Sea wave height prediction method and device, storage medium and computer program product
By training a regression model with historical sea wave features from satellite imagery, the method addresses the limitations of fixed-location monitoring, providing comprehensive and proactive sea wave height predictions.
Patent Information
- Application Number
- CN202510314593.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-17
- Publication Date
- 2025-07-15
AI Technical Summary
Traditional wave height monitoring methods can only collect data at fixed locations, resulting in limited monitoring range and inability to fully cover vast oceans. Especially in areas where wave activity is frequent or extreme weather events occur frequently, the problem of incomplete monitoring is prominent.
By extracting wave features from remote sensing images, training a regression model, establishing an annotated data set based on buoy data, predicting wave height, and decomposing the remote sensing images based on multiple resolutions to merge feature input models to achieve wave height prediction.
It realizes wave height monitoring in a vast ocean area, can predict wave height and evaluate sea surface hazard levels, provide timely alarms and emergency response measures, and improves the comprehensiveness and accuracy of wave monitoring.
Smart Images
Figure CN120318294A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of ocean engineering, and in particular, to a method and device for predicting wave height, a storage medium, and a computer program product. Background Art
[0002] Ocean environmental monitoring is an indispensable part of modern ocean engineering, climate research, coastal disaster prevention and mitigation, etc. Among them, the real-time monitoring of wave height is particularly important. Although traditional monitoring methods show high precision and reliability in wave height monitoring, they can only collect data at fixed positions and cannot comprehensively cover the vast sea area. For example, buoys are usually only deployed in key channels or specific research areas, and their early warning capabilities for large-scale wave change trends and extreme marine events are limited. This not only limits the scope of ocean environmental monitoring, but also makes it difficult to comprehensively understand the changing trend of wave height, especially in areas with frequent wave activities or frequent extreme weather events, the problem of incomplete monitoring is particularly prominent.
[0003] In view of the related art, traditional monitoring methods can only collect data at fixed positions, have limited monitoring range for waves, and are prone to the problem of incomplete monitoring. At present, no effective solution has been proposed.
[0004] Therefore, it is necessary to improve the related art to overcome the above-mentioned defects in the related art. Summary of the Invention
[0005] Embodiments of the present application provide a method and device for predicting wave height, a storage medium, and a computer program product, so as to at least solve the problem in the related art that traditional monitoring methods can only collect data at fixed positions, have limited monitoring range for waves, and are prone to the problem of incomplete monitoring.
[0006] According to an aspect of the embodiments of the present application, a method for predicting wave height is provided, including: determining first wave features related to wave height from a first remote sensing image in a first time period, and training a regression model through the first wave features to obtain a target regression model, where the first time period is a period of time before the current time; decomposing a second remote sensing image in a second time period according to a plurality of target resolutions to obtain a plurality of third remote sensing images, where the second time period is a period of time before the current time, and the second time period is later than the first time period, and the resolutions of the plurality of third remote sensing images correspond to the plurality of target resolutions one by one; obtaining a plurality of second wave features corresponding to the plurality of third remote sensing images one by one, combining the plurality of second wave features into a third wave feature, and inputting the third wave feature into the target regression model to predict the wave height at a target time, where the target time is a time point after the current time.
[0007] In an exemplary embodiment, training a regression model with the first sea wave feature to obtain a target regression model includes: determining a buoy position where a buoy is located within the first time period in the first remote sensing image, where the buoy is used to record the actual sea wave height at the buoy position; determining a plurality of fourth sea wave features corresponding to the plurality of buoy positions in the first sea wave feature, where the first sea wave feature includes the plurality of fourth sea wave features, and the plurality of buoy positions correspond to the plurality of fourth sea wave features one by one; establishing an annotation data set according to the one-to-one correspondence between the plurality of fourth sea wave features and the plurality of actual sea wave heights, and training the regression model with the annotation data set to obtain the target regression model.
