Weather phenomenon intelligent observation method, device and equipment and storage medium

Abstract

According to the intelligent observation method, device and equipment for the weather phenomenon and the storage medium provided by the invention, the collected multi-source data is intelligently identified and analyzed through machine learning models such as the cloud condition identification model, the rainfall identification model and the thunder and lightning identification model, and key meteorological elements such as cloud height, cloud category, rainfall intensity, rainfall type and thunder and lightning activity are automatically generated; manual intervention is reduced, and observation precision and automation level are improved. Moreover, a plurality of sensing devices, such as an image acquisition device, a weather radar, a rainfall sensor, a recording device and the like, are introduced to obtain multi-dimensional weather information and cover a plurality of weather phenomena such as cloud conditions, rainfall, thunder and lightning and the like, so that the problem of single data source of a traditional meteorological observation means is overcome, and the comprehensiveness and reliability of observation are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of weather phenomenon observation, and in particular to an intelligent observation method, device, equipment and storage medium for weather phenomena. Background Art

[0002] In the field of meteorological observation, traditional meteorological observation mainly relies on single physical sensors (such as rain gauges, anemometers, radars, etc.) and manual observation. Although these methods can provide relatively accurate weather data, there are problems such as one-sided information, lack of real-time performance and limited accuracy. In recent years, with the development of artificial intelligence technologies such as image processing technology, machine learning and deep learning, intelligent meteorological observation technology with multi-source data fusion has gradually emerged. By comprehensively processing images, audio and multiple sensor data, it can monitor and forecast weather phenomena in real time, comprehensively and accurately.

[0003] However, the existing meteorological monitoring system still faces the following problems: single data source, most systems rely only on meteorological radar or a single sensor, making it difficult to obtain comprehensive meteorological information; low fusion accuracy, when fusing multiple observation data, the system lacks an effective fusion algorithm, resulting in inaccurate observation results; insufficient real-time performance, the data processing efficiency of traditional systems is low, making it difficult to respond to weather changes in real time, affecting the timeliness of weather forecasts.

[0004] In summary, the problems existing in the prior art need to be solved urgently. Summary of the invention

[0005] The present invention provides an intelligent observation method, device, equipment and storage medium for weather phenomena, which are used to solve the defects in the prior art and realize automatic observation of weather phenomena.

[0006] The present invention provides an intelligent observation method for weather phenomena, comprising:

[0007] The image to be observed is obtained through a number of image acquisition devices, and the echo intensity data is obtained through the weather radar;

[0008] Inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data;

[0009] Get real-time precipitation data through rain sensors;

[0010] Inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0011] Acquire the audio to be detected through several recording and collection devices;

[0012] Inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data;

[0013] The cloud condition observation data, the precipitation observation data and the lightning observation data are integrated to obtain a weather observation result.

[0014] According to an intelligent observation method for weather phenomena provided by the present invention, after the step of inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data, the method further includes:

[0015] Acquire an image to be detected containing a fixed target object;

[0016] Visibility data is obtained according to the clarity of the image to be detected.

[0017] According to an intelligent observation method for weather phenomena provided by the present invention, the image acquisition device is a full-color fisheye camera, and the full-color fisheye camera is arranged on a pan-tilt platform. The step of acquiring the image to be observed by using a plurality of image acquisition devices specifically includes:

[0018] At preset time intervals, the pan / tilt is controlled to follow a preset acquisition route, and the full-color fisheye camera is controlled to shoot and acquire at a preset position to obtain the image to be observed.

[0019] According to an intelligent observation method of weather phenomena provided by the present invention, the rainfall sensor includes a piezoelectric rainfall sensor, a raindrop spectrometer and an optical rainfall sensor.

[0020] According to an intelligent observation method for weather phenomena provided by the present invention, the image to be observed includes a satellite cloud image and a cloud condition image, and the step of inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data is specifically as follows:

[0021] Inputting the satellite cloud image, the cloud shape image and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data;

[0022] The satellite cloud image is used for judging the cloud type and cloud height, and the cloud shape image is used for judging the cloud height.

[0023] According to an intelligent observation method for weather phenomena provided by the present invention, the image to be observed includes a precipitation image, and the step of inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data is specifically as follows:

[0024] Inputting the precipitation image and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0025] The precipitation observation data includes precipitation intensity and precipitation type.

