Sea surface temperature data processing method and device, electronic equipment and storage medium

Abstract

The invention relates to a sea surface temperature data processing method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining to-be-processed sea surface temperature data, time information and latitude information of the to-be-processed sea surface temperature data, and standard Gaussian noise of a target time step; performing interpolation processing on the to-be-processed sea surface temperature data and the latitude information to obtain the interpolated sea surface temperature data and the interpolated latitude information; establishing and training a diffusion model to obtain a trained diffusion model; inputting the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into a noise prediction module and a denoising module of a trained diffusion model to obtain predicted sea surface temperature residual data; and obtaining the target sea surface temperature data according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data, thereby improving the quality of the sea surface temperature data.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, electronic device and storage medium for processing sea surface temperature data. Background Art

[0002] Sea surface temperature is an important indicator in oceanography and climate science, directly affecting atmospheric circulation and global weather, such as El Niño and La Niña, which often lead to seasonal climate change and extreme weather. In addition, changes in sea surface temperature are also a signal of global climate change. Rising temperatures may intensify global warming and affect marine ecosystems and biodiversity, especially important habitats such as coral reefs. Therefore, obtaining high-resolution sea surface temperature data is critical for understanding climate change and its impacts, and protecting the health of marine ecosystems.

[0003] Most current sea surface temperature data have low spatial resolution, making it difficult to depict the actual situation in detail and unable to meet the needs of high-resolution applications. In addition, existing high-resolution remote sensing thermal infrared sea surface temperature data often suffer from data loss due to cloud cover, which also poses a challenge to practical applications. Summary of the invention

[0004] Based on this, the purpose of the present application is to provide a sea surface temperature data processing method, device, electronic device and storage medium, which can improve the spatial resolution of sea surface temperature data and make up for the lack of high-resolution sea surface temperature data.

[0005] According to a first aspect of an embodiment of the present application, a method for processing sea surface temperature data is provided, comprising the following steps: Obtaining sea surface temperature data to be processed, time information and latitude information of the sea surface temperature data to be processed, and standard Gaussian noise of a target time step; Performing interpolation processing on the sea surface temperature data and latitude information to be processed to obtain interpolated sea surface temperature data and interpolated latitude information; A diffusion model is established and trained to obtain a trained diffusion model; wherein the diffusion model includes a forward diffusion module and a reverse diffusion module, and the reverse diffusion module includes a noise prediction module and a denoising module; Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain the noisy sea surface temperature residual data at different time steps; the noise prediction module is used to perform noise prediction on the noisy sea surface temperature residual data at different time steps to obtain the predicted noise; the noise prediction module is iteratively trained through the added Gaussian noise and the predicted noise to obtain the trained diffusion model; Input the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module and the denoising module of the trained diffusion model to obtain the predicted sea surface temperature residual data; The target sea surface temperature data is obtained based on the predicted sea surface temperature residual data and the interpolated sea surface temperature data.

[0006] According to a second aspect of an embodiment of the present application, a sea surface temperature data processing device is provided, comprising: A sea surface temperature data acquisition module, used to acquire the sea surface temperature data to be processed, the time information and latitude information of the sea surface temperature data to be processed, and the standard Gaussian noise of the target time step; A sea surface temperature data interpolation module is used to interpolate the sea surface temperature data and latitude information to be processed to obtain the interpolated sea surface temperature data and the interpolated latitude information; A diffusion model training module is used to establish and train a diffusion model to obtain a trained diffusion model; wherein the diffusion model includes a forward diffusion module and a reverse diffusion module, and the reverse diffusion module includes a noise prediction module and a denoising module; Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain the noisy sea surface temperature residual data at different time steps; the noise prediction module is used to perform noise prediction on the noisy sea surface temperature residual data at different time steps to obtain the predicted noise; the noise prediction module is iteratively trained through the added Gaussian noise and the predicted noise to obtain the trained diffusion model; The sea surface temperature residual data prediction module is used to input the target time step, the standard Gaussian noise of the target time step, the latitude information of the interpolated sea surface temperature data and the time information into the noise prediction module and the denoising module of the trained diffusion model to obtain the predicted sea surface temperature residual data; The target sea surface temperature data acquisition module is used to obtain the target sea surface temperature data according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data.

[0007] According to a third aspect of an embodiment of the present application, there is provided an electronic device, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the steps of the method of the first aspect.