[0008] In an exemplary embodiment, after inputting the third sea wave feature into the target regression model to predict the sea wave height at a target time, the method further includes: when the sea wave height is less than a first preset value, determining that the danger level of the sea surface is a third-level danger; when the sea wave height is greater than the first preset value and less than a second preset value, determining that the danger level of the sea surface is a second-level danger, where the danger degree of the second-level danger is higher than that of the third-level danger; when the sea wave height is greater than the second preset value and less than a third preset value, determining that the danger level of the sea surface is a first-level danger, where the danger degree of the first-level danger is higher than that of the second-level danger.
[0009] In an exemplary embodiment, after inputting the third sea wave feature into the target regression model to predict the sea wave height at a target time, the method further includes: when the danger level is the first-level danger, issuing a first-level alarm, where the first-level alarm is used to remind a target object to initiate an emergency response; when the danger level is the second-level danger, issuing a second-level alarm, where the second-level alarm is used to remind the target object to prepare emergency supplies; when the danger level is the third-level danger, continuously monitoring the sea wave height.
[0010] In an exemplary embodiment, before determining the first sea wave feature related to the sea wave height from the first remote sensing image within the first time period, the method further includes: acquiring the first remote sensing image within the first time period and performing denoising processing on the first remote sensing image; determining the target pixel value f new (x, y) of the denoised first remote sensing image according to the following Gaussian filtering formula, where w(i, j) is the weight function of the Gaussian filter, m is the radius of the Gaussian filter, and f(x + i, y + j) is the initial pixel value of the first remote sensing image.
[0011] In an exemplary embodiment, merging the plurality of second sea wave features into a third sea wave feature includes: performing scale normalization on the plurality of second sea wave features to obtain a plurality of fifth sea wave features at the same scale; and merging the plurality of fifth sea wave features according to the weight coefficients corresponding to the plurality of fifth sea wave features to obtain the third sea wave feature.
[0012] According to another aspect of the embodiments of the present application, there is also provided a device for predicting sea wave height, including: a determination module, configured to determine first sea wave features related to sea wave height from a first remote sensing image within a first time period, and train a regression model through the first sea wave features to obtain a target regression model, where the first time period is a period of time before the current time; a decomposition module, configured to decompose a second remote sensing image within a second time period according to a plurality of target resolutions to obtain a plurality of third remote sensing images, where the second time period is a period of time before the current time and the second time period is later than the first time period, and the resolutions of the plurality of third remote sensing images correspond one-to-one to the plurality of target resolutions; and a prediction module, configured to obtain a plurality of second sea wave features corresponding one-to-one to the plurality of third remote sensing images, merge the plurality of second sea wave features into a third sea wave feature, and input the third sea wave feature into the target regression model to predict the sea wave height at a target time, where the target time is a time point after the current time.
[0013] According to yet another aspect of the embodiments of the present application, there is also provided a computer-readable storage medium storing a computer program, where the computer program is configured to execute the above-mentioned method for predicting sea wave height when running.
[0014] According to yet another aspect of the embodiments of the present application, there is also provided an electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the above-mentioned processor executes the above-mentioned method for predicting sea wave height through the computer program.
[0015] According to yet another aspect of the embodiments of the present application, there is also provided a computer program product including a computer program, where the steps of the methods in the various embodiments of the present application are implemented when the computer program is executed by a processor.
[0016] Through this application, first wave features related to wave height are determined from a first remote sensing image within a first time period, and a regression model is trained using the first wave features to obtain a target regression model, where the first time period is a period of time before the current time; the second remote sensing image within a second time period is decomposed according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond one-to-one to the multiple target resolutions; multiple second wave features corresponding to the multiple third remote sensing images are obtained, the multiple second wave features are combined into a third wave feature, and the third wave feature is input into the target regression model to predict the wave height at the target time, where the target time is a time point after the current time. Thereby, in the related art, the problem that traditional monitoring methods can only collect data at fixed positions, have a limited monitoring range for waves, and are prone to incomplete monitoring is solved. Brief Description of the Drawings
[0017] The drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.
