CN110826526A - Method for cloud detection radar to identify clouds - Google Patents

Method for cloud detection radar to identify clouds Download PDF

Info

Publication number
CN110826526A
CN110826526A CN201911135857.3A CN201911135857A CN110826526A CN 110826526 A CN110826526 A CN 110826526A CN 201911135857 A CN201911135857 A CN 201911135857A CN 110826526 A CN110826526 A CN 110826526A
Authority
CN
China
Prior art keywords
cloud
fuzzy logic
logic function
parameters
learning
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.)
Pending
Application number
CN201911135857.3A
Other languages
Chinese (zh)
Inventor
丁霞
王平
王海涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Radio Equipment Research Institute
Original Assignee
Shanghai Radio Equipment Research Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Radio Equipment Research Institute filed Critical Shanghai Radio Equipment Research Institute
Priority to CN201911135857.3A priority Critical patent/CN110826526A/en
Publication of CN110826526A publication Critical patent/CN110826526A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • 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
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/95—Radar or analogous systems specially adapted for specific applications for meteorological use
    • G01S13/955—Radar or analogous systems specially adapted for specific applications for meteorological use mounted on satellite
    • 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
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G01S7/418—Theoretical aspects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • 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

Abstract

The invention discloses a method for identifying clouds by a cloud detection radar, which comprises the following steps: establishing a cloud physical quantity learning library for autonomous learning and a matching rule that cloud characteristic parameters belong to a certain class of cloud; establishing a fuzzy logic function of each cloud characteristic parameter belonging to each type of cloud and a corresponding fuzzy logic function parameter; setting influence weight values, calculating the probability that different cloud characteristic parameters belong to certain cloud by adopting a weighted normalization method, and obtaining a calculation result of cloud identification according to a matching rule; calculating the error of the cloud identification result, and optimizing the parameters of the fuzzy logic function and the influence weight by utilizing partial derivative loop iterative operation; and calculating cloud identification results of the new cloud-measuring radar observation information, updating a cloud physical learning library, and realizing cyclic autonomous learning and updating of fuzzy logic function parameters and influence weights. The invention overcomes the defect that the threshold value method is easy to misjudge, solves the problem of parameter acquisition of fuzzy logic, and has self-adaptive learning optimization capability.

