Meteorological data processing methods, meteorological data-based alerting methods and devices

Abstract

This disclosure provides a meteorological data processing method, a meteorological data-based prompting method and apparatus, relating to deep learning technology and cloud computing technology. The specific implementation scheme includes: acquiring video sample data and meteorological sample data for a predetermined area; acquiring the correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time moment; and performing regression training on an initial video classification model based on the correspondence to obtain a meteorological data model.

Description

Technical Field

[0001] This disclosure relates to the field of image processing technology, and more particularly to the fields of deep learning technology and cloud computing technology, specifically to a meteorological data processing method, a prompting method and device based on meteorological data. Background Technology

[0002] Currently, the main ways to obtain meteorological information (such as rainfall information) are: weather forecasts, surveillance cameras, and meteorological station monitoring. Among these, weather forecasts can only predict the weather in the short term, but the granularity of the predicted area is relatively coarse, and it is difficult to be very precise in terms of time. Surveillance cameras can monitor the real-time situation, but it requires manual judgment of the rainfall intensity, and the error of human judgment is relatively large. Although meteorological stations can accurately monitor the rainfall intensity in real time, the distribution of meteorological stations is small and the coverage is insufficient, and the areas not covered are not given attention.

[0003] There is currently no effective solution to the problems of low accuracy and small coverage in weather forecasts. Summary of the Invention

[0004] This disclosure provides a meteorological data processing method, a meteorological data-based prompting method, and an apparatus.

[0005] According to one aspect of this disclosure, a meteorological data processing method is provided, comprising: acquiring video sample data and meteorological sample data for a predetermined area; acquiring a correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time moment; and performing regression training on an initial video classification model based on the correspondence to obtain a meteorological data model.

[0006] According to another aspect of this disclosure, a prompting method based on meteorological data is provided, comprising: acquiring video surveillance data of the area to be predicted; inputting the video surveillance data and a meteorological data model trained in any of the above-described meteorological data processing methods to obtain a meteorological forecast result output by the meteorological data model; and performing a prompting operation based on the meteorological forecast result.

[0007] According to another aspect of this disclosure, a meteorological data processing method apparatus is provided, comprising: a first acquisition module for acquiring video sample data and meteorological sample data for a predetermined area; a second acquisition module for acquiring a correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time moment; and a training module for performing regression training on an initial video classification model based on the correspondence to obtain a meteorological data model.

[0008] According to another aspect of this disclosure, a meteorological data-based prompting device is provided, comprising: a third acquisition module for acquiring video surveillance data of an area to be predicted; a processing module for processing the video surveillance data and a meteorological data model trained in the meteorological data processing device to obtain a meteorological forecast result output by the meteorological data model; and a prompting module for performing a prompting operation based on the meteorological forecast result.

[0009] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform any of the aforementioned meteorological data processing methods and any of the aforementioned meteorological data-based prompting methods.

[0010] According to another aspect of this disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to perform any of the above-described meteorological data processing methods and any of the above-described meteorological data-based prompting methods.

[0011] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements any of the above-described meteorological data processing methods and any of the above-described meteorological data-based prompting methods.

[0012] According to another aspect of this disclosure, a weather data-based alert product is provided, comprising: the electronic device described above.

[0013] In this embodiment, video sample data and meteorological sample data for a predetermined area are acquired; the correspondence between the meteorological sample data and the video sample data is obtained, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time moment; and the initial video classification model is regressed and trained according to the correspondence to obtain a meteorological data model. This achieves the purpose of constructing a meteorological data model based on video sample data and meteorological sample data, thereby improving the accuracy of meteorological data prediction and expanding the scope of meteorological data prediction. This solves the technical problems of low meteorological data prediction accuracy and small applicability in existing meteorological data prediction methods.

[0014] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0015] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0016] Figure 1 This is a flowchart of a meteorological data processing method according to the first embodiment of this disclosure;

[0017] Figure 2 This is a schematic diagram of an optional distribution of meteorological monitoring points and cameras according to the first embodiment of this disclosure;

[0018] Figure 3 This is a flowchart of an optional meteorological data model training method according to the first embodiment of this disclosure;

[0019] Figure 4 This is a flowchart of an optional meteorological data processing method according to the first embodiment of this disclosure;

[0020] Figure 5 This is a flowchart of another optional meteorological data model training according to the first embodiment of this disclosure;

[0021] Figure 6 This is a flowchart of another optional meteorological data processing method according to the first embodiment of this disclosure;

[0022] Figure 7 This is a flowchart of a meteorological data-based prompting method according to a second embodiment of the present disclosure;

[0023] Figure 8 This is a schematic diagram illustrating an optional meteorological data model deployment application according to a second embodiment of this disclosure;

[0024] Figure 9 This is a flowchart of an optional meteorological data-based prompting method according to a second embodiment of the present disclosure;

[0025] Figure 10 This is a schematic diagram of the structure of a meteorological data processing method apparatus according to the third embodiment of this disclosure;

[0026] Figure 11 This is a schematic diagram of the structure of a meteorological data-based alerting device according to the fourth embodiment of this disclosure;

[0027] Figure 12 This is a block diagram of an electronic device used to implement the meteorological data processing method or the meteorological data-based prompting method of the embodiments of this disclosure. Detailed Implementation

[0028] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Example 1

[0031] Heavy rainfall can easily trigger severe natural disasters, such as floods, which can lead to numerous casualties and huge economic losses.

