CN115511011B - Radar data correction method and system based on countermeasure generation network model - Google Patents
Radar data correction method and system based on countermeasure generation network model Download PDFInfo
- Publication number
- CN115511011B CN115511011B CN202211454168.0A CN202211454168A CN115511011B CN 115511011 B CN115511011 B CN 115511011B CN 202211454168 A CN202211454168 A CN 202211454168A CN 115511011 B CN115511011 B CN 115511011B
- Authority
- CN
- China
- Prior art keywords
- mask
- image
- corrected
- radar
- radar data
- 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
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/95—Lidar systems specially adapted for specific applications for meteorological use
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a radar data correction method and a system based on an countermeasure generation network model, which belong to the field of radar detection, and the radar data correction method based on the countermeasure generation network model comprises the following steps: inputting radar data to be corrected, wherein the radar data to be corrected is a radial wind speed field measured by a laser wind measuring radar; reconstructing the data to be corrected by adopting a gray conversion mode; performing shielding judgment on the reconstructed data by adopting a shielding detection program, and performing shielding mask processing on the data to be corrected according to the obtained shielding judgment result to obtain a shielded image and a mask; judging whether the shielding degree of the mask is within a preset threshold range or not; inputting the shielded image and the mask into a countermeasure generation network model obtained by training; the method and the device can output the corrected radar data by combining the correction model, and accurately provide various meteorological elements for atmospheric scientific research in time.
Description
Technical Field
The invention relates to the field of radar detection, in particular to a radar data correction method and system based on an antagonism generation network model.
Background
Due to the propagation characteristics of electromagnetic waves and the blocking of hard targets such as tall buildings and the like near the radar, the radar often has the phenomenon of beam blocking in actual detection, so that the radar data quality is low. In particular, in the case of radars arranged in urban areas, echo data are more susceptible to beam blocking, and even a slight degree of blocking, electromagnetic waves emitted by the radars cannot propagate forward completely, so that echoes are weaker or blocked completely, and the accuracy of radar data is affected. The development of radar detection technology has become one of the most important remote sensing technologies so far, is an indispensable existence in the field of atmospheric science, and can obtain the space-time distribution situation of various meteorological elements by means of the radar detection technology, thereby making great contribution to solving the problems in the field of atmospheric science. How to solve the beam blocking phenomenon in the radar detection process is a key difficult problem for improving the radar detection technology.
Currently, the correction for radar beam blocking mainly depends on three modes of a digital elevation model (Digital Elevation Model, DEM), an echo probability feature-based recognition algorithm and a spatial correlation-based beam blocking recognition algorithm. The inventors of the present invention have found through studies that: the digital elevation model can carry out echo correction on the area with beam blocking according to actual mapping data, but the method has certain limitation, because new buildings are continuously appeared along with social and economic development and urban construction, the beam blocking can be caused, and the mapping data are difficult to update in real time. Therefore, the beam blocking correction scheme based on the digital elevation model has difficulty in ensuring correction accuracy. The beam blocking recognition algorithm based on the spatial correlation does not need digital elevation data, is not influenced by the terrain and atmospheric refraction conditions, can better recognize and correct the beam blocking, but when the condition of large-range beam blocking occurs, the adjacent echo signals cannot be utilized, because the strong correlation of the radar echo large-span space cannot be ensured, and when zero-degree layer bright bands appear and are used for correcting, the condition of overhigh correction amount can be caused.
Disclosure of Invention
1. Technical problem to be solved
Aiming at the problems existing in the prior art, the invention aims to provide a radar data correction method and a system based on an antagonism generation network model, which can output corrected radar data according to the radar data and the correction model, and accurately provide various meteorological elements for atmospheric scientific research in time.
2. Technical proposal
In order to solve the problems, the invention adopts the following technical scheme.
