Sea ice thickness estimation method and device, storage medium and electronic equipment

Abstract

This disclosure relates to the field of sea ice monitoring technology, specifically to a sea ice thickness estimation method, apparatus, storage medium, and electronic device. The sea ice thickness estimation method includes: acquiring image data, GNSS signals, pile foundation data, and environmental parameters; segmenting the pile foundation area and ice layer area based on the image data to obtain the pile foundation area and ice layer area, and shielding the pile foundation area to extract image features of the ice layer area; calculating the satellite elevation angle based on the direct signal in the GNSS signal, and calculating the observations based on the reflected signal in the GNSS signal, to calculate signal features based on the satellite elevation angle and the observations; inputting the image features, signal features, pile foundation data, and environmental parameters into a pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model. The sea ice thickness estimation method provided by this disclosure can improve the accuracy of sea ice thickness estimation under marine photovoltaic background.

Description

Technical Field

[0001] This disclosure relates to the field of sea ice monitoring technology, specifically to a sea ice thickness estimation method, a sea ice thickness estimation device, a storage medium, and an electronic device. Background Technology

[0002] In recent years, with the gradual scarcity of land resources with good sunlight conditions, floating photovoltaic power generation systems in rivers, lakes, and reservoirs have developed rapidly. Offshore photovoltaics, due to its advantages such as abundant water resources, unobstructed views, high light reflectivity, suppression of temperature rise losses, proximity to load centers, and less dust, is becoming a research hotspot.

[0003] The thickness of sea ice directly affects the load design of photovoltaic support structures, and accurate prediction of ice thickness changes can optimize operation and maintenance cycles and reduce maintenance costs caused by ice disasters. Therefore, sea ice thickness estimation is a key technical requirement to ensure the reliability, economy and adaptability of offshore photovoltaic systems to complex marine environments.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this disclosure is to provide a sea ice thickness estimation method, a sea ice thickness estimation device, a storage medium, and an electronic device, aiming to solve the problem of poor sea ice thickness estimation accuracy under marine photovoltaic background.

[0006] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0007] According to one aspect of this disclosure, a method for estimating sea ice thickness is provided, comprising:

[0008] The system collects image data of the surrounding sea area of ​​the offshore photovoltaic pile foundation, GNSS signals of the sea ice surface to be detected, pile foundation data, and environmental parameters; wherein, the pile foundation data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters.

[0009] Based on the image data, image segmentation of the pile foundation area and the ice layer area is performed to obtain the pile foundation area and the ice layer area. The pile foundation area is then masked to extract image features of the ice layer area. These image features include sea ice texture feature values ​​and / or the straight-line spacing values ​​at the ice layer edges.

[0010] The satellite elevation angle is calculated based on the direct signal in the GNSS signal, and the observation is calculated based on the reflected signal in the GNSS signal. The signal characteristics are then calculated based on the satellite elevation angle and the observation.

[0011] The image features, signal features, pile foundation data, and environmental parameters are input into a pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model.

[0012] Optionally, the step of segmenting the pile foundation region and the ice layer region based on the image data to obtain the pile foundation region and the ice layer region includes:

[0013] Acquire training images and annotate the pile foundation area and ice layer area in the training images to obtain label information;

[0014] The training images and the label information are used to train the model to obtain a trained semantic segmentation model;

[0015] The image data is input into the semantic segmentation model to obtain the pile foundation area and ice layer area output by the semantic segmentation model.

[0016] Optionally, when the image features include sea ice texture feature values, the extraction of image features of the ice region includes:

[0017] The gray-level value distribution of adjacent pixel pairs in the image of the ice region is statistically analyzed to generate a gray-level co-occurrence matrix;

[0018] The sea ice texture feature values ​​are extracted based on the gray-level co-occurrence matrix; the sea ice texture feature values ​​include entropy values ​​and / or contrast values.

[0019] Optionally, when the image features include the straight-line spacing value of the ice layer edge, the extraction of image features of the ice layer region includes:

[0020] Edge points were obtained by performing edge detection on the ice region using the Canny operator.

[0021] The edge points are mapped to the Hough parameter space to determine the straight lines of the edges on both sides of the ice region.

[0022] The distance between the straight lines at the edge of the ice layer is calculated based on the straight lines at the edge.