Description

Method for cloud detection radar to identify clouds
Technical Field
The invention relates to the field of meteorological remote sensing and neural networks, in particular to a cyclic autonomous learning cloud identification method applying a cloud measuring radar.
Background
The cloud has an important regulation effect on the radiation balance and the water vapor circulation of the earth-atmosphere system, different clouds have different dynamic processes and physical characteristics, generate different radiation scattering characteristics, macroscopic parameters such as cloud thickness and cloud range, and the inversion of microscopic parameters such as particle size, shape, number concentration, ice water content and particle spectrum distribution in the cloud is closely related to the cloud phase state. Climate model studies show that the difference of cloud characteristic parameters of different types of clouds in the patterns can cause great difference between simulation results, so that unreasonable characterization of the clouds in the patterns becomes a main uncertainty source in climate change prediction. Therefore, correctly classifying the cloud is an important prerequisite for conducting cloud physics research.
The cloud classification method mainly comprises a threshold value method, a structural analysis method, a fuzzy logic method, a neural network method and the like, and cloud products are widely applied to meteorological research. The international satellite cloud climate program (ISCP) adopts a threshold value method, and utilizes air pressure and optical thickness to realize cloud identification. The observation data of the cloud measuring radar is important data in cloud research, before the cloud-Sat satellite-borne millimeter wave cloud radar appears, a satellite cloud picture occupies an important part of cloud classification research, and scholars at home and abroad use a threshold value method, histogram clustering and the like for cloud picture identification, but since the satellite image changes along with time, place and earth surface, universal threshold value standards and judgment conditions are difficult to find out. The structural analysis method achieves the purpose of classifying the cloud pictures based on the texture structure of the cloud.
In 2006, a CloudSat satellite transmitted by NASA in the united states carries a millimeter wave cloud radar, observation data of a global cloud vertical structure is realized, obtained cloud profile data is classified by adopting a threshold value method, whether cloud exists or not is determined by using radar echo data, different threshold values are set according to factors such as cloud height, temperature, effective reflectivity factors and spatial distribution, and the cloud is divided into eight types including cirrus (cirrus), high-layer cloud (altostratus), high-volume cloud (altoculus), layer cloud (stratus), cumulus, rainus (nimbostratus) and deep convection cloud (deep convection). The threshold method is a simple and easy-to-operate method, the classification result is very sensitive to the selection of the threshold, the selection of the threshold has strong locality, and the acquisition of the uniform threshold suitable for various regional and seasonal changes is difficult. In order to solve the problem, a fuzzy logic method is adopted to carry out cloud classification on the CloudSat satellite data, cloud parameters are divided into various grades, misjudgment occurring in a threshold value method is reduced according to a loose classification principle, a reasonable classification result can be obtained, the method has strong expandability and compatibility, the selection of the logic parameters is obtained based on historical observation data statistics, the workload is large, and the applicability is limited.
In summary, the existing cloud classification algorithm generally adopts a threshold method, which is simple and convenient to apply, and the selection of the threshold often causes great difference in classification results. The fuzzy logic method of the cloud profile can avoid the defect that the threshold value method excessively depends on the size of the threshold value, and can obtain a reasonable classification result, but the specific parameters of the logic rule are complex to obtain, and the classification result is directly influenced.
Disclosure of Invention
The invention provides a method for cloud identification according to cloud radar observation data, wherein the cloud radar observation data comprises radar reflectivity factors, longitude, latitude, height and temperature information. According to the method, remote sensing information of the cloud detection radar is comprehensively utilized, function parameters of various clouds of cloud characteristic parameters are obtained through autonomous learning, effective discrimination results are obtained, cyclic calculation and parameter updating are carried out according to observation data of the cloud detection radar, adaptive optimization of the parameters is achieved through cyclic autonomous learning, and therefore accurate cloud classification results are obtained. The method overcomes the defect that the threshold value method is easy to misjudge, solves the problem of parameter acquisition of fuzzy logic, and has self-adaptive learning optimization capability.
In order to achieve the aim, the invention provides a method for identifying clouds by a cloud detection radar, which comprises the following steps:
s1, extracting cloud characteristic parameters from the cloud measuring radar observation information sample data to serve as input quantity, using clouds as output quantity, establishing a cloud physical quantity learning library for autonomous learning, carrying out quantization processing on cloud classification values, and establishing a cloud characteristic parameter matching rule of each cloud according to the characteristic parameter distribution range of each cloud;
s2, establishing fuzzy logic functions of which each cloud characteristic parameter belongs to each type of cloud, and parameter vectors A and B corresponding to each fuzzy logic function;
s3, setting an influence weight W of a fuzzy logic function output value of a certain cloud characteristic parameter belonging to a certain cloud, calculating the probability of the cloud characteristic parameter belonging to the certain cloud by adopting a weighted normalization method, and obtaining a calculation result of cloud identification according to a matching rule;
s4, calculating errors of the cloud identification results, setting error convergence conditions according to quantization rules of the cloud identification results, and transmitting parameter vectors A, B and influence weights W corresponding to the optimized fuzzy logic function through error partial derivatives;
and S5, calculating cloud identification results of the new cloud detection radar observation information, and adding the new cloud detection radar observation information and the cloud identification results into a cloud physical quantity learning library to realize cyclic autonomous learning and updating of fuzzy logic function parameters and influence weights in matching rules.
Further, the cloud characteristic parameters comprise radar reflectivity factors, longitude, latitude, altitude and temperature; the cloud class comprises a rolling cloud, a high-layer cloud, a high-volume cloud, a layer cloud, a volume cloud, a rain layer cloud and a deep convection cloud, and cloud class information is converted into digital output to be used as an output value of the cloud physical quantity learning library.