A radar profile correction method based on a challenge-generating network model, comprising the steps of:
detecting a shielding portion based on the radar data;
training a countermeasure generation network model according to radar data; the countermeasure generation network model corrects radar data;
selecting preset parameters to classify radar data, and establishing a multi-radar data correction model;
acquiring real-time radar data;
matching the radar data with the established multi-radar data correction system to obtain a matched correction system; the beam blocking of the real-time radar data is identified and corrected by the correction system.
And inverting according to the corrected radar data to obtain real-time meteorological elements and outputting the real-time meteorological elements.
Further, detecting the masking portion from the radar material includes: converting radar data into a gray image, judging whether the value of each pixel point is within a preset threshold range, and if the value is outside the threshold range, outputting the value as a mask, wherein the point is a shielding area; if the value is within the threshold range, the point is an effective area and the image is outputted as a broken image.
Further, training the challenge-generating network model based on the radar data, comprising: and selecting a certain amount of radar data to extract masks in batches, judging whether the shielding degree of the masks is within a preset threshold range, selecting the masks within the threshold range, and classifying the masks according to different shielding degrees of the masks to be used as a mask set. And selecting a certain amount of radar data, judging whether the shielding degree of a mask is within a preset threshold range, and selecting the radar data within the threshold range as a background set. The mask set and the background set are respectively divided into a training set, a verification set and a test set of the mask and a training set, a verification set and a test set of the background according to preset proportion. Training the challenge-generating network model using the data, the trained model including an edge reconstruction network and an image restoration network. The edge reconstruction network reconstructs the edge information of the missing region according to the input radar data and the mask information, and the image restoration network reconstructs the image of the missing region under the adjustment of the edge information to obtain a restored image.
Further, selecting a predetermined parameter to classify the radar data, and establishing a multi-radar data correction model, comprising: training correction models applicable to different data according to different radar detection data; and inputting the radar data to be corrected into a correction system, classifying and judging the radar data to be corrected by adopting a probability algorithm, and selecting a proper correction model.
Further, acquiring real-time radar data includes: communication is established with the output end of the radar system, and the radar data output from the output end is received in real time.
Further, the radar data is matched with the established multi-radar data correction system, and a matched correction system is obtained; the beam blocking of the real-time radar data is identified and corrected by the correction system.
3. Advantageous effects
Compared with the prior art, the invention has the advantages that: according to the radar data, the corrected radar data can be output by combining with the correction model, and various meteorological elements can be accurately provided for atmospheric scientific research in time.
Drawings
FIG. 1 is a flowchart of a method for correcting radar data based on an countermeasure generation network model according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a method for establishing a multi-radar data correction according to an embodiment of the present invention;
FIG. 3 is a training flow chart of an countermeasure generation network model provided by an embodiment of the present invention;
FIG. 4 is a flow chart of a radar data correction system based on an countermeasure generation network model according to an embodiment of the present invention;
FIG. 5 is an image of a radial wind velocity field to be corrected provided by an embodiment of the present invention;
FIG. 6 is a radial wind velocity field image reconstructed after gray level conversion provided by an embodiment of the present invention;
FIG. 7 is a diagram of an image after masking and masking processes provided by an embodiment of the present invention;
FIG. 8 is an image of a corrected radial wind velocity field provided by an embodiment of the present invention.
Detailed Description
The drawings in the embodiments of the present invention will be combined; the technical scheme in the embodiment of the invention is clearly and completely described; obviously; the described embodiments are only a few embodiments of the present invention; but not all embodiments, are based on embodiments in the present invention; all other embodiments obtained by those skilled in the art without undue burden; all falling within the scope of the present invention.
In the description of the present invention, it should be noted that the positional or positional relationship indicated by the terms such as "upper", "lower", "inner", "outer", "top/bottom", etc. are based on the positional or positional relationship shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "configured to," "engaged with," "connected to," and the like are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
Example 1:
referring to fig. 1, a radar data correction method based on an countermeasure generation network model includes the following steps:
s1, inputting radar data to be corrected;
the radar data to be corrected in this embodiment refers to a radial wind speed field measured by the laser wind-finding radar, and is input into a radial wind speed field to be corrected in RGB color of 256×256×3.