[0023] Optionally, the step of calculating the satellite elevation angle based on the direct signal in the GNSS signal, and calculating the observations based on the reflected signal in the GNSS signal, and calculating signal characteristics based on the satellite elevation angle and the observations, includes:

[0024] The satellite elevation angle is obtained by performing positioning calculations on the direct signal; and

[0025] The phases of the left-hand and right-hand reflected signals are calculated based on the reflected signals, and the observations are calculated based on the phases of the left-hand and right-hand reflected signals.

[0026] Estimate the oscillation frequency of the observed quantity as a function of the satellite elevation angle to obtain the signal characteristics.

[0027] Optionally, before calculating the satellite elevation angle based on the direct signal in the GNSS signal, calculating the observations based on the reflected signal in the GNSS signal, and calculating signal characteristics based on the satellite elevation angle and the observations, the method further includes:

[0028] Phase noise caused by pile foundations in the GNSS signal is removed.

[0029] Optionally, removing phase noise caused by pile foundations from the GNSS signal includes:

[0030] A noise model was created based on the correlation between pile foundation vibration and phase noise;

[0031] The phase noise in the GNSS signal is filtered out using an adaptive filter based on the noise model.

[0032] According to a second aspect of this disclosure, a sea ice thickness estimation apparatus is provided, comprising:

[0033] The acquisition module is used to acquire image data of the sea area surrounding the offshore photovoltaic pile foundation, GNSS signals of the sea ice surface to be detected, pile foundation data, and environmental parameters; wherein, the pile foundation data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters.

[0034] An image module is used to segment the pile foundation area and the ice layer area based on the image data to obtain the pile foundation area and the ice layer area, and to perform masking processing on the pile foundation area to extract the image features of the ice layer area; wherein, the image features include sea ice texture feature values ​​and / or the straight line spacing values ​​of the ice layer edges; and

[0035] The signal module calculates the satellite elevation angle based on the direct signal in the GNSS signal, and calculates the observations based on the reflected signal in the GNSS signal, and calculates the signal characteristics based on the satellite elevation angle and the observations;

[0036] The estimation module is used to input the image features, the signal features, the pile foundation data and the environmental parameters into a pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model.

[0037] According to a third aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the sea ice thickness estimation method as described in the above embodiments.

[0038] According to a fourth aspect of the present disclosure, an electronic device is provided, characterized in that it includes: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the sea ice thickness estimation method as described in the above embodiments.

[0039] The exemplary embodiments disclosed herein may have some or all of the following beneficial effects:

[0040] In the technical solutions provided by some embodiments of this disclosure, when estimating sea ice thickness, image features of image data of the sea area surrounding the pile foundation, signal features of GNSS signals on the surface of the sea ice to be detected, as well as pile foundation data and environmental parameters are extracted for estimation. By fusing multi-source data, compared with the existing estimation based on a single signal feature, the error of a single data source can be reduced and the accuracy of estimation can be improved.

[0041] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0042] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0043] Figure 1 The schematic diagram illustrates a flowchart of a sea ice thickness estimation method according to an exemplary embodiment of the present disclosure;

[0044] Figure 2 This illustration schematically shows the correlation between the oscillation frequency and oscillation amplitude of an observation quantity and sea ice thickness in an exemplary embodiment of this disclosure;

[0045] Figure 3 This schematic diagram illustrates the composition of a sea ice thickness estimation device according to an exemplary embodiment of the present disclosure;

[0046] Figure 4 The schematic diagram illustrates the structure of a computer system of an electronic device according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0047] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0048] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0049] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0050] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily need to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0051] The implementation details of the technical solutions of the embodiments of this disclosure are described in detail below.

[0052] Figure 1 This illustration schematically shows a flowchart of a sea ice thickness estimation method according to an exemplary embodiment of this disclosure. Figure 1 As shown, the sea ice thickness estimation method includes steps S101 to S1,4:

[0053] Step S101: Collect image data of the sea area surrounding the offshore photovoltaic pile foundation, GNSS signals of the sea ice surface to be detected, pile foundation data, and environmental parameters; wherein, the GNSS signals include direct signals and reflected signals; the pile foundation data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters;

[0054] Step S102: Based on the image data, image segmentation is performed on the pile foundation area and the ice layer area to obtain the pile foundation area and the ice layer area. The pile foundation area is then masked to extract image features of the ice layer area. These image features include sea ice texture feature values ​​and / or the straight-line spacing values ​​at the ice layer edges.