[0008] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the method of the first aspect are implemented.

[0009] The embodiment of the present application obtains the sea surface temperature data to be processed, the time information and latitude information of the sea surface temperature data to be processed, and the standard Gaussian noise of the target time step; interpolates the sea surface temperature data and latitude information to be processed to obtain the interpolated sea surface temperature data and the interpolated latitude information; establishes and trains a diffusion model to obtain a trained diffusion model; wherein the diffusion model includes a forward diffusion module and a reverse diffusion module, and the reverse diffusion module includes a noise prediction module and a denoising module; Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain the noisy sea surface temperature data at different time steps Temperature residual data; noise prediction is performed on the noisy sea surface temperature residual data of different time steps through the noise prediction module to obtain the predicted noise; the noise prediction module is iteratively trained through the added Gaussian noise and the predicted noise to obtain the trained diffusion model; the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information are input into the noise prediction module and the denoising module of the trained diffusion model to obtain the predicted sea surface temperature residual data; the target sea surface temperature data is obtained according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data. The embodiment of the present application first interpolates the sea surface temperature data to be processed with low spatial resolution to obtain the interpolated sea surface temperature data with high spatial resolution. The diffusion model is trained by performing forward diffusion and reverse denoising processes on the sea surface temperature residual data between the interpolated sea surface temperature data with high spatial resolution and the real sea surface temperature data with high spatial resolution, so that the diffusion model can obtain the sea surface temperature residual data from the standard Gaussian noise prediction. Based on the high spatial resolution interpolated sea surface temperature data and the predicted sea surface temperature residual data, the target sea surface temperature data is obtained, and the reconstruction from low spatial resolution sea surface temperature data to high spatial resolution sea surface temperature data is realized. The reconstructed sea surface temperature data has high accuracy, improves the quality of sea surface temperature data, and can make up for the data loss of existing real sea surface temperature data. At the same time, the input of the diffusion model includes time information and latitude information, so that the reconstructed target sea surface temperature data has spatiotemporal continuity, which improves the quality of sea surface temperature data.

[0010] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application.

[0011] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of a flow chart of a method for processing sea surface temperature data provided by an embodiment of the present application; Figure 2A structural block diagram of a sea surface temperature data processing device provided in one embodiment of the present application; Figure 3 A schematic block diagram of the structure of an electronic device provided for one embodiment of the present application. DETAILED DESCRIPTION

[0013] In order to make the objectives, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0014] It should be clear that the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0015] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0016] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0017] In addition, in the description of this application, unless otherwise specified, "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0018] See also Figure 1 , which is a flow chart of a method for processing sea surface temperature data provided by an embodiment of the present application. The method for processing sea surface temperature data provided by an embodiment of the present application comprises the following steps: S10: Acquire the sea surface temperature data to be processed, the time information and latitude information of the sea surface temperature data to be processed, and the standard Gaussian noise of the target time step.

[0019] The sea surface temperature data to be processed is sea surface temperature data with low spatial resolution, and needs to be reconstructed into sea surface temperature data with high spatial resolution. Specifically, it can be sea surface temperature data of a certain geographical area.

[0020] The time information refers to the time when the sea surface temperature data is captured or recorded. For example, the time information is October 11, 2022. Latitude affects the spatial distribution of sea surface temperature, and the sea surface temperature decreases from low latitudes to high latitudes.

[0021] Among them, standard Gaussian noise refers to random noise whose probability density function follows a standard normal distribution. The time step is represented by time t, and the target time step is t=T. The value of T can be set manually according to actual needs. For example, T=1000.

[0022] In the embodiment of the present application, the sea surface temperature data to be processed is obtained from the OSTIA (The Operational Sea Surface Temperature and Ice Analysis) data set, and its spatial resolution is 0.05°. The OSTIA data set records the time information and latitude information of the sea surface temperature data, and the time information and latitude information of the corresponding sea surface temperature data can be extracted from the OSTIA data set. The standard Gaussian noise of the target time step refers to the standard Gaussian noise sampled at the target time step, which can be expressed as , .

[0023] S20: performing interpolation processing on the sea surface temperature data and latitude information to be processed to obtain interpolated sea surface temperature data and interpolated latitude information.

[0024] Among them, the interpolated sea surface temperature data is sea surface temperature data with high spatial resolution.