[0026] According to an intelligent observation method for weather phenomena provided by the present invention, the step of integrating the cloud condition observation data, the precipitation observation data and the lightning observation data to obtain a weather observation result specifically includes:

[0027] Performing weight allocation on the cloud condition observation data, the precipitation observation data, and the lightning observation data;

[0028] The cloud condition observation data, the precipitation observation data and the lightning observation data are weightedly fused to obtain a weather observation result.

[0029] The present invention also provides an intelligent observation device for weather phenomena, comprising:

[0030] A first data acquisition module is used to acquire the image to be observed through a plurality of image acquisition devices and to acquire echo intensity data through a weather radar;

[0031] A cloud condition recognition module, used for inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data;

[0032] A second data acquisition module is used to acquire real-time precipitation data through a rainfall sensor;

[0033] A precipitation recognition module, used for inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0034] A third data acquisition module is used to acquire the audio to be detected through a plurality of recording and collection devices;

[0035] A lightning recognition module, used for inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data;

[0036] The data integration module is used to integrate the cloud condition observation data, the precipitation observation data and the lightning observation data to obtain weather observation results.

[0037] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, an intelligent observation method for weather phenomena as described above is implemented.

[0038] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an intelligent observation method for weather phenomena as described in any one of the above.

[0039] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the intelligent observation method of any of the weather phenomena described above is implemented.

[0040] The intelligent observation method, device, equipment and storage medium of weather phenomena provided by the present invention intelligently identify and analyze the collected multi-source data through machine learning models such as cloud condition recognition model, precipitation recognition model and lightning recognition model, and automatically generate key meteorological elements such as cloud height, cloud category, precipitation intensity, precipitation type, and lightning activity, thereby reducing human intervention and improving the accuracy and automation level of observation. In addition, a variety of sensing devices are introduced, such as image acquisition equipment, weather radar, rainfall sensor, recording equipment, etc., to obtain multi-dimensional weather information, covering a variety of weather phenomena such as cloud conditions, precipitation, and lightning, overcoming the problem of a single data source in traditional meteorological observation methods and enhancing the comprehensiveness and reliability of observation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 It is a schematic diagram of the flow of the intelligent observation method of weather phenomena provided by the present invention;

[0043] Figure 2 It is a structural schematic diagram of the intelligent observation device for weather phenomena provided by the present invention;

[0044] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In order to solve the problems in the prior art, the present invention proposes an intelligent observation method for weather phenomena to achieve intelligent observation of weather phenomena. The intelligent observation method for weather phenomena is described below. Figure 1As shown, including but not limited to the following steps:

[0047] Step 110: Acquire the image to be observed through a plurality of image acquisition devices, and acquire echo intensity data through a weather radar.

[0048] In step 110, first, several image acquisition devices are deployed in the observation area, preferably full-color fisheye cameras, which are installed on the pan-tilt platform and can perform 360-degree panoramic observation. In order to ensure the breadth and accuracy of the observation, the system sets a predetermined acquisition line to control the pan-tilt platform to capture images in different directions to ensure that the image to be observed including the entire sky can be obtained.

[0049] At the same time, combined with weather radar equipment, real-time radar echo intensity data of the corresponding area is obtained. Weather radar is used to capture the dynamic changes of clouds. By analyzing the echo intensity, the height and shape changes of clouds can be obtained, thus providing an important basis for subsequent cloud condition identification.

[0050] Step 120: input the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data.

[0051] In step 120, the collected image to be observed and the echo intensity data obtained from the weather radar are input into the cloud condition recognition model. The model is based on deep learning technology and is trained with a large number of cloud images and radar data to automatically recognize and analyze cloud information in the image. Specifically, the model generates cloud height data and cloud category data based on different types of cloud characteristics and combined with radar echo intensity data.

[0052] Cloud height data indicates the observed height of clouds from the ground, while cloud category data includes information on different types of clouds such as cumulus, stratus, cirrus, etc. These data provide an important basis for weather forecasting and storm prediction.

[0053] Step 130: Acquire real-time precipitation data through a rain sensor.

[0054] In this step, a variety of rain sensors are deployed to obtain real-time precipitation data. Piezoelectric rain sensors, raindrop spectrometers, and optical rain sensors are preferably used. Piezoelectric rain sensors can accurately measure precipitation intensity by sensing the impact force of raindrops; raindrop spectrometers can capture the particle size and distribution of precipitation; optical rain sensors analyze the type and intensity of precipitation through the refraction and reflection of light. After obtaining real-time precipitation information, these sensors can accurately provide precipitation data for the current area.