[0055] Step S103: Calculate the satellite elevation angle based on the direct signal in the GNSS signal, and calculate the observations based on the reflected signal in the GNSS signal, and calculate the signal characteristics based on the satellite elevation angle and the observations;

[0056] Step S104: Input the image features, the signal features, the pile foundation data and the environmental parameters into the pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model.

[0057] In the technical solutions provided by some embodiments of this disclosure, when estimating sea ice thickness, image features of image data of the sea area surrounding the pile foundation, signal features of GNSS signals on the surface of the sea ice to be detected, as well as pile foundation data and environmental parameters are extracted for estimation. By fusing multi-source data, compared with the existing estimation based on a single signal feature, the error of a single data source can be reduced and the accuracy of estimation can be improved.

[0058] The following will describe in more detail each step of the sea ice thickness estimation method in this exemplary embodiment, with reference to the accompanying drawings and embodiments.

[0059] In step S101, image data of the sea area surrounding the offshore photovoltaic pile foundation, GNSS signals of the sea ice surface to be detected, pile foundation data, and environmental parameters are collected; wherein, the pile foundation data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters; the GNSS signals include direct signals and reflected signals.

[0060] Specifically, the first step is to collect image data of the sea area surrounding the offshore photovoltaic (PV) pile foundations. This can be achieved by deploying high-resolution optical cameras or synthetic aperture radar (SAR) equipment to obtain real-time images of the sea area surrounding the pile foundations, with a focus on the area where ice meets the pile foundations.

[0061] Secondly, GNSS signals need to be acquired. Installing a GNSS receiver against the backdrop of offshore photovoltaic systems to measure sea ice thickness can be achieved by combining the characteristics of pile foundations or floating structures. Typically, the receiver can be installed on top of the photovoltaic pile foundation or in a stable position on the floating platform, secured with high-strength supports to ensure unobstructed vertical flow. The GNSS receiver includes an antenna assembly for receiving direct satellite signals and detecting signals reflected from the sea surface. The antenna assembly includes a right-hand circularly polarized direct antenna, a left-hand circularly polarized reflective antenna, and a right-hand circularly polarized reflective antenna, along with an antenna support. The right-hand circularly polarized direct antenna is positioned at the top of the antenna support, while the left-hand and right-hand circularly polarized reflective antennas are positioned below and on either side of the right-hand circularly polarized direct antenna. The right-hand circularly polarized direct signal is acquired through the right-hand circularly polarized direct antenna, while the left-hand and right-hand circularly polarized signals are acquired through the left-hand and right-hand circularly polarized reflective antennas.

[0062] Secondly, it is also necessary to collect pile foundation data. In the context of offshore photovoltaic systems, the distribution of pile foundations may interfere with GNSS reflected signals through multipath effects, thereby affecting the accuracy of sea ice thickness estimation. Therefore, it is essential to collect pile foundation data. This data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters. Pile foundation parameters include, for example, pile length, pile diameter, pile type, and pile dimensions; pile distribution parameters include, for example, spacing and arrangement; and pile installation parameters include, for example, the relative position of the piles to the antenna and the depth data of the piles inserted into different seabed strata.

[0063] Finally, environmental parameters need to be collected. Different environmental conditions may affect the accuracy of sea ice thickness estimation, so collecting environmental parameters during the detection process can help eliminate the influence of the environment. Environmental parameters can include data such as tides, wind speed, and water temperature.

[0064] In step S102, the image of the pile foundation area and the ice layer area is segmented based on the image data to obtain the pile foundation area and the ice layer area, and the pile foundation area is masked to extract the image features of the ice layer area; wherein, the image features include sea ice texture feature values ​​and / or the straight line spacing values ​​of the ice layer edge.

[0065] Specifically, one aspect is the need to extract image features. When extracting image features, considering the pile foundations deployed in offshore photovoltaic systems, it is necessary to separate the pile foundation area from the ice layer area in the image. Then, the pile foundation area is masked using masking technology, and only the ice layer area is analyzed, thereby reducing the interference of the pile foundation area on image feature extraction.

