Wave monitoring method and system, storage medium and terminal

Abstract

The invention provides a wave monitoring method and system, a storage medium and a terminal.The wave monitoring method comprises the steps that firstly, a wave image is collected; and then extracting wave characteristics representing wave physical characteristics from the wave image. And finally, the neural network receives the wave characteristics and reasones the wave characteristics to obtain wave parameters of the wave, and the wave parameters comprise a wave height parameter, a wave direction parameter and a wave period parameter. According to the wave monitoring method, waves of all water areas of the whole wharf can be continuously monitored for a long time, wave parameter data are inferred, potential safety hazards can be warned in time, and effective safety guarantee is provided for ships in the berthing period of the wharf.

Description

Technical Field

[0001] The present invention relates to the field of ship safety technology, and in particular to a wave monitoring method, system, storage medium and terminal. Background Art

[0002] Waves are one of the key hydrodynamic processes that affect the safety of ship mooring. With the action of waves, the tension of mooring cables will change dramatically, especially when the wave height is large and the wave direction is perpendicular to the ship's direction, the cable is most likely to break. For port terminals located near estuaries, they are greatly affected by waves from the open sea, and this impact will become particularly significant under extreme weather conditions such as typhoons. The traditional means of wave observation is to use related equipment to directly observe the wave field, and commonly used ones include wave buoys, ADCPs, and radar wave meters.

[0003] The observation equipment used in traditional hydrological observation methods is relatively expensive and requires regular calibration, and the cost of purchasing, installing and maintaining the equipment is high. Moreover, due to the frequent traffic of ships on the waterway of the estuary terminal, it is difficult to conduct long-term continuous observations in the waters here, so it is impossible to continuously and effectively monitor the safety of ship moorings, and it is easy to miss safety hazards. In addition, under severe weather conditions, traditional observation equipment is very easy to be damaged, so the relevant observation work cannot be completed in this case, which is a huge gap for the safety of ship moorings. In addition, the observation data range of traditional observation equipment is only a single point or a limited angle of water, and cannot cover all berths in the entire port terminal, so it cannot guarantee the mooring safety of all ships. Summary of the invention

[0004] In view of the problems existing in the prior art mentioned above, the present application provides a wave monitoring method, system, storage medium and terminal, which can continuously monitor the waves in all waters of the entire dock for a long period of time, infer wave parameter data, help to timely warn of potential safety hazards, and provide effective safety protection for ships while they are moored at the dock.

[0005] To achieve the above and other related purposes, the present invention provides a wave monitoring method, comprising the following steps:

[0006] Collect wave images;

[0007] Extract wave features representing the physical properties of waves from wave images;

[0008] The neural network receives the wave features and infers the wave features to obtain the wave parameters of the wave, which include wave height parameters, wave direction parameters and wave period parameters.

[0009] Optionally, after collecting the wave images, the following steps are included:

[0010] Preprocess the wave images to reduce noise and interference in the wave images.

[0011] Optionally, preprocessing the wave images includes:

[0012] Perform denoising processing, contrast enhancement processing, color and brightness correction processing, sharpening and edge enhancement processing, lens distortion correction processing, perspective correction processing, water surface area segmentation processing, dynamic background removal processing, foreground enhancement and target extraction processing on the wave images.

[0013] Optionally, extracting wave features characterizing the physical properties of waves from the wave images includes the following steps:

[0014] The convolutional neural network receives the wave images and extracts wave features from the wave images to extract wave features characterizing the physical properties of waves.

[0015] Optionally, extracting wave features characterizing the physical properties of waves from the wave images includes:

[0016] Extract wave spatial features from the wave images;

[0017] Extract wave temporal features from the wave images;

[0018] Obtain wave spatio-temporal features based on the wave spatial features and the wave temporal features.

[0019] Optionally, extracting wave features characterizing the physical properties of waves from the wave images further includes the following steps:

[0020] Evaluate the importance of the wave features to screen out the wave features corresponding to the wave parameters.

[0021] On the other hand, the present invention provides a wave monitoring system, including:

[0022] A collection module for collecting wave images;

[0023] A feature extraction module for extracting wave features characterizing the physical properties of waves from the wave images;

[0024] A neural network module for receiving the wave features and reasoning on the wave features to obtain the wave parameters of the waves, where the wave parameters include wave height parameters, wave direction parameters, and wave period parameters.