[0025] In the embodiment of the present application, the sea surface temperature data and latitude information to be processed are interpolated using a preset interpolation algorithm to obtain the interpolated sea surface temperature data and the interpolated latitude information. The preset interpolation algorithm includes but is not limited to the nearest neighbor interpolation, bilinear interpolation and bicubic interpolation algorithms.

[0026] The spatial resolution of the interpolated sea surface temperature data is the same as that of the sea surface temperature data in the Himawari dataset. Specifically, the spatial resolution of the sea surface temperature data in the Himawari dataset is 0.02°.

[0027] S30: Establish and train a diffusion model to obtain a trained diffusion model; wherein the diffusion model includes a forward diffusion module and a reverse diffusion module, and the reverse diffusion module includes a noise prediction module and a denoising module; Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain the noisy sea surface temperature residual data at different time steps; the noise prediction module is used to perform noise prediction on the noisy sea surface temperature residual data at different time steps to obtain the predicted noise; the noise prediction module is iteratively trained through the added Gaussian noise and the predicted noise to obtain the trained diffusion model.

[0028] Among them, the forward diffusion module is the first step of the diffusion model, and its main task is to gradually transform the original data into noise. Specifically, the forward diffusion process refers to the process of gradually adding Gaussian noise from the original data until the data is completely transformed into standard Gaussian noise. The reverse diffusion module is the second step of the diffusion model, including the noise prediction module and the denoising module. Its main task is to restore the original data by gradually removing the noise starting from the standard Gaussian noise.

[0029] In an embodiment of the present application, the forward diffusion module uses the sample sea surface temperature residual data as initial data, gradually adds Gaussian noise according to the set noise addition schedule, and obtains the noisy sea surface temperature residual data at different time steps.

[0030] The noise prediction module predicts the Gaussian noise added between the initial data and the noisy sea surface temperature residual data at different time steps. The Gaussian noise added between the initial data and the noisy sea surface temperature residual data at different time steps is used as a label, and the noise prediction module can be trained based on the difference between the predicted Gaussian noise and the label.

[0031] The denoising module gradually denoises the noisy sea surface temperature residual data at different time steps to obtain the initial data.

[0032] S40: Input the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module and the denoising module of the trained diffusion model to obtain the predicted sea surface temperature residual data.

[0033] In the embodiment of the present application, the target time step T and the standard Gaussian noise of the target time step T are , interpolated sea surface temperature data, interpolated latitude information and time information are input into the noise prediction module of the trained diffusion model to obtain the predicted noise of the target time step T. The predicted noise of the target time step T and the standard Gaussian noise of the target time step T are Input to the denoising module to obtain the denoising result of time step T-1. Take time step T-1, the denoising result of time step T-1, the interpolated sea surface temperature data, the interpolated latitude information and the time information as the input of the noise prediction module, and output the predicted noise of time step T-1. Knowing the predicted noise of time step T-1, the denoising result of time step T-2 can be obtained. Traverse t from T to 1 until the predicted noise at time step t=1 is obtained. Knowing the predicted noise at time step t=1, the predicted sea surface temperature residual data can be obtained.

[0034] S50: Obtain target sea surface temperature data according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data.

[0035] In an embodiment of the present application, the interpolated sea surface temperature data is added to the predicted sea surface temperature residual data to obtain the target sea surface temperature data.

[0036] By using the embodiment of the present application, the sea surface temperature data to be processed with low spatial resolution is first interpolated to obtain the interpolated sea surface temperature data with high spatial resolution. The sea surface temperature residual data between the interpolated sea surface temperature data with high spatial resolution and the real sea surface temperature data with high spatial resolution are forward diffused and de-noised, and the diffusion model is trained so that the diffusion model can predict the sea surface temperature residual data from the standard Gaussian noise. Based on the interpolated sea surface temperature data with high spatial resolution and the predicted sea surface temperature residual data, the target sea surface temperature data is obtained, and the reconstruction from the sea surface temperature data with low spatial resolution to the sea surface temperature data with high spatial resolution is realized. The reconstructed sea surface temperature data has high accuracy, improves the quality of the sea surface temperature data, and can make up for the data loss of the existing real sea surface temperature data. At the same time, the input of the diffusion model includes time information and latitude information, so that the reconstructed target sea surface temperature data has spatiotemporal continuity, and improves the quality of the sea surface temperature data.

