Method and system for quickly and efficiently identifying mesoscale convection system in medium and low latitude areas

Abstract

The invention belongs to the field of image processing, and discloses a fast and efficient mesoscale convection system identification method and system for medium and low latitude areas, and the method comprises the steps: obtaining cloud top brightness temperature data; constructing a mesoscale convection system identification data set according to the cloud top brightness temperature data; constructing a mesoscale convection system identification model; training the model by using the constructed training set; and after a specified number of training rounds is reached, testing the model by using the test set to obtain a final mesoscale convection system identification result. According to the method, a fragment learning strategy is designed, so that the model can simultaneously learn the characteristics of the mesoscale convection system in different regions; a semi-residual multi-scale feature extractor is constructed to prevent the problem of small-scale feature loss in the feature extraction process; a semantic consistency regularization is designed to restrain feature asymmetry in redundant feature extraction and feature recovery processes. According to the method, the problem that the recognition speed is low when a traditional threshold value method is used in medium and low latitude areas is solved.

Description

Technical Field

[0001] The present invention belongs to, but is not limited to, the technical field of image processing, and particularly relates to a method and system for quickly and efficiently identifying mesoscale convective systems in mid - low latitude regions. Background Art

[0002] Deep moist convection is a common meteorological feature in many regions of the world and is usually formed by updrafts and downdrafts. Independent convective cells composed of these two airflows often interact with each other to form storm cells, storm lines, or storm clusters. This combination is called mesoscale convective systems (MCSs), which have a length scale greater than 100 kilometers in one or more directions and a duration of more than 3 hours. MCSs are the most important convective weather systems on Earth and are ubiquitous in tropical regions such as the western Pacific, tropical Africa, and the Amazon. They play an important role in the global energy and hydrological cycles because they vertically transport mass, momentum, and heat. Previous studies have shown that although they are only a small part of the wet convective systems, they contribute 50% - 90% of the tropical annual precipitation. In addition, MCSs are often closely related to disastrous weather events such as thunderstorms, floods, and strong winds. MCSs bring essential precipitation for agricultural production throughout the tropical and mid - latitude regions. Changes in the frequency of MCSs can directly affect agricultural production: too few MCSs can lead to a dry environment, while too many MCSs can lead to extreme weather events such as strong winds and floods. Given the hydrological importance and disastrousness of MCSs, it is extremely important to understand the regional distribution and spatial climatology characteristics of MCSs, especially in the context of climate change.

[0003] Due to the limitations of computing power, most of the previous studies on the identification of MCSs were regional or short-term. With the development of Earth-orbiting satellites in recent years, the high spatio-temporal resolution observations of clouds have made it possible to continuously capture the evolution of MCSs globally. Researchers have done a lot of work on identifying MCSs using two widely used identification methods. The first is to use geostationary satellite data to identify MCSs based on the characteristics of low cloud top brightness temperature. Generally, infrared satellite data in the 11-μm band is most commonly used to identify MCSs. Existing technologies identify MCSs in the tropical region (30°N - 30°S) based on temperature thresholds and area coverage thresholds of geostationary satellite data, and combine the area overlap method and the Kalman filtering method to track the evolution of MCSs. Further, existing technologies extend the identification area of MCSs to 60°N to 60°S, and use IMERG precipitation data to complete the identification and tracking of MCSs in the tropical and mid-latitude regions, supplementing the lack of information on the internal convective structure in geostationary satellite data. The focus of existing technologies is on the changes in MCS characteristics over land, ocean, and in the tropical and mid-latitude regions. The second is to use long-term satellite precipitation data to make additional measurements of cloud and precipitation structures to identify MCSs. A large number of previous studies have shown that the results of MCS identification and tracking obtained using GPM and IMERG precipitation data are consistent with those driven by radar. Existing technologies have identified and tracked MCSs in the IMERG precipitation field from 2014 to 2018, and tracked continuous regions larger than 1,000 km 2 as MCSs. In addition, existing technologies have used the Forward-in-Time algorithm (FiT) to track MCSs in the IMERG precipitation field in the global tropical region (30°N - 30°S) for 10 years (2011 - 2020).

[0004] Based on satellite data, both methods can accurately capture the evolution of short-term, local MCSs, including their germination, maturity, and demise. However, both methods are based on threshold methods, that is, by calculating the scale size and temperature of a single target to determine whether it is an MCS. When applied globally, especially in the tropical and mid-latitude regions, the threshold method takes a relatively long time to determine whether a target is an MCS. The huge time cost will slow down the speed of downstream tasks (tracking and forecasting), affecting the analysis of the climate characteristics of MCSs.

[0005] From the perspective of computer vision, the identification of mesoscale convective systems MCSs can be regarded as an image target recognition task. Visualize the cloud top brightness temperature data in the mid-latitude and low-latitude regions and the corresponding mesoscale convective system MCSs labels obtained based on the threshold method, such as Figure 1As shown. Target recognition is a basic task, that is, to identify the region of interest (foreground object) in the image from the image itself (background). The recognition of mesoscale convective systems (MCSs) can be regarded as a recognition task that combines salient object detection and camouflaged object detection.

[0006] The task of salient object detection aims to identify the most important and information-rich objects in a scene and then segment them at the pixel level. It has been applied to many vision problems, including image segmentation, object recognition, and visual tracking. Early deep learning methods for the SOD task extracted local patches of the image from single or multiple scales and classified each image pixel one by one into the salient or non-salient category. These methods usually produced rough output results because the fully connected layer could not effectively utilize the spatial information of the image. With the development of the fully convolutional neural network (FCN), FCN-based models have made significant progress in the SOD task. The prior art introduced R-dropout to construct an uncertain set of internal feature units, which enabled the model to learn deep uncertain convolutional features (UCFs), thus improving the robustness and accuracy of salient detection. The prior art constructed a deep level set network to generate a compact and uniform saliency map, thereby outputting a more precise boundary and a compact saliency. The prior art proposed a network that could combine salient prior knowledge and used a recursive architecture to enable the network to refine the saliency map by correcting previous errors. The prior art developed a new deep saliency network equipped with recursive aggregated deep features (RADF), which could more accurately detect salient objects in the image by leveraging complementary salient information captured by different layers. SAC-Net achieved salient object detection by fusing local and global image contexts inside, around, and outside the salient object. The prior art proposed a new progressive structure with a knowledge review network (PA-KRN) to simulate the global localization and local segmentation processes in human biology.

[0007] Different from the salient object detection task, the goal of camouflaged object detection is to identify objects from highly similar environments. The latest research has made remarkable progress in identifying camouflaged objects in a data-driven manner by leveraging the complex representation capabilities of deep learning models. The prior art proposed a two-branch (probability branch and main branch) network for camouflaged object detection, named Anabranch, which fuses the probability of camouflaged objects into the main branch for segmentation to improve segmentation accuracy. The prior art proposed a new framework that fully utilizes multiple visual cues (i.e., saliency and edges) to refine the prediction of camouflaged objects. The prior art established a two-stage model for camouflaged object detection, first finding the approximate region where camouflaged objects exist through a localization model, and then precisely segmenting the selected region through a segmentation model. The prior art proposed a new type of camouflaged object detection framework called CubeNet, which effectively fuses multi-layer features by introducing X connections. To address the problems of camouflaged objects with different scales and fuzzy appearances, the prior art proposed a hybrid-scale triple network - ZoomNet, which mimics the behavior of humans when observing blurred images. The prior art constructed a deep gradient network (DGNet), which is a new deep framework for camouflaged object detection using object gradient supervision. The prior art proposed a feature decomposition and edge reconstruction (FEDER) model for camouflaged object detection, which solves the problem of the inherent similarity between the foreground and the background by decomposing features into different frequency bands using learnable wavelets. The prior art proposed a novel learnable and separable frequency-aware mechanism to distinguish camouflaged objects from the background.