[0066] In one embodiment of this disclosure, step S102, which involves image segmentation of the pile foundation region and the ice layer region based on the image data to obtain the pile foundation region and the ice layer region, includes: acquiring a training image and labeling the pile foundation region and the ice layer region in the training image to obtain label information; using the training image and the label information to train a model to obtain a trained semantic segmentation model; and inputting the image data into the semantic segmentation model to obtain the pile foundation region and the ice layer region output by the semantic segmentation model.

[0067] Specifically, a deep learning-based semantic segmentation model can be pre-trained, and then used to perform image recognition and segmentation, thereby obtaining an image of the ice region. The specific content of the model training phase is as follows:

[0068] First, a labeled dataset containing the pile foundation and ice layer areas needs to be constructed. Image data of the surrounding sea area needs to be collected, and pixel-level labels for the two types of areas need to be manually annotated to form a training set. During annotation, pixel-level masks for the two types of areas must be distinguished. For the linear geometric features of the pile foundation, the Hough Transform or Canny edge detection algorithm can be used to extract the straight line contours and generate auxiliary feature maps.

[0069] In addition, data augmentation strategies can be designed to address reflection interference at the junction of ice and pile foundation. These strategies include basic enhancement methods such as mirroring, rotation, and brightness / contrast adjustment, or the introduction of synthetic noise to simulate ice surface reflection, such as Gaussian noise and specular highlight simulation.

[0070] Then, model training and optimization are performed. A semantic segmentation model with an encoder-decoder structure, such as DeepLabV3+ or U-Net, can be selected. The encoder part uses a VGG16 or ResNet pre-trained model to extract multi-scale features, and the decoder part introduces a geometric feature fusion module. That is, in the skip connection stage, the straight line feature map generated by edge detection is concatenated with the upsampled features of the decoder to enhance the recognition of straight structures of pipe piles.

[0071] The loss function combines Dice Loss with geometric constraints. The main loss function uses Dice Loss to optimize the segmentation boundary. The auxiliary loss function introduces a linear structural similarity metric (such as Hausdorff distance) to constrain the consistency between the prediction results and geometric features.

[0072] Based on the above method, a geometric feature-guided attention mechanism and CRF post-processing were used to achieve accurate segmentation of PHC pipe piles with straight structures. At the same time, a weighted loss function can be used to suppress ice reflection noise.

[0073] After the semantic segmentation model is trained, the collected image data is input into the model to obtain the output pixel classification results. The segmentation boundary is optimized by morphological operations or conditional random field (CRF) to finally separate the clear pile foundation and ice layer areas, and finally obtain the image of the ice layer area with the pile foundation area shielded.

[0074] After obtaining the image of the ice region, the next step is to extract the image features of the ice region.

[0075] In one embodiment of this disclosure, when the image features include sea ice texture feature values, the step of extracting the image features of the ice region includes: statistically analyzing the grayscale value distribution of adjacent pixel pairs in the image of the ice region to generate a grayscale co-occurrence matrix; extracting the sea ice texture feature values ​​based on the grayscale co-occurrence matrix; the sea ice texture feature values ​​include entropy values ​​and / or contrast values.

[0076] Specifically, the image of the ice region is first converted into a grayscale image. Then, the grayscale value distribution of adjacent pixel pairs is statistically analyzed to generate a grayscale co-occurrence matrix. For example, the pixel spacing and direction (such as horizontal or vertical) are set, and the frequency of occurrence of different grayscale combinations is statistically analyzed to form a matrix. Then, key indicators are extracted from the grayscale co-occurrence matrix, including entropy and / or contrast values. The entropy value reflects the degree of texture disorder; the higher the value, the more complex the ice surface texture. The contrast value measures the difference between pixels; the higher the value, the more drastic the texture change.

[0077] In one embodiment of this disclosure, when the image features include the distance between straight lines at the ice layer edges, the step of extracting the image features of the ice layer region includes: performing edge detection on the ice layer region using the Canny operator to obtain edge points; mapping the edge points to the Hough parameter space to determine the straight lines at the edges on both sides of the ice layer region; and calculating the distance between straight lines at the ice layer edges based on the straight lines at the edges.

