Ocean wave direction detection method and system based on extended maximum likelihood method
Through the combined method of extended maximum likelihood method and deep learning model, the accuracy and adaptability of wave wave detection in marine environmental monitoring of terahertz radar are solved, and efficient and accurate wave wave detection is achieved.
Patent Information
- Application Number
- CN202510508216.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-22
- Publication Date
- 2025-08-08
AI Technical Summary
The existing terahertz radar technology is difficult to accurately extract wave wave information in marine environment monitoring, especially in the context of high noise, and the ability to extract weak signals is limited, and building an effective training set requires a large amount of high-quality labeled data. The existing algorithms perform poorly in complex sea conditions.
The wave direction detection method based on the extended maximum likelihood method is adopted, combined with terahertz radar and deep learning model, and the wave direction spectrum is generated through preprocessing, Fourier transform, and cross-spectral matrix, and the deep learning model is used to automatically identify and classify echo signals.
It improves the accuracy and robustness of wave direction detection, can be adaptable and fast in complex marine environments, and captures tiny wave changes information that are difficult to detect by traditional radars.
Smart Images
Figure CN120446893A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of ocean monitoring, and in particular relates to a method and system for detecting ocean wave direction based on an extended maximum likelihood method. Background Art
[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.
[0003] With global climate change and increased maritime activity, accurate monitoring of the marine environment, especially wave conditions, has become increasingly important. Traditional wave monitoring methods, such as buoys and satellite remote sensing, have limitations in terms of real-time performance, coverage, and cost-effectiveness. In recent years, the development of radar technology, particularly terahertz (THz) radar, has provided new solutions for ocean monitoring. THz waves, with their high resolution, strong penetration, and ability to provide rich material information, have made them an ideal tool for ocean monitoring. However, existing THz radar technology still faces challenges in its application to marine environmental monitoring. On the one hand, the ocean surface is complex and ever-changing, and radar echo signals contain a large amount of interference information. On the other hand, accurately extracting wave direction information from this complex echo data remains a technical challenge. Traditional methods, such as the fast Fourier transform (FFT), can convert time series data into frequency-amplitude spectra, but are limited in their effectiveness when processing non-stationary signals and are unable to adapt to the precise measurement requirements under varying sea conditions. Furthermore, existing algorithms perform poorly in complex sea conditions, particularly with limited ability to extract weak signals against high noise backgrounds.
[0004] New methods combining machine learning and deep learning algorithms can now more accurately extract useful information from complex radar echo data. However, existing deep learning-based methods still face several challenges in practical application. For example, constructing an effective training set requires a large amount of high-quality annotated data, which is often difficult to achieve in practice. Furthermore, selecting an appropriate model structure and optimization algorithm to ensure real-time processing capabilities is a key issue. Therefore, the development of an efficient, accurate, and adaptable wave direction detection algorithm is particularly urgent. Summary of the Invention
[0005] To overcome the shortcomings of the above-mentioned existing technologies, the present invention provides a method and system for detecting wave direction based on the extended maximum likelihood method. By utilizing advanced frequency domain analysis technology and deep learning models, the accuracy and robustness of wave direction detection are significantly improved, providing strong technical support for ocean monitoring.
[0006] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:
[0007] The first aspect of the present invention provides a method for detecting the direction of ocean waves based on the extended maximum likelihood method;
[0008] A method for detecting wave direction based on an extended maximum likelihood method, comprising:
[0009] Obtain the time series data of wave height at the measurement point and perform preprocessing;
[0010] Invert the pre-processed wave height time series data to obtain the wave height data;
[0011] Constructing a cross-spectrum matrix based on the wave height data; obtaining position information of the measurement points and setting a transfer function of the measurement points;
[0012] Based on the cross-spectral matrix, the location information of the measurement points and the transfer function, the extended maximum likelihood method is used to generate the wave directional spectrum;
[0013] A heat map is drawn according to the wave direction spectrum, and the main wave direction of the wave is determined by the color depth changes in the heat map.