[0008] From the perspective of computer vision, the identification of MCSs in meteorology can be regarded as an image binary classification task, that is, separating MCSs from cloud-top brightness temperature data. However, the identification of MCSs globally is not a simple binary classification task. Specifically, for low-latitude regions (near the equator), MCSs have clearer contours and are well distinguishable from the background. We can consider the identification of MCSs in low-latitude regions as an SOD task. While in mid-latitude regions, the contours of MCSs are fuzzy and the distinguishability is low. We can regard the identification of MCSs in mid-latitude regions as a COD task. In addition, most MCSs in low-latitude regions are relatively scattered and the distance between adjacent MCSs is very close. While the monitoring in mid-latitude regions appears as a whole. Globally, MCSs in both mid-latitude and low-latitude regions have the characteristics of large-scale variations.

[0009] In view of the above analysis, the technical problems urgently needed to be solved in the prior art are: (1)Based on satellite data, traditional methods can accurately capture the evolution of MCSs in the short term and local areas, including their germination, maturity, and extinction. However, these methods are all based on thresholds, that is, by calculating the scale size and temperature of a single target to determine whether it is an MCS. When applied globally, especially in tropical and mid-latitude regions, the time required for the threshold method to determine whether a target is an MCS is quite long. The huge time cost will lead to slow progress in downstream tasks (tracking and prediction), affecting the analysis of the climatological characteristics of MCSs.

[0010] (2)Deep learning models have made significant progress in the field of target recognition, while the research in the field of mesoscale convection recognition is relatively blank. MCSs recognition is a mixed task, and most existing deep learning models are only for single significant target recognition or camouflaged target recognition tasks. Using a single significant target recognition (camouflaged target recognition) model to recognize MCSs will lead to unstable gradient backpropagation. How to use a single model to quickly and accurately recognize MCSs is the primary problem. In addition, the characteristics of large scale differences and close distances of MCSs also pose great challenges to recognition. Summary of the Invention

[0011] Aiming at the problems existing in the prior art, the present invention provides a method and system for quickly and efficiently identifying mesoscale convective systems in the mid-latitude and low-latitude regions.

[0012] The present invention is implemented as follows. A method for quickly and efficiently identifying mesoscale convective systems in the mid-latitude and low-latitude regions, characterized in that the method specifically includes: S1. Obtain the cloud top brightness temperature data of the mid-latitude and low-latitude regions; S2. Construct a mesoscale convective system recognition data set for the cloud top brightness temperature data, and the data set includes a training set and a test set; S3. Construct a mesoscale convective system recognition model; S4. Use the training set to train the mesoscale convective system recognition model; S5. Based on the trained mesoscale convective system recognition model, obtain the mesoscale convective system recognition result; S6. After S4 and S5 reach the specified number of training times in sequence, input the test set into the trained mesoscale convective system recognition model, and determine whether the current mesoscale convective system recognition model meets the index requirements. If so, save the parameters of the current model and enter S7; otherwise, return to S4; S7. Input the test set into the mesoscale convective system recognition model that meets the index requirements to obtain the final mesoscale convective system recognition result.

[0013] Further, the S2 includes the following steps: S201. Set the temperature threshold for mesoscale convective systems (MCSs), and mark the areas with temperatures lower than the threshold from the original cloud top brightness temperature data; S202. Set the regional coverage threshold for mesoscale convective systems (MCSs). For the areas marked in S201, further screen them using the regional coverage threshold. Select the areas in S201 with an area larger than the regional coverage threshold, and mark the places that meet the requirements of both thresholds as mesoscale convective system (MCSs) areas. Among them, the mesoscale convective system (MCSs) areas are marked as 1, and the non - mesoscale convective systems (MCSs) are marked as 0. After the above processing, the mesoscale convective system (MCSs) label is obtained, and the label size is 5143×1715; S203. Fill the null values in the cloud top brightness temperature data using the interpolation method. In addition, process the outliers in the data. Specifically, for the values in the cloud top brightness temperature data greater than 345, replace them with 345, and replace the values less than 170 with 170. The obtained cloud top brightness temperature data size is 5143×1715; S204. Perform edge padding on the mesoscale convective system (MCSs) label obtained in S202 and the cloud top brightness temperature data obtained in S203. Specifically, fill the cloud top brightness temperature data with a value of 300, and fill the label data with a value of 0. The sizes of both after padding are 5632×2048.

[0014] S205. According to the padded mesoscale convective system (MCSs) label and cloud top brightness temperature data obtained in S204, divide them into a training set and a test set. Among them, both the training set and the test set include cloud top brightness temperature data and mesoscale convective system (MCSs) labels, and the storage format is numpy.

[0015] Furthermore, the S3 includes the following steps: S301. Process the cloud top brightness temperature data obtained in S204 using max - min normalization and convert it to the tensor format, where the maximum value is taken as 345 and the minimum value is taken as 170. The MCS label data obtained in S204 remains unchanged; S302. Construct a mesoscale convective system recognition model according to the normalization result of S301.

[0016] Furthermore, the S4 includes the following steps: S401. For the cloud top brightness temperature data after normalization in S301, with a size of 5632×2048. First, split the data to obtain 11×4 data with a size of 512×512. Divide every 11 groups of data into a batch, and a total of 4 batch data are obtained, denoted as B i . Among them, the first batch B1 With the fourth batch B 4 is the cloud top brightness temperature data in the mid-latitude region, the second B 2 and the third batch B 3 are data in the low-latitude region; S402. Input the first batch B 1 into the multi-scale feature extractor SRFE of the semi-residual to extract hierarchical multi-scale pyramid features, and output the feature map Ei of each layer of the encoder; S403. In the decoder part, splice the feature map Ei output by the encoder and the feature map Di output by the decoder in the channel dimension and then input it into the convolution and perform upsampling to obtain Di, where Di represents the feature restored by the i-th layer of the decoder; S404. Input the output of the last layer of the decoder into the convective recognition head SegHead to obtain the recognition result of the mesoscale convective system; S405. When optimizing the model by backpropagation, we use semantic consistency regularization to optimize the model; S406. Repeat S401 - S405 until the 4 batches of data B i are all input into the model.

[0017] Furthermore, the model includes: An encoder for extracting hierarchical pyramid features after being normalized by S301. Among them, the encoder includes four multi-scale feature extractors SRFE of semi-residuals; the multi-scale feature extractor SRFE of semi-residuals is used to extract multi-scale features to prevent small-scale features from being lost during downsampling. The semi-residual design can prevent the forward propagation of redundant features; A decoder for decoding and restoring pyramid features at different levels. Among them, the decoder includes a convolution and an upsampling module.