The output value of a fuzzy logic function of a certain cloud characteristic parameter belonging to a certain type of cloud is between 0 and 1, and the fuzzy logic function is used for realizing fuzzification of observation information of the input cloud measuring radar, and the parameter vectors A and B corresponding to all the fuzzy logic functions are respectively as follows:
Figure BDA0002279579730000031
where m is 5, n is 8, and the vector a and the vector B respectively represent the fuzzy logic function parameters of each cloud feature parameter belonging to each cloud class.
Further, the fuzzy logic function can be set as a gaussian function, a beta function, a bell-shaped function and a trapezoid function, and the selection of the function form reflects the value range and the probability distribution condition of the radar observation value when the cloud is of a certain class, so that the cloud identification result is influenced significantly.
In step S3, the probability that the cloud characteristic parameter belongs to a certain type of cloud is calculated by applying a weighted average calculation rule to the output value of the fuzzy logic function of the cloud characteristic parameter belonging to the certain type of cloud:
Figure BDA0002279579730000032
wherein i is the number of cloud characteristic parameters, j is the number of clouds, CijFuzzy logic function output value omega for representing that certain cloud characteristic parameter belongs to certain cloudijThe influence weight of the fuzzy logic function output value of a certain cloud characteristic parameter belonging to a certain type of cloud, and the corresponding influence weight vector W is as follows:
the step S4 includes the following steps:
s4.1, calculating the error of the cloud identification result;
the error function is expressed as:
Figure BDA0002279579730000042
y is a calculation result of cloud identification in the cloud radar observation information sample data, and R is a real result of cloud identification in the cloud radar observation information sample data;
s4.2, setting an error convergence condition according to a quantization rule of the cloud identification result;
the error convergence condition is as follows:
E<<0.125；
s4.3, optimizing fuzzy logic function parameters and influence weights through error partial derivative inverse transfer, and realizing the autonomous learning of the fuzzy logic function parameters and the influence weights;
and (3) performing loop iteration partial derivative correction operation on the fuzzy logic function parameter vector A, B and the influence weight parameter vector W by the error function until the errors of the cloud identification result and the real result meet the error convergence condition.
The partial derivative correction operation is as follows: error partial derivative function is respectively according to
Figure BDA0002279579730000043
Andthe parameter vectors A, B and W are modified in a gradient descent manner, where k is the learning efficiency.
In step S5, the cloud identification result of the new cloud radar observation information is calculated, which includes the following steps:
s5.1, extracting five cloud characteristic parameters from the new observation information of the cloud measuring radar;
s5.2, calculating fuzzy logic function values of five cloud characteristic quantities of new cloud radar observation information belonging to various clouds according to the fuzzy logic function parameters and the fuzzy logic functions obtained by the autonomous learning in the step S4;
s5.3, calculating the probability that the cloud characteristic quantity of the new cloud detection radar observation information belongs to each cloud according to the influence weight and the weighted average calculation rule obtained by the autonomous learning in the step S4;
and S5.4, obtaining a new cloud identification result of the observation information of the cloud measuring radar according to the matching rule that the cloud characteristic parameters belong to each cloud.
According to the cloud classification method, remote sensing information of the cloud radar is comprehensively utilized, the cloud characteristic parameters belong to function parameters of various clouds and are obtained through autonomous learning, effective judgment results are obtained, cyclic calculation and parameter updating are carried out according to observation data of the cloud radar, adaptive optimization of the parameters is achieved through cyclic autonomous learning, and therefore accurate cloud classification results are obtained. The method overcomes the defect that the threshold value method is easy to misjudge, solves the problem of parameter acquisition of fuzzy logic, and has self-adaptive learning optimization capability.
Drawings
Fig. 1 is a flowchart of a method for identifying cloud class by a cloud radar according to the present invention.
Detailed Description
An embodiment of the present invention is described in detail below with reference to fig. 1.
As shown in fig. 1, the method for identifying cloud by using a cloud radar provided by the present invention includes the following steps:
step S1, extracting cloud characteristic parameters from the cloud measuring radar observation information sample data as input quantity, using clouds as output quantity, establishing a cloud physical quantity learning library for autonomous learning, carrying out quantization processing on cloud classification values, and establishing a cloud characteristic parameter matching rule of each cloud according to the characteristic parameter distribution range of each cloud;
and obtaining cloud classification products obtained by information such as vertical and horizontal cloud characteristics of the cloud, precipitation, cloud body temperature and the like according to the cloud measuring radar remote sensing data. Considering that the physical properties of the cloud are changed greatly in different space-time ranges, and meanwhile, radar observation data are changed along with the changes of atmospheric background and geographic positions, five parameters including radar reflectivity factors, longitude, latitude, altitude and temperature are extracted from sample data of cloud-measuring radar observation information and are characteristic parameters of the cloud. Clouds include rolling clouds, high-layer clouds, high-lying clouds, layer clouds, layered clouds, lying clouds, rainlayer clouds, deep convection clouds. Establishing a cloud physical quantity learning database for autonomous learning, wherein the input quantity is a cloud characteristic parameter vector X ═ X1,...,xm) The output quantities are cloud classes, which are used for converting the cloud classes into corresponding numerical values 2, 3, 4, 5, 6, 7, 8 and 9, and the actual output cloud class identification vector R ═ 5 (R ═ is1,...,rn) And n is 8. According to statistical analysisAnd establishing a cloud characteristic parameter matching rule of each type of cloud within the distribution range of the characteristic parameters of each type of cloud, wherein the specific parameters in the matching rule can be set as initial values according to the statistical analysis result of the historical data.
Step S2, establishing a fuzzy logic function of each cloud characteristic parameter belonging to each type of cloud, and parameter vectors A and B corresponding to each fuzzy logic function;