[0014] As a further technical solution, the method of obtaining time series data of wave height at a measurement point includes:
[0015] The time series data of the wave height of the ocean waves at the measurement points are obtained by using terahertz ocean radars, and the terahertz ocean radars are arranged in an inverted isosceles triangle.
[0016] As a further technical solution, the preprocessing process includes noise reduction and signal enhancement of the acquired wave height time series data.
[0017] As a further technical solution, the pre-processed wave height time series data is inverted to obtain the wave height data, including:
[0018] Perform Fourier transform on the pre-processed wave height time series data to generate the time domain and frequency domain waveforms of the wave height;
[0019] The wave height data of the ocean waves are obtained based on the time domain and frequency domain waveform diagrams of the wave height.
[0020] As a further technical solution, the process of constructing a cross-spectrum matrix based on the wave height data is as follows:
[0021]
[0022] Where, Φ MN (ω) is the cross-spectral matrix; Φ mn (ω), m, n∈[1,3] are the two point information measured by the radar, and the calculation process is shown in the following formula:
[0023]
[0024] Where T is the signal duration; is the conjugate of the Fourier transform of the received signal at position m; F n (ω) is the Fourier transform of the signal received at position n; h m (t) is the wave height data.
[0025] As a further technical solution, the transfer function of the measurement point is:
[0026] H(ω,θ)=h(ω)cos α θsin β θ;
[0027] Where H(ω,θ) is the transfer function; h(ω) is the wave energy spectrum; α and β are parameters related to the observed quantities.
[0028] As a further technical solution, based on the cross-spectral matrix, the location information of the measurement points and the transfer function, the process of generating the wave directional spectrum using the extended maximum likelihood method is as follows:
[0029]
[0030] Where, is the directional spectrum; is the (m, n)th element of the inverse matrix of the cross-spectrum matrix, where m and n represent the two measured points respectively; is the transfer function of the measurement point n, is the conjugate of the transfer function at measurement point m, and K is the proportionality coefficient.
[0031] A second aspect of the present invention provides a wave direction detection system based on the extended maximum likelihood method.
[0032] A wave direction detection system based on extended maximum likelihood method, comprising:
[0033] The preprocessing module is configured to: obtain the time series data of the wave height at the measurement point and perform preprocessing;
[0034] The wave height data acquisition module is configured to: invert the pre-processed wave height time series data to obtain the wave height data of the ocean waves;
[0035] The directional spectrum generation module is configured to: construct a cross-spectrum matrix based on the wave height data; obtain the location information of the measurement points and set the transfer function of the measurement points; and generate the wave directional spectrum using the extended maximum likelihood method based on the cross-spectrum matrix, the location information of the measurement points and the transfer function;
[0036] The wave direction detection module is configured to draw a heat map according to the wave direction spectrum, and determine the main wave direction of the wave by the color depth change in the heat map.
[0037] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the ocean wave direction detection method based on the extended maximum likelihood method as described in the first aspect of the present invention.
[0038] The fourth aspect of the present invention provides an electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps of the method for detecting the direction of ocean waves based on the extended maximum likelihood method as described in the first aspect of the present invention are implemented.
[0039] One or more of the above technical solutions have the following beneficial effects:
[0040] This method leverages the high-resolution nature of terahertz waves to capture subtle ocean wave variations that are difficult to detect with traditional radar. It also incorporates cross-spectrum into the extended maximum likelihood method (EMLM) to capture the relationship between multiple measurement points, providing additional phase information and overcoming the limitations of the traditional maximum likelihood method (MLM) in high-noise or data-deficient conditions. A deep learning model is introduced to automatically identify and classify different types of echo signals, improving processing speed and enhancing adaptability to complex ocean environments.
[0041] A deep learning model was developed in Python. Specifically, a multi-output convolutional neural network model was built based on the VGGNet framework. The model was trained and optimized to classify radar echo signals and singular signals.
