Low-altitude observation method for seawater suspended sediment concentration

Abstract

The application relates to the technical field of marine hydrological monitoring, and discloses a seawater suspended sediment concentration low-altitude observation method, which uses a low-altitude observation platform to carry digital photography equipment, obtains seawater suspended sediment concentration images, collects water samples in the photographed sea area, positions the water samples through a global satellite positioning system, uses a filtering method to measure the suspended sediment content, pre-processes the obtained seawater suspended sediment concentration images, realizes automatic splicing of the images and automatic imaging of orthographic images, trains the processed image data and the measured suspended sediment concentration data based on an optimized artificial neural network algorithm, establishes a seawater suspended sediment concentration inversion model, and evaluates the model precision. The method uses the low-altitude observation platform to obtain high-resolution images, uses the optimized artificial neural network model to invert the suspended sediment concentration, has the advantages of high precision, low cost, flexibility, timeliness and the like, and can effectively make up for the deficiency of satellite remote sensing.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of marine hydrology monitoring, and particularly relates to a method for low-altitude observation of seawater suspended sediment concentration. BACKGROUND

[0002] The suspended sediment is one of important elements of marine hydrology and has an extremely important influence on human activities. The suspended sediment has extremely strong adsorption, is a carrier of nutrient salts and organic matters, and adsorbs heavy metals and toxic waste, thereby playing an important role in the migration and circulation of pollutants. The suspended sediment significantly reduces the light transmission intensity and the efficiency of photosynthesis of water body through effects such as light refraction and scattering, thereby affecting the marine primary productivity and the type, quantity and distribution of phytoplankton and benthic organisms, and having different degrees of influence on the growth of algae, the absorption of nutrient salts and population succession. The suspended sediment controls the transport process of nutrient salts to some extent, leading to the unevenness of the distribution of nutrient salts in time and space, which is one of the reasons why the red tide presents seasonal and regional characteristics. The suspended sediment has a close relationship with the changes of coastal topography and geomorphology, the siltation of port channels and the scouring and silting changes around buildings, and is often related to the planning and layout of port site selection, the dredging of channels, the selection of mud throwing areas, the erosion of coastlines, the safety of foundations of marine structures, breakwaters, land reclamation, bridge erection and the like.

[0003] Since the suspended sediment plays an important role in the migration and circulation of pollutants, the marine biochemical cycle, carbon cycle, light transmission intensity of seawater, coastal erosion and siltation, ports and channels, and coastal engineering, it is the focus of multidisciplinary attention. A large number of observation and researches on the suspended sediment concentration have been carried out at home and abroad based on shore-based, ship-based, submarine-based and seabed-based methods using traditional water sampling and filtering methods, optical conversion methods and acoustic inversion methods. These methods are usually based on point or line observation, and it is difficult to meet the needs of rapid acquisition of suspended sediment concentration in a large area of seawater. The development of satellite remote sensing technology makes the inversion technology of seawater suspended sediment concentration based on space-based technology mature, and provides a rapid and large-area remote sensing method for the observation of seawater suspended sediment concentration. A large number of previous observations of seawater suspended sediment concentration based on satellite remote sensing technology make the satellite remote sensing inversion method of seawater suspended sediment concentration quite mature. However, the satellite has a certain orbit period, and the observation time is limited by the reentry period. At the same time, the satellite is far from the sea surface, the signal propagation distance is far, the image spatial resolution is low, and it is difficult to accurately monitor small-scale sea areas. Although the unmanned aerial vehicle carrying multi-spectral or hyperspectral cameras can compensate for some of the shortcomings of satellite remote sensing, the multi-spectral or hyperspectral cameras are expensive, which limits their large-scale application.

[0004] In view of this, we propose a low-altitude image of seawater suspended sediment concentration obtained by a low-altitude observation platform, based on artificial neural network, the seawater suspended sediment concentration image data and the measured seawater suspended sediment concentration are trained, and the seawater suspended sediment concentration inversion model is established, and the low-altitude observation method of seawater suspended sediment concentration is realized to solve the above problems. SUMMARY

[0005] The present application aims to solve the limitations of the above-mentioned prior art suspended sediment concentration observation method in small-scale sea area large-area rapid acquisition. Satellite remote sensing provides a remote sensing means, but it is limited by revisit period and image resolution, and it is difficult to accurately monitor small-scale sea area. The unmanned aerial vehicle remote sensing technology can make up for the deficiency of satellite remote sensing, but the equipment cost is high, which limits its wide application.