[0037] In one embodiment, step S30 includes steps S31 to S37, which are specifically as follows: S31: Acquire a sea surface temperature sample data set; the sea surface temperature sample data set includes a first sea surface temperature data set and a second sea surface temperature data set, the first sea surface temperature data set includes a plurality of first sea surface temperature data, the second sea surface temperature data set includes a plurality of second sea surface temperature data, and the spatial resolution of the first sea surface temperature data is less than the spatial resolution of the second sea surface temperature data.

[0038] In the embodiment of the present application, the first sea surface temperature data set is the OSTIA data set, and the second sea surface temperature data set is the Himawari data set. The spatial resolution of the first sea surface temperature data is 0.05°, and the spatial resolution of the second sea surface temperature data is 0.02°.

[0039] S32: interpolating the first sea surface temperature data to obtain third sea surface temperature data; interpolating the latitude information of the first sea surface temperature data to obtain the latitude information of the third sea surface temperature data; the spatial resolution of the third sea surface temperature data is consistent with the spatial resolution of the second sea surface temperature data.

[0040] In an embodiment of the present application, the first sea surface temperature data is interpolated using a preset interpolation algorithm to obtain the third sea surface temperature data. The latitude information of the first sea surface temperature data is interpolated using a preset interpolation algorithm to obtain the latitude information of the third sea surface temperature data. The preset interpolation algorithm includes but is not limited to nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation algorithms. The spatial resolution of the third sea surface temperature data is 0.02°.

[0041] S33: Obtaining sample sea surface temperature residual data according to the second sea surface temperature data and the third sea surface temperature data; wherein the sample sea surface temperature residual data is the sample sea surface temperature residual data of a preset time step.

[0042] In the embodiment of the present application, the difference between the third sea surface temperature data and the second sea surface temperature data is calculated to obtain the sample sea surface temperature residual data, wherein the preset time step is t=0.

[0043] S34: Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain noisy sea surface temperature residual data at several different time steps.

[0044] In the embodiment of the present application, the noisy sea surface temperature residual data at different time steps can be expressed as:

[0045] in, represents the noisy sea surface temperature residual data at time step t, represents the sample sea surface temperature residual data, , , represents the set noise addition schedule, , , T=1000, represents Gaussian noise.

[0046] S35: Input the current time step, the noisy sea surface temperature residual data of the current time step, the third sea surface temperature data, the time information and the latitude information of the third sea surface temperature data into the noise prediction module to obtain predicted noise; wherein the current time step is any time step sampled from several different time steps.

[0047] In the embodiment of the present application, the current time step is any time step t from 1 to T. The current time step t and the noised sea surface temperature residual data of the current time step t are converted into , the third sea surface temperature data, the time information of the third sea surface temperature data and the latitude information are input into the noise prediction module, and the predicted noise is output .

[0048] S36: Calculate the loss function value according to the Gaussian noise of the current time step, the predicted noise and the preset loss function.

[0049] In the embodiment of the present application, in order to measure the difference between the predicted noise and the real noise, the mean square error loss function is used as the preset loss function. The mean square error loss function will give a greater penalty to larger deviations, prompting the network to learn more accurately. Specifically, the calculation formula of the loss function value is as follows:

[0050] Where E represents the expected value of the random variable. represents the Gaussian noise at the current time step, represents the prediction noise.

[0051] S37: Iteratively train the noise prediction module according to the loss function value until a preset training condition is met, thereby obtaining a trained noise prediction module.

[0052] Among them, the preset training condition is that the loss function value is less than a preset threshold.

[0053] In an embodiment of the present application, the network parameters of the noise prediction module are adjusted according to the loss function value until the loss function value is less than a preset threshold, thereby obtaining a trained noise prediction module.

[0054] By training the noise prediction module through the current time step, the noisy sea surface temperature residual data of the current time step, the third sea surface temperature data, the time information and the latitude information of the third sea surface temperature data, the trained noise prediction module can be automatically and quickly obtained.

[0055] In one embodiment, step S33 includes steps S331 to S334, which are specifically as follows: S331: Obtain the longitude and latitude information of the second sea surface temperature data.

[0056] In the embodiment of the present application, the longitude coordinates and the latitude coordinates of the second sea surface temperature data are extracted from the Himawari data set.