S2, reconstructing the data to be corrected by adopting a gray conversion mode, wherein a radial wind speed field image obtained by reconstruction after gray conversion is shown in FIG. 6;
the method specifically comprises the following steps: converting the radial wind speed field diagram into a gray image, then sequentially judging whether the value of each pixel point is within a preset threshold range, if the value is outside the threshold range, blocking the beam of the point, setting the pixel value of the point as 255, and taking the pixel value as a shielding area; if the value is within the threshold range, the spot is not blocked and the spot pixel value is set to 0, which is the active area.
S3, performing shielding judgment on the reconstructed data by adopting a shielding detection program, and performing shielding mask processing on the data to be corrected according to the obtained shielding judgment result to obtain a shielded image and a mask, as shown in FIG. 7;
in one embodiment, the masking detection is performed by using the reconstructed gray-scale image in S2, and if the pixel value of a certain point is 255, the point is marked as a mask; otherwise, the effective area is marked, and after the shielding detection is completed, a mask is generated and an image after shielding is output respectively.
S4, judging whether the shielding degree of the mask is within a preset threshold range;
calculating the shielding degree (the ratio of the size of the mask to the size of the radial wind speed field to be corrected) by adopting the mask generated in the step S3, wherein the image of the radial wind speed field to be corrected is shown in fig. 5, and if the shielding degree is within a preset threshold range, the input data to be corrected is available according to the current input, and the input data to be corrected is input into the step S5; otherwise, the correction is ended.
S5, inputting the shielded image and the mask into a countermeasure generation network model obtained through training;
in one embodiment, the mask and masked image generated in S4 are input into a trained countermeasure generation network model, which includes an edge reconstruction network and an image restoration network. The edge reconstruction network can reconstruct the edge of the missing part of the image according to the input edge information of the shaded image. Then the image restoration network restores the image of the missing area under the adjustment of the edge information, so that the restoration effect is better.
S6, performing a radar data correction process, and outputting a correction result, as shown in FIG. 8.
In one embodiment, the radial wind speed field shielded by the mask is repaired by using the countermeasure generation network model, so that the reconstructed radial wind speed field is obtained, and the radar data correction is completed.
The countermeasure generation network model is mainly a technology for restoring missing parts in an image based on the input existing information of the radial wind speed field to be corrected. Therefore, after the mask and the masked image are obtained by the countermeasure generation network model, the image can be repaired by using the countermeasure generation network model, so that the reconstructed radial wind speed field is obtained.
The radar data correction system based on the countermeasure generation network model comprises a gray conversion module, a shielding detection module, a probability algorithm module, a correction model selection module and a radar data correction module, as shown in fig. 4.
As shown in fig. 3, the training process of the countermeasure generation network model is as follows:
s1, extracting a corresponding mask by utilizing radar data to generate a mask set, selecting more complete radar data as a background field and generating a background set;
in this embodiment, the radar data refers to a radial wind speed field measured by the laser wind-finding radar. And detecting the mask by using a shielding detection module, extracting a corresponding mask, and generating a mask set. The shielding detection module screens the radar data according to a preset threshold value, and if the radar data is within the threshold value range, the radar data is selected as a background field and a background set is generated.
S2, inputting a mask set and a background field into a countermeasure generation network model, performing shielding treatment on the background field by using the mask set by the model to generate a damaged image and a missing region image, wherein the missing region image is a shielding part of a mask in the background, and then obtaining a repair image by using the model;
s3, determining a loss value of the countermeasure generation network model according to the background, the repair image, the mask and the missing region image;
and S4, if the loss value is higher than a preset threshold value, updating the model.