[0032] However, real-time meteorological information obtained through weather forecasts can only provide advance predictions, but cannot achieve accurate real-time monitoring of meteorological information. Furthermore, other methods of obtaining meteorological information, such as real-time monitoring of rainfall based on camera footage, require manual judgment of rainfall intensity, which is prone to significant errors. Additionally, while real-time monitoring of rainfall intensity through meteorological stations can provide accurate and real-time monitoring, the limited distribution of these stations results in insufficient coverage, leaving unattended areas unaddressed.

[0033] Based on the above problems, this disclosure provides an embodiment of a meteorological data processing method. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0034] Figure 1 This is a flowchart of a meteorological data processing method according to the first embodiment of this disclosure, such as... Figure 1 As shown, the method includes the following steps:

[0035] Step S102: Obtain video sample data and meteorological sample data for the predetermined area;

[0036] Step S104: Obtain the correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time.

[0037] Step S106: Based on the above correspondence, perform regression training on the initial video classification model to obtain the meteorological data model.

[0038] In this embodiment, video sample data and meteorological sample data for a predetermined area are obtained; the correspondence between the meteorological sample data and the video sample data is acquired, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time moment; and the initial video classification model is trained by regression based on the correspondence to obtain a meteorological data model. This achieves the goal of constructing a meteorological data model based on video sample data and meteorological sample data, thereby improving the accuracy of meteorological data prediction and expanding the scope of meteorological data prediction. This solves the technical problems of low meteorological data prediction accuracy and small applicability in existing meteorological data prediction methods.

[0039] Optionally, the aforementioned predetermined area is the area within the target distance range of the meteorological monitoring point (i.e., the meteorological station), and the aforementioned video sample data is real-time video sample data acquired by multiple cameras within the predetermined area. The aforementioned video sample data includes rainfall information within the aforementioned predetermined area, for example, in... Figure 2 In the schematic diagram showing the distribution of meteorological monitoring points and cameras, triangles represent the locations of meteorological monitoring points. Multiple cameras are selected within a radius of 1 km from the meteorological monitoring points in the diagram. For example, video information collected by the camera at the location corresponding to the larger icon is selected as the video sample data.

[0040] Optionally, the above meteorological sample data is the rainfall data obtained from the above meteorological monitoring points. The above rainfall data may be, but is not limited to, rainfall intensity. For example, the rainfall data of the above meteorological monitoring points is shown in Table 1. Table 1 records the rainfall at the minute level at each time point. The rainfall amount divided by the duration gives the above rainfall intensity.

[0041] Table 1

[0042]

[0043]

[0044] Optionally, there is a unique correspondence between the above video sample data and the above meteorological sample data, that is, the above video sample data and the above meteorological sample data correspond one-to-one in terms of time (moment).

[0045] It should be noted that the key to obtaining the above meteorological data model lies in acquiring the training data. The video sample data needs to be calibrated with the above meteorological sample data, that is, the video sample data must correspond to the rainfall intensity. Therefore, selecting video sample data collected by cameras within a predetermined area (i.e., near the meteorological monitoring point) can be considered as corresponding to the above meteorological sample data in time, thus obtaining the above-mentioned calibrated video sample data.

[0046] In one optional embodiment, the initial video classification model is trained by regression based on the above correspondence to obtain a meteorological data model, including:

[0047] Step S204: The video sample data is calibrated according to the above correspondence to obtain calibrated video sample data;

[0048] Step S206: Divide the above-calibrated video sample data into multiple video sample sub-data, wherein the time length of different video sample sub-data is the same.

[0049] Step S208: Use at least a portion of the above-mentioned video sample sub-data to perform regression training on the above-mentioned initial video classification model to obtain the above-mentioned meteorological data model.

[0050] Optionally, the calibrated video sample data can be divided into multiple video sample sub-data of a fixed time length (e.g., 10 seconds).

[0051] Optionally, based on the convolutional neural network (CNN) algorithm, the initial video classification model is trained by regression using at least a portion of the aforementioned video sample sub-data to obtain a meteorological data model.

[0052] Optionally, the above regression training results are calculated using a fully connected computation method, with the specific formula being: O∑w i ·x i +b, where x i w represents an intermediate feature i b and b represent the fully connected parameters to be learned and the bias, respectively, and O represents the meteorological data prediction value, i.e. the rainfall intensity prediction value.