[0057] S332: Extracting target third sea surface temperature data having the same longitude and latitude information as the second sea surface temperature data from the third sea surface temperature data according to the longitude and latitude information of the second sea surface temperature data.

[0058] In an embodiment of the present application, target third sea surface temperature data having the same longitude coordinates and latitude coordinates as the second sea surface temperature data is matched from the OSTIA data set, so that the target third sea surface temperature data and the second sea surface temperature data are sea surface temperature data of the same area.

[0059] S333: Scan the second sea surface temperature data and the target third sea surface temperature data respectively according to the preset scanning window to obtain a plurality of second sea surface temperature data blocks and a plurality of target third sea surface temperature data blocks.

[0060] The preset scanning window can be manually set according to actual needs. For example, the preset scanning window is 64*64.

[0061] In the embodiment of the present application, the second sea surface temperature data and the target third sea surface temperature data are scanned according to a preset window to obtain a plurality of second sea surface temperature data blocks and a plurality of target third sea surface temperature data blocks. For example, if the size of the second sea surface temperature data is 256*256 and the preset scanning window is 64*64, the number of the second sea surface temperature data blocks is 16.

[0062] S334: Calculate the residual between each second sea surface temperature data block and the corresponding target third sea surface temperature data block to obtain sample sea surface temperature residual data.

[0063] In the embodiment of the present application, for each second sea surface temperature data block, a target third sea surface temperature data block with the same longitude coordinate and latitude coordinate is obtained. The sea surface temperature data of each second sea surface temperature data block is subtracted from the sea surface temperature data of the corresponding target third sea surface temperature data block to obtain sample sea surface temperature residual data.

[0064] By matching the longitude and latitude information and scanning the preset scanning window, the sample sea surface temperature residual data can be automatically obtained.

[0065] In one embodiment, step S35 includes steps S351 to S354, which are specifically as follows: S351: Vectorize the current time step to obtain a first vector.

[0066] In the embodiment of the present application, a preset vector encoding algorithm is used to convert the current time step into a first vector, wherein the preset vector encoding algorithm includes but is not limited to a bag-of-words model and word embedding.

[0067] S352: vectorizing the noisy sea surface temperature residual data and the third sea surface temperature data corresponding to the current time step respectively to obtain a second vector and a third vector; and concatenating the second vector and the third vector to obtain a concatenated vector.

[0068] In the embodiment of the present application, a preset vector encoding algorithm is used to convert the noisy sea surface temperature residual data corresponding to the current time step into a second vector, and the third sea surface temperature data into a third vector. The preset vector encoding algorithm includes but is not limited to a bag-of-words model and word embedding.

[0069] S353: Input the time information and latitude information of the third sea surface temperature data into the space-time encoder respectively to obtain the fourth vector and the fifth vector.

[0070] Among them, the space-time encoder is an automatic encoder used to compress the input data into a low-dimensional vector.

[0071] In the embodiment of the present application, the time information and the latitude information of the third sea surface temperature data are respectively encoded by a space-time encoder to obtain a fourth vector and a fifth vector.

[0072] S354: Input the first vector, the concatenated vector, the fourth vector and the fifth vector into a noise prediction module to obtain predicted noise.

[0073] In the embodiment of the present application, the first vector, the concatenated vector, the fourth vector and the fifth vector are respectively used as input quantities of a noise prediction module, and the noise prediction module outputs predicted noise.

[0074] The embodiment of the present application can automatically determine the input amount of the noise prediction module and automatically and quickly obtain the predicted noise by vectorizing the current time step, the noisy sea surface temperature residual data corresponding to the current time step, the third sea surface temperature data, the time information of the third sea surface temperature data, and the latitude information.

[0075] In one embodiment, step S40 includes steps S41 to S45, which are specifically as follows: S41: Input the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the time information and the latitude information into the noise prediction module to obtain the predicted noise of the target time step.

[0076] In an embodiment of the present application, the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information are respectively vectorized, and the corresponding vectors are input into the noise prediction module to obtain the predicted noise of the target time step.

[0077] The vectorization process may refer to steps S351 to S354, which will not be described in detail here.

[0078] S42: Input the predicted noise of the target time step and the standard Gaussian noise of the target time step into a denoising module to obtain a denoising result of the time step before the target time step.