[0078] Specifically, the Canny operator is first used to detect edges in the image, preserving key points of the ice layer contour and reducing irrelevant noise. Then, the edge points are mapped to the Hough parameter space, where ρ is the distance from the line to the origin and θ is the angle. A voting mechanism is used to find the line parameters corresponding to the peak values, thereby determining the lines on both sides of the ice layer edge. The two parallel lines are represented by polar coordinate parameters as (ρ1,θ) and (ρ2,θ). Finally, parallel lines with similar angles are selected, and the actual distance is calculated based on the difference in their ρ values ​​and θ values, i.e., h = |ρ1 - ρ2| * cosθ.

[0079] Based on the above method, the Hough transform converts the edge points in the image into a parameter space through a parameter space mapping mechanism, and determines the line parameters by statistically analyzing the peak values ​​through an accumulator. This global search characteristic makes it highly tolerant to noise and local edge breaks, and can improve the anti-interference ability and robustness of calculating the straight line spacing value of ice layer edges.

[0080] In step S103, the satellite elevation angle is calculated based on the direct signal in the GNSS signal, and the observation is calculated based on the reflected signal in the GNSS signal. The signal characteristics are then calculated based on the satellite elevation angle and the observation.

[0081] For sea ice thickness detection technology, the sea ice thickness inversion method is adopted. GNSS signals can penetrate sea ice of a certain thickness and are reflected twice at the air-sea ice surface and the sea ice-sea water surface, respectively. A three-layer medium model of air-sea ice-sea water is established, and then the sea ice thickness is inverted using the phase of the GNSS reflection signal.

[0082] In one embodiment of this disclosure, the GNSS signal includes a direct signal and a reflected signal. The steps of calculating the satellite elevation angle based on the direct signal in the GNSS signal, calculating the observation based on the reflected signal in the GNSS signal, and calculating signal characteristics based on the satellite elevation angle and the observation include: performing positioning calculations on the direct signal to obtain the satellite elevation angle; calculating the left-hand and right-hand reflected signal phases based on the reflected signal to calculate the observation based on the left-hand and right-hand reflected signal phases; and estimating the oscillation frequency of the observation with respect to the satellite elevation angle to obtain the signal characteristics.

[0083] Specifically, pre-deploying antenna components to receive direct satellite signals and detect sea surface reflected signals allows for the collection of both direct and reflected signals.

[0084] The system performs positioning calculations on the direct signal to obtain relevant information about the satellite elevation angle θ, and synchronizes the direct signal by transmitting code phase and carrier Doppler information to the reflection channel to achieve synchronization of the reflected signal.

[0085] The phase of the GNSS reflected signal can be obtained by the following formula:

[0086]

[0087] In the formula, I max.r and Q max.r These are the complex time delay correlation waveforms Y. p The real and imaginary parts of the peak complex correlation value of (τ).

[0088] The phase of a GNSS reflected signal consists of the reflection coefficient phase and the phase difference between the direct and reflected signals. Generally, because the phase difference between the direct and reflected signals oscillates violently with elevation angle, and the oscillation frequency increases with antenna height, it often overwhelms the phase of the reflection coefficient in the reflected signal phase. Therefore, to detect sea ice thickness, it is necessary to first eliminate the phase difference between the direct and reflected signals. One method to eliminate the phase difference is to use dual-polarization observation for sea ice thickness observation. In a shore-based scenario, since the left and right circularly polarized reflected signals experience the same propagation space, their path differences relative to the direct signal are approximately equal. Therefore, the phase difference between the direct and reflected signals can be eliminated through left and right circularly polarized observation.

[0089] Therefore, the phases of the left-handed reflected signals are first obtained according to equation (1). and the phase of the right-hand reflection signal Then the observation M(θ) of the inverted sea ice thickness is calculated as follows:

[0090]

[0091] The phase difference caused by the time delay difference between direct and reverse signals is eliminated by phase subtraction. Since the phase provided by the reflection coefficient in the right-hand reflection signal fluctuates around 0, the resulting phase difference is approximately the phase provided by the left-hand reflection coefficient.