[0018] Furthermore, S402 includes the following steps: S4021. Input the first batch B 1 into the first multi-scale feature extractor SRFE of semi-residuals of the encoder. First, duplicate the input B 1 4 times in the channel dimension. Use a convolution with a size of 1×1 to extract features from the first copy to obtain the first branch feature map F 1 , use a convolution with a size of 3×3 to extract features from the second copy to obtain the second branch feature map F 2 , use a convolution with a size of 5×5 to extract features from the third copy to obtain the third branch feature map F 3 , use a convolution with a size of 7×7 to extract features from the fourth copy to obtain the fourth branch feature map F 4 ; S4022. Concatenate the four feature maps F obtained in S4021 along the channel dimension, and then perform multi-scale feature fusion using a 1×1 convolution to obtain a preliminary fusion feature map M 1 ,F 2 ,F 3 ,F 4 ; 1 ; S4023. Split the preliminary fusion feature map M into four parts along the channel dimension. Use a 1×1 convolution to extract features from the first part to obtain a first branch feature map F 1 ,use a 3×3 convolution to extract features from the second part to obtain a second branch feature map F 11 ,use a 5×5 convolution to extract features from the third part to obtain a third branch feature map F 22 ,and use a 7×7 convolution to extract features from the fourth part to obtain a fourth branch feature map F 33 ; 44 ; S4024. Concatenate the four feature maps F obtained in S4023 along the channel dimension, and then perform multi-scale feature fusion using a 1×1 convolution to obtain a final fusion feature map M 11 ,F 22 ,F 33 ,F 44 ,add M 11 and M 11 to obtain the final output E of the first half-residual multi-scale feature extractor SRFE 1 ; 1 ; S4025. Input the final output E1 of the first half-residual multi-scale feature extractor SRFE into the next half-residual multi-scale feature extractor SRFE, and perform the operations of S4021 - S4025 to obtain the final output E2 of the second half-residual multi-scale feature extractor SRFE

[0019] S4026. Repeat S4025 twice to obtain the final outputs E 3 and E 4 of the third and fourth half-residual multi-scale feature extractors SRFE; Further, S405 includes the following steps: S4051. For the output Ei of each layer of the encoder, calculate the L2 norm of each channel, reorder Ei by channel according to the value size, and represent the reordered result as Es_i; S4052. Divide the re - ordered result Es_i into two parts on the channel. The first half is the part with large feature differences, and the second half is the part with small feature differences. By calculating the losses of these two parts, the model is constrained to extract more diverse features. The formula is as follows; Among them, represents the intra - class semantic consistency regularization loss, represents the result of the first half of the features extracted by the i - th layer decoder after re - ordering, represents the result of the second half of the features extracted by the i - th layer decoder after re - ordering, and N represents the total number of decoder layers.

[0020] S4053. Downsample the last - layer feature D4 restored by the S403 decoder so that the scale of D4 is the same as the scale of the features Ei extracted by each layer of the encoder. The four downsampled features are respectively denoted as d 1 , d 2 , d 3 , d 4 ; S4054. For the feature Ei extracted by each layer of the encoder, perform random channel selection so that the number of channels of Ei is the same as that of D4. The features after random channel selection are denoted as e1, e2, e3, e4; S4055. After S4053 and S4054, d 1 , d 2 , d 3 ,d 4 and e 1 , e 2 , e 3 , e 4 will be obtained, which have the same spatial dimension and channel dimension. By calculating the mean square error of their corresponding positions, the symmetry of the features extracted by the model and the features in the decoding process is constrained. The formula is as follows; Among them, represents the encoder - decoder semantic consistency regularization loss, represents the result of the output of each layer of the encoder after random channel selection, represents the results of D4 after four - time downsampling respectively, and N represents the total number of encoder layers.

[0021] Furthermore, the expression of the loss function of the model is as follows: Among them, represents the total loss, Represents the regular loss of codec semantic consistency, Represents the regular loss of intra-class semantic consistency, Represents the weighted binary cross-entropy loss function, Represents the dice coefficient loss function. Output represents the recognition result of the mesoscale convective image, and mask represents the labeled data.

[0022] Another object of the present invention is to provide a fast and efficient mesoscale convective system recognition system in the mid-low latitude region. The system specifically includes: An input module for obtaining cloud top brightness temperature data in the mid-low latitude region; A preprocessing module for constructing a mesoscale convective system recognition dataset and preprocessing it; A model construction module for constructing a mesoscale convective system recognition model; A training module for training the mesoscale convective system recognition model; An output module for inputting the test set into the mesoscale convective system recognition model that meets the index requirements to obtain the final mesoscale convective system recognition result.

[0023] Combined with the above technical solutions and solved technical problems, the advantages and positive effects of the technical solutions to be protected by the present invention are: First, a fast and efficient mesoscale convective system recognition method provided by the present invention solves the problems of large computational resource consumption, slow recognition speed, and small recognition area in the recognition of mesoscale convective systems MCSs by traditional methods, and solves the problem that most existing deep learning models are only for single significant target recognition or camouflaged target recognition tasks.

[0024] The present invention constructs a model for rapidly and efficiently identifying mesoscale convective systems in the mid - low latitude region, and uses the constructed dataset for identifying mesoscale convective systems (MCSs) as the training set to train the model. First, a multi - scale feature extractor with four semi - residuals (SRFE) is used to extract hierarchical multi - scale features, which improves the model's perception ability of MCSs at different scales with fewer parameters, prevents the loss of small - scale features during downsampling, and provides sufficient guidance for accurately identifying MCSs at different scales. In addition, the design of semi - residuals can prevent the forward propagation of redundant features. The present invention proposes a piece - wise learning strategy. By segmenting the MCSs in the mid - latitude and low - latitude regions and inputting the segments into the model in sequence for learning, the model can learn the features of MCSs in the mid - latitude (low - latitude) region without being affected by the features of MCSs in the low - latitude (mid - latitude) region. In this way, the model can stably learn the feature expressions of MCSs in different regions. Finally, we propose semantic consistency regularization to constrain the features of the model in terms of the diversity of feature extraction, and to constrain the model to extract features symmetric to those obtained in the encoding and decoding processes.

[0025] Second, as the creative auxiliary evidence of the claims of the present invention, it is also reflected in the following important aspects: (1) The technical solution of the present invention fills the technical gaps in the domestic and international industries: Currently, most methods in the field of MCSs identification are based on thresholds, and the threshold method has a slow speed for MCSs identification. The present invention uses deep learning technology for MCSs identification, greatly improving the speed of MCSs identification.

[0026] (2) The technical solution of the present invention solves the technical problems that people have been eager to solve but have never succeeded in: Limited by computing resources and the time cost of MCSs identification, most existing MCSs identifications are limited to local areas. There is less research on the identification of MCSs in a large range (mid - low latitude region). The technical solution of the present invention can meet the identification requirements of MCSs in a large range. Description of the Drawings

[0027] Figure 1 It is a schematic diagram of the dataset for identifying mesoscale convective systems (MCSs) in the mid - low latitude region provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the network structure provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of the multi - scale feature extractor with semi - residuals (SRFE) provided by an embodiment of the present invention; Figure 4 It is a flowchart of the method for rapidly and efficiently identifying mesoscale convective systems in the mid - low latitude region provided by an embodiment of the present invention; Figure 5 It is a schematic diagram of the structure of the decoder provided by an embodiment of the present invention; Figure 6 It is a schematic structural diagram of the calculation method of semantic consistency regularization provided by an embodiment of the present invention; Figure 7 It is a module diagram of a fast and efficient mesoscale convective system identification system in the mid - low latitude region provided by an embodiment of the present invention; Figure 8 It is a precision - recall curve and a threshold - F - measure curve on the mesoscale convective system MCSs identification dataset in the mid - low latitude region provided by an embodiment of the present invention; Figure 9 It is a qualitative result diagram on the mesoscale convective system MCSs identification dataset in the mid - low latitude region provided by an embodiment of the present invention; Figure 10 It is a spatio - temporal qualitative result diagram on the mesoscale convective system MCSs identification dataset in a local area provided by an embodiment of the present invention; Figure 11 It is another spatio - temporal qualitative result diagram on the mesoscale convective system MCSs identification dataset in a local area provided by an embodiment of the present invention. Detailed implementation manners

[0028] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0029] Before describing the present invention, the following terms are explained: MCSs: Mesoscale Convective Systems, mesoscale convective systems.