[0042] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.
[0044] Figure 1 This is a flow chart of the method of the first embodiment.
[0045] Figure 2 This is a schematic diagram of the arrangement of the terahertz ocean radar in the first embodiment.
[0046] Figure 3 This is a system structure diagram of the second embodiment. DETAILED DESCRIPTION
[0047] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.
[0048] It should be noted that the terms used herein are for describing particular embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.
[0049] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.
[0050] The present invention utilizes the high-resolution characteristics of terahertz waves to capture information on tiny ocean wave changes, uses the extended maximum likelihood method (EMLM) to obtain the wave direction spectrum, and draws a related heat map to determine the wave direction. The above method significantly improves the accuracy and robustness of wave direction detection, providing strong technical support for ocean monitoring.
[0051] Example 1
[0052] This embodiment discloses a method for detecting the direction of ocean waves based on the extended maximum likelihood method;
[0053] like Figure 1 As shown, a method for detecting the direction of ocean waves based on the extended maximum likelihood method comprises:
[0054] Step S1, obtaining the time series data of the wave height at the measurement point and performing preprocessing;
[0055] In step S1, the time series data of the wave height at the measurement point is obtained by using the terahertz ocean radar. The terahertz ocean radar can capture the tiny wave change information. Since a single radar is difficult to observe the wave information from multiple angles, the measurement of wave direction must rely on multiple radars to observe simultaneously to obtain more comprehensive direction information. Therefore, combined with Figure 2 In this embodiment, terahertz oceanographic radars are arranged in an inverted isosceles triangle, located at the three vertices of the triangle, to acquire time-series data on wave heights. The symmetry of the isosceles triangle layout simplifies data processing and reduces computational complexity, while ensuring consistent measurement accuracy in both primary directions, thereby improving overall measurement accuracy and reliability.
[0056] In this embodiment, in order to identify and eliminate abnormal or singular signals (such as outliers caused by human interference, reflection distortion, and equipment abnormalities) from the wave echo signals received by the terahertz ocean radar, a simple lightweight convolutional neural network (VGGNet) is used as the basic framework of the classification model, and a binary cross-entropy loss function is used to evaluate and optimize the model accuracy. It automatically distinguishes "valid wave echo signals" from "singular abnormal signals", further improves the accuracy of directional spectrum estimation, and realizes the classification and elimination of wave echo signals.
[0057] Furthermore, the acquired wave height time series data is preprocessed, wherein the preprocessing process includes steps such as DC component and filtering.
[0058] First, the DC component is used to remove the constant offset in the signal to avoid its interference with subsequent processing. Then, the intermediate frequency signal is extracted through intermediate frequency filtering. Finally, the Butterworth filter is used to further remove high-frequency noise or smooth the signal.
[0059] Step S2, inverting the pre-processed wave height time series data to obtain the wave height data;
[0060] In step S2, the pre-processed wave height time series data is subjected to Fourier transform to generate wave height waveforms in the time domain and frequency domain;
[0061] The wave height data of the ocean waves are obtained based on the time domain and frequency domain waveform diagrams of the wave height.
[0062] Step S3, constructing a cross spectrum matrix based on the wave height data; obtaining the position information of the measurement point and setting the transfer function of the measurement point;
[0063] The acquired wave height data is recorded as h m (t), construct the cross-spectral matrix as follows:
[0064]
[0065] Where, Φ MN (ω) is the cross-spectral matrix; Φ mn (ω), m, n∈[1,3] are two points measured by the radar, and the calculation process is shown in the following formula:
[0066]
[0067] Where T is the signal duration; is the conjugate of the Fourier transform of the received signal at position m; F n (ω) is the Fourier transform of the signal received at position n; h m (t) is the wave height data.