[0079] In the embodiment of the present application, the denoising module denoises the standard Gaussian noise of the target time step according to the predicted noise of the target time step using a preset denoising formula to obtain the denoising result of the time step before the target time step. The preset denoising formula is as follows:

[0080] in, represents the denoising result at time step t-1, represents the noisy sea surface temperature residual data at time step t, z represents the noise sampled from a normal distribution, .

[0081] Let t = T, and use the above preset denoising formula to get the denoising result of the time step before the target time step.

[0082] S43: The time step before the target time step, the denoising result of the time step before the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information are input into the noise prediction module to obtain the predicted noise of the time step before the target time step.

[0083] In an embodiment of the present application, the time step before the target time step, the denoising result of the time step before the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information are respectively vectorized, and the corresponding vectors are input into the noise prediction module to obtain the predicted noise of the time step before the target time step.

[0084] S44: inputting the predicted noise of the previous time step of the target time step and the denoising result of the previous time step of the target time step into a denoising module until a denoising result of a preset time step is obtained.

[0085] In the embodiment of the present application, the preset time step is t=1, and the denoising result of the preset time step can be obtained by repeating steps S41 to S43 using the preset denoising formula.

[0086] S45: Using the denoising result of the preset time step as the residual data of the predicted sea surface temperature.

[0087] In the embodiment of the present application, by gradually denoising the standard Gaussian noise of the target time step, the predicted sea surface temperature residual data can be obtained.

[0088] In one embodiment, after step S50, steps S51 to S52 are included, which are specifically as follows: S51: Acquire target sea surface temperature data of the target area.

[0089] Among them, the target area can be an area for marine heat wave monitoring.

[0090] In an embodiment of the present application, target sea surface temperature data of the target area is extracted based on the latitude and longitude coordinates of the target area.

[0091] S52: When the target sea surface temperature data of the target area indicates that the sea surface temperature value of the target area is greater than a preset temperature value and the duration is greater than or equal to the preset duration, it is determined that an ocean heat wave occurs in the target area.

[0092] Among them, marine heatwaves refer to short-term temperature extreme events that occur when the sea surface temperature exceeds the seasonally adjusted 90th percentile threshold and lasts for at least five days.

[0093] Among them, the preset temperature value is the 90th percentile threshold of seasonal adjustment, and the preset duration is five days.

[0094] In the embodiment of the present application, when the target sea surface temperature data of the target area indicates that the sea surface temperature value of the target area is greater than the seasonally adjusted 90th percentile threshold and the duration is greater than or equal to five days, it is determined that a marine heat wave has occurred in the target area. When the target sea surface temperature data of the target area indicates that the sea surface temperature value of the target area is greater than the seasonally adjusted 90th percentile threshold and the duration is less than five days, it is determined that no marine heat wave has occurred in the target area.

[0095] The embodiment of the present application can accurately determine whether an ocean heat wave occurs in the target area based on the target sea surface temperature data of the target area.

[0096] The following is an embodiment of the device of the present application, which can be used to execute the content of the method in the embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the content of the method in the embodiment of the present application.

[0097] See also Figure 2 , which shows a schematic diagram of the structure of the sea surface temperature data processing device provided in an embodiment of the present application. The sea surface temperature data processing device 7 provided in an embodiment of the present application comprises: The sea surface temperature data acquisition module 71 is used to acquire the sea surface temperature data to be processed, the time information and latitude information of the sea surface temperature data to be processed, and the standard Gaussian noise of the target time step; A sea surface temperature data interpolation module 72 is used to perform interpolation processing on the sea surface temperature data and latitude information to be processed to obtain the interpolated sea surface temperature data and the interpolated latitude information; The diffusion model training module 73 is used to establish and train the diffusion model to obtain the trained diffusion model; wherein the diffusion model includes a forward diffusion module and a reverse diffusion module, and the reverse diffusion module includes a noise prediction module and a denoising module; Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain the noisy sea surface temperature residual data at different time steps; the noise prediction module is used to perform noise prediction on the noisy sea surface temperature residual data at different time steps to obtain the predicted noise; the noise prediction module is iteratively trained through the added Gaussian noise and the predicted noise to obtain the trained diffusion model; The sea surface temperature residual data prediction module 74 is used to input the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module and the denoising module of the trained diffusion model to obtain the predicted sea surface temperature residual data; The target sea surface temperature data acquisition module 75 is used to obtain the target sea surface temperature data according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data.