[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.
[0019] Figure 1 It is a hardware structure block diagram of a computer terminal for a method of predicting wave height according to an embodiment of this application;
[0020] Figure 2 It is a flowchart of a method of predicting wave height according to an embodiment of this application;
[0021] Figure 3 It is a structure block diagram of a device for predicting wave height according to an embodiment of this application. Detailed Description of the Embodiments
[0022] To enable those skilled in the art of this technology to better understand the solutions of this application, the following will clearly and completely describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts should fall within the scope of protection of this application.
[0023] It should be noted that the terms "first", "second", etc. in the description, claims and the above-mentioned drawings of this application are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order different from those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.
[0024] The method embodiments provided in the embodiments of this application can be executed on a computer terminal or a similar computing device. Taking the operation on a computer terminal as an example, Figure 1 is a hardware structure block diagram of a computer terminal for a method of predicting sea wave height according to an embodiment of this application. As Figure 1 shown, the computer terminal may include one or more ( Figure 1 only one is shown in the figure) processors 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor (Central Processing Unit, MCU) or a field programmable gate array (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data. Among them, the above-mentioned computer terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those of ordinary skill in the art can understand that Figure 1 the structure shown is only schematic and does not limit the structure of the above-mentioned computer terminal. For example, the computer terminal may further include more or fewer components than those shown in Figure 1 the figure, or have a different configuration from that shown in Figure 1 the figure.
[0025] The memory 104 can be used to store computer programs. For example, software programs and modules of application software, such as the computer program corresponding to the method of predicting sea wave height in the embodiments of this application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above-mentioned method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely disposed relative to the processor 102, and these remote memories can be connected to the computer terminal through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and their combinations.
[0026] A wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, abbreviated as NIC), which can be connected to other network devices through a base station and thus communicate with the Internet. In one example, the transmission device 106 can be a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.
[0027] In this embodiment, a method for predicting the sea wave height is provided, which is applied to the above computer terminal. Figure 2 It is a flowchart of a method for predicting the sea wave height according to an embodiment of the present application, as Figure 2 shown. The process includes the following steps:
[0028] Step S202, determine the first sea wave features related to the sea wave height from the first remote sensing image in the first time period, and train a regression model through the first sea wave features to obtain a target regression model, where the first time period is a period of time before the current time;
[0029] Step S204, decompose the second remote sensing image in the second time period according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time, and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond to the multiple target resolutions one by one;
[0030] Step S206, obtain the multiple second sea wave features corresponding to the multiple third remote sensing images one by one, merge the multiple second sea wave features into a third sea wave feature, and input the third sea wave feature into the target regression model to predict the sea wave height at the target time, where the target time is a time point after the current time.
[0031] Through the above steps, determine the first wave characteristics related to the wave height from the first remote sensing image in the first time period, and train a regression model with the first wave characteristics to obtain a target regression model, where the first time period is a period of time before the current time; decompose the second remote sensing image in the second time period according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond one-to-one to the multiple target resolutions; obtain multiple second wave characteristics corresponding one-to-one to the multiple third remote sensing images, merge the multiple second wave characteristics into a third wave characteristic, and input the third wave characteristic into the target regression model to predict the wave height at the target time, where the target time is a time point after the current time. Thus, it solves the problem in the related art that traditional monitoring methods can only collect data at fixed positions, have a limited monitoring range for waves, and are prone to incomplete monitoring.
[0032] In an exemplary embodiment, training a regression model with the first wave characteristics to obtain a target regression model includes: determining the buoy position where the buoy is located in the first time period in the first remote sensing image, where the buoy is used to record the actual wave height at the buoy position; determining multiple fourth wave characteristics corresponding to the multiple buoy positions in the first wave characteristics according to the buoy position, where the first wave characteristics include the multiple fourth wave characteristics, and the multiple buoy positions correspond one-to-one to the multiple fourth wave characteristics; establishing an annotation data set according to the one-to-one correspondence between the multiple fourth wave characteristics and the multiple actual wave heights, and training the regression model with the annotation data set to obtain the target regression model.