[0053] In an optional embodiment, the above-mentioned initial video classification model is trained by regression using at least a portion of the above-mentioned video sample sub-data to obtain the above-mentioned meteorological data model, including:

[0054] The initial video classification model was trained by using the gradient descent backpropagation optimization algorithm and at least a portion of the aforementioned video sample sub-data to obtain the aforementioned meteorological data model.

[0055] It should be noted that during the training of meteorological data models, there is often a certain gap between the predicted results and the actual values ​​of meteorological data. Figure 3 As shown, the loss function can characterize the degree of difference between the meteorological data prediction results and the actual meteorological data. The gradient descent backpropagation optimization algorithm and the above video sample sub-data are used to perform regression training on the above initial video classification model. The purpose is to reduce / eliminate the difference between the meteorological data prediction results and the actual meteorological data, thereby achieving the technical effect of further improving the accuracy of the meteorological data model.

[0056] Optionally, the loss function can be used to characterize the difference between the predicted meteorological data and the actual meteorological data. The loss function used in the regression training of the initial video classification model can be, but is not limited to, the minimum mean square error (MSE) loss function, specifically expressed as: MSE = 0.5(GT-O). 2 , where GT represents the above meteorological sample data, that is, the rainfall intensity obtained from the above meteorological monitoring points, and O represents the rainfall intensity obtained after model regression training.

[0057] In one optional embodiment, the initial video classification model is trained by regression based on the above correspondence to obtain a meteorological data model, including:

[0058] Step S302: The video sample data is calibrated according to the above correspondence to obtain calibrated video sample data;

[0059] Step S304: Obtain optical flow sample data corresponding to the above-calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the above-calibrated video sample data;

[0060] Step S306: The above-mentioned optical flow sample data is used to perform regression training on the above-mentioned initial video classification model to obtain the above-mentioned meteorological data model.

[0061] Optionally, the duration of the optical flow sample data is the same as the duration of the video sample sub-data.

[0062] Optionally, the target objects mentioned above may include, but are not limited to, motion characteristics such as rain, snow, hail, frost, clouds, and the sun.

[0063] It should be noted that optical flow information can reflect the motion characteristics in the video, that is, it can better capture the vertical motion characteristics of raindrops falling during rain. Based on the regression training of the initial video classification model using the above video sample sub-data, preprocessing to extract optical flow sample data and adding a branch, and using the above optical flow sample data to perform regression training on the above initial video classification model, can improve the algorithm performance and thus improve the accuracy of the meteorological data model.

[0064] It should be noted that after obtaining the above meteorological data model, it can be deployed on any server that can receive video surveillance data, thereby expanding the monitoring range and getting rid of the limitation that rainfall intensity can only be monitored in real time near meteorological stations.

[0065] Optionally, the dense optical flow algorithm is used to obtain the optical flow sample data, and the initial video classification model is trained by regression based on the optical flow sample data to obtain the meteorological data model, which can improve the recognition strength of target objects (such as the motion features of rain, snow, hail, frost, white clouds, sun, etc.).

[0066] As an optional embodiment, Figure 4 This is a flowchart of an optional meteorological data processing method according to the first embodiment of this disclosure, such as... Figure 4 As shown, the initial video classification model is trained using regression based on the above correspondence to obtain a meteorological data model, including:

[0067] Step S501: The video sample data is calibrated according to the above correspondence to obtain calibrated video sample data;

[0068] Step S502: Divide the above-calibrated video sample data into multiple video sample sub-data, wherein the time length of different video sample sub-data is the same.

[0069] Step S503: Use at least a portion of the above-mentioned video sample sub-data to perform regression training on the initial video classification model to obtain the first prediction model;

[0070] Step S504: Obtain optical flow sample data corresponding to the above-calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the above-calibrated video sample data;

[0071] Step S505: Use the aforementioned optical flow sample data to perform regression training on the initial video classification model to obtain the second prediction model;

[0072] Step S506: Based on the first prediction model and the second prediction model, the meteorological data model is obtained.

[0073] Optionally, the calibrated video sample data is divided into multiple video sample sub-data of a fixed time length (e.g., 10 seconds); the time length of the optical flow sample data is the same as the time length of the video sample sub-data; the target object may include, but is not limited to, motion features such as rain, snow, hail, frost, white clouds, sun, etc.

[0074] It should be noted that optical flow information can reflect the motion characteristics in a video, that is, it can better capture the vertical motion characteristics of raindrops falling during rain. In this embodiment, based on the regression training of the initial video classification model using the above-mentioned video sample sub-data to obtain the first prediction model, optical flow sample data is preprocessed and extracted, an additional branch is added, and the above-mentioned optical flow sample data is used to perform regression training of the above-mentioned initial video classification model to obtain the second prediction model. This can achieve the goal of improving the accuracy of the meteorological data model, thereby realizing the technical effect of improving the accuracy of meteorological data prediction.