[0030] SRFE: Semi - residual Feature Extractor, semi - residual multi - scale feature extractor.

[0031] The present invention proposes a mesoscale convective system identification model (MCSeg), Figure 2Schematic diagram of the MCSeg network structure proposed by the present invention. Its input data is Input, and the output is the recognition result map Output. Among them, the resolution of Input is C×W×H, where C represents the number of channels, H represents the height of the image, and W represents the width of the image. The resolution sizes of the dataset are: 1×5143×1715. The present invention proposes a shard learning strategy, a semi-residual multi-scale feature extractor SRFE, and semantic consistency regularization. The shard learning strategy shards the MCSs in the middle and low latitudes and inputs the shards into the model in sequence for learning, so that the model can learn the features of the middle (low) latitude MCSs without being affected by the features of the low (middle) latitude MCSs. In this way, the model can stably learn the feature expressions of different regions of MCSs, improve the stability during model training, and accelerate model convergence. The semi-residual multi-scale feature extraction SRFE ( Figure 3 ) extracts hierarchical multi-scale features, improves the model's perception ability of MCSs at different scales with fewer parameters, prevents small-scale features from being lost during downsampling, and provides sufficient guidance for accurately identifying MCSs at different scales. Semantic consistency regularization is used to constrain the model's diverse features during feature extraction and to constrain the model to extract features symmetric to those obtained during the decoding process.

[0032] As Figure 4 shown, an embodiment of the present invention provides a method for quickly and efficiently identifying mesoscale convective systems in the middle and low latitude regions. The method specifically includes: S1. Obtain the cloud top brightness temperature data of the middle and low latitude regions; S2. Construct a mesoscale convective system recognition dataset for the cloud top brightness temperature data. The dataset includes a training set and a test set; S3. Construct a mesoscale convective system recognition model; S4. Use the training set to train the mesoscale convective system recognition model; S5. Based on the trained mesoscale convective system recognition model, obtain the mesoscale convective system recognition result; S6. Based on S4 and S5, when the input data volume reaches 1 / 5 of the total data volume, the test set is passed into the mesoscale convective system recognition model for testing. Determine whether the current Dice index of the network test is the highest. If so, save the current model parameters and enter S7. Otherwise, return to S4; S7. Input the test set into the mesoscale convective system recognition model that meets the index requirements to obtain the final mesoscale convective system recognition result.

[0033] Retrieve the cloud top brightness temperature data of the mid - low latitude region through meteorological satellite remote sensing technology. The cloud top brightness temperature reflects the intensity and scope of the convective system. After processing the spatial and temporal resolutions of these data, they are used to construct an identification dataset for mesoscale convective systems. The dataset is divided into a training set and a test set for model training and validation respectively.

[0034] Construct a deep learning model suitable for the identification of mesoscale convective systems, such as a convolutional neural network (CNN). This model can extract spatial and texture features from the input cloud top brightness temperature data. Use the data in the training set to train the model. The model continuously adjusts its parameters through the backpropagation algorithm to minimize the loss function. During the training process, the model learns how to distinguish mesoscale convective systems from other weather systems.

[0035] During the model training process, whenever the input data volume reaches 1 / 5 of the total data volume, input the test set into the model for testing and calculate the Dice metric. The Dice metric is used to evaluate the accuracy and consistency of the model in segmenting mesoscale convective systems. If the Dice metric of the current test reaches the highest value, save the current model parameters; otherwise, return to continue training and optimize the model parameters to ensure the continuous improvement of the model's performance on the test set.

[0036] When the Dice metric of the model on the test set meets the requirements, input the entire test set into the trained identification model for comprehensive identification of mesoscale convective systems. The model accurately identifies the location and scope of mesoscale convective systems by extracting features and classifying the input data, and generates an identification result map. The final result provides an efficient and accurate analysis tool for meteorological research and disaster warning.

[0037] In step S1, retrieve the original cloud top brightness temperature data. In this embodiment, retrieve the original global cloud top brightness temperature data (European Union Cloud Archive User Service (CLAUS) project dataset), and the data range is (60°S–60°N, 180°W–180°E).

[0038] Step S2 includes: S201: Set the temperature threshold for mesoscale convective systems (MCSs) and mark the areas with temperatures lower than the threshold from the original cloud top brightness temperature data. S202. Set the regional coverage threshold for mesoscale convective systems (MCSs). For the regions marked in S201, further screening is carried out using the regional coverage threshold. Select the regions in S201 with an area larger than the regional coverage threshold, and mark the places that meet the requirements of both thresholds as the regions of mesoscale convective systems (MCSs). Among them, the regions of mesoscale convective systems (MCSs) are marked as 1, and the non-mesoscale convective systems (MCSs) are marked as 0. After the above processing, the mesoscale convective system (MCSs) label is obtained, and the label size is 5143×1715; S203. Fill the null values in the cloud top brightness temperature data using the interpolation method. In addition, process the outliers in the data. Specifically, those with a value greater than 345 in the cloud top brightness temperature data are replaced with 345, and those with a value less than 170 are replaced with 170. The obtained cloud top brightness temperature data size is 5143×1715; S204. Perform edge filling on the mesoscale convective system (MCSs) label obtained in S202 and the cloud top brightness temperature data obtained in S203. Specifically, fill the cloud top brightness temperature data with a value of 300, and fill the label data with a value of 0. The size after filling for both is 5632×2048.

[0039] S205. Divide the filled mesoscale convective system (MCSs) label and the cloud top brightness temperature data obtained in S204 into a training set and a test set. Among them, both the training set and the test set include the cloud top brightness temperature data and the mesoscale convective system (MCSs) label, and the storage format is numpy.

[0040] Through the above design, the present invention can construct the data set required for training the described mesoscale convective system recognition model.

[0041] In this embodiment, the specific implementation steps of the mesoscale convective system recognition model (MCSeg) proposed by the present invention are as follows: A1. Identify mesoscale convective systems (MCSs) from the original cloud top brightness temperature data. In this step, the threshold method is used to identify and label the mesoscale convective systems (MCSs) in the cloud top brightness temperature data. First, set the temperature threshold for mesoscale convective systems (MCSs) to 233K, and preliminarily label the regions with a temperature less than 233K.

[0042] A2. Set the regional coverage threshold for mesoscale convective systems (MCSs) to 5000 km 2 , and further screen the regions marked in A1. Select the regions with a coverage area larger than 5000 km 2 , and mark these regions as mesoscale convective systems (MCSs). Among them, the regions of mesoscale convective systems (MCSs) are marked as 1, and the regions of non-mesoscale convective systems (MCSs) are marked as 0.