[0033] In the remote sensing image data acquisition stage, a first remote sensing image containing ocean surface information is obtained. The first remote sensing image covers a wide geographical area and contains rich wave features. However, the remote sensing image itself cannot directly provide the accurate value of the wave height, so it needs to be combined with the buoy data on site. The buoy can record the wave height at its location in real time and provide a reference standard for on-site measurement. Through geographic information system technology, the geographical location of each buoy within the first time period can be determined in the first remote sensing image, that is, the buoy position. After determining the buoy position, methods such as gray level co-occurrence matrix and Canny edge detection are used to extract the fourth wave features related to each buoy position. Since the remote sensing image covers multiple buoy positions, multiple fourth wave features will be extracted from the first remote sensing image, and each fourth wave feature corresponds to the wave height data of an actual buoy position. A labeled data set is constructed based on the actual wave height data of the buoy position and the corresponding fourth wave features. The labeled data set is used to train a regression model. The purpose of the regression model is to learn the relationship between the input features and the target output (i.e., the wave height at the target time) so as to make predictions for new inputs. In this embodiment, a support vector machine regression model is adopted, which finds a hyperplane by optimizing an objective function, making the training data points as close to this hyperplane as possible and maintaining the prediction accuracy for unseen data. After the model training is completed, a target regression model will be obtained, and this target regression model can predict the wave height based on the wave features in the remote sensing image.
[0034] In an exemplary embodiment, after inputting the third wave feature into the target regression model to predict the wave height at the target time, the method further includes: when the wave height is less than a first preset value, determining that the danger level of the sea surface is level-three danger; when the wave height is greater than the first preset value and less than a second preset value, determining that the danger level of the sea surface is level-two danger, where the danger level of level-two danger is higher than that of level-three danger; when the wave height is greater than the second preset value and less than a third preset value, determining that the danger level of the sea surface is level-one danger, where the danger level of level-one danger is higher than that of level-two danger.
[0035] When the predicted wave height is less than the first preset value (e.g., 0.5 meters), the computer terminal evaluates the sea surface danger level as level three, which indicates that the sea surface is relatively calm and has little impact on navigation and offshore operations. For example, on a calm day, the wave height may generally be lower than 0.5 meters, and most areas of the sea surface are marked as level three danger level at this time. For the level three danger level, the computer terminal does not trigger a special warning mechanism. When the wave height is greater than the first preset value (e.g., 0.5 meters) but less than the second preset value (e.g., 2 meters), the computer terminal raises the danger level to level two, indicating that the sea surface condition begins to become complex and may have a certain impact on ships or offshore facilities. For example, in a weather with moderate wind and waves, the wave height may be between 0.5 meters and 2 meters, and the affected sea surface area is marked as level two danger level at this time. For the level two danger level, the computer terminal sends a navigation warning to relevant ships, reminding them to pay attention to the wave changes and adopt appropriate navigation strategies. When the wave height is greater than the second preset value (e.g., 2 meters) and less than the third preset value (e.g., 5 meters), the sea surface danger level is raised to level one, which indicates that the sea surface condition is very bad and poses a serious threat to navigation and maritime safety. For example, in bad weather conditions, such as strong storms or frequent wave activities, the wave height may reach between 2 meters and 5 meters. At this time, the computer terminal marks the danger area as level one danger level and triggers an emergency warning mechanism, sending a strong wave warning to all nearby ships and suggesting that the ships immediately take evasive measures, such as changing the course or finding a safe haven.