[0025] Optionally, the wave monitoring system further includes:

[0026] A preprocessing module for preprocessing the wave images to reduce noise and interference in the wave images.

[0027] On the other hand, the present invention provides a storage medium with a computer program stored thereon, and when the program is executed by a processor, the above-mentioned wave monitoring method is implemented.

[0028] On the other hand, the present invention provides a terminal, including:

[0029] A memory for storing a computer program;

[0030] A processor for executing the computer program stored in the memory, so that the terminal executes the above-mentioned wave monitoring method.

[0031] As described above, the wave monitoring method, system, storage medium and terminal provided by the present invention at least have the following beneficial technical effects:

[0032] The wave monitoring method of the present invention first collects wave images. Then, wave features characterizing the physical properties of the waves are extracted from the wave images. Finally, the neural network receives the wave features and infers the wave features to obtain wave parameters, including wave height parameters, wave direction parameters and wave period parameters. The wave monitoring method of the present invention can continuously monitor the waves in all waters of the entire wharf for a long time, infer the wave parameter data, which helps to timely warn of potential safety hazards and provide effective safety guarantees for ships during berthing at the wharf. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It shows a schematic flow chart of the wave monitoring method provided in Embodiment 1 of the present invention.

[0034] Figure 2 It shows a schematic module diagram of the wave monitoring system provided in Embodiment 2 of the present invention.

[0035] Figure 3 It shows a schematic structural diagram of the terminal provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0036] The following specific examples illustrate the embodiments of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0037] It should be noted that the illustrations provided in this embodiment only schematically illustrate the basic concept of the present invention. Although only the components related to the present invention are shown in the illustrations and they are not drawn according to the number, shape, and size of the components in actual implementation, the form, number, positional relationship, and proportion of each component in actual implementation can be arbitrarily changed on the premise of implementing the technical solution of the present invention, and the layout form of its components may also be more complex.

[0038] Embodiment 1

[0039] This embodiment provides a wave monitoring method. As Figure 1 shown, the method includes the following steps:

[0040] S100: Collect wave images;