[0055] Step 140: input the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data.

[0056] The image to be observed and the real-time precipitation data obtained in the above steps are input into the precipitation recognition model. The model can accurately identify the precipitation intensity and precipitation type (such as light rain, showers, heavy rain, etc.) by analyzing the raindrop morphology and density and combining the precipitation characteristics in the image. The precipitation recognition model is trained with a large number of rain scene images and meteorological data to achieve high-precision precipitation observation.

[0057] Precipitation observation data includes not only the intensity of precipitation (such as mm / hour), but also the type of precipitation. By combining different types of precipitation data, the system can provide decision support for flood warnings, crop irrigation, etc.

[0058] Step 150: Acquire the audio to be detected through a plurality of recording and collection devices.

[0059] Next, several recording and acquisition devices are deployed in the observation area to obtain the environmental audio to be detected. Lightning activities are usually accompanied by violent sounds, so these devices are mainly used to capture audio features such as lightning and thunder during lightning. By analyzing the audio data, it is possible to determine whether there is lightning activity and its intensity.

[0060] Step 160: input the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data.

[0061] The collected images to be observed and the audio to be detected are input into the lightning recognition model. The model combines the lightning features in the image and the thunder spectrum in the audio to perform multimodal fusion analysis to identify the frequency, intensity and location of lightning activities. The model can effectively analyze the time, location and intensity information of lightning through a large amount of training on historical lightning images and audio data.

[0062] Step 170: Integrate the cloud condition observation data, the precipitation observation data and the lightning observation data to obtain a weather observation result.

[0063] After completing the individual identification of cloud conditions, precipitation and lightning, the system integrates these observation data together, using algorithms such as weighted fusion or voting fusion to finally generate comprehensive weather observation results. By means of weight distribution, the observation results are adjusted in combination with the weights of different weather phenomena to ensure that the final weather observation results can fully reflect the current meteorological conditions.

[0064] The observation results can include cloud types, heights, precipitation types and intensity, as well as the time, location and intensity of lightning. The final weather observation results can be used in many fields such as weather forecasting, natural disaster warnings and flight scheduling, greatly improving the accuracy and response speed of meteorological observations.

[0065] As a further optional embodiment, after the step of inputting the image to be observed and the audio to be detected into the lightning recognition model to obtain lightning observation data, the method further includes:

[0066] Acquire an image to be detected containing a fixed target object;

[0067] Visibility data is obtained according to the clarity of the image to be detected.

[0068] In this embodiment, after completing the acquisition of lightning observation data, the system will further acquire images to be detected containing fixed targets. These images are collected by image acquisition devices set at specific locations. Fixed targets can be known landmark buildings, towers, trees, etc. Such objects have clear outlines or logos in the image and are used as references for visibility analysis.

[0069] To ensure the accuracy of the analysis, the image acquisition device preferably uses a high-resolution full-color camera and regularly captures images containing fixed targets according to a preset acquisition strategy. The system will repeat the acquisition at a certain time interval to obtain a continuous image data stream.

[0070] Next, the system performs a clarity analysis on the acquired image containing the fixed target. Specifically, the system uses image processing technology to analyze the fixed target's outline, edge clarity, and image contrast, brightness and other parameters. By calculating the image blur of the fixed target, the system can determine the visibility of the observation area.

[0071] Visibility data indicates the air transparency in the observation area, usually in meters or kilometers. The decrease in visibility is usually related to weather phenomena such as fog, haze, rain and snow, so this data is of great reference significance for predicting traffic safety and flight scheduling.

[0072] As a further optional embodiment, the image acquisition device is a full-color fisheye camera, and the full-color fisheye camera is arranged on a pan / tilt platform. The step of acquiring the image to be observed by using a plurality of image acquisition devices specifically includes:

[0073] At preset time intervals, the pan / tilt is controlled to follow a preset acquisition route, and the full-color fisheye camera is controlled to shoot and acquire at a preset position to obtain the image to be observed.

[0074] In this embodiment, the image acquisition device uses a full-color fisheye camera. The fisheye camera can provide an ultra-wide-angle field of view, usually covering an observation range of more than 180 degrees, and can capture weather phenomena in a wide range. By using the full-color mode, the camera can capture richer image details, including information such as color and light intensity, which helps to improve the recognition accuracy of weather phenomena such as cloud conditions and precipitation.