[0092] Figure 2 This illustration schematically shows the correlation between the oscillation frequency and amplitude of an observation quantity and sea ice thickness in an exemplary embodiment of this disclosure. For example... Figure 2 As shown, the oscillation frequency and amplitude of the observed measurements differ for different sea ice thicknesses. Regarding the oscillation frequency, the observed oscillation frequency and sea ice thickness show a linear relationship in the range of 0–10 m. Regarding the oscillation amplitude, the oscillation amplitude remains approximately 1 in the range of 0–1 m, while it decreases inversely with sea ice thickness in the range of 1–10 m.

[0093] As shown above, compared to oscillation amplitude, oscillation frequency has a wider detection range for sea ice thickness. Furthermore, since it exhibits a linear relationship with sea ice thickness, oscillation frequency can be used as a signal feature for sea ice thickness detection. To estimate the oscillation frequency of the observation M(θ) as a function of the satellite elevation angle θ, a cosine function can be used to fit the observation sequence, i.e.:

[0094]

[0095] In the formula, a, b, and c are the fitting parameters, representing the oscillation amplitude a, oscillation frequency b, and initial phase c of the observed quantity.

[0096] It should be noted that, as a metallic structure, the pile foundation may alter the signal propagation path through reflection or diffraction, resulting in an additional delay difference in the carrier phase. Therefore, before calculating the satellite elevation angle based on the direct signal in the GNSS signal, calculating the observations based on the reflected signal in the GNSS signal, and calculating the signal characteristics based on the satellite elevation angle and the observations, the method further includes: removing phase noise caused by the pile foundation in the GNSS signal.

[0097] In one embodiment of this disclosure, the step of removing phase noise caused by pile foundation in the GNSS signal includes: creating a noise model based on the correlation between pile foundation vibration and phase noise; and using an adaptive filter to filter out the phase noise in the GNSS signal according to the noise model.

[0098] Specifically, the vibration characteristics of the pile foundation are first analyzed to establish a noise model. This can be achieved by deploying vibration sensors to monitor the mechanical vibration frequency of the pile foundation structure (such as steel pipe piles and PHC pipe piles), and combining this with phase gradient diagram analysis to establish a correlation model between pile foundation vibration and phase noise.

[0099] Based on this, an adaptive filter is designed to filter out phase interference at specific frequencies according to the noise model. The adaptive filter dynamically adjusts its parameters based on real-time acquired satellite elevation angle information. A transition threshold is set for the phase difference map of the reflected signal, and regions with phase abrupt changes exceeding the threshold are set to zero, generating a denoising mask.

[0100] In step S104, the image features, the signal features, the pile foundation data, and the environmental parameters are input into a pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model.

[0101] In one embodiment of this disclosure, the sea ice thickness estimation model can employ random forest or convolutional neural network (CNN) to complete the training of the multimodal fusion sea ice thickness estimation model using a training dataset including image features, signal features, pile foundation data, environmental parameters, and sea ice thickness values.

[0102] The inputs to the trained sea ice thickness estimation model are image features, signal features, pile foundation data, and environmental parameters, and the output of the model is the estimated sea ice thickness value.

[0103] Figure 3 This schematic diagram illustrates the composition of a sea ice thickness estimation device according to an exemplary embodiment of the present disclosure, such as... Figure 3 As shown, the sea ice thickness estimation device 300 may include an acquisition module 301, an image module 302, a signal module 303, and an estimation module 304. Wherein:

[0104] The acquisition module 301 is used to acquire image data of the sea area surrounding the offshore photovoltaic pile foundation, GNSS signals of the sea ice surface to be detected, pile foundation data, and environmental parameters; wherein, the GNSS signals include direct signals and reflected signals; the pile foundation data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters;

[0105] Image module 302 is used to perform image segmentation of the pile foundation region and the ice layer region based on the image data to obtain the pile foundation region and the ice layer region, and to perform masking processing on the pile foundation region to extract image features of the ice layer region; wherein, the image features include sea ice texture feature values ​​and / or the straight line spacing values ​​of the ice layer edges; and

[0106] Signal module 303 calculates the satellite elevation angle based on the direct signal in the GNSS signal, and calculates the observations based on the reflected signal in the GNSS signal, and calculates the signal characteristics based on the satellite elevation angle and the observations;

[0107] The estimation module 304 is used to input the image features, the signal features, the pile foundation data and the environmental parameters into a pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model.