[0041] The wave images within the monitored water area are acquired through a camera device, which serves as the basis for subsequent processing and analysis. To improve the overall reliability and the accuracy of wave feature extraction, it is necessary to ensure the high quality of the captured images as much as possible. Therefore, the performance, installation, and settings of the camera need to meet the following requirements. Specifically, since the characteristic details of waves are relatively complex, especially under different lighting conditions and weather changes, the wave form may change significantly. Thus, the resolution of the camera is crucial. The higher the resolution of the camera, the more accurate the extraction of wave features will be, but at the same time, it also increases the burden of processing and storage. When selecting the resolution of the camera, a balance needs to be found between image quality and system load. To ensure that sufficient details of the image can be captured, the selected resolution of the camera needs to be greater than or equal to 1080P. Since waves are a dynamic hydrodynamical process and their form changes continuously over time, the frame rate of the camera determines whether sufficient continuous details of wave changes can be captured. If the frame rate of the camera is too low, it may lead to the loss of important wave details, affecting subsequent feature extraction and wave parameter estimation. To ensure that the system can capture the dynamic changes of waves without generating excessive amounts of data, the frame rate of the camera is between 10 and 30 frames per second. This frame rate is suitable for the real-time processing of waves. The lighting conditions in the hydrographic environment are complex, and the camera will face strong sunlight reflections, shadows, strong light contrasts during sunrise and sunset, etc. Through the wide dynamic range function, the camera can process very bright and very dark areas in the same frame image, ensuring that clear wave forms can be captured regardless of how the light changes. Therefore, the camera must have the wide dynamic range function to maintain clear image quality in high-contrast scenarios. To achieve continuous 24-hour monitoring of waves, the camera needs to have night vision capabilities and be able to continue to work effectively under low-light conditions at night or in adverse weather conditions. The camera uses a CMOS sensor that can still form images under low-light conditions to ensure normal acquisition of wave images at night, in the early morning, or at dusk. The environment in the dock area is humid and variable. Especially for docks in estuary areas, the environmental conditions are very harsh, and salt spray and dust may have an adverse impact on the normal operation of the camera. Moreover, it is necessary to ensure the normal operation of the camera under extreme weather conditions such as typhoons. Therefore, the camera needs to have waterproof and dustproof performance, and the housing material with corrosion resistance should be selected. The protection level of the camera is greater than or equal to IP67. The distance between the camera and the monitored water area should be adjusted according to the coverage range of the camera and the focal length of the camera. An overly wide viewing angle may result in insufficient wave details in the image, while an overly narrow viewing angle may not cover a sufficient water area. The higher the installation height, the larger the horizontal range that the camera can cover. However, an overly high installation position may cause the wave details in the image to be compressed, affecting the monitoring effect. Therefore, appropriate installation height and viewing angle parameters of the camera need to be selected to ensure that the entire monitoring area can be covered while retaining sufficient wave details.The pitch angle of the camera should be adjusted to the optimal angle suitable for observing waves, reducing the influence of reflected light and specular effects on the image quality. Usually, a slightly downward tilt angle helps to reduce the proportion of the sky in the image, thus focusing more on the waves. Using a seismic mounting base to ensure the stability of the camera is a very important measure, especially in the water area monitoring scenario, because the camera may be disturbed by wind, water wave vibrations generated by passing ships or other external factors. Therefore, it is necessary to use a seismic mounting base to ensure the stability of the camera and avoid the impact of camera jitter or movement on image acquisition. In this embodiment, when collecting the image data of waves, an image acquisition strategy combining continuous shooting and time-interval shooting is adopted. Continuous shooting can capture the more complete dynamic changes of waves, which helps to extract wave features more accurately, but it will generate a large amount of data and requires powerful computing and storage capabilities. The acquisition strategy can be adjusted according to actual needs. Select interval acquisition when the wave field is relatively stable, and switch to continuous acquisition mode when the waves change violently. During sunrise and sunset phases, the lighting conditions change rapidly, and the acquisition frequency can be increased. During storms or other extreme weather, increase the image acquisition frequency to obtain more data to support emergency response measures. To improve the accuracy of wave monitoring, multiple cameras are arranged in the same monitoring area to collect images from different angles, so as to obtain the three-dimensional information of the waves and improve the accuracy of wave parameter estimation. Preliminary image compression and encoding are performed at the camera end to reduce the data volume and the pressure on the transmission bandwidth while ensuring the image quality, and to ensure the stability and timeliness of data transmission. Under different weather conditions, such as in fog, heavy rain or strong wind weather, the image quality may be affected. Therefore, the settings of the camera may need to be adjusted by adjusting parameters such as exposure, contrast, ISO, etc. to adapt to environmental changes. At the same time, the camera should be equipped with an automatic windshield wiper or a protective cover to keep the lens clean in bad weather.

[0042] S200: Extract wave features representing the physical properties of waves from the wave images;

[0043] To remove or reduce the noise and interference in the collected wave images and enhance the saliency of wave features, it is necessary to preprocess the wave images. The preprocessing includes denoising processing, contrast enhancement processing, color and brightness correction processing, sharpening and edge enhancement processing, lens distortion correction processing, perspective correction processing, water surface area segmentation processing, dynamic background removal processing, foreground enhancement and target extraction processing.

