CN110824481A - Quantitative precipitation prediction method based on radar reflectivity extrapolation - Google Patents
Quantitative precipitation prediction method based on radar reflectivity extrapolation Download PDFInfo
- Publication number
- CN110824481A CN110824481A CN201911030327.2A CN201911030327A CN110824481A CN 110824481 A CN110824481 A CN 110824481A CN 201911030327 A CN201911030327 A CN 201911030327A CN 110824481 A CN110824481 A CN 110824481A
- Authority
- CN
- China
- Prior art keywords
- radar echo
- neural network
- radar
- region
- layer
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000001556 precipitation Methods 0.000 title claims abstract description 33
- 238000002310 reflectometry Methods 0.000 title claims abstract description 25
- 238000013213 extrapolation Methods 0.000 title claims abstract description 23
- 230000003287 optical effect Effects 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims description 40
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000002592 echocardiography Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
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
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/40—Means for monitoring or calibrating
- G01S7/4052—Means for monitoring or calibrating by simulation of echoes
-
- 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
-
- 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
- G01S7/35—Details of non-pulse systems
- G01S7/352—Receivers
- G01S7/354—Extracting wanted echo-signals
-
- 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)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a quantitative precipitation prediction method based on radar reflectivity extrapolation, which relates to the technical field of atmospheric science, and comprises the following steps of S1: collecting radar base data for prediction, and generating a radar echo reflectivity image by using a reflectivity factor in the radar base data; s2: solving a pre-constructed optical flow equation based on the radar echo reflectivity image to obtain a motion field of the radar echo; s3: identifying a strong convection region in a motion field of a radar echo by using a pre-constructed identification model, and tracking and positioning the strong convection region; s4: constructing a radar echo extrapolation model, training the radar echo extrapolation model by using a strong convection area, and predicting radar echo data; s5: the method has the advantages of simplicity and high prediction accuracy.
Description
Technical Field
The invention relates to the technical field of atmospheric science, in particular to a quantitative precipitation prediction method based on radar reflectivity extrapolation.
Background
Currently, there are two methods for short-term rainfall forecasting:
one is a numerical weather pattern forecasting method, although the resolution and the precision of the numerical weather forecasting pattern are high at present, particularly the precision of a medium-short term flow field and a shape potential field of the numerical weather forecasting pattern is high, the short rainfall error of forecasting by using the numerical weather pattern is large due to the spin-up problem of the numerical weather pattern.
The other method is a weather radar echo diagram extrapolation method, a proximity prediction technology based on weather radar detection data plays a very positive and effective role in short-time proximity precipitation prediction, the technology is researched from the last 60 th century, and with the continuous updating of radar hardware infrastructure and detection technology, the weather radar-based precipitation prediction method also makes a long-term progress.
From the point of mathematical statistics, the weather radar echo diagram extrapolation method effectively solves the prediction problem of the development and change of radar echoes in the future, but the development and evolution process of actual weather is often more complex, and from echo images with continuous time, it can be seen that the water-degradable areas represented by the echo images are continuously consumed, developed, moved and changed from whole to local at every moment.
Disclosure of Invention
The invention aims to: the invention provides a quantitative rainfall prediction method based on radar reflectivity extrapolation, and aims to solve the problems that the existing rainfall prediction based on a linear motion relation has larger deviation from the actual condition and lower rainfall prediction accuracy when a long-time echo image is predicted.
The invention specifically adopts the following technical scheme for realizing the purpose:
a method of quantitative precipitation prediction based on radar reflectivity extrapolation, comprising:
s1: collecting radar base data for prediction, and generating a radar echo reflectivity image by using a reflectivity factor in the radar base data;
s2: solving a pre-constructed optical flow equation based on the radar echo reflectivity image to obtain a motion field of the radar echo;
s3: identifying a strong convection region in a motion field of a radar echo by using a pre-constructed identification model, and tracking and positioning the strong convection region;
s4: constructing a radar echo extrapolation model, training the radar echo extrapolation model by using a strong convection area, and predicting radar echo data;
s5: and converting the radar echo data by using the trained quantitative precipitation neural network prediction model to obtain the predicted precipitation.
Further, when solving the optical flow equation in S2, local constraint is performed on the optical flow equation by using the Lucas-Kanade method, so as to ensure smooth change of the optical flow.