[0075] As an optional embodiment, Figure 5 This is a flowchart of another optional meteorological data-based prompting method according to the first embodiment of this disclosure, such as... Figure 5 As shown, the acquired video surveillance data of the target area is segmented according to a fixed time, that is, it is segmented into continuous video with a duration of S seconds. The first video segmentation model is then trained using the aforementioned continuous video with a duration of S seconds to obtain the first predicted rainfall intensity. Furthermore, multiple S-second optical flows corresponding to the aforementioned continuous video with a duration of S seconds are acquired, and the aforementioned multiple S-second optical flows are then used to train the second video segmentation model to obtain the second predicted rainfall intensity.

[0076] As an optional embodiment, Figure 6 This is a flowchart of another optional meteorological data processing method according to the first embodiment of this disclosure, such as... Figure 6 As shown, the above-mentioned calibrated video sample data is used to perform regression training on the initial video classification model to obtain the meteorological data model, which includes:

[0077] Step S602: The video sample data is calibrated according to the above correspondence to obtain calibrated video sample data;

[0078] Step S604: Divide the above-calibrated video sample data into multiple video sample sub-data, wherein the time length of different video sample sub-data is the same.

[0079] Step S606: Use at least a portion of the above-mentioned video sample sub-data to perform regression training on the initial video classification model to obtain the first prediction model;

[0080] Step S608: Obtain optical flow sample data corresponding to the above-calibrated video sample data, wherein the above-calibrated optical flow sample data is used to reflect the motion characteristics of the target object in the above-calibrated video sample data;

[0081] Step S610: The first prediction model is trained by regression using the optical flow sample data to obtain the meteorological data model.

[0082] Optionally, the calibrated video sample data is divided into multiple video sample sub-data of a fixed time length (e.g., 10 seconds); the time length of the optical flow sample data is the same as the time length of the video sample sub-data; the target object may include, but is not limited to, motion features such as rain, snow, hail, frost, white clouds, sun, etc.

[0083] It should be noted that optical flow information can reflect the motion characteristics in the video, that is, it can better capture the vertical motion characteristics of raindrops falling during rain. In this embodiment of the present disclosure, based on the regression training of the initial video classification model using the above-mentioned video sample sub-data to obtain the first prediction model, optical flow sample data is preprocessed and extracted. The above-mentioned optical flow sample data is then used to perform regression training of the above-mentioned first prediction model to obtain the above-mentioned meteorological data model. This can achieve the goal of improving the accuracy of the meteorological data model, thereby realizing the technical effect of improving the accuracy of meteorological data prediction.

[0084] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the above-described vehicle information prompting method embodiments, and will not be repeated here. The acquisition, storage, and application of user personal information involved in the technical solutions of this disclosure all comply with relevant laws and regulations and do not violate public order and good morals.

[0085] Example 2

[0086] According to an embodiment of this disclosure, an embodiment of a prompting method based on meteorological data is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0087] Figure 7 This is a flowchart of a meteorological data-based prompting method according to a second embodiment of the present disclosure, as follows: Figure 7 As shown, the method includes the following steps:

[0088] Step S112: Obtain video surveillance data of the area to be predicted;

[0089] Step S114: Input the above video surveillance data and the meteorological data model trained by any of the above meteorological data processing methods to obtain the meteorological forecast result output by the above meteorological data model.

[0090] Step S116: Perform a prompt operation based on the above meteorological forecast results.

[0091] Optionally, the meteorological data model mentioned above is a meteorological data model trained by inputting the video surveillance data and the meteorological data processing method in any of the above embodiments 1, and the meteorological prediction result output by the meteorological data model is obtained; for example, the video sample data after calibration processing of meteorological sample data is used to regress the initial video classification model, and the meteorological prediction result is used to describe the current meteorological data and meteorological change data of the area to be predicted.

[0092] Optionally, the above meteorological forecast results may be, but are not limited to, rainfall intensity forecast values. If the above rainfall intensity forecast value exceeds the target threshold, a prompting operation may be performed based on the above meteorological forecast results, such as outputting alarm information based on the above meteorological forecast results.

[0093] In this embodiment of the disclosure, video surveillance data of the area to be predicted is acquired; the video surveillance data and the meteorological data model trained in any of the above-mentioned meteorological data processing methods are input to obtain the meteorological forecast result output by the meteorological data model; and a prompting operation is performed based on the meteorological forecast result. This achieves the purpose of making a meteorological forecast for the target area based on the video surveillance data and making a prompting operation based on the meteorological forecast result. This achieves the technical effect of expanding the meteorological forecast range and improving the accuracy of meteorological forecasts and meteorological prompts, thereby solving the technical problems of low meteorological forecast accuracy and small forecast range in existing meteorological forecasting methods, and making it difficult to effectively perform accurate prompting operations.

[0094] It should be noted that after obtaining the above meteorological data model, it can be deployed on any server capable of receiving video surveillance data, such as... Figure 8 As shown, this expands the monitoring range and eliminates the limitation that rainfall intensity can only be monitored in real time near meteorological stations.