[0006] To achieve the above purpose, the present application provides the following technical scheme:

[0007] A low-altitude observation method of seawater suspended sediment concentration, comprising the following steps:

[0008] S1, with the help of a low-altitude observation platform, the seawater suspended sediment concentration image of the target sea area is obtained by digital photography, the obtained image can be stored on a chip memory card to download or transmit to a computer workstation for subsequent processing; water sample collection and global satellite positioning are carried out at the same time of photography, the collected water sample is used to determine the suspended sediment content by filtering method, and the collected water sample is divided into modeling samples and verification samples;

[0009] S2, the obtained seawater suspended sediment concentration image is preprocessed, including mean filtering, geographic registration, angle correction, automatic splicing and orthographic image processing, and image data for further analysis is generated;

[0010] S3, according to the collected suspended sediment concentration image data and the measured suspended sediment concentration data, an artificial neural network is constructed, based on the optimization of artificial neural network algorithm, the artificial neural network training is carried out by using the seawater suspended sediment concentration image data and the measured suspended sediment concentration data, the network parameters are adjusted, the model performance is optimized, and the seawater suspended sediment concentration low-altitude inversion model is established;

[0011] S4, the seawater suspended sediment concentration low-altitude inversion model is verified by using the remote sensing suspended sediment concentration inversion model and the measured verification sample, the precision of the seawater suspended sediment concentration inversion model is evaluated, and the model is adjusted and optimized according to the evaluation result;

[0012] S5, the seawater suspended sediment concentration low-altitude inversion model is applied to the target sea area image data, and the low-altitude observation result of seawater suspended sediment concentration is obtained by inversion.

[0013] As preferred, in step S1, the low-altitude observation platform is a platform with a certain height capable of low-altitude photography and obtaining digital images of seawater suspended sediment concentration, including but not limited to unmanned aerial vehicles, airships, coastal highlands, structures, offshore platforms, and ships.

[0014] As preferred, in step S1, the digital photography includes high-resolution visible light digital cameras such as smartphones and cameras.

[0015] As preferred, in step S1, when obtaining the seawater suspended sediment concentration image by digital photography, a fixed angle or a normal angle is adopted through angle correction.

[0016] As preferred, in step S2, the specific implementation method of geographic registration when processing the seawater suspended sediment concentration image is as follows:

[0017] Geographic registration: through the global satellite positioning system installed on the camera system, real-time dynamic carrier phase difference is adopted for spatial positioning, or feature ground objects are selected, global satellite positioning system is used for spatial positioning, ground object latitude and longitude are obtained, feature ground object coordinate points and image pixel points are combined to perform geographic registration by using ArcGIS software, and if there is no feature ground object in the observation area, artificial marking is performed.

[0018] The specific implementation method of angle correction when processing the seawater suspended sediment concentration image is as follows:

[0019] A single-color liquid is tested under the same light source, the camera is used to take pictures at different angles from 20° to 90°, the RED, GREEN, and BLUE band values of the camera image are extracted by using Matlab software, the different angles of the camera and the band values of the camera at the same position are functionally fitted, and the angle correction equation is obtained as follows: C X =13*λ*C i / 1000+λ, wherein C X is the corrected band value, C i is the uncorrected band value, and λ is the angle between the camera and the horizontal plane.

[0020] The RED, GREEN, and BLUE band values are corrected to normal camera angles by using the angle correction equation, and the corrected RED, GREEN, and BLUE band values are as follows: C R =13*λ*C r / 1000+λ;

[0021] C G =13*λ*C g / 1000+λ; C b =13*λ*C b / 1000+λ.

[0022] As preferred, based on the artificial neural network algorithm, the image band value of seawater suspended sediment concentration and the measured suspended sediment concentration are used for artificial neural network training to establish a low-altitude inversion model of seawater suspended sediment concentration; the artificial neural network is a BP neural network model containing an input layer, two hidden layers and an output layer.

[0023] As preferred, a neural network model containing an input layer, a hidden layer and an output layer is constructed by using Matlab, the input layer is the RGB band data extracted according to the camera image, the output layer is the suspended sediment concentration data, the band data of each pixel point is the input, and the suspended sediment concentration is the corresponding output; the neural network algorithm training data includes two parts, which are input data and output data, and the two parts of data are combined into a data pair, 70% of the measured seawater suspended sediment concentration data and the processed corrected image band RGB data pair are used as the training data set, and the remaining data pair is used as the verification test data set;

[0024] The input data and the output data are normalized to the interval [0, 1] by using the mapminmax function; a neural network is constructed by using the feedforwardnet function; the tansig activation function is used for the hidden layer, the number of neurons in the hidden layer is set through experimental comparison and analysis, the number of neurons in the first hidden layer is 12, and the number of neurons in the second hidden layer is 7; the purelin linear activation function is used for the output layer;

[0025] The trainlm algorithm is selected to train the neural network, and the training parameters are set, including the maximum training times 1000 times, the learning rate 0.01, and the training target error 10 -6 , after optimization, 5-fold cross-validation and L2 regularization are used to avoid overfitting, and the neural network parameters are optimized.