Further, the identifying module in S3 identifies a strong convection region in the motion field of the radar echo, specifically including:
traversing all pixel points of the motion field of the radar echo, if the reflectivity intensity of the current pixel point is greater than a set threshold value, recording the intensity flag bit of the pixel point as 1, otherwise, recording as 0;
searching an eight-connected region of a region where the pixel point with the strength zone bit marked as 1 is located, if the area of the eight-connected region is not less than 20 pixel points, keeping the eight-connected region, and if not, deleting the eight-connected region;
detecting all the reserved eight-connected areas, judging whether the eight-connected areas are non-precipitation echoes caused by super refraction, if not, reserving the eight-connected areas, and otherwise, deleting the eight-connected areas;
processing the reserved eight-connected region by utilizing an opening operation, eliminating false connection and obtaining a final eight-connected region image;
and extracting the characteristics of the final eight-connected region image, and deciding the extracted characteristics based on a set decision method to obtain a strong convection region.
Further, in S3, the SCIT algorithm is used to track and locate the strong convection region.
Further, the radar echo extrapolation model adopts a network structure of ST-LSTM-RNN.
Furthermore, the ST-LSTM-RNN network structure comprises an ST-LSTM unit, a time memory module and a space memory module;
inputting data of a strong convection region into a plurality of ST-LSTM units which are connected in sequence, and outputting predicted radar echo data;
the time memory module and the space memory module are added into each ST-LSTM unit, the time memory module is used for memorizing and accumulating the first m moments of the neurons in the current layer, and the space memory module is used for memorizing and superposing the first m moments of the neurons in different layers.
Furthermore, the quantitative precipitation neural network prediction model comprises an input layer, a second layer neural network, a third layer neural network and a fourth layer neural network which are connected in sequence, wherein the second layer neural network comprises two depth separable convolutional layers, and the output ends of the two depth separable convolutional layers are connected with the input end of the third layer neural network; the third layer of neural network comprises two depth separable convolutional layers, and the output ends of the two depth separable convolutional layers are connected with the input end of the fourth layer of neural network; the fourth layer of neural network includes two convolutional layers.
Further, a large amount of radar echo data and actual precipitation data are used as a training data set to train the quantitative precipitation neural network prediction model, and network parameters are optimized and adjusted based on a preset loss function and an optimizer until the network converges, so that the trained quantitative precipitation neural network prediction model is obtained.
The invention has the following beneficial effects:
1. according to the method, the quantitative rainfall is predicted by combining the identification model, the radar echo extrapolation model and the quantitative rainfall neural network prediction model, after the radar reflectivity is extrapolated to obtain a preliminary prediction value, the radar echo data is converted by using the quantitative rainfall neural network prediction model, so that the predicted rainfall approaches to a true rainfall value, the problem of low accuracy of rainfall prediction based on linear motion is solved, and the prediction accuracy is improved.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention.
Detailed Description
For a better understanding of the present invention by those skilled in the art, the present invention will be described in further detail below with reference to the accompanying drawings and the following examples.
Example 1
As shown in fig. 1, the present embodiment provides a method for predicting quantitative precipitation based on radar reflectivity extrapolation, including:
s1: collecting radar base data for prediction, and generating a radar echo reflectivity image by using a reflectivity factor in the radar base data;
s2: solving a pre-constructed optical flow equation based on a radar echo reflectivity image to obtain a motion field of a radar echo, wherein when the optical flow equation is solved, a Lucas-Kanade method is used for carrying out local constraint on the optical flow equation to ensure smooth change of the optical flow;
s3: the method includes the steps that a pre-constructed identification model is used for identifying a strong convection area in a motion field of a radar echo, and a SCIT algorithm is used for tracking and positioning the strong convection area, wherein the identification model identifies the strong convection area in the motion field of the radar echo, and the method specifically comprises the following steps:
traversing all pixel points of the motion field of the radar echo, if the reflectivity intensity of the current pixel point is greater than a set threshold value, recording the intensity flag bit of the pixel point as 1, otherwise, recording as 0;
searching an eight-connected region of a region where the pixel point with the strength zone bit marked as 1 is located, if the area of the eight-connected region is not less than 20 pixel points, keeping the eight-connected region, and if not, deleting the eight-connected region;
detecting all the reserved eight-connected areas, judging whether the eight-connected areas are non-precipitation echoes caused by super refraction, if not, reserving the eight-connected areas, and otherwise, deleting the eight-connected areas;
processing the reserved eight-connected region by utilizing an opening operation, eliminating false connection and obtaining a final eight-connected region image;
extracting the characteristics of the final eight-connected region image, and making a decision on the extracted characteristics based on a set decision method to obtain a strong convection region;