[0095] As an optional embodiment, Figure 9 This is a flowchart of an optional meteorological data-based prompting method according to a second embodiment of the present disclosure, such as... Figure 9 As shown, the above video surveillance data is input into the above meteorological data model to obtain the meteorological forecast results output by the above meteorological data model, including:

[0096] Step S212: Input the above video surveillance data into the first prediction model in the above meteorological data model to obtain the first prediction result output by the first prediction model;

[0097] Step S214: Input the above video surveillance data into the second prediction model in the above meteorological data model to obtain the second prediction result output by the second prediction model;

[0098] Step S216: The first forecast result and the second forecast result are weighted to obtain the meteorological forecast result.

[0099] Optionally, the first prediction result is used to predict the current meteorological data of the video surveillance data at different times; wherein, the first prediction result is used to predict the motion characteristics of the target object in the video surveillance data at different times.

[0100] Optionally, the target objects mentioned above may include, but are not limited to, motion characteristics such as rain, snow, hail, frost, clouds, and the sun.

[0101] Optionally, the first prediction model is obtained by using the video sample sub-data to perform regression training on the initial video classification model; the second prediction model is obtained by acquiring the optical flow sample data corresponding to the calibrated video sample data and using the optical flow sample data to perform regression training on the initial video classification model, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data.

[0102] Optionally, by assigning different weights to the first forecast result and the second forecast result, different meteorological forecast results can be obtained. For example, if the weights of the first forecast result O1 and the second forecast result O2 are w1 and w2 respectively, where w1 = w2 = 0.5, then the meteorological forecast result O2... final It can be represented as O final = w1·O1+w2·O2.

[0103] In this embodiment of the disclosure, by inputting the video surveillance data into the first prediction model in the meteorological data model, a first prediction result output by the first prediction model is obtained;

[0104] By inputting the aforementioned video surveillance data into the second prediction model within the aforementioned meteorological data model, a second prediction result is obtained from the output of the second prediction model. By weighting the aforementioned first prediction result and the aforementioned second prediction result to obtain the aforementioned meteorological prediction result, the accuracy of the meteorological prediction result can be improved.

[0105] As an optional embodiment, it remains as follows Figure 4As shown, the acquired video surveillance data of the target area is segmented according to a fixed time interval, that is, divided into continuous video segments with a duration of S seconds. The first video segmentation model is then trained using the aforementioned continuous video segments with a duration of S seconds to obtain the first predicted rainfall intensity. Furthermore, multiple S-second optical flows corresponding to each of the aforementioned continuous video segments with a duration of S seconds are acquired, and the aforementioned multiple S-second optical flows are then used to train the second video segmentation model to obtain the second predicted rainfall intensity. The first predicted rainfall intensity and the second predicted rainfall intensity are then weighted to obtain the final predicted rainfall intensity of the target area.

[0106] In an optional embodiment, the above method further includes:

[0107] Step S312: When the above meteorological forecast result meets one of the following conditions, determine to perform the prompt operation: Based on the above meteorological forecast result, determine that the current meteorological data meets the first meteorological warning condition, wherein the above first meteorological warning condition includes at least: the rainfall intensity value is greater than or equal to the rainfall warning threshold;

[0108] Step S214: Based on the above meteorological forecast results, determine that the current meteorological change data meets the second meteorological warning conditions, wherein the above-mentioned second meteorological warning conditions include at least: the change value of rainfall intensity is greater than or equal to the rainfall change warning threshold.

[0109] Optionally, the first meteorological warning condition mentioned above may include, but is not limited to, rainfall intensity values ​​greater than or equal to the rainfall warning threshold (e.g., greater than 70 mm / h), snowfall intensity values ​​greater than or equal to the snowfall warning threshold, etc.; the second meteorological warning condition mentioned above may include, but is not limited to, sudden weather changes, such as a sudden increase or excessive change in rainfall or snowfall.

[0110] Optional, as before Figure 8 As shown, the above meteorological data model is deployed on any server that can receive video surveillance data, and the above alarm information is pushed to relevant personnel to remind them to pay attention to the alarm information and take appropriate countermeasures in a timely manner.

[0111] In this embodiment of the disclosure, if the meteorological forecast results determine that the current meteorological data meets the first meteorological warning condition, for example, the rainfall intensity value is greater than or equal to the rainfall warning threshold; or if the meteorological forecast results determine that the current meteorological change data meets the second meteorological warning condition, for example, the rainfall intensity change value is greater than or equal to the rainfall change warning threshold, the purpose of issuing timely warning information for different weather deterioration conditions is achieved, thereby helping to take timely countermeasures against different natural disasters.

[0112] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the above-described vehicle information prompting method embodiments, and will not be repeated here. The acquisition, storage, and application of user personal information involved in the technical solutions of this disclosure all comply with relevant laws and regulations and do not violate public order and good morals.