the number of cloud characteristic parameters is 5, including: radar reflectivity factor, longitude, latitude, altitude, temperature, cloud category is 8, including: the cloud computing system comprises rolling clouds, high-level clouds, high-integrating clouds, layer clouds, laminating clouds, integrating clouds, raining layer clouds and deep convection clouds, one fuzzy logic function corresponds to a computing rule that a characteristic parameter belongs to a certain type of clouds, and the number of the fuzzy logic functions of each cloud characteristic parameter of each type of clouds is 40. The form of the fuzzy logic function can be set as a Gaussian function, a beta function, a bell-shaped function, a trapezoid function and the like, and the selection of the function form reflects the value range and the probability distribution condition of the radar observation value when the cloud is of a certain type, so that the cloud type identification result is influenced significantly. In this embodiment, a gaussian function is taken as an example to describe the implementation process of the method. The fuzzy logic function of a certain cloud characteristic parameter belonging to a certain cloud is as follows:
the output value of the formula (1) is between 0 and 1, and the formula is used for realizing fuzzification of the observation information of the input cloud measuring radar, wherein aijAnd bijFor undetermined parameters of the fuzzy logic function, corresponding parameter vectors are respectively A and B:
the vector A and the vector B respectively represent fuzzy logic function parameter values of each cloud characteristic parameter belonging to each type of cloud.
S3, setting an influence weight W of a fuzzy logic function output value of a certain cloud characteristic parameter belonging to a certain cloud, calculating the probability of the cloud characteristic parameter belonging to the certain cloud by adopting a weighted normalization method, and obtaining a calculation result of cloud identification according to a matching rule;
calculating the probability that the cloud characteristic parameters belong to a certain cloud by adopting a weighted average calculation rule for the output value of the fuzzy logic function of which the cloud characteristic parameters belong to the certain cloud according to the number of the clouds:
Figure BDA0002279579730000071
wherein i is the number of cloud characteristic parameters, j is the number of clouds, CijFuzzy logic function output value omega for representing that certain cloud characteristic parameter belongs to certain cloudijThe influence weight of the fuzzy logic function output value of a certain cloud characteristic parameter belonging to a certain type of cloud, and the corresponding influence weight vector is as follows:
Figure BDA0002279579730000072
the output of the formula (2) is the probability that the cloud characteristic parameter belongs to a certain cloud, the calculation result that the cloud characteristic parameter belongs to the certain cloud is obtained according to the matching rule, and the output vector of the formula (2) is Y ═ Y1,...,yn) And the calculation result of cloud class identification of the cloud radar observation information sample data is represented.
Step S4, calculating errors of the cloud identification results, setting error convergence conditions according to quantization rules of the cloud identification results, and transmitting parameter vectors A, B and influence weights W corresponding to the optimized fuzzy logic function through error partial derivatives;
according to a cloud identification calculation result Y of the cloud radar observation information sample data, calculating an error of the cloud identification result, wherein an error function is as follows:
Figure BDA0002279579730000073
wherein, R is the real result of cloud class identification in the cloud radar observation information sample data, and E is the square error between the calculation result and the real result of cloud class identification.
Setting an error function judgment threshold as an error convergence condition, and setting the error convergence condition as E according to the quantization rule of the cloud identification result<<0.125, calculating partial derivatives of the error function to the fuzzy logic function parameter vector A, B and the weight-affected parameter vector W, respectivelyAnd
Figure BDA0002279579730000075
correcting the parameter vectors A, B and W in a gradient descent mode, wherein k is learning efficiency, and obtaining a parameter vector A after correction1、B1And W1Repeating the calculation of the formula (1), the formula (2) and the formula (3) to obtain the corrected error E1Judgment E1Whether the error convergence condition is met.
If the error E1If the error convergence condition is not met, the iterative error partial derivative correction operation is started until the error of the calculation result and the real result of cloud identification meets the error convergence condition, and the undetermined parameter a of the fuzzy logic function is realizedij、bijAnd influence weight ωijThe cycle of (2) learning autonomously.
And step S5, calculating cloud identification results of the new cloud-measuring radar observation information, adding the new cloud-measuring radar observation information and the cloud identification results into a cloud physical quantity learning library, and realizing cyclic autonomous learning and updating of fuzzy logic function parameters and influence weights in matching rules.
For the new cloud radar observation information, five cloud characteristic parameters (radar reflectivity factor, longitude, latitude, height and temperature) are extracted, the fuzzy logic function values of various clouds belonging to the five cloud characteristic parameters of the new cloud radar observation information are calculated according to the fuzzy logic function parameters obtained through the step S4 of the cyclic self-learning and the formula (1), and then cloud identification results are obtained through calculation according to the influence weight obtained through the step S4 of the cyclic self-learning, the formula (2) and the matching rule.
And adding new cloud radar observation information and identification results into a cloud physical quantity learning library, repeating the steps S1-S4, updating undetermined parameters and influence weights of the fuzzy logic function through cyclic autonomous learning, and realizing adaptive learning and optimization of the undetermined parameters and the influence weights of the fuzzy logic function in the matching rule so as to obtain more accurate cloud identification results.
According to the cloud classification method, remote sensing information of the cloud radar is comprehensively utilized, the cloud characteristic parameters belong to function parameters of various clouds and are obtained through autonomous learning, effective judgment results are obtained, cyclic calculation and parameter updating are carried out according to observation data of the cloud radar, adaptive optimization of the parameters is achieved through cyclic autonomous learning, and therefore accurate cloud classification results are obtained. The method overcomes the defect that the threshold value method is easy to misjudge, solves the problem of parameter acquisition of fuzzy logic, and has self-adaptive learning optimization capability.
While the present invention has been described in detail with reference to the preferred embodiments, it should be understood that the above description should not be taken as limiting the invention. Various modifications and alterations to this invention will become apparent to those skilled in the art upon reading the foregoing description. Accordingly, the scope of the invention should be determined from the following claims.