[0026] As preferred, in step S4, the low-altitude inversion model of seawater suspended sediment concentration is verified by comparing with the measured suspended sediment concentration and the remote sensing inversion suspended sediment concentration, and the precision of the suspended sediment concentration inversion model is evaluated by using three indexes of mean square error MSE, root mean square error RMSE and mean absolute error MAE, and the formulas are as follows:

[0027]

[0028]

[0029]

[0030] Wherein, y i is the measured value, is the model prediction value, and n is the sample number.

[0031] As preferred, the low-altitude inversion model of seawater suspended sediment concentration is applicable to seawater, lakes and reservoirs.

[0032] Compared with the prior art, the technical effects and advantages of the present application are:

[0033] The seawater suspended sediment concentration low-altitude observation method takes advantage of the low-altitude observation platform, obtains seawater suspended sediment concentration information closer to seawater, realizes low-altitude observation of seawater suspended sediment concentration, has the advantages of high resolution, high precision, low cost, rapid efficiency, flexibility, low risk, repeatability, and is suitable for small-scale sea area observation.

[0034] The seawater suspended sediment concentration low-altitude observation method uses an optimized neural network algorithm to set the number of hidden layer neurons to optimize the neural network algorithm, and constructs more suitable seawater suspended sediment concentration inversion parameters for complex water bodies, thereby improving the suitability and reliability of the model.

[0035] The seawater suspended sediment concentration low-altitude observation method uses bands and water body characteristics in the seawater suspended sediment concentration image to invert the suspended sediment concentration, making up for the deficiency of using spectral characteristics to invert seawater suspended sediment concentration. The present application can accurately describe, predict and evaluate the movement and change of suspended sediment, is conducive to promoting the ecological environment protection of nearshore sea areas and the scientific development of coastal areas, and has a broad application prospect in the field of marine hydrological observation. BRIEF DESCRIPTION OF DRAWINGS

[0036] Fig. 1 The flowchart of the seawater suspended sediment concentration low-altitude observation method of the present application;

[0037] Fig. 2 The verification chart of the seawater suspended sediment concentration inversion result of the present application;

[0038] Fig. 3 The chart of different seawater suspended sediment concentration inversion results of the present application. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0040] The following will be described in detail with reference to the drawings in the embodiments of the present application, Figs. 1-3 The present application will be described in further detail,

[0041] The present application discloses a seawater suspended sediment concentration low-altitude observation method, comprising the following steps:

[0042] S1, data collection: select a suitable low-altitude observation platform, prepare for observation, low-altitude observation platform includes unmanned aerial vehicle, airship, coastal high ground, structure, offshore platform or ship and other platforms with certain height can carry out low-altitude photography, obtain seawater suspended sediment concentration digital image platform;

[0043] The digital photography is used to obtain the seawater suspended sediment concentration image, and the obtained image can be stored on the chip storage card, which can be downloaded or transmitted to the computer workstation for subsequent processing. The digital photography includes high-resolution visible light digital camera such as smart phone and camera;

[0044] When the digital photography is used to obtain the seawater suspended sediment concentration image, a fixed angle or a normal angle is adopted through angle correction, and water sample collection and global satellite positioning are carried out in the photographic sea area. The collected water sample is used to determine the suspended sediment content by filtration method. The collected water sample is divided into training sample and verification sample.

[0045] S2, image processing: the obtained seawater suspended sediment concentration image is processed. Since the obtained seawater suspended sediment concentration image is affected by factors such as shooting angle, light intensity and sea wave fluctuation, the collected image data needs to be corrected, including mean filter, geographic registration and angle correction, in order to reduce noise and error, improve data quality, and more accurately reflect the information of seawater suspended sediment concentration. Through automatic splicing and orthographic image processing, image data which can be used for further analysis is generated. The specific implementation method is:

[0046] (1) Mean filter. The resolution of seawater suspended sediment concentration image obtained by low-altitude is much higher than remote sensing. The small changes on the surface of seawater can be identified. Since the fluctuation of seawater will cause the fluctuation of seawater surface, the image will present wave when shooting at a certain angle. The image needs to be processed by mean filter, and the image null value needs to be interpolated to reduce error.