[0098] It should be noted that the sea surface temperature data processing device provided in the above embodiment only uses the division of the above functional modules as an example when executing the sea surface temperature data processing method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the sea surface temperature data processing device and the sea surface temperature data processing method provided in the above embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0099] The following is an embodiment of the device of the present application, which can be used to execute the content of the method in the embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the content of the method in the embodiment of the present application.

[0100] See also Figure 3 The present application also provides an electronic device 300, which may be a computer, a mobile phone, a tablet computer, etc. In an exemplary embodiment of the present application, the electronic device 300 is a computer, which may include: at least one processor 301, at least one memory 302, at least one display, at least one network interface 303, a user interface 304, and at least one communication bus 305.

[0101] The user interface 304 is mainly used to provide an input interface for the user and obtain data input by the user. Optionally, the user interface may also include a standard wired interface or a wireless interface.

[0102] The network interface 303 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0103] The communication bus 305 is used to realize the connection and communication between these components.

[0104] Among them, the processor 301 may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the entire electronic device, and executes various functions of the electronic device and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed by the display layer; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor, but implemented separately through a chip.

[0105] Among them, the memory 302 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory can be used to store instructions, programs, codes, code sets or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory may also be optionally at least one storage device located away from the aforementioned processor. As Figure 3 As shown, the memory as a computer storage medium may include an operating system, a network communication module, a user interface module, and an operating application program.

[0106] The processor can be used to call the application of the sea surface temperature data processing method stored in the memory, and specifically execute the method steps of the above-mentioned embodiment. The specific execution process can refer to the specific description shown in the embodiment, which will not be repeated here.

[0107] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0108] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for processing sea surface temperature data, characterized in that: The steps include: Acquire sea surface temperature data to be processed, time information and latitude information of the sea surface temperature data to be processed, and standard Gaussian noise of a target time step; Performing interpolation processing on the sea surface temperature data to be processed and the latitude information to obtain interpolated sea surface temperature data and interpolated latitude information; Establish and train a diffusion model to obtain a trained diffusion model; wherein the diffusion model includes a forward diffusion module and a reverse diffusion module, and the reverse diffusion module includes a noise prediction module and a denoising module; Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain the noisy sea surface temperature residual data at different time steps; the noise prediction module is used to perform noise prediction on the noisy sea surface temperature residual data at different time steps to obtain predicted noise; the noise prediction module is iteratively trained by adding the Gaussian noise and the predicted noise to obtain the trained diffusion model; Inputting the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module and the denoising module of the trained diffusion model to obtain predicted sea surface temperature residual data; Target sea surface temperature data are obtained according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data.

2. The method for processing sea surface temperature data according to claim 1, characterized in that: The step of inputting the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module and the denoising module of the trained diffusion model to obtain the predicted sea surface temperature residual data includes: Inputting the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module to obtain the predicted noise of the target time step; Inputting the predicted noise of the target time step and the standard Gaussian noise of the target time step into the denoising module to obtain a denoising result of a time step before the target time step; Inputting the previous time step of the target time step, the denoising result of the previous time step of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module to obtain the predicted noise of the previous time step of the target time step; Inputting the predicted noise of the previous time step of the target time step and the denoising result of the previous time step of the target time step into the denoising module until the denoising result of the preset time step is obtained; The denoising result of the preset time step is used as the predicted sea surface temperature residual data.

3. The method for processing sea surface temperature data according to claim 1, characterized in that: The step of establishing and training a diffusion model to obtain a trained diffusion model includes: Acquire a sea surface temperature sample data set; the sea surface temperature sample data set includes a first sea surface temperature data set and a second sea surface temperature data set, the first sea surface temperature data set includes a plurality of first sea surface temperature data, the second sea surface temperature data set includes a plurality of second sea surface temperature data, and the spatial resolution of the first sea surface temperature data is less than the spatial resolution of the second sea surface temperature data; interpolating the first sea surface temperature data to obtain third sea surface temperature data; interpolating the latitude information of the first sea surface temperature data to obtain the latitude information of the third sea surface temperature data; the spatial resolution of the third sea surface temperature data is consistent with the spatial resolution of the second sea surface temperature data; The sample sea surface temperature residual data is obtained according to the second sea surface temperature data and the third sea surface temperature data; wherein the sample sea surface temperature residual data is the sample sea surface temperature residual data of a preset time step; Stepwise adding Gaussian noise to the sample sea surface temperature residual data through the forward diffusion module to obtain a number of noisy sea surface temperature residual data at different time steps; Inputting the current time step, the noisy sea surface temperature residual data of the current time step, the third sea surface temperature data, and the time information and latitude information of the third sea surface temperature data into the noise prediction module to obtain predicted noise; wherein the current time step is any time step sampled from the plurality of different time steps; Calculate the loss function value according to the Gaussian noise of the current time step, the predicted noise and a preset loss function; The noise prediction module is iteratively trained according to the loss function value until a preset training condition is met, thereby obtaining a trained noise prediction module.