Example 2:
referring to fig. 2, a radar data correction method based on an countermeasure generation network model includes the following steps:
s1, inputting radar data to be corrected, wherein the radar data to be corrected is a radial wind speed field measured by a laser anemometer radar;
s2, classifying and judging the radar data to be corrected by adopting a maximum probability algorithm, and selecting a proper correction model;
s3, reconstructing the data to be corrected by adopting a gray level conversion mode;
s4, performing shielding judgment on the reconstructed data by adopting a shielding detection program, and performing shielding mask processing on the data to be corrected according to the obtained shielding judgment result to obtain a shielded image and a mask;
s5, judging whether the shielding degree of the mask is within a preset threshold range;
s6, inputting the shielded image and the mask into a countermeasure generation network model obtained through training;
s7, performing a radar data correction process and outputting a correction result.
The above; is only a preferred embodiment of the present invention; the scope of the invention is not limited in this respect; any person skilled in the art is within the technical scope of the present disclosure; equivalent substitutions or changes are made according to the technical proposal of the invention and the improved conception thereof; are intended to be encompassed within the scope of the present invention.
Claims (4)
1. A radar data correction method based on a countermeasure generation network model is characterized in that: the method comprises the following steps:
s1, inputting radar data to be corrected, wherein the radar data to be corrected is a radial wind speed field measured by a laser anemometer radar, and the radar data to be corrected comprises a radial wind speed field to be corrected of RGB color which is input into 256 multiplied by 3;
s2, reconstructing the data to be corrected by adopting a gray level conversion mode;
the specific method comprises the following steps: converting the radial wind speed field map into a gray level image, then sequentially judging whether the value of each pixel point is within a preset threshold range, and if the value is outside the threshold range, blocking a beam according to the point and setting the pixel value of the point as 255 to be a shielding area; when the value is within the threshold range, the point is not blocked by the beam and the pixel value of the point is set to be 0, and the point is an effective area;
s3, performing shielding judgment on the reconstructed data by adopting a shielding detection program, and performing shielding mask processing on the data to be corrected according to the obtained shielding judgment result to obtain a shielded image and a mask;
performing shielding detection by adopting the reconstructed gray level image obtained in the step S2, and marking a certain point as a mask when the pixel value of the certain point is 255; otherwise, marking the mask as an effective area, and respectively generating a mask and outputting a masked image after the masking detection is completed;
s4, judging whether the shielding degree of the mask is within a preset threshold range; calculating the shielding degree by adopting the mask generated in the step S3, and inputting the shielding degree into the step S5 according to the availability of the input data to be corrected when the shielding degree is within a preset threshold range; otherwise, ending the correction;
s5, inputting the shielded image and the mask into a countermeasure generation network model obtained through training, wherein the countermeasure generation network model comprises an edge reconstruction network and an image restoration network, the edge reconstruction network reconstructs the edge of the missing part of the image according to the input edge information of the shielded image, and the image of the missing region is restored by the post-image restoration network under the regulation of the edge information;
s6, performing a radar data correction process and outputting a correction result.
2. A method of radar profile correction based on an countermeasure generation network model as claimed in claim 1, wherein: the training process of the countermeasure generation network model comprises the following steps:
s1, extracting a corresponding mask by utilizing radar data to generate a mask set, selecting more complete radar data as a background field and generating a background set;
s2, inputting a mask set and a background field into a countermeasure generation network model, wherein the model utilizes the mask set to mask the background field to generate a damaged image and a missing region image, the missing region image is a mask masking part in the background, and then a repair image is obtained by the model;
s3, determining a loss value of the countermeasure generation network model according to the background, the repair image, the mask and the missing region image;
and S4, if the loss value is higher than a preset threshold value, updating the model.
3. A method of radar profile correction based on an countermeasure generation network model as claimed in claim 1, wherein: before the data to be corrected is rebuilt in a gray conversion mode, the radar data to be corrected is classified and judged by adopting a maximum probability algorithm, and a proper correction model is selected.