[0113] Example 3

[0114] According to embodiments of this disclosure, an apparatus embodiment for implementing the above-described meteorological data processing method is also provided. Figure 10 This is a schematic diagram of the structure of a meteorological data processing method apparatus according to the third embodiment of this disclosure, as shown below. Figure 10 As shown, the above-mentioned meteorological data processing method apparatus includes: a first acquisition module 700, a second acquisition module 702, and a training module 704, wherein:

[0115] The first acquisition module 700 is used to acquire video sample data and meteorological sample data for a predetermined area; the second acquisition module 702 is used to acquire the correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time; the training module 704 is used to perform regression training on the initial video classification model according to the correspondence to obtain a meteorological data model.

[0116] In this embodiment, a first acquisition module 700 is set up to acquire video sample data and meteorological sample data for a predetermined area; a second acquisition module 702 is set up to acquire the correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time; and a training module 704 is set up to perform regression training on an initial video classification model according to the correspondence to obtain a meteorological data model. This achieves the purpose of constructing a meteorological data model based on video sample data and meteorological sample data, thereby improving the accuracy of meteorological data prediction and expanding the scope of meteorological data prediction. This solves the technical problems of low meteorological data prediction accuracy and small applicability in existing meteorological data prediction methods.

[0117] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0118] It should be noted that the first acquisition module 700, the second acquisition module 702, and the training module 704 mentioned above correspond to steps S102 to S106 in Embodiment 1. The instances and application scenarios implemented by the above modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should also be noted that the above modules, as part of the device, can run on a computer terminal.

[0119] Optionally, the training module further includes: a first calibration module, used to calibrate the video sample data according to the above correspondence to obtain calibrated video sample data; a first segmentation module, used to segment the calibrated video sample data into multiple video sample sub-data, wherein different video sample sub-data have the same time length; and a first training sub-module, used to perform regression training on the initial video classification model using at least some of the video sample sub-data to obtain the meteorological data model.

[0120] Optionally, the first training submodule further includes a second training submodule, used to perform regression training on the initial video classification model using the gradient descent backpropagation optimization algorithm and at least a portion of the video sample subdata to obtain the meteorological data model.

[0121] Optionally, the training module further includes: a second calibration module, used to calibrate the video sample data according to the above correspondence to obtain calibrated video sample data; a first acquisition submodule, used to acquire optical flow sample data corresponding to the calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data; and a third training submodule, used to perform regression training on the initial video classification model using the optical flow sample data to obtain the meteorological data model.

[0122] Optionally, the training module further includes: a third calibration module, used to calibrate the video sample data according to the above correspondence to obtain calibrated video sample data; a second segmentation module, used to segment the calibrated video sample data into multiple video sample sub-data, wherein different video sample sub-data have the same time length; a fourth training sub-module, used to perform regression training on the initial video classification model using at least some of the video sample sub-data to obtain a first prediction model; a second acquisition sub-module, used to acquire optical flow sample data corresponding to the calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data; a fifth training sub-module, used to perform regression training on the initial video classification model using the optical flow sample data to obtain a second prediction model; and a third acquisition sub-module, used to obtain the meteorological data model based on the first prediction model and the second prediction model.

[0123] Optionally, the training module further includes: a fourth calibration module, used to calibrate the video sample data according to the above correspondence to obtain calibrated video sample data; a third segmentation module, used to segment the calibrated video sample data into multiple video sample sub-data, wherein different video sample sub-data have the same time length; a sixth training sub-module, used to perform regression training on the initial video classification model using at least some of the video sample sub-data to obtain a first prediction model; a fourth acquisition sub-module, used to acquire optical flow sample data corresponding to the calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data; and a seventh training sub-module, used to perform regression training on the first prediction model using the optical flow sample data to obtain the meteorological data model.

[0124] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant description in Embodiment 1, and will not be repeated here. The acquisition, storage, and application of user personal information involved in the technical solution of this disclosure all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0125] Example 4

[0126] According to embodiments of this disclosure, an apparatus embodiment for implementing the above-described meteorological data-based alerting method is also provided. Figure 11 This is a schematic diagram of the structure of a meteorological data-based alerting device according to the fourth embodiment of this disclosure, as shown below. Figure 11 As shown, the above-mentioned meteorological data-based alerting device includes: a third acquisition module 412, a processing module 414, and an alerting module 416, wherein:

[0127] The third acquisition module 412 is used to acquire video surveillance data of the area to be predicted; the processing module 414 is used to input the video surveillance data and the meteorological data model trained in any of the above-mentioned meteorological data processing devices to obtain the meteorological forecast result output by the meteorological data model; the prompting module 416 is used to perform a prompting operation based on the above-mentioned meteorological forecast result.