[0108] According to an exemplary embodiment of this disclosure, the image module 302 is further configured to acquire training images and annotate the pile foundation region and ice layer region in the training images to obtain label information; use the training images and the label information to train the model to obtain a trained semantic segmentation model; and input the image data into the semantic segmentation model to obtain the pile foundation region and ice layer region output by the semantic segmentation model.

[0109] According to an exemplary embodiment of the present disclosure, when the image features include sea ice texture feature values, the image module 302 is further configured to statistically analyze the grayscale value distribution of adjacent pixel pairs in the ice region image to generate a grayscale co-occurrence matrix; extract the sea ice texture feature values ​​based on the grayscale co-occurrence matrix; the sea ice texture feature values ​​include entropy values ​​and / or contrast values.

[0110] According to an exemplary embodiment of the present disclosure, when the image features include the distance between straight lines at the ice layer edges, the image module 302 is further configured to perform edge detection on the ice layer region using the Canny operator to obtain edge points; map the edge points to the Hough parameter space to determine the straight lines at the edges on both sides of the ice layer region; and calculate the distance between straight lines at the ice layer edges based on the straight lines at the edges.

[0111] According to an exemplary embodiment of this disclosure, the signal module 303 is further configured to perform positioning calculations on the direct signal to obtain the satellite elevation angle; and to calculate the left-handed and right-handed reflected signal phases based on the reflected signal, so as to calculate the observations based on the left-handed and right-handed reflected signal phases; and to estimate the oscillation frequency of the observations with respect to the satellite elevation angle to obtain the signal characteristics.

[0112] According to an exemplary embodiment of the present disclosure, the signal module 303 further includes a signal correction unit, configured to calculate the satellite elevation angle based on the direct signal in the GNSS signal, and calculate the observation based on the reflected signal in the GNSS signal, and remove phase noise caused by the pile foundation in the GNSS signal before calculating signal characteristics based on the satellite elevation angle and the observation.

[0113] According to an exemplary embodiment of this disclosure, the signal correction unit is further configured to create a noise model based on the correlation between pile foundation vibration and phase noise; and to use an adaptive filter to filter out phase noise in the GNSS signal according to the noise model.

[0114] The specific details of each module in the aforementioned sea ice thickness estimation device 300 have been described in detail in the corresponding sea ice thickness estimation method, so they will not be repeated here.

[0115] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0116] In exemplary embodiments of this disclosure, a storage medium capable of implementing the above-described methods is also provided. It may be a portable compact disc read-only memory (CD-ROM) and include program code, and can run on a terminal device, such as a mobile phone. However, the program product of this disclosure is not limited thereto. In this document, a readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0117] In an exemplary embodiment of this disclosure, an electronic device capable of implementing the above-described method is also provided. Figure 4 The schematic diagram illustrates the structure of a computer system of an electronic device according to an exemplary embodiment of the present disclosure.

[0118] It should be noted that, Figure 4The computer system 400 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0119] like Figure 4 As shown, the computer system 400 includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 402 or programs loaded from storage section 408 into Random Access Memory (RAM) 403. The RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An Input / Output (I / O) interface 405 is also connected to the bus 404.

[0120] The following components are connected to I / O interface 405: an input section 406 including a keyboard, mouse, etc.; an output section 407 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to I / O interface 405 as needed. A removable medium 411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 410 as needed so that computer programs read from it can be installed into storage section 408 as needed.

[0121] In particular, according to embodiments of this disclosure, the processes described below with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411. When the computer program is executed by central processing unit (CPU) 401, it performs various functions defined in the system of this disclosure.

[0122] It should be noted that the computer-readable medium shown in the embodiments of this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such transmitted data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0123] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0124] The units described in the embodiments of this disclosure can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the unit itself.

[0125] In another aspect, this disclosure also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to perform the methods described in the above embodiments.

[0126] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0127] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this disclosure.

[0128] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein.