4. A radar profile correcting system based on an countermeasure-generating network model for implementing the radar profile correcting method based on an countermeasure-generating network model according to claim 1, characterized in that: the system comprises a gray conversion module, a shielding detection module, a probability algorithm module, a correction model selection module and a radar data correction module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211454168.0A CN115511011B (en) | 2022-11-21 | 2022-11-21 | Radar data correction method and system based on countermeasure generation network model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211454168.0A CN115511011B (en) | 2022-11-21 | 2022-11-21 | Radar data correction method and system based on countermeasure generation network model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115511011A CN115511011A (en) | 2022-12-23 |
| CN115511011B true CN115511011B (en) | 2023-06-09 |
Family
ID=84514166
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211454168.0A Active CN115511011B (en) | 2022-11-21 | 2022-11-21 | Radar data correction method and system based on countermeasure generation network model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115511011B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115358151A (en) * | 2022-08-25 | 2022-11-18 | 兰州大学 | Correction method for near-stratum wind speed product of numerical weather forecast |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8493262B2 (en) * | 2011-02-11 | 2013-07-23 | Mitsubishi Electric Research Laboratories, Inc. | Synthetic aperture radar image formation system and method |
| CN111861901A (en) * | 2020-06-05 | 2020-10-30 | 西安工程大学 | Edge generation image restoration method based on GAN network |
| CN113311436B (en) * | 2021-04-30 | 2022-07-12 | 中国人民解放军国防科技大学 | Method for correcting wind measurement of motion attitude of laser wind measuring radar on mobile platform |
-
2022
- 2022-11-21 CN CN202211454168.0A patent/CN115511011B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115358151A (en) * | 2022-08-25 | 2022-11-18 | 兰州大学 | Correction method for near-stratum wind speed product of numerical weather forecast |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115511011A (en) | 2022-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112965146B (en) | Quantitative precipitation estimation method combining meteorological radar and rainfall barrel observation data | |
| WO2020003586A1 (en) | Data generation device, image identification device, data generation method, and storage medium | |
| US20200410710A1 (en) | Method for measuring antenna downtilt based on linear regression fitting | |
| CN110837006B (en) | Satellite lightning detection evaluation method based on satellite-ground synchronous observation and comparison | |
| CN108229274B (en) | Method and device for training multilayer neural network model and recognizing road characteristics | |
| CN117423002B (en) | Small-scale tidal flat image processing method based on DEM | |
| KR102248963B1 (en) | Method and apparatus for predicting sea fog | |
| CN113887337A (en) | Seal identification method and system with bent text and storage medium | |
| CN116310883B (en) | Agricultural disaster prediction method based on remote sensing image space-time fusion and related equipment | |
| CN113644996A (en) | Cell-level RSRP estimation method based on deep learning | |
| Lumban-Gaol et al. | Satellite-derived bathymetry using convolutional neural networks and multispectral sentinel-2 images | |
| CN115511011B (en) | Radar data correction method and system based on countermeasure generation network model | |
| CN117664874B (en) | Estuary area salt tide tracing monitoring method and system based on space-earth combination | |
| CN109886980A (en) | A kind of infrared image cirrus detection method based on neighborhood intensity texture coding | |
| CN117670981A (en) | Laser spot centroid positioning method and system integrating physical information deep learning | |
| CN114252871B (en) | Radar measurement precision compensation method based on machine learning | |
| CN116973946A (en) | Navigation satellite positioning method for deep learning and fisheye camera image segmentation | |
| JP2020173160A (en) | Tsunami height and tsunami arrival time prediction system | |
| CN114187409A (en) | Method for building ship model based on video image and laser radar point cloud fusion | |
| CN110223250B (en) | SAR geometric correction method based on homography transformation | |
| CN113743373A (en) | High-resolution remote sensing image cropland change detection device and method based on deep learning | |
| CN112017108A (en) | Satellite ortho-image color relative correction method based on independent model method block adjustment | |
| CN117523321B (en) | Optical shallow water classification method based on passive remote sensing spectral image application neural network | |
| CN112699710A (en) | GF2 remote sensing image dense target identification method and system based on deep learning | |
| CN114020725B (en) | Window sliding GPM data correction method considering spatial distribution |
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 |