[0043] A3. Convert the original cloud top brightness temperature data and the labeled data obtained in A2 into the numpy format. The size of a single image is 5143×1715, and there are a total of 4392 groups of data.

[0044] A4. Replace the values in the cloud top brightness temperature data whose median is greater than 345 with 345, and replace the values whose median is less than 170 with 170. The size of the obtained cloud top brightness temperature data is 5143×1715.

[0045] A5. Divide the training set and the test set. The data volumes of the finally obtained training set and test set are 2920 and 1472 respectively.

[0046] S3. Build a mesoscale convective system recognition model, and its implementation method is as follows: S301. Process the cloud top brightness temperature data obtained in S204 using max-min normalization and convert it into the tensor format, where the maximum value is taken as 345 and the minimum value is taken as 170. The MCS label data obtained in S204 remains unchanged; S302. Build a mesoscale convective system recognition model according to the normalization result of S301.

[0047] The present invention preprocesses the constructed data set to improve the stability during model training and accelerate model convergence.

[0048] In this embodiment, the mesoscale convective system recognition model includes: An encoder for extracting hierarchical pyramid features after the normalization process of S301. Among them, the encoder includes four semi-residual multi-scale feature extractors SRFE; the semi-residual multi-scale feature extractor SRFE is used to extract multi-scale features to prevent small-scale features from being lost during downsampling. The semi-residual design can prevent the forward propagation of redundant features; A decoder for decoding and restoring pyramid features at different levels. Among them, the decoder includes a convolution and an upsampling module. The structural schematic diagram of the decoder is as Figure 5 shown.

[0049] In this embodiment, build and initialize the mesoscale convective system recognition model, and send the training set preprocessed by S2 into the small target recognition network, specifically including: B1. Convert the cloud top brightness temperature data into the tensor format tensor, and normalize all pixel values in the cloud top brightness temperature data to 0 to 1 using max-min normalization; B3. Mesoscale convective system recognition model; In this embodiment, the mesoscale convective system recognition model includes an encoder with a multi-scale feature extractor SRFE embedded with semi-residuals and a decoder, and the overall model is an encoder-decoder architecture. The encoder of the mesoscale convective system recognition model consists of four multi-scale feature extractors SRFE with semi-residuals, which are used to extract multi-scale hierarchical pyramid features of the input data. The decoder is stacked by convolutional and upsampling modules and an object recognition head, which is used to decode and recover the pyramid features at different levels step by step, and the recognition head is used for final recognition.

[0050] In this embodiment, the input of the mesoscale convective system recognition model is the cloud top brightness temperature data obtained from S205, denoted as Input. First, Input is split to obtain 11×4 data with a size of 512×512. Every 11 groups of data are divided into a batch, and a total of 4 batch data are obtained, denoted as B i . Among them, the first batch B 1 and the fourth batch B 4 are cloud top brightness temperature data in the mid-latitude region, and the second B 2 and the third batch B 3 are data in the low-latitude region. B i is passed into the encoder to extract multi-scale feature information, and the feature maps of each layer are obtained, which are respectively denoted as feature map Ei, where i represents the i-th layer of the encoder. Subsequently, the feature map Ei is passed into the decoder. The decoder takes the output E4 of the last layer of the encoder as the input, and obtains the output D1 of the first layer of the decoder. The input of the second layer of the decoder comes from the output D1 of the first layer of the decoder and the output E3 of the third layer of the encoder. The input of the third layer of the decoder comes from the output D2 of the second layer of the decoder and the output E2 of the second layer of the encoder. The input of the fourth layer of the decoder comes from the output D3 of the third layer of the decoder and the output E1 of the first layer of the encoder. Finally, the output of the fourth layer of the decoder is input into the recognition head to obtain the final mesoscale convective system MCSs recognition map Output. Finally, the difference degree between the predicted map Output and the original label mask is constrained by the Dice coefficient loss function, binary cross-entropy loss, and semantic consistency regularization loss, and the calculated loss value is passed into the network for backpropagation to automatically adjust the parameters of the network.

[0051] Figure 6 It is a schematic structural diagram of the calculation method of the semantic consistency regularization proposed by the present invention.

[0052] In this embodiment, when the input data volume reaches 1 / 5 of the total data volume, the test set is passed into the mesoscale convective system recognition model for testing to determine whether the Dice index of the current network test is the highest. If so, the current model parameters are saved; otherwise, training continues. Dice represents the Dice coefficient.

[0053] S4. Use the training set to train the mesoscale convective system recognition model, and its implementation method is as follows: S401. For the cloud top brightness temperature data after normalization in S301, the size is 5632×2048. First, split the data to obtain 11×4 data with a size of 512×512. Divide every 11 groups of data into a batch, and a total of 4 batches of data are obtained, denoted as B i . Among them, the first batch B 1 and the fourth batch B 4 are the cloud top brightness temperature data in the mid-latitude region, and the second B 2 and the third batch B 3 are the data in the low-latitude region; S402. Input the first batch B 1 into the semi-residual multi-scale feature extractor SRFE to extract hierarchical multi-scale pyramid features, and output the feature maps of each layer of the encoder , and its implementation method is as follows: S4021. Input the first batch B 1 into the first semi-residual multi-scale feature extractor SRFE of the encoder. First, duplicate the input B 1 4 times in the channel dimension. Use a convolution with a size of 1×1 to extract features from the first copy to obtain the first branch feature map F 1 , use a convolution with a size of 3×3 to extract features from the second copy to obtain the second branch feature map F 2 , use a convolution with a size of 5×5 to extract features from the third copy to obtain the third branch feature map F 3 , and use a convolution with a size of 7×7 to extract features from the fourth copy to obtain the fourth branch feature map F 4 ; S4022. Concatenate the four feature maps F 1 , F 2 , F 3 , F 4 obtained in S4021 in the channel dimension, and then use a 1×1 convolution for multi-scale feature fusion to obtain the preliminary fusion feature map M 1 S4023. Split the preliminary fusion feature map M 1 into 4 copies in the channel dimension. Use a convolution with a size of 1×1 to extract features from the first copy to obtain the first branch feature map F 11 , use a convolution with a size of 3×3 to extract features from the second copy to obtain the second branch feature map F 22 , use a convolution with a size of 5×5 to extract features from the third copy to obtain the third branch feature map F 33, the fourth part is used to extract features through a 7×7 convolution to obtain the fourth branch feature map F 44 ; S4024. Concatenate the four feature maps F 11 , F 22 , F 33 , F 44 on the channel dimension and then use a 1×1 convolution for multi-scale feature fusion to obtain the final fusion feature map M 11 . Add M 11 to M 1 to obtain the final output E of the first half-residual multi-scale feature extractor SRFE 1 .

[0054] S4025. Input the final output E1 of the first half-residual multi-scale feature extractor SRFE into the next half-residual multi-scale feature extractor SRFE, and perform the operations of S4021 - S4025 to obtain the final output E of the second half-residual multi-scale feature extractor SRFE 2 .

[0055] S4026. Repeat S4025 twice to obtain the final outputs E 3 and E 4 ; S403. In the decoder part, upsample the feature map Di output by the decoder, concatenate it with the feature map Ei output by the encoder on the channel dimension, and then input it into a convolution and perform upsampling to obtain Di, where Di represents the feature restored by the i-th layer of the decoder; S404. Input the output of the last layer of the decoder into the convective recognition head SegHead to obtain the recognition result of the mesoscale convective system.