[0075] In order to enhance the flexibility and dynamism of the image acquisition device, the full-color fisheye camera is installed on a pan-tilt platform. The pan-tilt platform is driven by a motor and can achieve multi-angle and multi-directional rotation, so that the camera can shoot in different directions according to needs, ensuring full coverage of weather data.

[0076] During the image acquisition process, the system presets the acquisition time interval and acquisition route. For example, at a certain interval (such as every 5 minutes), the PTZ is controlled to rotate according to the set route, and the full-color fisheye camera is pointed to the predetermined observation direction for shooting. This process can ensure multi-angle coverage of the observation area and avoid missing important weather phenomena.

[0077] By collecting multiple sets of observation images, the system can obtain weather image data at different times and locations, which provide diverse and sufficient inputs for subsequent cloud recognition, precipitation analysis, lightning detection and other models. At the same time, the wide-angle field of view and high-definition images provided by the fisheye camera ensure the capture of weather phenomena in detail, which helps to improve the accuracy and stability of model recognition results.

[0078] This embodiment uses a full-color fisheye camera and combines it with the dynamic adjustment function of the pan / tilt to enhance the spatial coverage and image acquisition efficiency of the weather observation system. By properly designing the acquisition time interval and acquisition line, high-quality observation image data in a large area can be regularly acquired, providing rich input for subsequent weather phenomenon analysis, thereby improving the overall observation accuracy and timeliness.

[0079] As a further optional embodiment, the rain sensor includes a piezoelectric rain sensor, a raindrop spectrometer and an optical rain sensor.

[0080] In this embodiment, the rainfall sensor in the system is not limited to a single sensing device, but integrates a piezoelectric rainfall sensor, a raindrop spectrometer and an optical rainfall sensor. These three types of sensors work together to comprehensively and accurately monitor different types of precipitation conditions.

[0081] As a further optional embodiment, the image to be observed includes a satellite cloud image and a cloud condition image, and the step of inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data is specifically as follows:

[0082] Inputting the satellite cloud image, the cloud shape image and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data;

[0083] The satellite cloud image is used for judging the cloud type and cloud height, and the cloud shape image is used for judging the cloud height.

[0084] In this step, the system combines a variety of observation methods, combines satellite cloud images, cloud images and radar echo data, and inputs them into the cloud recognition model to determine the cloud height and cloud category. The details are as follows:

[0085] Satellite cloud images: Cloud images captured by satellite remote sensing technology provide large-scale cloud distribution information. Satellite cloud images are mainly used to determine cloud types and cloud heights. Their wide coverage helps to determine cloud types, such as cumulonimbus clouds, high-level clouds, etc., and helps identify the formation of different meteorological phenomena.

[0086] Cloud condition image: An image of local cloud conditions obtained by ground image acquisition equipment (such as a full-color fisheye camera). Cloud condition images can provide detailed information on changes in cloud morphology, and especially provide accurate supplementary data for judging cloud height.

[0087] Echo intensity data: The echo intensity data provided by weather radar reflects the density of clouds and the intensity of precipitation. This data combined with image data helps to more accurately determine the height and structure of clouds.

[0088] By inputting the above three types of data into the cloud condition recognition model, the system can make judgments on cloud conditions based on multi-dimensional information and obtain observation data including cloud height and cloud category, thereby improving the observation accuracy of weather phenomena.

[0089] As a further optional embodiment, the image to be observed includes a precipitation image, and the step of inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data is specifically as follows:

[0090] Inputting the precipitation image and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0091] The precipitation observation data includes precipitation intensity and precipitation type.

[0092] In this step, the precipitation recognition model uses multi-source data, including precipitation images and real-time precipitation data obtained by rain sensors, as follows:

[0093] Precipitation images: Precipitation scene images acquired by image acquisition equipment, used to identify the visual characteristics of precipitation, such as the density, shape and distribution of raindrops. Precipitation images can provide visual precipitation feature information and help identify precipitation types, such as showers, drizzle, and heavy rain.

[0094] Real-time precipitation data: Real-time data collected by rain sensors, providing physical measurements of precipitation intensity and precipitation rate. Real-time precipitation data is obtained through a variety of rain sensors, such as:

[0095] Piezoelectric rain sensor: measures precipitation intensity by sensing the electrical signal generated by raindrop impact.