s4: constructing a radar echo extrapolation model, training the radar echo extrapolation model by using a strong convection area, and predicting radar echo data;
the radar echo extrapolation model adopts an ST-LSTM-RNN network structure, and the ST-LSTM-RNN network structure comprises an ST-LSTM unit, a time memory module and a space memory module;
inputting data of a strong convection region into a plurality of ST-LSTM units which are connected in sequence, and outputting predicted radar echo data;
the time memory module and the space memory module are added into each ST-LSTM unit, the time memory module is used for memorizing and accumulating the first m moments of the neurons in the current layer, and the space memory module is used for memorizing and accumulating the first m moments of the neurons in different layers;
s5: converting radar echo data by using a trained quantitative precipitation neural network prediction model to obtain a predicted precipitation amount;
the quantitative precipitation neural network prediction model comprises an input layer, a second layer neural network, a third layer neural network and a fourth layer neural network which are sequentially connected, wherein the second layer neural network comprises two depth separable convolutional layers, and the output ends of the two depth separable convolutional layers are connected with the input end of the third layer neural network; the third layer of neural network comprises two depth separable convolutional layers, and the output ends of the two depth separable convolutional layers are connected with the input end of the fourth layer of neural network; the fourth layer of neural network comprises two convolutional layers; and training the quantitative precipitation neural network prediction model by using a large amount of radar echo data and actual precipitation data as a training data set, and optimizing and adjusting network parameters based on a preset loss function and an optimizer until the network converges to obtain the trained quantitative precipitation neural network prediction model.
The quantitative precipitation is predicted by combining the identification model, the radar echo extrapolation model and the quantitative precipitation neural network prediction model, after the radar reflectivity is extrapolated to obtain a primary predicted value, the radar echo data is converted by using the quantitative precipitation neural network prediction model, so that the predicted precipitation approaches to a true precipitation value, the problem of low accuracy of precipitation prediction based on linear motion is solved, and the prediction accuracy is improved.
The above description is only a preferred embodiment of the present invention, and not intended to limit the present invention, the scope of the present invention is defined by the appended claims, and all structural changes that can be made by using the contents of the description and the drawings of the present invention are intended to be embraced therein.
Claims (8)
1. A quantitative precipitation prediction method based on radar reflectivity extrapolation is characterized by comprising the following steps:
s1: collecting radar base data for prediction, and generating a radar echo reflectivity image by using a reflectivity factor in the radar base data;
s2: solving a pre-constructed optical flow equation based on the radar echo reflectivity image to obtain a motion field of the radar echo;
s3: identifying a strong convection region in a motion field of a radar echo by using a pre-constructed identification model, and tracking and positioning the strong convection region;
s4: constructing a radar echo extrapolation model, training the radar echo extrapolation model by using a strong convection area, and predicting radar echo data;
s5: and converting the radar echo data by using the trained quantitative precipitation neural network prediction model to obtain the predicted precipitation.
2. The method of claim 1, wherein when solving the optical flow equation in S2, the method of Lucas-Kanade is used to locally constrain the optical flow equation to ensure smooth change of optical flow.
3. The method of claim 1, wherein the identification model in S3 identifies strong convection regions in the motion field of the radar returns, and comprises:
traversing all pixel points of the motion field of the radar echo, if the reflectivity intensity of the current pixel point is greater than a set threshold value, recording the intensity flag bit of the pixel point as 1, otherwise, recording as 0;
searching an eight-connected region of a region where the pixel point with the strength zone bit marked as 1 is located, if the area of the eight-connected region is not less than 20 pixel points, keeping the eight-connected region, and if not, deleting the eight-connected region;
detecting all the reserved eight-connected areas, judging whether the eight-connected areas are non-precipitation echoes caused by super refraction, if not, reserving the eight-connected areas, and otherwise, deleting the eight-connected areas;
processing the reserved eight-connected region by utilizing an opening operation, eliminating false connection and obtaining a final eight-connected region image;
and extracting the characteristics of the final eight-connected region image, and deciding the extracted characteristics based on a set decision method to obtain a strong convection region.
4. The method of claim 1, wherein in step S3, the SCIT algorithm is used to track and locate the areas of strong convection.
5. The method of claim 1, wherein the model for radar echo extrapolation adopts a network structure of ST-LSTM-RNN.
6. The method of claim 5, wherein the method comprises the steps of: the ST-LSTM-RNN network structure comprises an ST-LSTM unit, a time memory module and a space memory module;
inputting data of a strong convection region into a plurality of ST-LSTM units which are connected in sequence, and outputting predicted radar echo data;
the time memory module and the space memory module are added into each ST-LSTM unit, the time memory module is used for memorizing and accumulating the first m moments of the neurons in the current layer, and the space memory module is used for memorizing and superposing the first m moments of the neurons in different layers.