[0047] (2) Geographic registration. Select feature ground objects, use global satellite positioning system for spatial positioning, and obtain the latitude and longitude of the ground objects. Combine the feature ground object coordinate points with the pixel points on the image to use ArcGIS software for geographic registration. If there is no feature ground object in the observation area, it can be marked manually.

[0048] (3) Angle correction. When collecting seawater suspended sediment concentration image data, the band value of the image will be different due to different shooting angles, which needs to be corrected. Single color liquid is tested under the same light source. The camera is used to shoot at different angles from 20° to 90°. The RED, GREEN and BLUE band values of the camera image are extracted. The function fitting of different camera angles and the same position camera band values is carried out, and the angle correction equation is obtained as follows: X =13*λ*C i / 1000+λ, where C X is the corrected band value, C i is the uncorrected band value, and λ is the angle between the camera and the horizontal plane; the RED, GREEN, and BLUE band values are corrected to a positive camera angle using an angle correction equation, and the corrected RED, GREEN, and BLUE band values are: R = 13 * λ * C r / 1000+λ; C G = 13 * λ * C g / 1000+λ; C b = 13 * λ * C b / 1000+λ.

[0049] S3, Model establishment and training: according to the collected suspended sediment concentration image data and the measured suspended sediment concentration data, an artificial neural network including an input layer, a hidden layer, and an output layer is constructed. The neural network is trained using the suspended sediment concentration image data and the corresponding measured data, the network parameters are adjusted, and the model performance is optimized.

[0050] Based on the artificial neural network algorithm, the artificial neural network is trained using the image band values of the seawater suspended sediment concentration and the measured suspended sediment concentration data, and a seawater suspended sediment concentration low-altitude inversion model is established. A neural network model including an input layer, a hidden layer, and an output layer is constructed using Matlab. The input layer of this embodiment is the RGB band data extracted from the camera image, and the output layer is the suspended sediment concentration data. Specifically, the band data of each pixel point is input, and the suspended sediment concentration is the corresponding output. The neural network algorithm training data includes two parts, which are input data and output data. The two parts of data are combined to form a data pair, and 70% of the measured seawater suspended sediment concentration data and the processed corrected image band RGB data pair are used as the training data set, and the remaining data pair is used as the verification test data set. The input data and the output data are normalized to the [0, 1] interval using the mapminmax function. A neural network is constructed using the feedforwardnet function. The tansig activation function is used in the hidden layer, and the number of neurons in the hidden layer is set through experimental comparison and analysis. The number of neurons in the first hidden layer is 12, and the number of neurons in the second hidden layer is 7. The output layer uses the linear activation function purelin.

[0051] The trainlm algorithm is selected to train the neural network. The training parameters are set, including the maximum number of training times (1000 times), the learning rate (0.01), the training target error (10 -6 ), 5-fold cross-validation and L2 regularization are used to avoid overfitting and optimize the neural network parameters. After training, the training results of each group tend to be stable and meet the error requirements. The accuracy of the model for complex water body suspended sediment concentration inversion is improved.

[0052] S4, model verification and precision evaluation: using remote sensing suspended sediment concentration inversion model and measured verification sample to verify the seawater suspended sediment concentration low-altitude inversion model, using mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE) to evaluate the precision of seawater suspended sediment concentration inversion model;

[0053]

[0054]

[0055]

[0056] Wherein, y i is the measured value, is the model prediction value, n is the sample number.

[0057] The mean square error (MSE) of this embodiment is 120.04, the root mean square error (RMSE) is 10.95, the mean absolute error (MAE) is 7.04, and the model inversion accuracy is 89.75%. Fig. 2 In order to compare the low-altitude inversion results of seawater suspended sediment concentration with the inversion precision of satellite remote sensing data, the low-altitude observation precision is significantly improved;

[0058] S5, result application: applying the seawater suspended sediment concentration low-altitude inversion model to the image data of the target sea area, and obtaining the low-altitude observation results of seawater suspended sediment concentration.

[0059] Using the inversion model to observe the seawater suspended sediment concentration, Fig. 3 In order to use the mobile phone to shoot the seawater suspended sediment concentration image of different concentrations in Haizhou Bay, a1, a2 and a3 are the original images of the camera, which are seawater images of different suspended sediment concentrations, a1 is low-concentration water body, a2 is high-concentration water body, and a3 is medium-high-concentration water body. b1, b2 and b3 are the corresponding seawater suspended sediment concentration inversion results. It can be seen from the inverted image that the suspended sediment concentration is between 50-75 mg / l, and the distribution of suspended sediment is consistent with the measured results.