[0036] In an exemplary embodiment, after inputting the third wave feature into the target regression model to predict the wave height at the target time, the method further includes: in the case where the danger level is the first-level danger, issuing a first-level alarm, where the first-level alarm is used to remind the target object to initiate an emergency response; in the case where the danger level is the second-level danger, issuing a second-level alarm, where the second-level alarm is used to remind the target object to prepare emergency supplies; in the case where the danger level is the third-level danger, continuously monitoring the wave height.
[0037] When it is determined that the wave height is below the first preset value and the danger level is level three, it indicates that the sea surface conditions are relatively stable and will not pose a direct threat to the normal operation of the wind farm and the safety of maintenance personnel. For example, when the wave height is 0.3 meters, which is lower than the preset 0.5 meters, the computer terminal will continuously monitor the wave height and will not issue an alarm. The operation and maintenance team can carry out daily operation and maintenance and inspections according to normal operations. If the computer terminal detects that the wave height rises above the first preset value but is below the second preset value, such as when the wave height reaches 1.5 meters (assuming the first preset value is 0.5 meters and the second preset value is 2 meters), the danger level will be upgraded to level two. At this time, the computer terminal will automatically issue a level-two alarm to remind the operation and maintenance team to prepare emergency supplies. The emergency supplies include additional safety equipment, repair tools, waterproof equipment, etc., to deal with possible equipment damage or personnel safety problems caused by moderate waves. At the same time, the operation and maintenance team may need to adjust the work plan, such as postponing non-essential offshore operations, to ensure personnel safety. When the wave height further increases and exceeds the second preset value (such as reaching 3.5 meters, assuming the second preset value is 2 meters and the third preset value is 5 meters), the computer terminal evaluates the danger level as level one, which means that the sea surface conditions are extremely harsh and may cause significant damage to the wind farm or threaten the safety of personnel. In this case, a level-one alarm is immediately issued to remind the target object to initiate a comprehensive emergency response. After receiving the level-one alarm, the operation and maintenance team needs to immediately execute the emergency response plan, which includes, but is not limited to, evacuating offshore personnel, ensuring that all equipment is in a safe state, maintaining close contact with the ocean forecasting center to obtain the latest developments, and taking all necessary measures to protect the assets and personnel safety of the wind farm.
[0038] In an exemplary embodiment, before determining the first wave feature related to the wave height from the first remote sensing image in the first time period, the method further includes: acquiring the first remote sensing image in the first time period and performing denoising processing on the first remote sensing image; determining the target pixel value f new (x,y) of the denoised first remote sensing image according to the following Gaussian filtering formula where w(i,j) is the weight function of the Gaussian filter, m is the radius of the Gaussian filter, and f(x+i,y+j) is the initial pixel value of the first remote sensing image.
[0039] Optionally, the target pixel value f new (x,y) of the denoised first remote sensing image can also be determined by the following median filtering formula, f new (x,y) = median(f(x , ,y , )) where f(x , ,y ,) is the value of the pixel points within N(x, y), where N(x, y) is a rectangular or circular area centered at (x, y), and its size is determined by the radius m of the filter. Specifically, the values of all pixel points within N(x, y) are extracted from the first remote sensing image to form a set. All the pixel values in this set are sorted, either from smallest to largest or from largest to smallest. The pixel value at the middle position is selected from the sorted set as the denoised pixel value. If the number of elements in the set is odd, the median is the value in the exact middle; if the number of elements in the set is even, the median is the average of the two middle values. The above process is repeated for each pixel point in the first remote sensing image until the entire first remote sensing image is processed.
[0040] In an exemplary embodiment, merging the multiple second sea wave features into a third sea wave feature includes: performing scale normalization on the multiple second sea wave features to obtain multiple fifth sea wave features at the same scale; merging the multiple fifth sea wave features according to the weight coefficients corresponding to the multiple fifth sea wave features to obtain the third sea wave feature.