[0056] S405. When optimizing the model through backpropagation, we use semantic consistency regularization to optimize the model, and the implementation method is as follows: S4051. For the output Ei of each layer of the encoder, calculate the L2 norm of each channel, reorder the channels according to the value size, and represent the reordered result as Es_i; S4052. Divide the reordered result Es_i into two parts on the channel. The first half is the part with large feature differences, and the second half is the part with small feature differences. Calculate the losses of these two parts to constrain the model to extract more diverse features; Among them, represents the intra-class semantic consistency regularization loss, It represents the first half of the result after reordering the features extracted by the i-th layer decoder. It represents the second half of the result after reordering the features extracted by the i-th layer decoder, and N represents the total number of layers of the decoder.

[0057] S4053. Downsample the last layer of features D4 restored by the S403 decoder so that the scale of D4 is the same as the scale of the features Ei extracted by each layer of the encoder. The four downsampled features are respectively denoted as d 1 , d 2 , d 3 , d 4 ; S4054. For the features Ei extracted by each layer of the encoder, perform random channel selection so that the number of channels of Ei is the same as that of D4. The features after random channel selection are denoted as d 1 , d 2 , d 3 , d 4 ; S4055. After S4053 and S4054, d 1 , d 2 , d 3 ,d 4 and e 1 , e 2 , e 3 , e 4 will be obtained. The symmetry of the features extracted by the model and the features in the decoding process is constrained by calculating the mean square error of their corresponding positions; Among them, represents the regular loss of the semantic consistency of the encoder-decoder, represents the result after random channel selection of the output of each layer of the encoder, represents the results of D4 being downsampled four times respectively, and N represents the total number of layers of the encoder.

[0058] S406. Repeat S402 - S405 until the 4 batches of data B i are all input into the model.

[0059] The present invention prevents the loss of small-scale features in the feature extraction process by designing a semi-residual multi-scale feature extractor SRFE. The semi-residual method can enable the model to more effectively transmit the information flow and prevent the transmission of redundant features.

[0060] Constrains the redundant feature extraction and the feature asymmetry in the feature recovery process through the intra-class semantic consistency regular loss and the encoder-decoder semantic consistency regular loss.

[0061] In this embodiment, the loss between the final predicted segmentation map Output of calculation step S5 and the label data mask in step S302 is calculated. The expression of the loss function of the mesoscale convective system recognition model is as follows: Among them, represents the total loss, represents the encoder-decoder semantic consistency regularization loss, represents the intra-class semantic consistency regularization loss, represents the weighted binary cross-entropy loss function, represents the dice coefficient loss function, Output represents the mesoscale convective image recognition result, and mask represents the label data.

[0062] Through the above loss function, the present invention can optimize the mesoscale convective system recognition model and accelerate the model convergence.

[0063] In this embodiment, the mesoscale convective system recognition model with the best performance during the test process is taken out, and then the test set is sequentially fed into this network. Seven quantitative indicators are used to evaluate the performance of this model: average dice coefficient mDice, average intersection over union mIoU, weighted F-index , structural similarity , enhanced alignment index E, mean absolute error , and maximum F-index , and the MCSs recognition map Output is saved. It is used to evaluate the average value of the difference between the predicted value and the true value. The smaller the difference, the closer the predicted result is to the true result. represents the best performance of the segmentation result. is the weighted average of precision and recall, and evaluates the overall performance of the recognition result by balancing precision and recall. It is used to evaluate the structural similarity between the recognition result and the true result. E measures the average structural similarity between the recognition result and the true result. mIoU is used to compare the similarity and diversity of the sample set. mDice is used to calculate the similarity of two samples. Except for M, all the above indicators are closer to 1, indicating a better recognition result. The specific calculation formulas of the above indicators are as follows: Among them, represents the set of network prediction results of the mesoscale convective system identification model MCSeg, represents the set of corresponding true labels, K represents the number of categories, P represents the precision rate, and R represents the recall rate. and respectively represent the regional structural similarity and the structural similarity of the object. represents the enhanced alignment matrix, and w and h respectively represent the width and height of the output result.

[0064] As Figure 7 shown, an efficient mesoscale convective system identification system in the mid - low latitude region provided by an embodiment of the present invention specifically includes: An input module for obtaining the cloud top brightness temperature data of the mid - low latitude region; A pre - processing module for constructing a mesoscale convective system identification data set and pre - processing it; A model construction module for constructing a mesoscale convective system identification model; A training module for training the mesoscale convective system identification model; An output module for inputting the test set into the mesoscale convective system identification model that meets the index requirements to obtain the final mesoscale convective system identification result.

[0065] The present invention belongs to the field of image processing, and discloses a method and system for efficiently identifying mesoscale convective systems in the mid - low latitude region, including obtaining cloud top brightness temperature data; constructing a mesoscale convective system identification data set based on the cloud top brightness temperature data; constructing a mesoscale convective system identification model; training the model using the constructed training set; after reaching the specified number of training rounds, testing the model with the test set to obtain the final mesoscale convective system identification result.

[0066] To verify the effectiveness of the method of the present invention, the proposed method is compared with other advanced methods. For a fair comparison, the official released codes of other methods are used and the experimental settings are followed, where all methods are implemented in the same computing environment and quantitative and qualitative analyses are carried out simultaneously. Table 1 gives the quantitative comparison results of seven indicators on 11 different network structures of the mesoscale convective systems MCSs data set.

[0067] Table 1 As shown in Table 1, this table illustrates the performance of our model in MCSs recognition. It can be seen that the model (MCSeg) proposed in the present invention is superior to other models in all metrics. Specifically, on the MCSs recognition dataset, MCSeg performs excellently in each of the seven metrics: it is 0.0016 in and 0.9692 in , 0.9866 in E, 0.9809 in , 0.9624 in , 0.9119 in IoU, and 0.9539 in Dice. In addition, in this embodiment, a precision-recall curve and an F-measure curve are also provided to further evaluate the performance of all methods on the MCSs recognition dataset, as shown in Figure 8 . It can be seen that the red solid line represents MCSeg, which is superior to other models at most thresholds. The excellent performance of this model is attributed to the carefully designed semi-residual encoder and semantic consistency regularity constraint. The semi-residual encoder enhances the multi-scale feature extraction ability of the model without increasing the number of parameters. At the same time, the semantic consistency regularity further improves the ability of the model to extract diverse features.

[0068] In this embodiment, in order to verify the generalization ability of our model, we performed MCSs recognition on the brightness temperature data for a total of 4 years from 2011 to 2014, and the results are shown in Tables 2, 3, 4, and 5. All 7 metrics are similar to the results obtained from the test set, which indicates that our model has good generalization ability. Although the model was only trained with the data in 2022, it still performs well in the convective recognition results from 2011 to 2014, which is a large time span. It can be concluded that our model accurately learns the feature distribution of MCSs, and the performance of the model does not decline over the years.

[0069] Table 2 Table 3 Table 4 Table 5 Table 6 In this embodiment, in order to verify the effectiveness of the module design, learning strategy, and loss function, we conducted ablation experiments, as shown in Table 6. In the table, M represents the multi-scale residual semantic encoder (MRSE), S represents the sharding learning strategy, and L represents the semantic consistency regularization. Taking the [indicator] as an example, after adding the multi-scale residual semantic encoder (MRSE), the performance of the baseline model improved from 0.0027 to 0.0019. After adding the slice training strategy, the [indicator] was further improved to 0.0020, and the addition of semantic consistency regularization further increased it to 0.0024. By gradually integrating these components into the baseline model, the overall performance of the model was significantly improved.