[0068] Obtain the location information of the measurement point and set the transfer function of the measurement point;
[0069] In this embodiment, three terahertz ocean radars are used to obtain wave height data, and are located at the three vertices of the triangle. Figure 2 , the position coordinates of the three terahertz ocean radars are set as
[0070] Assume that the transfer function between the measurement points is:
[0071] H(ω,θ)=h(ω)cos α θsin β θ;
[0072] Where H(ω,θ) is the transfer function; h(ω) is the wave energy spectrum; α and β are parameters related to the observed quantity. In this embodiment, the observed quantity is the wave height. If the values of α and β are 0, the transfer function H(ω,θ) is 1.
[0073] Step S4, generating the wave directional spectrum using the extended maximum likelihood method based on the cross-spectral matrix, the position information of the measurement points, and the transfer function;
[0074] The basic idea of the extended maximum likelihood method is that the cross spectrum obtained from the observation contains noise, and obtaining the directional spectrum is a problem of detecting the signal from the noise. Its core calculation formula is shown below:
[0075]
[0076] Where, is the (m, n)th element of the inverse matrix of the cross spectrum matrix, where m and n represent the two measured points respectively. is the transfer function of the measurement point n, is the conjugate of the transfer function at the measurement point m, K is a proportional coefficient obtained by experimental measurement, is the directional spectrum to be measured.
[0077] The cross-spectrum matrix obtained in step S3 is inverted to obtain the inverse matrix of the cross-spectrum matrix. The inverse matrix of the cross-spectrum matrix, the position coordinates of the three terahertz ocean radars and the transfer function are substituted into the calculation formula of the extended maximum likelihood method to obtain:
[0078]
[0079] After the above process, the cross spectrum is calculated, and the directional spectrum is solved according to the proportional coefficient K determined experimentally, that is:
[0080]
[0081] Among them, U(θ,ω) is the direction distribution function;
[0082]
[0083] The polar angle θ is the wave vector The angle with the x-axis, that is, the propagation direction of the wave, ranges from [0,2π). Indicates the average wavelength of ocean waves.
[0084] Step S5: draw a heat map according to the wave direction spectrum, and determine the main wave direction of the wave by the color depth change in the heat map.
[0085] According to the acquired directional spectrum of the waves, the values are represented in the θ-ω plane using colors of varying shades. For example, in this embodiment, red can be used to represent high values and blue to represent low values. At the same time, the darker the color, the larger the value, and the lighter the color, the smaller the value. The darkest red is the direction of the "main wave" of the wave.
[0086] Example 2
[0087] This embodiment discloses a wave direction detection system based on the extended maximum likelihood method;
[0088] like Figure 3 As shown, a wave direction detection system based on the extended maximum likelihood method includes:
[0089] The preprocessing module is configured to: obtain the time series data of the wave height at the measurement point and perform preprocessing;
[0090] The wave height data acquisition module is configured to: invert the pre-processed wave height time series data to obtain the wave height data of the ocean waves;
[0091] The directional spectrum generation module is configured to: construct a cross-spectrum matrix based on the wave height data; obtain the location information of the measurement points and set the transfer function of the measurement points; and generate the wave directional spectrum using the extended maximum likelihood method based on the cross-spectrum matrix, the location information of the measurement points and the transfer function;
[0092] The wave direction detection module is configured to draw a heat map according to the wave direction spectrum, and determine the main wave direction of the wave by the color depth change in the heat map.
[0093] Example 3
[0094] The purpose of this embodiment is to provide a computer-readable storage medium.
[0095] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for detecting the direction of ocean waves based on an extended maximum likelihood method as described in Example 1.
[0096] Example 4
[0097] The purpose of this embodiment is to provide an electronic device.
[0098] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the processor implements the steps of a method for detecting the direction of ocean waves based on an extended maximum likelihood method as described in Example 1.
[0099] The steps involved in the apparatuses of Examples 2, 3, and 4 above correspond to those of Method Example 1. For detailed implementation, please refer to the relevant description of Example 1. The term "computer-readable storage medium" should be understood to mean a single medium or multiple media containing one or more instruction sets; it should also be understood to include any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and causing the processor to perform any method of the present invention.