[0041] Optionally, determine a common scale range, such as mapping all feature values to between [0, 1] or [-1, 1], and calculate the minimum and maximum values of each second sea wave feature. Use min-max normalization or other normalization techniques, such as Z-Score normalization, to normalize each feature value into the determined scale range. The normalized features are called fifth sea wave features, and a weight coefficient is assigned to each fifth sea wave feature according to the importance of the feature or the weights learned during the model training process. The multiple fifth sea wave features are weighted and summed according to their corresponding weight coefficients to obtain a comprehensive feature, that is, the third sea wave feature.
[0042] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present application.
[0043] In this embodiment, a device for predicting the sea wave height is further provided. This device is used to implement the above-mentioned embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.
[0044] Figure 3 FIG. is a structural block diagram of a device for predicting the sea wave height according to an embodiment of the present application. The device includes:
[0045] A determination module 30, configured to determine a first sea wave feature related to the sea wave height from a first remote sensing image within a first time period, and train a regression model through the first sea wave feature to obtain a target regression model, where the first time period is a period of time before the current time;
[0046] A decomposition module 32, configured to decompose a second remote sensing image within a second time period according to a plurality of target resolutions to obtain a plurality of third remote sensing images, where the second time period is a period of time before the current time, and the second time period is later than the first time period, and the resolutions of the plurality of third remote sensing images correspond to the plurality of target resolutions one by one;
[0047] A prediction module 34, configured to obtain a plurality of second sea wave features corresponding to the plurality of third remote sensing images one by one, merge the plurality of second sea wave features into a third sea wave feature, and input the third sea wave feature into the target regression model to predict the sea wave height at a target time, where the target time is a time point after the current time.
[0048] Through the above device, a first sea wave feature related to the sea wave height is determined from a first remote sensing image within a first time period, and a regression model is trained through the first sea wave feature to obtain a target regression model, where the first time period is a period of time before the current time; the second remote sensing image within a second time period is decomposed according to a plurality of target resolutions to obtain a plurality of third remote sensing images, where the second time period is a period of time before the current time, and the second time period is later than the first time period, and the resolutions of the plurality of third remote sensing images correspond to the plurality of target resolutions one by one; a plurality of second sea wave features corresponding to the plurality of third remote sensing images one by one are obtained, the plurality of second sea wave features are merged into a third sea wave feature, and the third sea wave feature is input into the target regression model to predict the sea wave height at a target time, where the target time is a time point after the current time. Thus, the problem in the related art that traditional monitoring methods can only collect data at fixed positions, have a limited monitoring range for sea waves, and are prone to incomplete monitoring is solved.
[0049] In an exemplary embodiment, the determining module 30 is further configured to determine the buoy position where the buoy is located within the first time period in the first remote sensing image, where the buoy is used to record the actual sea wave height at the buoy position; determine a plurality of fourth sea wave features corresponding to the plurality of buoy positions in the first sea wave features, where the first sea wave features include the plurality of fourth sea wave features, and the plurality of buoy positions correspond to the plurality of fourth sea wave features one by one; establish an annotation data set according to the one-to-one correspondence between the plurality of fourth sea wave features and the plurality of actual sea wave heights, and train the regression model through the annotation data set to obtain the target regression model.
[0050] In an exemplary embodiment, the prediction module 34 is further configured to determine that the danger level of the sea surface is a third-level danger when the sea wave height is less than a first preset value; determine that the danger level of the sea surface is a second-level danger when the sea wave height is greater than the first preset value and less than a second preset value, where the danger level of the second-level danger is higher than that of the third-level danger; determine that the danger level of the sea surface is a first-level danger when the sea wave height is greater than the second preset value and less than a third preset value, where the danger level of the first-level danger is higher than that of the second-level danger.
[0051] In an exemplary embodiment, the prediction module 34 is further configured to issue a first-level alarm when the danger level is the first-level danger, where the first-level alarm is used to remind the target object to initiate an emergency response; issue a second-level alarm when the danger level is the second-level danger, where the second-level alarm is used to remind the target object to prepare emergency supplies; continuously monitor the sea wave height when the danger level is the third-level danger.