[0128] In this embodiment, a second acquisition module 412 is provided to acquire video surveillance data of the area to be predicted; a processing module 414 is provided to input the video surveillance data and a meteorological data model trained in any of the meteorological data processing devices mentioned above to obtain the meteorological forecast result output by the meteorological data model; and a prompting module 416 is provided to perform a prompting operation based on the meteorological forecast result. This achieves the purpose of making a meteorological forecast for the target area based on video surveillance data and performing a prompting operation based on the meteorological forecast result. This achieves the technical effect of expanding the meteorological forecast range and improving the accuracy of meteorological forecasts and meteorological prompts, thereby solving the technical problems of low meteorological forecast accuracy and small forecast range in existing meteorological forecasting methods, and difficulty in effectively performing accurate prompting operations.

[0129] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0130] It should be noted that the third acquisition module 412, processing module 414, and prompting module 416 mentioned above correspond to steps S112 to S116 in Embodiment 1. The instances and application scenarios implemented by the above modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules, as part of the device, can run in a computer terminal.

[0131] Optionally, the above processing module further includes: a first input module, used to input the video surveillance data into a first prediction model in the meteorological data model to obtain a first prediction result output by the first prediction model, wherein the first prediction result is used to predict the current meteorological data of the video surveillance data at different times; a second input module, used to input the video surveillance data into a second prediction model in the meteorological data model to obtain a second prediction result output by the second prediction model, wherein the first prediction result is used to predict the motion characteristics of the target object in the video surveillance data at different times; and a weighted processing module, used to perform weighted processing on the first prediction result and the second prediction result to obtain the meteorological prediction result.

[0132] Optionally, the above device further includes: a first prompting submodule, configured to determine to perform a prompting operation when the above meteorological forecast result meets one of the following conditions: determining that the current meteorological data meets a first meteorological warning condition based on the above meteorological forecast result, wherein the first meteorological warning condition includes at least: the rainfall intensity value is greater than or equal to a rainfall warning threshold; and a determining module, configured to determine that the current meteorological change data meets a second meteorological warning condition based on the above meteorological forecast result, wherein the second meteorological warning condition includes at least: the rainfall intensity change value is greater than or equal to a rainfall change warning threshold.

[0133] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant description in Embodiment 1, and will not be repeated here. The acquisition, storage, and application of user personal information involved in the technical solution of this disclosure all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0134] Example 5

[0135] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0136] Figure 12 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0137] like Figure 12 As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 802 or a computer program loaded from storage unit 808 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0138] Multiple components in device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 807, such as various types of monitors, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0139] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as methods for acquiring video sample data and meteorological sample data. For example, in some embodiments, the methods for acquiring video sample data and meteorological sample data may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the methods described above for acquiring video sample data and meteorological sample data may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured by any other suitable means (e.g., by means of firmware) to perform a method to acquire video sample data and meteorological sample data.

[0140] This disclosure also includes a weather data-based alert product, comprising at least the electronic device described above.

[0141] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0142] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0143] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0144] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0145] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0146] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0147] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0148] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A meteorological data processing method, comprising: Acquire video sample data and meteorological sample data for a predetermined area, wherein the predetermined area is the area within the target distance range of a meteorological station, the video sample data is video sample data acquired in real time by multiple cameras within the predetermined area, the video sample data includes rainfall information within the predetermined area, and the meteorological sample data is used to represent the rainfall intensity obtained based on the meteorological station; Obtain the correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time moment; The initial video classification model is trained by regression based on the correspondence to obtain a meteorological data model, wherein the meteorological data model is used to determine the predicted value of rainfall intensity. The step of performing regression training on the initial video classification model based on the correspondence to obtain a meteorological data model includes: calibrating the video sample data according to the correspondence to obtain calibrated video sample data; dividing the calibrated video sample data into multiple video sample sub-data, wherein different video sample sub-data have the same time length; using at least a portion of the video sample sub-data to perform regression training on the initial video classification model to obtain a first prediction model; acquiring optical flow sample data corresponding to the calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data; and using the optical flow sample data to perform regression training on the first prediction model to obtain the meteorological data model. Alternatively, the step of performing regression training on the initial video classification model based on the correspondence to obtain a meteorological data model includes: calibrating the aforementioned video sample data according to the correspondence to obtain calibrated video sample data; dividing the calibrated video sample data into multiple video sample sub-data, wherein different video sample sub-data have the same time length; using at least a portion of the video sample sub-data to perform regression training on the initial video classification model to obtain a first prediction model; acquiring optical flow sample data corresponding to the calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data; using the optical flow sample data to perform regression training on the initial video classification model to obtain a second prediction model; and obtaining the meteorological data model based on the first prediction model and the second prediction model.

2. A meteorological data-based alerting method, comprising: Obtain video surveillance data for the area to be predicted; The video surveillance data is input into the meteorological data model trained in the meteorological data processing method of claim 1 to obtain the meteorological forecast result output by the meteorological data model; The prompt operation will be executed based on the weather forecast results.