[0100] Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computer device. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.
[0101] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.
Claims
1. A method for detecting wave direction based on extended maximum likelihood method, characterized in that: include: Obtain the time series data of wave height at the measurement point and perform preprocessing; Invert the pre-processed wave height time series data to obtain the wave height data; constructing a cross-spectral matrix based on the wave height data; Obtain the location information of the measurement point and set the transfer function of the measurement point; Based on the cross-spectral matrix, the location information of the measurement points and the transfer function, the extended maximum likelihood method is used to generate the wave directional spectrum; A heat map is drawn according to the wave direction spectrum, and the main wave direction of the wave is determined by the color changes in the heat map.
2. The method for detecting the direction of ocean waves based on the extended maximum likelihood method according to claim 1, wherein: The method of obtaining time series data of wave height at a measurement point includes: The time series data of the wave height of the ocean waves at the measurement points are obtained by using terahertz ocean radars, and the terahertz ocean radars are arranged in an inverted isosceles triangle.
3. The method for detecting the direction of ocean waves based on the extended maximum likelihood method according to claim 1, wherein: The pre-processing process includes performing DC component and filtering processing on the acquired ocean wave height time series data.
4. The method for detecting the direction of ocean waves based on the extended maximum likelihood method according to claim 1, wherein: The pre-processed wave height time series data is inverted to obtain the wave height data, including: Perform Fourier transform on the pre-processed wave height time series data to generate time domain and frequency domain waveforms of the wave height; The wave height data of the ocean waves are obtained based on the time domain and frequency domain waveform diagrams of the wave height.
5. The method for detecting the direction of ocean waves based on the extended maximum likelihood method according to claim 1, wherein: The cross spectrum matrix constructed according to the wave height data is: Where, Φ MN (ω) is the cross-spectral matrix; Φ mn (ω), m, n∈[1,3] are two points measured by the radar, and the calculation process is shown in the following formula: Where T is the signal duration; is the conjugate of the Fourier transform of the received signal at position m; F n (ω) is the Fourier transform of the signal received at position n; h m (t) is the wave height data.
6. The method for detecting the direction of ocean waves based on the extended maximum likelihood method according to claim 1, wherein: The transfer function of the measurement point is: H(ω,θ)=h(ω)cos α θsin β I; Where H(ω,θ) is the transfer function; h(ω) is the wave energy spectrum; α and β are parameters related to the observed quantities.
7. The method for detecting the direction of ocean waves based on the extended maximum likelihood method according to claim 1, wherein: Based on the cross-spectral matrix, the location information of the measurement points and the transfer function, the process of generating the wave directional spectrum using the extended maximum likelihood method is as follows: Where, is the directional spectrum; is the (m, n)th element of the inverse matrix of the cross-spectrum matrix, where m and n represent the two measured points respectively; is the transfer function of the measurement point n, is the conjugate of the transfer function at measurement point m, and K is the proportionality coefficient.
8. A wave direction detection system based on the extended maximum likelihood method, characterized by: include: The preprocessing module is configured to: obtain the time series data of the wave height at the measurement point and perform preprocessing; The wave height data acquisition module is configured to: invert the pre-processed wave height time series data to obtain the wave height data of the ocean waves; A directional spectrum generation module is configured to: construct a cross spectrum matrix according to the wave height data; Obtain the location information of the measurement point and set the transfer function of the measurement point; generate the wave directional spectrum using the extended maximum likelihood method based on the cross-spectral matrix, the location information of the measurement point and the transfer function; The wave direction detection module is configured to draw a heat map according to the wave direction spectrum, and determine the main wave direction of the wave by the color depth change in the heat map.
9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps of the ocean wave direction detection method based on the extended maximum likelihood method as described in any one of claims 1 to 7 are implemented.
10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the ocean wave direction detection method based on the extended maximum likelihood method as described in any one of claims 1 to 7 are implemented.