[0096] Raindrop spectrometer: Analyzes the size distribution and falling speed of raindrops to provide accurate precipitation information.

[0097] Optical rain sensor: detects the number and intensity of precipitation particles through the refraction and reflection of light.

[0098] By inputting precipitation images and real-time precipitation data into the precipitation recognition model, the model can combine visual features with physical measurements to obtain more accurate precipitation observation data. Specific precipitation observation data includes:

[0099] Precipitation intensity: Indicates the actual intensity of precipitation, such as rainfall per hour (mm / h).

[0100] Precipitation type: Distinguish different types of precipitation, such as light rain, heavy rain, rainstorm, snow, etc.

[0101] This implementation method effectively improves the accuracy and comprehensiveness of precipitation observations by integrating multi-source data, ensuring the reliability of weather observation results.

[0102] As a further optional embodiment, the step of integrating the cloud condition observation data, the precipitation observation data and the lightning observation data to obtain a weather observation result specifically includes:

[0103] Performing weight allocation on the cloud condition observation data, the precipitation observation data, and the lightning observation data;

[0104] The cloud condition observation data, the precipitation observation data and the lightning observation data are weightedly fused to obtain a weather observation result.

[0105] In this step, weights are assigned to cloud condition observation data, precipitation observation data, and lightning observation data according to the degree of influence of each type of data on different weather phenomena. Different weights reflect the importance of each type of data to the final weather phenomenon identification result. For example:

[0106] Cloud condition observation data may have a greater weight in judging weather conditions such as cloudy and clear.

[0107] Precipitation observation data plays a more important role in the judgment of rainy days and therefore has a higher weight when it rains.

[0108] Lightning observation data plays a key role in the identification of thunderstorms or severe convective weather, and therefore has a higher weight in lightning weather.

[0109] The specific weight allocation can be dynamically adjusted to adapt to different weather types based on historical data and weather model training results.

[0110] The weighted fusion process combines the data after the above weight assignment to finally generate weather observation results. The fusion model uses the following steps:

[0111] Weighted processing: Weight cloud conditions, precipitation, and lightning data according to the assigned weights to amplify the impact of important data.

[0112] Data fusion: These multi-dimensional data are integrated through weighted averaging, rule fusion or machine learning algorithms to generate a single weather observation result.

[0113] Output weather phenomena: After weighted fusion, the results can be used to obtain the current comprehensive weather phenomena, such as sunny, cloudy, light rain, heavy rain, thunderstorm, etc.

[0114] This weighted fusion method can make full use of the advantages of multi-source observation data, combine the characteristics and weights of different data, obtain accurate weather observation results, and improve the reliability and accuracy of the system. This implementation method ensures that the system can make reasonable judgments under various weather conditions and is applicable to various environments.

[0115] The intelligent observation device for weather phenomena provided by the present invention is described below. As shown in FIG. x, the intelligent observation device for weather phenomena described below and the intelligent observation method for weather phenomena described above can be referenced to each other.

[0116] An intelligent observation device for weather phenomena, comprising:

[0117] The first data acquisition module 210 is used to acquire the image to be observed through a plurality of image acquisition devices and acquire the echo intensity data through the weather radar;

[0118] A cloud condition recognition module 220, configured to input the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data;

[0119] The second data acquisition module 230 is used to acquire real-time precipitation data through a rainfall sensor;

[0120] A precipitation recognition module 240 is used to input the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0121] The third data acquisition module 250 is used to acquire the audio to be detected through a plurality of recording and collection devices;

[0122] A lightning recognition module 260, configured to input the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data;

[0123] The data integration module 270 is used to integrate the cloud condition observation data, the precipitation observation data and the lightning observation data to obtain weather observation results.

[0124] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the intelligent observation method of the weather phenomenon, which includes:

[0125] The image to be observed is obtained through a number of image acquisition devices, and the echo intensity data is obtained through the weather radar;

[0126] Inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data;

[0127] Get real-time precipitation data through rain sensors;

[0128] Inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0129] Acquire the audio to be detected through several recording and collection devices;

[0130] Inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data;

[0131] The cloud condition observation data, the precipitation observation data and the lightning observation data are integrated to obtain a weather observation result.

[0132] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM, Read-On l yMemory), a random access memory (RAM, Random Access Memory), a disk or an optical disk.