[0044] Specifically, changes in ambient light, electronic noise, signal interference in data transmission, etc. may cause more noise in the image. Therefore, it is necessary to denoise the wave image. By using bilateral filtering technology to remove the noise in the image, the noise in the background, such as the reflected light on the water surface and floating objects on the water surface, can be effectively removed, while the morphological characteristics of the waves are retained. After the noise reduction process, the denoising effect is evaluated by calculating indicators such as the signal-to-noise ratio and mean square error of the images before and after. To facilitate the subsequent extraction of wave feature data, enable the morphology of the waves to be fully displayed under different lighting conditions, and at the same time make the features of the waves more prominent, it is necessary to enhance the light and dark contrast in the image. Therefore, it is necessary to perform contrast enhancement processing on the image. While maintaining the overall contrast improvement, it is also necessary to avoid noise amplification or image distortion caused by excessive enhancement in local areas. In an alternative embodiment of this Embodiment 1, the adaptive histogram equalization technology is used to divide the image into multiple blocks, perform histogram equalization on each block separately, and then combine the blocks into a complete image through interpolation. To ensure the image quality and the consistency of subsequent processing, it is necessary to perform color and brightness correction processing on the image. Brightness correction makes the light distribution of the image more uniform by adjusting the global or local brightness of the image. Color correction ensures that the color of the image is consistent with the actual environment by correcting the ratio of the RGB channels or performing tone mapping. Through brightness and color correction, it is ensured that the images collected at different times and under different weather conditions are consistent, and the robustness of wave monitoring is improved. Perform sharpening and edge enhancement processing on the image: By enhancing the high-frequency components of the image, the contour of the waves can be made clearer. The methods for sharpening the image include, but are not limited to, the Laplace operator, Sobel operator, and Canny edge detection. After the image edge enhancement processing, it not only helps to improve the visibility of the waves but also helps the neural network better identify the morphological characteristics of the waves. To avoid geometric distortion introduced by the camera, causing the straight lines in the image to bend and affecting the presentation of the wave morphology, resulting in deviations in the extraction of wave features, it is necessary to perform lens distortion correction processing on the image to ensure that the wave morphology in the image reflects the actual situation. For example, calibration can be achieved through camera calibration and inverse mapping, that is, by taking a calibration plate image, calculating the radial distortion and tangential distortion coefficients of the lens, and then correcting the distorted image to a normal image through inverse mapping. Since the image may have perspective distortion, making the waves farther from the camera look smaller and the waves closer look larger, resulting in the height and morphology of the waves being inconsistent with the actual situation, it is necessary to perform perspective correction processing on the wave image to eliminate the perspective effect in the wave image through geometric transformation. Specifically, the perspective parameters of the wave image can be deduced by using the horizontal line as a geometric reference, and then the wave image can be corrected through affine transformation or projective transformation to ensure that the geometric morphology of the waves is accurately presented in the wave image, thereby improving the accuracy of wave parameter inference.In order to classify and label different regions in an image, it is necessary to separate the wavy part in the image from the backgrounds such as the sky, ships, and buildings, so that the neural network can focus on analyzing the characteristics of the waves. Therefore, it is necessary to perform water surface area segmentation on the image. The segmentation methods include, but are not limited to, threshold-based segmentation methods, edge-based segmentation methods, region-based segmentation methods, and machine learning-based segmentation methods. Perform dynamic background removal on the image: In practical applications, wave monitoring may encounter situations where the background changes dynamically, such as overcast clouds. These dynamic backgrounds may interfere with the extraction and analysis of wave characteristics. Therefore, it is necessary to eliminate them through background removal techniques. The dynamic background removal methods include, but are not limited to, frame difference method, mixture Gaussian model, and optical flow method. Dynamic background removal can effectively reduce background interference and improve the accuracy of wave feature extraction. After completing the water surface area segmentation and background removal processing, the next step is to perform foreground enhancement and target extraction on the wave region, that is, it is necessary to perform foreground enhancement and target extraction processing on the image. Foreground enhancement can make the shape of the waves clearer by performing contrast adjustment, brightness enhancement, and edge sharpening on the wave region. Target extraction is to extract the wave region in the wave image and generate wave features suitable for input into the neural network. The extraction methods include traditional image processing techniques and deep learning.