Claims (2)

1. A method for identifying cloud class by a cloud detection radar is characterized by comprising the following steps:
step S1, extracting cloud characteristic parameters from the cloud measuring radar observation information sample data as input quantity, using clouds as output quantity, establishing a cloud physical quantity learning library for autonomous learning, carrying out quantization processing on cloud classification values, and establishing a cloud characteristic parameter matching rule of each cloud according to the characteristic parameter distribution range of each cloud;
step S2, establishing a fuzzy logic function of each cloud characteristic parameter belonging to each type of cloud, and parameter vectors A and B corresponding to each fuzzy logic function;
s3, setting an influence weight W of a fuzzy logic function output value of a certain cloud characteristic parameter belonging to a certain cloud, calculating the probability of the cloud characteristic parameter belonging to the certain cloud by adopting a weighted normalization method, and obtaining a calculation result of cloud identification according to a matching rule;
step S4, calculating errors of the cloud identification results, setting error convergence conditions according to quantization rules of the cloud identification results, and transmitting parameter vectors A, B and influence weights W corresponding to the optimized fuzzy logic function through error partial derivatives;
and step S5, calculating cloud identification results of the new cloud-measuring radar observation information, adding the new cloud-measuring radar observation information and the cloud identification results into a cloud physical quantity learning library, and realizing cyclic autonomous learning and updating of fuzzy logic function parameters and influence weights in matching rules.
2. The method as claimed in claim 1, wherein the cloud characteristic parameters include radar reflectivity factor, longitude, latitude, altitude, temperature; the cloud class comprises a rolling cloud, a high-layer cloud, a high-volume cloud, a layer cloud, a volume cloud, a rain layer cloud and a deep convection cloud, and cloud class information is converted into digital output to be used as an output value of the cloud physical quantity learning library.
CN201911135857.3A 2019-11-19 2019-11-19 Method for cloud detection radar to identify clouds Pending CN110826526A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911135857.3A CN110826526A (en) 2019-11-19 2019-11-19 Method for cloud detection radar to identify clouds

