 | 2019-06-18 | 中国人民解放军战略支援部队航天工程大学 | A method of it quickly generates and simulation ISAR echo data |
CN112597428A (en) * | 2020-12-22 | 2021-04-02 | 同济大学 | Flutter detection correction method based on beam adjustment and image resampling of RFM model |
CN113065277A (en) * | 2021-03-11 | 2021-07-02 | 自然资源部国土卫星遥感应用中心 | Method for high-resolution remote sensing satellite flutter detection and modeling in cooperation with multi-load data |
CN113483879A (en) * | 2021-06-28 | 2021-10-08 | 同济大学 | Small satellite flutter high-speed video measurement method |
-
2022
- 2022-02-24 CN CN202210172160.9A patent/CN114545959B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003284074A (en) * | 2002-03-25 | 2003-10-03 | Toshiba Corp | Apparatus and method for performing changeover processing of digital broadcast signal |
CN103093186A (en) * | 2011-07-28 | 2013-05-08 | 施乐公司 | Systems and methods for improving image recognition |
CN103438865A (en) * | 2013-08-15 | 2013-12-11 | 航天东方红卫星有限公司 | Obtaining method of optical remote sensing satellite image in-scene relative distortion indexes |
CN103791899A (en) * | 2014-02-14 | 2014-05-14 | 同济大学 | Satellite attitude fluttering detection method based on imaging sensor parallax error |
CN104864852A (en) * | 2015-05-08 | 2015-08-26 | 同济大学 | High resolution satellite attitude fluttering detection method based on intensive control points |
CN105136164A (en) * | 2015-08-13 | 2015-12-09 | 航天恒星科技有限公司 | Staring imaging simulation and quality evaluation method and device taking regard of satellite comprehensive motion |
CN109901166A (en) * | 2019-03-28 | 2019-06-18 | 中国人民解放军战略支援部队航天工程大学 | A method of it quickly generates and simulation ISAR echo data |
CN112597428A (en) * | 2020-12-22 | 2021-04-02 | 同济大学 | Flutter detection correction method based on beam adjustment and image resampling of RFM model |
CN113065277A (en) * | 2021-03-11 | 2021-07-02 | 自然资源部国土卫星遥感应用中心 | Method for high-resolution remote sensing satellite flutter detection and modeling in cooperation with multi-load data |
CN113483879A (en) * | 2021-06-28 | 2021-10-08 | 同济大学 | Small satellite flutter high-speed video measurement method |
Non-Patent Citations (5)
Title |
---|
RUIDA YE, 等: "Bearing Fault Detection Based on Convolutional Self-Attention Mechanism", ICCASIT, pages 869 - 876 * |
ZHU SU,等: "Vibration characteristic and flutter analysis of elastically restrained stiffened functionally graded plates in thermal environment", INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES, pages 872 * |
刘彦丽, 曹东晶: "高分辨率敏捷卫星颤振对成像的影响分析方法", 航天返回与遥感 第35 卷 第2 期 46 SPACECRAFT RECOVERY & REMOTE, vol. 35, no. 2, pages 46 - 53 * |
赵薇薇,等: "线加速度计在轨检测卫星影像颤振的可行性分析", 航天返回与遥感, vol. 39, no. 5, pages 49 - 56 * |
闰利,胡晓斌: "基于MIF滤波的卫星颤振图像恢复研究", 遥感信息, vol. 28, no. 5, pages 12 - 17 * |
Also Published As
Publication number | Publication date |
---|---|
CN114545959B (en) | 2024-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102722697B (en) | Unmanned aerial vehicle autonomous navigation landing visual target tracking method | |
CN109509230A (en) | A kind of SLAM method applied to more camera lens combined type panorama cameras | |
CN106780543A (en) | A kind of double framework estimating depths and movement technique based on convolutional neural networks | |
CN110070025B (en) | Monocular image-based three-dimensional target detection system and method | |
CN111340868B (en) | Unmanned underwater vehicle autonomous decision control method based on visual depth estimation | |
CN112434655A (en) | Gait recognition method based on adaptive confidence map convolution network | |
CN111563878A (en) | Space target positioning method | |
CN113392584B (en) | Visual navigation method based on deep reinforcement learning and direction estimation | |
CN106875403B (en) | A kind of imitative hawkeye visual movement object detection method for air refuelling | |
CN105100546A (en) | Movement estimation method and device | |
CN112633220A (en) | Human body posture estimation method based on bidirectional serialization modeling | |
CN114283325A (en) | Underwater target identification method based on knowledge distillation | |
CN104504691B (en) | Camera position and posture measuring method on basis of low-rank textures | |
CN116434088A (en) | Lane line detection and lane auxiliary keeping method based on unmanned aerial vehicle aerial image | |
CN106097277B (en) | A kind of rope substance point-tracking method that view-based access control model measures | |
CN117252928B (en) | Visual image positioning system for modular intelligent assembly of electronic products | |
CN114545959A (en) | Remote sensing satellite platform control based on flutter information and image correction method thereof | |
CN117115252A (en) | Bionic ornithopter space pose estimation method based on vision | |
Caruso et al. | 3d reconstruction of non-cooperative resident space objects using instant ngp-accelerated nerf and d-nerf | |
CN112975968B (en) | Mechanical arm imitation learning method based on third visual angle variable main body demonstration video | |
CN115984443A (en) | Space satellite target image simulation method of visible light camera | |
CN104978721A (en) | Atmospheric disturbance image recovering method based on variation regularization | |
CN106920249A (en) | The fast track method of space maneuver target | |
CN114219901A (en) | Three-dimensional chassis projection method based on projection consistency and twin transform | |
CN111680552B (en) | Feature part intelligent recognition method |
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 |