[0045] Next, wave features that can characterize the physical properties of waves are extracted from the wave image data, including the following steps: The convolutional neural network receives the wave image and extracts wave features from the wave image to extract wave features that characterize the physical properties of waves. In this embodiment, the wave features are extracted by the convolutional neural network model in the neural network. Specifically, the size and number of convolutional kernels determine the extraction ability of the wave feature data in the convolutional layer. Smaller convolutional kernels can capture the detailed wave features in the wave image, while larger convolutional kernels can extract more global wave features. For example, a stacked structure of multiple convolutional layers can be designed, and each convolutional layer extracts wave features of different complexities. Through the step-by-step processing of multi-layer convolution, the complex and dynamically changing wave field is dealt with to ensure that the convolutional neural network can capture the multi-level wave features of the waves. A pooling layer can be added after the convolutional layer to reduce the dimension of the wave features, reduce the size of the wave feature image, retain the main wave features at the same time, and improve the robustness while reducing the computational amount. The pooling methods include but are not limited to max pooling and average pooling. Next, the convolutional layer and the fully connected layer are connected to the activation function to introduce non-linearity and simulate more complex functional relationships. The activation functions include but are not limited to ReLU, Leaky ReLU, Sigmoid, and Tanh, etc. The wave forms and motions of various parts in the wave field have multi-scale wave features. In order to extract wave feature data of different scales, fuse these wave features, and improve the generalization ability of wave feature extraction, a multi-scale analysis mechanism can be introduced into the convolutional neural network. Multiple convolutional kernels are used to extract wave features in parallel, and the wave features at different levels are merged through the residual network structure.

[0046] Waves are dynamic phenomena with temporal continuity. When extracting wave features, not only spatial features but also temporal features need to be considered. Temporal analysis models are used to capture the temporal dynamic features of waves, including but not limited to recurrent neural networks and long short-term memory neural networks, etc. Then, the temporal features are fused with the spatial features to obtain the spatio-temporal features of the waves. Specifically, the spatial features of each frame of the image are first extracted through a convolutional neural network, and then these features are input into a long short-term memory neural network to extract their dynamic temporal features, and the temporal features are fused with the spatial features. Since high-dimensional feature representations are usually obtained after wave feature extraction, direct use may lead to an increase in computational complexity and even the problem of dimensionality disaster. Therefore, it is necessary to optimize and reduce the dimension of the extracted wave features to reduce the computational complexity while retaining the key wave features. Commonly used dimensionality reduction techniques include but are not limited to principal component analysis, linear discriminant analysis, and non-linear dimensionality reduction methods based on autoencoders, etc. Dimensionality reduction can compress the feature space and remove redundant information. To prevent the convolutional neural network model from overfitting, it is necessary to filter out the noise features in the wave features and enhance the generalization ability of the convolutional neural network model by controlling the complexity of the convolutional neural network model through regularization terms. Commonly used regularization methods include but are not limited to L1 regularization, L2 regularization, and dropout, etc. In addition, the expressive ability of the wave features is increased through feature enhancement and expansion to generate new wave feature dimensions to capture more complex wave patterns. Related methods include but are not limited to feature combination, interactive feature generation, and feature transformation methods, etc. After the wave feature extraction and optimization are completed, it is necessary to evaluate the importance of the wave features and select the wave features corresponding to the wave parameters for subsequent wave parameter inference. Inference methods include but are not limited to correlation analysis, variance analysis, and feature visualization methods based on deep learning. Specifically, after evaluating the importance of the wave features, the optimal wave feature subset is selected for wave parameter estimation. Commonly used wave feature selection strategies include but are not limited to forward selection, backward elimination, and recursive feature elimination, etc.

[0047] S300: The neural network receives the wave features and infers the wave features to obtain the wave parameters of the wave, where the wave parameters include wave height parameters, wave direction parameters, and wave period parameters.

[0048] Specifically, the wave height is inferred based on the regression model in the neural network. When inferring the wave height, common loss functions include but are not limited to mean square error, root mean square error, and mean absolute error, etc. To evaluate the accuracy of the regression model, the coefficient of determination is generally calculated as an evaluation index. The coefficient of determination measures the correlation between the predicted value and the actual value of the model, and its value ranges from 0 to 1. The closer the coefficient of determination is to 1, the higher the correlation between the predicted value and the actual value of the regression model, that is, the better the inference accuracy of the regression model. Regarding the inference of wave direction, the wave direction is usually represented by 0 to 360°. The inference of wave direction can be regarded as a classification problem. The wave direction is divided into several angular intervals, such as every 10° as an interval, and then the propagation direction of the wave is inferred through the regression model. The wave period is essentially a time dynamic feature. In the process of inferring the wave period, common loss functions include but are not limited to mean square error, root mean square error, and mean absolute error, and common optimization algorithms include but are not limited to Adam and RMSprop.