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911135857.3A CN110826526A (en) 2019-11-19 2019-11-19 Method for cloud detection radar to identify clouds

Publications (1)

Publication Number Publication Date
CN110826526A true CN110826526A (en) 2020-02-21

Family

ID=69556924

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911135857.3A Pending CN110826526A (en) 2019-11-19 2019-11-19 Method for cloud detection radar to identify clouds

Country Status (1)

Country Link
CN (1) CN110826526A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112949726A (en) * 2021-03-11 2021-06-11 易天气(北京)科技有限公司 ISCP cloud classification method, system, medium and terminal based on FY-4A satellite
CN113237801A (en) * 2021-05-13 2021-08-10 北京市人工影响天气办公室 Method for identifying sand and dust cloud mixture
CN114708279A (en) * 2022-04-11 2022-07-05 西安邮电大学 Cloud microparticle data area extraction method
CN112949726B (en) * 2021-03-11 2024-05-14 易天气(北京)科技有限公司 ISCCP cloud classification method, system, medium and terminal based on FY-4A satellite

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103792524A (en) * 2014-03-06 2014-05-14 兰州大学 Radar quantitative precipitation estimation method based on cloud classification
CN105891833A (en) * 2016-04-15 2016-08-24 兰州大学 Method of identifying warm cloud precipitation rate based on Doppler radar information
CN106547840A (en) * 2016-10-13 2017-03-29 国家卫星气象中心 A kind of parsing of global three-dimensional atmospheric data and management method
CN107015229A (en) * 2017-05-22 2017-08-04 新疆维吾尔自治区人工影响天气办公室 Artificial Hail Suppression operation command method based on dual-polarization weather radar
CN108983183A (en) * 2018-06-28 2018-12-11 浙江大学 A kind of adaptive radar sea clutter forecast system
CN109035223A (en) * 2018-07-11 2018-12-18 北京航空航天大学 A kind of intelligent evaluation method for satellite remote sensing images availability
CN109061646A (en) * 2018-06-26 2018-12-21 成都信息工程大学 A kind of Doppler radar quantitative precipitation estimation method based on Precipitation Clouds identification
CN110346844A (en) * 2019-07-15 2019-10-18 南京恩瑞特实业有限公司 Quantitative Precipitation estimating and measuring method of the NRIET based on cloud classification and machine learning

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103792524A (en) * 2014-03-06 2014-05-14 兰州大学 Radar quantitative precipitation estimation method based on cloud classification
CN105891833A (en) * 2016-04-15 2016-08-24 兰州大学 Method of identifying warm cloud precipitation rate based on Doppler radar information
CN106547840A (en) * 2016-10-13 2017-03-29 国家卫星气象中心 A kind of parsing of global three-dimensional atmospheric data and management method
CN107015229A (en) * 2017-05-22 2017-08-04 新疆维吾尔自治区人工影响天气办公室 Artificial Hail Suppression operation command method based on dual-polarization weather radar
CN109061646A (en) * 2018-06-26 2018-12-21 成都信息工程大学 A kind of Doppler radar quantitative precipitation estimation method based on Precipitation Clouds identification
CN108983183A (en) * 2018-06-28 2018-12-11 浙江大学 A kind of adaptive radar sea clutter forecast system
CN109035223A (en) * 2018-07-11 2018-12-18 北京航空航天大学 A kind of intelligent evaluation method for satellite remote sensing images availability
CN110346844A (en) * 2019-07-15 2019-10-18 南京恩瑞特实业有限公司 Quantitative Precipitation estimating and measuring method of the NRIET based on cloud classification and machine learning