[0070] In this embodiment, Figure 9 shows the recognition results of MCSs in the mid-low latitude region at 00:00 on March 1, 2023. Overall, there is little difference between the results of the MCSs recognized by the model proposed in the present invention and the threshold-based method. Based on the slice training strategy we designed, the model can well learn the characteristics of MCSs in different latitudes. The cross-learning method enables the model to balance the low-latitude region and the mid-latitude region, avoiding the model forgetting the features learned previously. To more clearly observe the differences, we plotted a difference map ( Figure 9 ). In the difference map, the green area represents the over-recognized area (non-MCS is recognized as MCS), and the red area represents the under-recognized area (MCS is not accurately recognized). We selected four regions (including mid-latitude and low-latitude regions) for magnification, and some over-recognition and under-recognition situations can be observed. In region 1, most of the misrecognized areas are located at the edge of the main body, which is the result of too large a difference in the convection pattern. In region 2, most of the MCSs are isolated from each other. In regions 3 and 4, the misrecognized areas are independent of the main body, which can be attributed to the too large scale difference of MCSs. Since a single image is too large, we did not compare the recognition results of the mid-low latitude regions of other models here.

[0071] In this embodiment, Figure 10 and Figure 11 show the spatio-temporal recognition performance of all models (recognition of consecutive time periods). The selected time period is 00:00 - 21:00 on March 1, 2023, and there are a total of eight frames of convection data. From the graph, the recognition results of our model are closer to the true labels. Some relatively good models, such as SwinU and CPD, have better edge recognition ability for MCSs, but have more serious recognition errors for small-scale MCSs. The remaining models (C2FNet, HitNet, Poly-PVT, PraNet, SINet) can only recognize the main part of MCSs and are not detailed enough in the recognition of the edge region. In the edge recognition of MCSs, the problems of over-recognition and adjacent MCSs sticking to each other are more serious. Compared with these methods, our model can accurately locate and recognize the content and morphological structure of MCSs in complex scenarios.

[0072] In terms of both quantitative and qualitative results, our model has performed excellently. Regarding the recognition results, the MCSs identified by the model are very similar to those obtained by the threshold-based method. The use of deep learning technology has made our method much faster than the threshold-based method in terms of recognition speed. It takes 2 hours and 42 minutes to identify MCSs with a time span of one month using the threshold method, while it only takes 1 minute and 40 seconds using our method, and the recognition speed has increased by 200 times.

[0073] In summary, based on satellite data, traditional threshold methods can accurately capture the evolution of MCSs in short-term and local regions, including their germination, maturity, and extinction. However, these methods are all based on thresholds, that is, by calculating the scale size and temperature of a single target to determine whether it is an MCS. When applied globally, especially in tropical and mid-latitudes regions, the time required for the threshold method to determine whether a target is an MCS is quite long. The huge time cost will lead to slow progress in downstream tasks (tracking and prediction), affecting the analysis of the climatological characteristics of MCSs. The present invention constructs a mesoscale convective system recognition model (MCSeg) for identifying mesoscale convective systems MCSs in the mid-latitudes and low-latitudes. In addition, the present invention also constructs a mid-latitudes and low-latitudes MCSs recognition dataset for the training and testing of the mesoscale convective system recognition model. The present invention proposes a piecewise learning strategy, a semi-residual multi-scale feature extractor SRFE, and semantic consistency regularization. The piecewise learning strategy slices the MCSs in the mid-latitudes and low-latitudes, and inputs the slices into the model in sequence for learning, so that the model can learn the characteristics of mid-latitudes (low-latitudes) MCSs without being affected by the characteristics of low-latitudes (mid-latitudes) MCSs. In this way, the model can stably learn the feature expressions of MCSs in different regions, improve the stability during the model training process, and accelerate the model convergence. The semi-residual multi-scale feature extraction SRFE extracts hierarchical multi-scale features, improving the model's perception ability of MCSs with different scales with fewer parameters, preventing small-scale features from being lost during the downsampling process, and providing sufficient guidance for accurately identifying MCSs with different scales. The semantic consistency regularization is used to constrain the features of the model in terms of feature extraction diversity, and to constrain the model to extract features symmetric to those obtained in the feature extraction and decoding processes. Finally, it is proved on the constructed dataset that the present invention can quickly and accurately identify mesoscale convective systems MCSs in the mid-latitudes and low-latitudes. Compared with other deep learning models, the mesoscale convective system recognition model proposed by the present invention has better performance, solving the problems of large computational resource consumption, slow recognition speed, and small recognition area in the recognition of mesoscale convective systems MCSs by traditional methods.

[0074] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated designed hardware. Those of ordinary skill in the art can understand that the above devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code is provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and their modules of the present invention can be implemented by hardware circuits of programmable hardware devices such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above hardware circuits and software such as firmware.

[0075] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall all be covered by the protection scope of the present invention.

Claims

1. A fast and efficient method for identifying mesoscale convective systems in mid- and low-latitude regions, characterized in that: The method specifically includes: S1. Obtain cloud top brightness temperature data in mid- and low-latitude areas; S2. Construct a mesoscale convective system identification dataset based on cloud top brightness temperature data. The dataset includes a training set and a test set. S3, construct a mesoscale convective system identification model; S4, using the training set to train the mesoscale convective system identification model; S5. Based on the trained mesoscale convective system identification model, a mesoscale convective system identification result is obtained; S6, after reaching the specified number of training times in S4 and S5, input the test set into the trained mesoscale convective system identification model to determine whether the current mesoscale convective system identification model meets the index requirements. If so, save the parameters of the current model and enter S7, otherwise, return to S4; S7. Input the test set into the mesoscale convective system identification model that meets the index requirements to obtain the final mesoscale convective system identification result.

2. The method for quickly and efficiently identifying mesoscale convective systems in mid- and low-latitude regions as claimed in claim 1 is characterized in that: The S2 comprises the following steps: S201, setting a temperature threshold of the mesoscale convective system MCSs, and marking the area with a temperature less than the temperature threshold from the original cloud top brightness temperature data; S202, set the regional coverage threshold of the mesoscale convective system MCSs, further screen the area marked in S201 using the regional coverage threshold, select the area marked in S201 with an area greater than the regional coverage threshold, and mark the area that meets both threshold requirements as the mesoscale convective system MCSs area, where the mesoscale convective system MCSs area is marked as 1, and the non-mesoscale convective system MCSs area is marked as 0. After the mesoscale convective system MCSs area marking process, the mesoscale convective system MCSs label is obtained, and the label size is 5143×1715; S203, fill the empty values ​​in the cloud top brightness temperature data by using the interpolation method, process the abnormal values ​​in the data, replace the cloud top brightness temperature data with a median value greater than 345 by 345, and replace the cloud top brightness temperature data with a median value less than 170 by 170, and obtain the cloud top brightness temperature data size of 5143×1715; S204, edge filling is performed on the mesoscale convective system MCSs label obtained in S202 and the cloud top brightness temperature data obtained in S203, the cloud top brightness temperature data is filled with a value of 300, and the label data is filled with a value of 0, and the size of the two after filling is 5632×2048; S205. Divide the filled mesoscale convective system MCSs labels and cloud top brightness temperature data obtained in S204 into a training set and a test set, wherein both the training set and the test set include cloud top brightness temperature data and mesoscale convective system MCSs labels, and the data is saved in npy format, which is a data format.