3. The method according to claim 2, wherein, The video surveillance data is input into the meteorological data model to obtain the meteorological forecast results output by the meteorological data model, including: The video surveillance data is input into the first prediction model in the meteorological data model to obtain the first prediction result output by the first prediction model, wherein the first prediction result is used to predict the current meteorological data of the video surveillance data at different times; The video surveillance data is input into the second prediction model in the meteorological data model to obtain the second prediction result output by the second prediction model. The second prediction result is used to predict the motion characteristics of the target object in the video surveillance data at different times. The first prediction result and the second prediction result are weighted to obtain the meteorological prediction result.

4. The method according to claim 2, wherein, The method further includes: The prompt operation is determined to be executed when the weather forecast result meets one of the following conditions: the current weather data is determined to meet the first weather warning condition based on the weather forecast result, wherein the first weather warning condition includes at least: the rainfall intensity value is greater than or equal to the rainfall warning threshold; Based on the meteorological forecast results, it is determined that the current meteorological change data meets the second meteorological warning conditions, wherein the second meteorological warning conditions include at least: the change value of rainfall intensity is greater than or equal to the rainfall change warning threshold.

5. A meteorological data processing device, comprising: The first acquisition module is used to acquire video sample data and meteorological sample data for a predetermined area, wherein the predetermined area is an area within the target distance range of a meteorological station, the video sample data is video sample data acquired in real time by multiple cameras within the predetermined area, the video sample data includes rainfall information within the predetermined area, and the meteorological sample data is used to represent the rainfall intensity obtained based on the meteorological station. The second acquisition module is used to acquire the correspondence between the meteorological sample data and the video sample data, wherein the correspondence is used to indicate the meteorological sample data corresponding to the video sample data at each time moment; The training module is used to perform regression training on the initial video classification model based on the correspondence to obtain the meteorological data model; The training module further includes: a fourth calibration module, used to calibrate the video sample data according to the correspondence to obtain calibrated video sample data; a third segmentation module, used to segment the calibrated video sample data into multiple video sample sub-data, wherein different video sample sub-data have the same time length; a sixth training sub-module, used to perform regression training on an initial video classification model using at least some of the video sample sub-data to obtain a first prediction model; a fourth acquisition sub-module, used to acquire optical flow sample data corresponding to the calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data; and a seventh training sub-module, used to perform regression training on the first prediction model using the optical flow sample data to obtain the meteorological data model. Alternatively, the training module further includes: a third calibration module, used to calibrate the video sample data according to the correspondence to obtain calibrated video sample data; a second segmentation module, used to segment the calibrated video sample data into multiple video sample sub-data, wherein the different video sample sub-data have the same time length; a fourth training sub-module, used to perform regression training on the initial video classification model using at least some of the video sample sub-data to obtain a first prediction model; a second acquisition sub-module, used to acquire optical flow sample data corresponding to the calibrated video sample data, wherein the optical flow sample data is used to reflect the motion characteristics of the target object in the calibrated video sample data; a fifth training sub-module, used to perform regression training on the initial video classification model using the optical flow sample data to obtain a second prediction model; and a third acquisition sub-module, used to obtain the meteorological data model based on the first prediction model and the second prediction model.

6. A weather data-based alerting device, comprising: The third acquisition module is used to acquire video surveillance data of the area to be predicted. The processing module is used to input the video surveillance data into the meteorological data processing device of claim 5, which has been trained to obtain the meteorological forecast result output by the meteorological data model. The prompting module is used to perform prompting operations based on the weather forecast results.

7. The apparatus according to claim 6, wherein, The processing module further includes: The first input module is used to input the video surveillance data into the first prediction model in the meteorological data model to obtain the first prediction result output by the first prediction model, wherein the first prediction result is used to predict the current meteorological data of the video surveillance data at different times; The second input module is used to input the video surveillance data into the second prediction model in the meteorological data model to obtain the second prediction result output by the second prediction model, wherein the second prediction result is used to predict the motion characteristics of the target object in the video surveillance data at different times; The weighted processing module is used to perform weighted processing on the first prediction result and the second prediction result to obtain the meteorological prediction result.

8. The apparatus according to claim 6, wherein, The device further includes: The first prompting submodule is used to determine to perform a prompting operation when the weather forecast result meets one of the following conditions: the current weather data is determined to meet a first weather warning condition based on the weather forecast result, wherein the first weather warning condition includes at least: the rainfall intensity value is greater than or equal to the rainfall warning threshold; The determination module is used to determine, based on the meteorological forecast results, whether the current meteorological change data meets the second meteorological warning condition, wherein the second meteorological warning condition includes at least: the change value of rainfall intensity is greater than or equal to the rainfall change warning threshold.

9. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the meteorological data processing method of any one of claims 1 and the meteorological data-based prompting method of any one of claims 2-4.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the meteorological data processing method according to any one of claims 1, and the meteorological data-based prompting method according to any one of claims 2-4.

11. A computer program product comprising a computer program that, when executed by a processor, implements the meteorological data processing method according to any one of claims 1 and the meteorological data-based prompting method according to any one of claims 2-4.

12. A weather data-based alert product, comprising: The electronic device as described in claim 9.

Images