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
任建奇，严卫，杨汉乐，施健康: "基于模糊逻辑的CloudSat 卫星资料云分类算法", vol. 12, no. 12, pages 91 - 96 *
任建奇;严卫;杨汉乐;施健康;: "基于模糊逻辑的CloudSat卫星资料云分类算法", 解放军理工大学学报(自然科学版), no. 01, pages 91 - 96 *
李峰;李柏;吴蕾;杨荣康;邢毅;黄兴友;肖辉;王斌;: "WMO第八届阳江国际探空比对辅助遥感综合试验", no. 08, pages 917 - 924 *
李红斌;方德贤;侯国成;张殿刚;陈农;濮文耀;: "基于模块化设计的数字雷达人工增雨作业指挥系统", no. 06, pages 197 - 204 *
蔡桂芳: "基于模糊神经网络的故障预测方法研究" *
蔡桂芳: "基于模糊神经网络的故障预测方法研究", 机电产品开发与创新, vol. 21, no. 21, pages 35 - 39 *
陈刚毅，丁旭羲，赵丽妍: "用模糊神经网络自动识别云的技术研究", 大气科学, vol. 29, no. 5, pages 838 - 844 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112949726A (en) * 2021-03-11 2021-06-11 易天气(北京)科技有限公司 ISCP cloud classification method, system, medium and terminal based on FY-4A satellite
CN112949726B (en) * 2021-03-11 2024-05-14 易天气(北京)科技有限公司 ISCCP cloud classification method, system, medium and terminal based on FY-4A satellite
CN113237801A (en) * 2021-05-13 2021-08-10 北京市人工影响天气办公室 Method for identifying sand and dust cloud mixture
CN113237801B (en) * 2021-05-13 2022-06-28 北京市人工影响天气中心 Method for identifying sand cloud mixture
CN114708279A (en) * 2022-04-11 2022-07-05 西安邮电大学 Cloud microparticle data area extraction method

Similar Documents

Publication Publication Date Title
CN110516596B (en) Octave convolution-based spatial spectrum attention hyperspectral image classification method
CN112766199B (en) Hyperspectral image classification method based on self-adaptive multi-scale feature extraction model
Nikam et al. Modeling rainfall prediction using data mining method: A Bayesian approach
CN111414991B (en) Meteorological frontal surface automatic identification method based on multiple regression
CN112101278A (en) Hotel point cloud classification method based on k nearest neighbor feature extraction and deep learning
CN114254561A (en) Waterlogging prediction method, waterlogging prediction system and storage medium
CN109946762B (en) Method and system for short-time rainfall forecast based on probability distribution
CN110929607A (en) Remote sensing identification method and system for urban building construction progress
CN103914705B (en) Hyperspectral image classification and wave band selection method based on multi-target immune cloning
CN112541508A (en) Fruit segmentation and recognition method and system and fruit picking robot
CN113837499A (en) Ultra-short-term wind power prediction method and system
CN112950780B (en) Intelligent network map generation method and system based on remote sensing image
CN110826526A (en) Method for cloud detection radar to identify clouds
CN109671019A (en) A kind of remote sensing image sub-pixed mapping drafting method based on multi-objective optimization algorithm and sparse expression
CN115659853A (en) Nonlinear mixed-effect strain coefficient downscaling method and system
CN114973019A (en) Deep learning-based geospatial information change detection classification method and system
Dungan Toward a comprehensive view of uncertainty in remote sensing analysis
CN104463252A (en) Foundation cloud classification method based on self-adaptive extreme learning machine
Liu et al. Soil water content forecasting by ANN and SVM hybrid architecture
CN114169502A (en) Rainfall prediction method and device based on neural network and computer equipment
CN116760017A (en) Prediction method for photovoltaic power generation
CN116822716A (en) Typhoon prediction method, system, equipment and medium based on space-time attention
CN115393631A (en) Hyperspectral image classification method based on Bayesian layer graph convolution neural network
CN113066537B (en) Compound classification method based on graph neural network
CN112668615B (en) Satellite cloud picture prediction method based on depth cross-scale extrapolation fusion

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