[0133] On the other hand, the present invention further provides a computer program product, the computer program product comprising a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the intelligent observation method of weather phenomena provided by the above methods, the method comprising:

[0134] The image to be observed is obtained through a number of image acquisition devices, and the echo intensity data is obtained through the weather radar;

[0135] Inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data;

[0136] Get real-time precipitation data through rain sensors;

[0137] Inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0138] Acquire the audio to be detected through several recording and collection devices;

[0139] Inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data;

[0140] The cloud condition observation data, the precipitation observation data and the lightning observation data are integrated to obtain a weather observation result.

[0141] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the intelligent observation method of weather phenomena provided by the above methods, the method comprising:

[0142] The image to be observed is obtained through a number of image acquisition devices, and the echo intensity data is obtained through the weather radar;

[0143] Inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data;

[0144] Get real-time precipitation data through rain sensors;

[0145] Inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data;

[0146] Acquire the audio to be detected through several recording and collection devices;

[0147] Inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data;

[0148] The cloud condition observation data, the precipitation observation data and the lightning observation data are integrated to obtain a weather observation result.

[0149] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0150] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent observation method for weather phenomena, characterized in that: include: The image to be observed is obtained through a number of image acquisition devices, and the echo intensity data is obtained through the weather radar; Inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data; Get real-time precipitation data through rain sensors; Inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data; Acquire the audio to be detected through several recording and collection devices; Inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data; The cloud condition observation data, the precipitation observation data and the lightning observation data are integrated to obtain a weather observation result.

2. The intelligent observation method of weather phenomena according to claim 1, characterized in that: After the step of inputting the image to be observed and the audio to be detected into the lightning recognition model to obtain lightning observation data, the method further includes: Acquire an image to be detected containing a fixed target object; Visibility data is obtained according to the clarity of the image to be detected.

3. The intelligent observation method of weather phenomena according to claim 1, characterized in that: The image acquisition device is a full-color fisheye camera, and the full-color fisheye camera is arranged on a pan-tilt platform. The step of acquiring the image to be observed by using a plurality of image acquisition devices specifically includes: At preset time intervals, the pan / tilt is controlled to follow a preset acquisition route, and the full-color fisheye camera is controlled to shoot and acquire at a preset position to obtain the image to be observed.

4. The intelligent observation method of weather phenomena according to claim 1, characterized in that: The rain sensor comprises a piezoelectric rain sensor, a raindrop spectrometer and an optical rain sensor.

5. The intelligent observation method of weather phenomena according to claim 1, characterized in that: The image to be observed includes a satellite cloud image and a cloud condition image. The step of inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data is specifically as follows: Inputting the satellite cloud image, the cloud shape image and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data; The satellite cloud image is used for judging the cloud type and cloud height, and the cloud shape image is used for judging the cloud height.

6. The intelligent observation method of weather phenomena according to claim 1, characterized in that: The image to be observed includes a precipitation image. The step of inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data is specifically as follows: Inputting the precipitation image and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data; The precipitation observation data includes precipitation intensity and precipitation type.

7. The intelligent observation method of weather phenomena according to claim 1, characterized in that: The step of integrating the cloud condition observation data, the precipitation observation data and the lightning observation data to obtain a weather observation result specifically includes: Performing weight allocation on the cloud condition observation data, the precipitation observation data, and the lightning observation data; The cloud condition observation data, the precipitation observation data and the lightning observation data are weightedly fused to obtain a weather observation result.

8. An intelligent observation device for weather phenomena, characterized in that: include: A first data acquisition module is used to acquire the image to be observed through a plurality of image acquisition devices and to acquire echo intensity data through a weather radar; A cloud condition recognition module, used for inputting the image to be observed and the echo intensity data into a cloud condition recognition model to obtain cloud condition observation data, wherein the cloud condition observation data includes cloud height data and cloud category data; A second data acquisition module is used to acquire real-time precipitation data through a rainfall sensor; A precipitation recognition module, used for inputting the image to be observed and the real-time precipitation data into a precipitation recognition model to obtain precipitation observation data; A third data acquisition module is used to acquire the audio to be detected through a plurality of recording and collection devices; A lightning recognition module, used for inputting the image to be observed and the audio to be detected into a lightning recognition model to obtain lightning observation data; The data integration module is used to integrate the cloud condition observation data, the precipitation observation data and the lightning observation data to obtain weather observation results.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the intelligent observation method of the weather phenomenon as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the intelligent observation method of the weather phenomenon as described in any one of claims 1 to 7 is implemented.

Images