7. The method of claim 1, wherein the quantitative precipitation neural network prediction model comprises an input layer, a second layer neural network, a third layer neural network and a fourth layer neural network which are connected in sequence, wherein the second layer neural network comprises two depth separable convolutional layers, and output ends of the two depth separable convolutional layers are connected with input ends of the third layer neural network; the third layer of neural network comprises two depth separable convolutional layers, and the output ends of the two depth separable convolutional layers are connected with the input end of the fourth layer of neural network; the fourth layer of neural network includes two convolutional layers.
8. The method of claim 7, wherein the quantitative rainfall neural network prediction model is trained using a large amount of radar echo data and actual rainfall data as a training data set, and the network parameters are optimally adjusted based on a preset loss function and an optimizer until the network converges to obtain the trained quantitative rainfall neural network prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911030327.2A CN110824481A (en) | 2019-10-28 | 2019-10-28 | Quantitative precipitation prediction method based on radar reflectivity extrapolation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911030327.2A CN110824481A (en) | 2019-10-28 | 2019-10-28 | Quantitative precipitation prediction method based on radar reflectivity extrapolation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110824481A true CN110824481A (en) | 2020-02-21 |
Family
ID=69550957
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911030327.2A Pending CN110824481A (en) | 2019-10-28 | 2019-10-28 | Quantitative precipitation prediction method based on radar reflectivity extrapolation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110824481A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111352113A (en) * | 2020-04-01 | 2020-06-30 | 易天气(北京)科技有限公司 | Strong convection weather short-term forecasting method and system, storage medium and terminal |
| CN112363140A (en) * | 2020-11-05 | 2021-02-12 | 南京叁云科技有限公司 | Thermodynamic constraint extrapolation objective correction method based on cyclic neural network |
| CN112698427A (en) * | 2020-12-09 | 2021-04-23 | 最美天气(上海)科技有限公司 | Short-term forecasting method and system based on space-time forecasting model |
| CN113327301A (en) * | 2021-05-25 | 2021-08-31 | 成都信息工程大学 | Strong convection extrapolation method and system based on depth analogy network under multi-dimensional radar data |
| CN115047463A (en) * | 2022-08-16 | 2022-09-13 | 成都远望科技有限责任公司 | Collaborative scanning scheduling method considering radar echo attenuation in rainfall area |
| CN115792847A (en) * | 2022-11-08 | 2023-03-14 | 江西师范大学 | Quantitative precipitation estimation method based on neural network and echo vertical information |
| CN117129963A (en) * | 2023-10-09 | 2023-11-28 | 北京大学重庆大数据研究院 | Radar echo extrapolation method based on migration prediction and readable storage medium thereof |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106707767A (en) * | 2017-03-13 | 2017-05-24 | 山东农业大学 | System and method for integrally and intelligently controlling water and fertilizer in field based on multi-source information fusion |
| CN107121679A (en) * | 2017-06-08 | 2017-09-01 | 湖南师范大学 | Recognition with Recurrent Neural Network predicted method and memory unit structure for Radar Echo Extrapolation |
| CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
| CN108508505A (en) * | 2018-02-05 | 2018-09-07 | 南京云思创智信息科技有限公司 | Heavy showers and thunderstorm forecasting procedure based on multiple dimensioned convolutional neural networks and system |
| CN109145983A (en) * | 2018-08-21 | 2019-01-04 | 电子科技大学 | A kind of real-time scene image, semantic dividing method based on lightweight network |
-
2019
- 2019-10-28 CN CN201911030327.2A patent/CN110824481A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106707767A (en) * | 2017-03-13 | 2017-05-24 | 山东农业大学 | System and method for integrally and intelligently controlling water and fertilizer in field based on multi-source information fusion |
| CN107121679A (en) * | 2017-06-08 | 2017-09-01 | 湖南师范大学 | Recognition with Recurrent Neural Network predicted method and memory unit structure for Radar Echo Extrapolation |
| CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
| CN108508505A (en) * | 2018-02-05 | 2018-09-07 | 南京云思创智信息科技有限公司 | Heavy showers and thunderstorm forecasting procedure based on multiple dimensioned convolutional neural networks and system |
| CN109145983A (en) * | 2018-08-21 | 2019-01-04 | 电子科技大学 | A kind of real-time scene image, semantic dividing method based on lightweight network |