3. The method for quickly and efficiently identifying mesoscale convective systems in mid- and low-latitude regions as claimed in claim 1 is characterized in that: The S3 comprises the following steps: S301, the cloud top brightness temperature data obtained in S204 is processed by maximum and minimum normalization and converted into a tensor format, wherein the maximum value is 345 and the minimum value is 170, and the MCS label data obtained in S204 remains unchanged; S302. According to the normalization processing result of S301, a mesoscale convective system identification model is constructed.

4. The method for quickly and efficiently identifying mesoscale convective systems in mid- and low-latitude regions as claimed in claim 1 is characterized in that: The mesoscale convective system identification model includes: An encoder, used to extract hierarchical pyramid features after the normalization process in S301, wherein the encoder includes four semi-residual multi-scale feature extractors SRFE; the semi-residual multi-scale feature extractors SRFE are used to extract multi-scale features; The decoder is used to decode and restore the pyramid features at different levels, wherein the decoder includes convolution and upsampling modules.

5. The method for quickly and efficiently identifying mesoscale convective systems in mid- and low-latitude regions as claimed in claim 1 is characterized in that: The S4 comprises the following steps: S401: For the cloud top brightness temperature data after normalization in S301, the size is 5632×2048. First, the data is split into 11×4 data of size 512×512. Each 11 groups of data are divided into a batch, and a total of 4 batches of data are obtained, which are represented as B. i , where the first batch B1 and the fourth batch B4 are cloud top brightness temperature data in mid-latitude regions, and the second batch B2 and the third batch B3 are data in low-latitude regions; S402, inputting the first batch B1 into the semi-residual multi-scale feature extractor SRFE to extract hierarchical multi-scale pyramid features, and outputting each layer feature map Ei of the encoder; S403, in the decoder part, the feature map Ei output by the encoder and the feature map Di output by the decoder are spliced ​​in the channel dimension and input into the convolution and upsampled to obtain Di, where Di represents the feature restored by the i-th layer decoder; S404, input the output of the last layer of the decoder into the convection recognition head SegHead to obtain the recognition result of the mesoscale convection system; S405, when back-propagating the optimization model, the semantic consistency regularization is used to optimize the model; S406, repeat S401-S405 until 4 batches of data B i are input into the model.

6. The method for quickly and efficiently identifying mesoscale convective systems in mid- and low-latitude regions as claimed in claim 1, characterized in that: The S402 includes the following steps: S4021, input the first batch B1 into the multi-scale feature extractor SRFE of the first half-residual of the encoder, first copy the input B1 in the channel dimension 4 copies, use the convolution of size 1×1 to extract features of the first copy to obtain the first branch feature map F1, use the convolution of size 3×3 to extract features of the second copy to obtain the second branch feature map F2, use the convolution of size 5×5 to extract features of the third copy to obtain the third branch feature map F3, and use the convolution of size 7×7 to extract features of the fourth copy to obtain the fourth branch feature map F4; S4022, concatenating the four feature maps F1, F2, F3, and F4 obtained in S4012 in the channel dimension, and then performing multi-scale feature fusion using 1×1 convolution to obtain a preliminary fused feature map M1; S4023, split the preliminary fusion feature map M1 into 4 parts in the channel dimension, and extract the features of the first part using a convolution of size 1×1 to obtain the first branch feature map F 11 , the second part uses a 3×3 convolution to extract features to obtain the second branch feature map F 22 , the third part is extracted using a 5×5 convolution to obtain the third branch feature map F 33 , the fourth part uses a 7×7 convolution to extract features to obtain the fourth branch feature map F 44 ; S4024, the four feature graphs F obtained in S4023 11 , F 22 , F 33 , F 44 After splicing in the channel dimension, 1×1 convolution is used to perform multi-scale feature fusion to obtain the final fused feature map M 11 , M 11 Added to M1, the final output E1 of the first semi-residual multi-scale feature extractor SRFE is obtained; S4025, inputting the final output E1 of the first half-residual multi-scale feature extractor SRFE into the next half-residual multi-scale feature extractor SRFE, and performing the operations of S4021-S4025 to obtain the final output E2 of the second half-residual multi-scale feature extractor SRFE; S4026. Repeat S4025 twice to obtain the final outputs E3 and E4 of the third and fourth semi-residual multi-scale feature extractors SRFE.

7. The method for quickly and efficiently identifying mesoscale convective systems in mid- and low-latitude regions as claimed in claim 1, characterized in that: The S405 comprises the following steps: S4051. For the output Ei of each layer of the encoder, calculate the L2 norm of each channel, reorder Ei by channel according to the value, and represent the reordered result as Es_i; S4052, the reordered result Es_i is divided into two parts in the channel, the first half is the part with large feature differences, and the second half is the part with small feature differences. The loss of these two parts is calculated to constrain the model to extract more diverse features. The formula is as follows; in, represents the regularization loss of semantic consistency within the class, represents the first half of the results after reordering the features extracted by the i-th layer decoder, represents the result of the second half of the features extracted by the i-th layer decoder after reordering, and N represents the total number of decoder layers; S4053, down-sampling the last layer feature D4 recovered by the decoder in S403, so that the scale of D4 is the same as the scale of the feature Ei extracted by each layer of the encoder, and the four features after downsampling are represented as d1, d2, d3, and d4 respectively; S4054, for the feature Ei extracted by each layer of the encoder, random channel selection is performed so that the number of channels of Ei is the same as that of D4, and the features after random channel selection are represented as e1, e2, e3, e4; S4055. After S4053 and S4054, d1, d2, d3, d4 and e1, e2, e3, e4 with the same spatial dimension and channel dimension are obtained. The symmetry between the features extracted by the model and the features in the decoding process is constrained by calculating the loss of the corresponding positions of the two. The formula is as follows; in, represents the regular loss of encoding and decoding semantic consistency, It represents the result after random channel selection of the output of each layer encoder. It represents the result of D4 being downsampled four times, and N represents the total number of encoder layers.

8. The method for quickly and efficiently identifying mesoscale convective systems in mid- and low-latitude regions as claimed in claim 1, characterized in that: The loss function of the mesoscale convective system identification model is expressed as follows: in, represents the total loss, represents the regular loss of encoding and decoding semantic consistency, represents the regularization loss of semantic consistency within the class, represents the weighted binary cross entropy loss function, represents the dice coefficient loss function, Output represents the mesoscale convection image recognition result, and mask represents the label data.

9. A system for quickly and efficiently identifying mesoscale convective systems in low and mid-latitude regions that implements the method for quickly and efficiently identifying mesoscale convective systems in low and mid-latitude regions as described in any one of claims 1 to 8, characterized in that: The fast and efficient mesoscale convective system identification system in mid- and low-latitude regions specifically includes: Input module, used to obtain cloud top brightness temperature data in mid- and low-latitude areas; The preprocessing module is used to construct and preprocess the mesoscale convective system identification dataset; Model building module, used to build a mesoscale convective system identification model; Training module, used to train the mesoscale convective system identification model; The output module is used to input the test set into the mesoscale convective system identification model that meets the index requirements to obtain the final mesoscale convective system identification results.

Images