4. The method for processing sea surface temperature data according to claim 3, characterized in that: The step of obtaining sample sea surface temperature residual data according to the second sea surface temperature data and the third sea surface temperature data comprises: Obtaining the latitude and longitude information of the second sea surface temperature data; extracting, from the third sea surface temperature data, target third sea surface temperature data having the same longitude and latitude information as the second sea surface temperature data, according to the longitude and latitude information of the second sea surface temperature data; According to a preset scanning window, the second sea surface temperature data and the target third sea surface temperature data are scanned respectively to obtain a plurality of second sea surface temperature data blocks and a plurality of target third sea surface temperature data blocks; The residual between each of the second sea surface temperature data blocks and the corresponding target third sea surface temperature data block is calculated to obtain sample sea surface temperature residual data.

5. The method for processing sea surface temperature data according to claim 3, characterized in that: The step of inputting the current time step, the noisy sea surface temperature residual data corresponding to the current time step, the third sea surface temperature data, and the time information and latitude information of the third sea surface temperature data into the noise prediction module to obtain the predicted noise includes: Vectorizing the current time step to obtain a first vector; Vectorizing the noisy sea surface temperature residual data corresponding to the current time step and the third sea surface temperature data respectively to obtain a second vector and a third vector; splicing the second vector and the third vector to obtain a spliced ​​vector; Inputting the time information and the latitude information of the third sea surface temperature data into the space-time encoder respectively to obtain a fourth vector and a fifth vector; The first vector, the concatenated vector, the fourth vector and the fifth vector are input into a noise prediction module to obtain predicted noise.

6. The method for processing sea surface temperature data according to any one of claims 1 to 5, characterized in that: After the step of obtaining target sea surface temperature data according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data, the method further comprises: Obtain target sea surface temperature data of the target area; When the target sea surface temperature data of the target area indicates that the sea surface temperature value of the target area is greater than a preset temperature value and lasts for a duration greater than or equal to a preset duration, it is determined that an ocean heat wave occurs in the target area.

7. A sea surface temperature data processing device, characterized in that: include: A sea surface temperature data acquisition module, used to acquire the sea surface temperature data to be processed, the time information and latitude information of the sea surface temperature data to be processed, and the standard Gaussian noise of the target time step; A sea surface temperature data interpolation module, used to perform interpolation processing on the sea surface temperature data to be processed and the latitude information to obtain interpolated sea surface temperature data and interpolated latitude information; A diffusion model training module, used for establishing and training a diffusion model to obtain a trained diffusion model; wherein the diffusion model includes a forward diffusion module and a reverse diffusion module, and the reverse diffusion module includes a noise prediction module and a denoising module; Gaussian noise is gradually added to the sample sea surface temperature residual data through the forward diffusion module to obtain the noisy sea surface temperature residual data at different time steps; the noise prediction module is used to perform noise prediction on the noisy sea surface temperature residual data at different time steps to obtain the predicted noise; the noise prediction module is iteratively trained by adding the Gaussian noise and the predicted noise to obtain the trained diffusion model; A sea surface temperature residual data prediction module, used for inputting the target time step, the standard Gaussian noise of the target time step, the interpolated sea surface temperature data, the interpolated latitude information and the time information into the noise prediction module and the denoising module of the trained diffusion model to obtain predicted sea surface temperature residual data; The target sea surface temperature data acquisition module is used to obtain target sea surface temperature data according to the predicted sea surface temperature residual data and the interpolated sea surface temperature data.

8. An electronic device, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the steps of the sea surface temperature data processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the sea surface temperature data processing method according to any one of claims 1 to 6 are implemented.

Images