[0052] In an exemplary embodiment, the determining module 30 is further configured to obtain the first remote sensing image within the first time period and perform denoising processing on the first remote sensing image; determine the target pixel value f new (x,y) of the denoised first remote sensing image according to the following Gaussian filtering formula where w(i,j) is the weight function of the Gaussian filter, m is the radius of the Gaussian filter, and f(x+i,y+j) is the initial pixel value of the first remote sensing image.
[0053] In an exemplary embodiment, the prediction module 34 is further configured to perform scale normalization on the multiple second sea wave features to obtain multiple fifth sea wave features at the same scale; and combine the multiple fifth sea wave features according to the weight coefficients corresponding to the multiple fifth sea wave features to obtain the third sea wave feature.
[0054] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, and the computer program is configured to execute the steps in any one of the above method embodiments when running.
[0055] Optionally, in this embodiment, the above storage medium may be configured to store a computer program for performing the following steps:
[0056] S1, determining first sea wave features related to sea wave height from a first remote sensing image in a first time period, and training a regression model through the first sea wave features to obtain a target regression model, where the first time period is a period of time before the current time;
[0057] S2, decomposing a second remote sensing image in a second time period according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time, and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond one-to-one to the multiple target resolutions;
[0058] S3, obtaining multiple second sea wave features corresponding to the multiple third remote sensing images one by one, combining the multiple second sea wave features into a third sea wave feature, and inputting the third sea wave feature into the target regression model to predict the sea wave height at a target time, where the target time is a time point after the current time.
[0059] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: various media such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disc that can store a computer program.
[0060] Specific examples in this embodiment may refer to the examples described in the above embodiments and exemplary embodiments, and will not be repeated here.
[0061] An embodiment of the present application further provides an electronic device, including a memory and a processor, where a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.
[0062] Optionally, in this embodiment, the above-mentioned processor may be configured to perform the following steps by a computer program:
[0063] S1. Determine first wave features related to wave height from a first remote sensing image within a first time period, and train a regression model with the first wave features to obtain a target regression model, where the first time period is a period of time before the current time;
[0064] S2. Decompose a second remote sensing image within a second time period according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time, and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond one-to-one to the multiple target resolutions;
[0065] S3. Obtain multiple second wave features corresponding to the multiple third remote sensing images one by one, merge the multiple second wave features into a third wave feature, and input the third wave feature into the target regression model to predict the wave height at a target time, where the target time is a time point after the current time.
[0066] In an exemplary embodiment, the above-mentioned electronic device may further include a transmission device and an input / output device, where the transmission device is connected to the above-mentioned processor, and the input / output device is connected to the above-mentioned processor.
[0067] An embodiment of the present application also provides a computer program product, including a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium stores a computer program product, and when the computer program is executed by a processor, it implements the steps of the methods in the various embodiments of the present application.
[0068] Optionally, in this embodiment, the above-mentioned computer program may be configured to implement the following steps when executed by a processor:
[0069] S1. Determine first wave features related to wave height from a first remote sensing image within a first time period, and train a regression model with the first wave features to obtain a target regression model, where the first time period is a period of time before the current time;
[0070] S2. Decompose a second remote sensing image within a second time period according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time, and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond one-to-one to the multiple target resolutions;
[0071] S3. Obtain multiple second wave features corresponding to the multiple third remote sensing images one by one, merge the multiple second wave features into a third wave feature, and input the third wave feature into the target regression model to predict the wave height at the target time, where the target time is a time point after the current time.
[0072] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary embodiments, and details thereof are not repeated herein.
[0073] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present application can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. They can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order from here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps among them can be made into a single integrated circuit module to implement. In this way, the present application is not limited to any specific combination of hardware and software.
[0074] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.