Non-Patent Citations (3)
| Title |
|---|
| 刘海: "强对流天气识别与预报方法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 林青: "光流法以及结合机器学习在临近预报中的应用", 《海峡科学》 * |
| 韩丰: "循环神经网络在雷达临近预报中的应用", 《应用气象学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111352113A (en) * | 2020-04-01 | 2020-06-30 | 易天气(北京)科技有限公司 | Strong convection weather short-term forecasting method and system, storage medium and terminal |
| CN112363140A (en) * | 2020-11-05 | 2021-02-12 | 南京叁云科技有限公司 | Thermodynamic constraint extrapolation objective correction method based on cyclic neural network |
| CN112363140B (en) * | 2020-11-05 | 2024-04-05 | 南京叁云科技有限公司 | Thermodynamic constraint extrapolation objective correction method based on cyclic neural network |
| CN112698427A (en) * | 2020-12-09 | 2021-04-23 | 最美天气(上海)科技有限公司 | Short-term forecasting method and system based on space-time forecasting model |
| CN113327301A (en) * | 2021-05-25 | 2021-08-31 | 成都信息工程大学 | Strong convection extrapolation method and system based on depth analogy network under multi-dimensional radar data |
| CN113327301B (en) * | 2021-05-25 | 2023-04-07 | 成都信息工程大学 | Strong convection extrapolation method and system based on depth analogy network under multi-dimensional radar data |
| CN115047463A (en) * | 2022-08-16 | 2022-09-13 | 成都远望科技有限责任公司 | Collaborative scanning scheduling method considering radar echo attenuation in rainfall area |
| CN115047463B (en) * | 2022-08-16 | 2022-11-01 | 成都远望科技有限责任公司 | Collaborative scanning scheduling method considering radar echo attenuation in rainfall area |
| CN115792847A (en) * | 2022-11-08 | 2023-03-14 | 江西师范大学 | Quantitative precipitation estimation method based on neural network and echo vertical information |
| CN117129963A (en) * | 2023-10-09 | 2023-11-28 | 北京大学重庆大数据研究院 | Radar echo extrapolation method based on migration prediction and readable storage medium thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110824481A (en) | Quantitative precipitation prediction method based on radar reflectivity extrapolation | |
| Yang et al. | Automatic pixel‐level crack detection and measurement using fully convolutional network | |
| CN114693615A (en) | Deep learning concrete bridge crack real-time detection method based on domain adaptation | |
| CN112561191A (en) | Prediction model training method, prediction method, device, apparatus, program, and medium | |
| Ni et al. | A generative adversarial learning strategy for enhanced lightweight crack delineation networks | |
| CN115879535A (en) | Training method, device, equipment and medium for automatic driving perception model | |
| CN109409307B (en) | Online video behavior detection method based on space-time context analysis | |
| CN108830423B (en) | Annual runoff prediction method based on forecast domain | |
| CN112489081A (en) | Visual target tracking method and device | |
| US20230342938A1 (en) | Adaptive Semi-Supervised Image Segmentation Method Based on Uncertainty Knowledge Domain and System thereof | |
| Zhang et al. | Integrated APC-GAN and AttuNet framework for automated pavement crack pixel-level segmentation: A new solution to small training datasets | |
| CN112130086A (en) | Method and system for predicting residual life of power battery | |
| CN114283285A (en) | Cross consistency self-training remote sensing image semantic segmentation network training method and device | |
| CN111105093B (en) | Near precipitation prediction method based on CNN and BilSTM | |
| CN116758421A (en) | Remote sensing image directed target detection method based on weak supervised learning | |
| Sun et al. | Roadway crack segmentation based on an encoder-decoder deep network with multi-scale convolutional blocks | |
| Dong et al. | Pixel-level intelligent segmentation and measurement method for pavement multiple damages based on mobile deep learning | |
| Kumar et al. | Feasibility analysis of convolution neural network models for classification of concrete cracks in Smart City structures | |
| CN116682043B (en) | SimCLR-based unsupervised depth contrast learning abnormal video cleaning method | |
| CN117155806A (en) | Communication base station flow prediction method and device | |
| CN116862315A (en) | Low-efficiency industrial land performance scoring method, device, equipment and storage medium | |
| CN110532868A (en) | A method of prediction free space semanteme boundary | |
| CN115661694A (en) | Intelligent detection method, system, storage medium and electronic equipment for light-weight main transformer focusing on key characteristics | |
| CN115808690A (en) | Intelligent revision method and system for weather forecast | |
| CN114693776A (en) | Cable position information determining method, device, equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200221 |