[0049] By combining image recognition and neural network technologies, the present invention can accurately infer the wave height, wave direction, and wave period of waves based on wave images. The low-cost, wide-coverage, uninterrupted, and high-resolution wave monitoring method of the present invention makes up for the defects of traditional observation equipment, continuously monitors key parameters such as wave height, wave period, and wave direction in all waters of the entire wharf for a long time, timely warns of potential safety hazards, and provides effective safety guarantees for large ships during berthing at the wharf.

[0050] Embodiment 2

[0051] This embodiment provides a wave monitoring system, as Figure 2 shown. The wave monitoring system of this embodiment includes an acquisition module, a feature extraction module, and a neural network module. The acquisition module is used to acquire wave images. The feature extraction module is used to extract wave features representing the physical characteristics of waves from the wave images. The neural network module is used to receive the wave features and infer the wave features to obtain wave parameters of the waves. The wave parameters include wave height parameters, wave direction parameters, and wave period parameters.

[0052] The wave monitoring system further includes a preprocessing module, and the preprocessing module is used to preprocess the wave images to reduce noise and interference in the wave images.

[0053] Embodiment 3

[0054] This embodiment provides a storage medium. A computer program is stored on the storage medium of this embodiment, and when the program is executed by a processor, it implements the wave monitoring method described in Embodiment 1. The storage medium includes various media such as ROM, RAM, magnetic disk, USB flash drive, memory card, or optical disc that can store program codes.

[0055] Embodiment 4

[0056] This embodiment provides a terminal. As Figure 3 shown, the terminal of this embodiment includes a memory and a processor. The memory is used to store computer programs. Preferably, the memory includes various media that can store program codes, such as ROM, RAM, magnetic disks, USB flash drives, memory cards, or optical discs. The processor is connected to the memory and is used to execute the computer programs stored in the memory, so that the terminal executes the wave monitoring method described in Embodiment 1. Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it can also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0057] The above embodiments are only illustrative of the principles and effects of the present invention, and are not used to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical idea disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. A wave monitoring method, characterized in that: The following steps are involved: Collect wave images; Extracting wave features representing wave physical properties from the wave image; The neural network receives the wave characteristics and infers the wave characteristics to obtain wave parameters of the wave, wherein the wave parameters include wave height parameters, wave direction parameters and wave period parameters.

2. The wave monitoring method according to claim 1, characterized in that: After acquiring the wave image, the following steps are included: The wave image is preprocessed to reduce noise and interference in the wave image.

3. The wave monitoring method according to claim 2, characterized in that: Preprocessing the wave image includes: The wave image is subjected to denoising, contrast enhancement, color and brightness correction, sharpening and edge enhancement, lens distortion correction, perspective correction, water surface area segmentation, dynamic background removal, foreground enhancement and target extraction.

4. The wave monitoring method according to claim 1, characterized in that: Extracting wave features representing wave physical properties from the wave image comprises the following steps: The convolutional neural network receives the wave image and performs wave feature extraction on the wave image to extract the wave features representing the physical properties of the wave.

5. The wave monitoring method according to claim 1, characterized in that: Extracting wave features representing wave physical properties from the wave image includes: Extracting wave spatial features from the wave image; Extracting wave time features from the wave image; The wave space-time characteristics are obtained based on the wave space characteristics and the wave time characteristics.

6. The wave monitoring method according to claim 1, characterized in that: Extracting wave features representing wave physical properties from the wave image also includes the following steps: The importance of the wave characteristics is evaluated to screen out the wave characteristics corresponding to the wave parameters.

7. A wave monitoring system, characterized in that: include: An acquisition module, used for acquiring wave images; A feature extraction module, used to extract wave features representing physical properties of waves from the wave image; The neural network module is used to receive the wave characteristics and infer the wave characteristics to obtain wave parameters of the wave, wherein the wave parameters include wave height parameters, wave direction parameters and wave period parameters.

8. The wave monitoring system according to claim 7, characterized in that: Also includes: The preprocessing module is used to preprocess the wave image to reduce noise and interference in the wave image.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the wave monitoring method according to any one of claims 1 to 6 is implemented.

10. A terminal, characterized in that: include: Memory for storing computer programs; A processor is used to execute the computer program stored in the memory so that the terminal executes the wave monitoring method according to any one of claims 1 to 6.

Images