[0129] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for estimating sea ice thickness, characterized in that, include: The system collects image data of the surrounding sea area of ​​the offshore photovoltaic pile foundation, GNSS signals of the sea ice surface to be detected, pile foundation data, and environmental parameters; wherein, the GNSS signals include direct signals and reflected signals; the pile foundation data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters. Based on the image data, image segmentation of the pile foundation area and the ice layer area is performed to obtain the pile foundation area and the ice layer area. The pile foundation area is then masked to extract image features of the ice layer area. These image features include sea ice texture feature values ​​and / or the straight-line spacing values ​​at the ice layer edges. A noise model is created based on the correlation between pile foundation vibration and phase noise; an adaptive filter is used to filter out phase noise in the GNSS signal according to the noise model; the satellite elevation angle is calculated based on the direct signal in the GNSS signal after removing phase noise, and the observations are calculated based on the reflected signal in the GNSS signal; and the signal characteristics are calculated based on the satellite elevation angle and the observations. The image features, signal features, pile foundation data, and environmental parameters are input into a pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model.

2. The sea ice thickness estimation method according to claim 1, characterized in that, The process of segmenting the pile foundation area and the ice layer area based on the image data to obtain the pile foundation area and the ice layer area includes: Acquire training images and annotate the pile foundation area and ice layer area in the training images to obtain label information; The training images and the label information are used to train the model to obtain a trained semantic segmentation model; The image data is input into the semantic segmentation model to obtain the pile foundation area and ice layer area output by the semantic segmentation model.

3. The sea ice thickness estimation method according to claim 1, characterized in that, When the image features include sea ice texture feature values, the extraction of image features of the ice region includes: The gray-level value distribution of adjacent pixel pairs in the image of the ice region is statistically analyzed to generate a gray-level co-occurrence matrix; The sea ice texture feature values ​​are extracted based on the gray-level co-occurrence matrix; the sea ice texture feature values ​​include entropy values ​​and / or contrast values.

4. The sea ice thickness estimation method according to claim 1, characterized in that, When the image features include the straight-line spacing value of the ice layer edge, the extraction of image features of the ice layer region includes: Edge points are obtained by performing edge detection on the ice layer region using the Canny operator. The edge points are mapped to the Hough parameter space to determine the straight lines of the edges on both sides of the ice region. The distance between the straight lines at the edge of the ice layer is calculated based on the straight lines at the edge.

5. The sea ice thickness estimation method according to claim 1, characterized in that, The calculation of the satellite elevation angle based on the direct signal in the GNSS signal, and the calculation of the observations based on the reflected signal in the GNSS signal, and the calculation of signal characteristics based on the satellite elevation angle and the observations, include: The satellite elevation angle is obtained by performing positioning calculations on the direct signal; and The phases of the left-hand and right-hand reflected signals are calculated based on the reflected signals, and the observations are calculated based on the phases of the left-hand and right-hand reflected signals. Estimate the oscillation frequency of the observed quantity as a function of the satellite elevation angle to obtain the signal characteristics.

6. A sea ice thickness estimation device, characterized in that, include: The acquisition module is used to acquire image data of the surrounding sea area of ​​the offshore photovoltaic pile foundation, GNSS signals of the sea ice surface to be detected, pile foundation data, and environmental parameters; wherein, the GNSS signals include direct signals and reflected signals; the pile foundation data includes one or more of the following: pile foundation parameters, pile distribution parameters, and pile installation parameters; An image module is used to segment the pile foundation area and the ice layer area based on the image data to obtain the pile foundation area and the ice layer area, and to perform masking processing on the pile foundation area to extract the image features of the ice layer area; wherein, the image features include sea ice texture feature values ​​and / or the straight line spacing values ​​of the ice layer edges; and The signal module is used to create a noise model based on the correlation between pile foundation vibration and phase noise; use an adaptive filter to filter out phase noise in the GNSS signal according to the noise model; calculate the satellite elevation angle based on the direct signal in the GNSS signal after removing phase noise, and calculate the observations based on the reflected signal in the GNSS signal, and calculate the signal characteristics based on the satellite elevation angle and the observations; The estimation module is used to input the image features, the signal features, the pile foundation data and the environmental parameters into a pre-trained sea ice thickness estimation model to obtain the sea ice thickness value output by the sea ice thickness estimation model.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the sea ice thickness estimation method as described in any one of claims 1 to 5.

8. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the sea ice thickness estimation method as described in any one of claims 1 to 5.

Images