Claims
1. A method for predicting the wave height, characterized in that, Including: Determine first wave features related to wave height from a first remote sensing image within a first time period, and train a regression model using the first wave features to obtain a target regression model, where the first time period is a period of time before the current time; Decompose a second remote sensing image within a second time period according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond one-to-one to the multiple target resolutions; Obtain multiple second wave features corresponding to the multiple third remote sensing images one by one, merge the multiple second wave features into a third wave feature, and input the third wave feature into the target regression model to predict the wave height at a target time, where the target time is a time point after the current time.
2. The method for predicting the wave height according to claim 1, wherein Training a regression model using the first wave features to obtain a target regression model includes: Determine a buoy position where a buoy is located within the first time period in the first remote sensing image, where the buoy is used to record the actual wave height at the buoy position; Determine multiple fourth wave features corresponding to the multiple buoy positions in the first wave features according to the buoy position, where the first wave features include the multiple fourth wave features, and the multiple buoy positions correspond one-to-one to the multiple fourth wave features; Establish an annotation data set according to the one-to-one correspondence between the multiple fourth wave features and the multiple actual wave heights, and train the regression model using the annotation data set to obtain the target regression model.
3. The method for predicting the wave height according to claim 1, characterized in that, After inputting the third wave feature into the target regression model to predict the wave height at the target time, the method further includes: when the wave height is less than a first preset value, determining that the danger level of the sea surface is level three danger; When the wave height is greater than the first preset value and less than a second preset value, determining that the danger level of the sea surface is level two danger, where the danger level of level two danger is higher than that of level three danger; When the wave height is greater than the second preset value and less than a third preset value, determining that the danger level of the sea surface is level one danger, where the danger level of level one danger is higher than that of level two danger.
4. The prediction method of wave height according to claim 3, characterized in that, After inputting the third wave feature into the target regression model to predict the wave height at the target time, the method further includes: when the danger level is level one danger, issuing a level one alarm, where the level one alarm is used to remind the target object to initiate an emergency response; When the danger level is level two danger, issuing a level two alarm, where the level two alarm is used to remind the target object to prepare emergency supplies; When the danger level is level three danger, continuously monitor the wave height.
5. The prediction method of the sea wave height according to claim 1, wherein Before determining the first wave feature related to the wave height from the first remote sensing image in the first time period, the method further includes: acquiring the first remote sensing image in the first time period and performing denoising processing on the first remote sensing image; Determine the target pixel value f of the denoised first remote sensing image according to the following Gaussian filtering formula new (x, y), where w(i, j) is the weight function of the Gaussian filter, m is the radius of the Gaussian filter, and f(x + i, y + j) is the initial pixel value of the first remote sensing image.
6. The prediction method of wave height according to claim 1, wherein Merging the multiple second wave features into a third wave feature includes: Performing scale normalization on the multiple second wave features to obtain multiple fifth wave features at the same scale; Merging the multiple fifth wave features according to the weight coefficients corresponding to the multiple fifth wave features to obtain the third wave feature.
7. A device for predicting the height of ocean waves, characterized in that, Includes: A determination module, configured to determine a first wave feature related to the wave height from a first remote sensing image in a first time period, and train a regression model through the first wave feature to obtain a target regression model, where the first time period is a period of time before the current time; A decomposition module, configured to decompose a second remote sensing image in a second time period according to multiple target resolutions to obtain multiple third remote sensing images, where the second time period is a period of time before the current time and the second time period is later than the first time period, and the resolutions of the multiple third remote sensing images correspond one-to-one to the multiple target resolutions; A prediction module, configured to acquire multiple second wave features corresponding one-to-one to the multiple third remote sensing images, merge the multiple second wave features into a third wave feature, and input the third wave feature into the target regression model to predict the wave height at a target time, where the target time is a time point after the current time.
8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, where the program, when running, executes the method according to any one of claims 1 to 6.
9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 6.