[0060] The present application is a kind of seawater suspended sediment concentration low-altitude observation method, which uses low-altitude observation platforms such as unmanned aerial vehicles, airships, coastal highlands, structures, offshore platforms or ships, etc. to obtain high-resolution image data of seawater suspended sediment concentration closer to seawater, thereby improving the observation accuracy of suspended sediment concentration. Compared with satellite remote sensing and multispectral cameras and hyperspectral cameras, the digital photography equipment (such as smartphones and cameras) used in this method has low cost and is easy to promote and apply on a large scale. This method can quickly obtain suspended sediment concentration information of large areas of sea, thereby improving the observation efficiency. With the help of low-altitude observation platforms, the observation sea area and observation time can be adjusted at any time according to needs, and the method has strong flexibility. Compared with traditional observation methods, low-altitude observation platforms have lower risks and are suitable for complex sea conditions. Low-altitude observation can be quickly repeated, which is convenient for dynamic monitoring of changes in suspended sediment concentration.

[0061] The seawater suspended sediment concentration low-altitude observation method effectively makes up for the deficiencies of the prior art in retrieving seawater suspended sediment concentration, the model is more accurate, and the timeliness, accuracy and flexibility of the observation area of seawater suspended sediment concentration observation are improved, thereby realizing efficient observation of seawater suspended sediment concentration in small-scale sea areas.

[0062] Finally, it should be noted that: the above only describes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent replacements to some technical features, as long as they are within the spirit and principles of the present application. Any modification, equivalent replacement, improvement, etc. shall be included in the protection scope of the present application.

Claims

1. A low-altitude observation method for suspended sediment concentration in seawater, characterized in that: The steps include: S1. Using a low-altitude observation platform, digital photography is used to obtain images of suspended sediment concentration in the target sea area. The images can be stored on a chip memory card for downloading or transferring to a computer workstation for subsequent processing. Water samples are collected and global satellite positioning is performed simultaneously in the imaging sea area. The collected water samples are filtered to determine the suspended sediment content. The collected water samples are divided into modeling samples and verification samples. S2. Preprocess the acquired seawater suspended sediment concentration images, including mean filtering, georeferencing, and angle correction. Through automatic stitching and orthophoto processing, generate image data for further analysis; S3. Based on the collected suspended sediment concentration image data and the measured suspended sediment concentration data, an artificial neural network is constructed. Based on the optimized artificial neural network algorithm, the artificial neural network is trained using the seawater suspended sediment concentration image data and the measured suspended sediment concentration data. The network parameters are adjusted to optimize the model performance and establish a low-altitude inversion model for seawater suspended sediment concentration. S4. Use the remote sensing suspended sediment concentration inversion model and field-measured verification samples to verify the low-altitude inversion model for seawater suspended sediment concentration, evaluate the accuracy of the seawater suspended sediment concentration inversion model, and adjust and optimize the model based on the evaluation results; S5. Apply the low-altitude inversion model of seawater suspended sediment concentration to the target sea area image data to obtain the low-altitude observation results of seawater suspended sediment concentration; In step S2, the specific implementation method of georeferencing during the seawater suspended sediment concentration image processing is as follows: Geo-reference: The camera system is equipped with a global satellite positioning system, and real-time dynamic carrier phase difference is used for spatial positioning. Alternatively, characteristic features are selected and spatially positioned using the global satellite positioning system to obtain the longitude and latitude of the features. The coordinate points of the characteristic features are combined with the pixel points on the image to perform geo-reference using ArcGIS software. If there are no characteristic features in the observation area, they are manually marked. The specific implementation method of angle correction in seawater suspended sediment concentration image processing is as follows: The test was conducted using a monochromatic liquid under the same light source. The camera was used to shoot at different angles from 20° to 90°. The RED, GREEN, and BLUE band values ​​of the camera image were extracted using Matlab software. The function was fitted between the different camera inclination angles and the camera band values ​​at the same position. The angle correction equation was obtained as follows: X =13*λ*C / 1000+λ, where C X is the corrected band value, C is the uncorrected band value, and λ is the angle between the camera and the horizontal plane; Use the angle correction equation to correct the RED, GREEN and BLUE band values ​​to the positive angle. The corrected RED, GREEN and BLUE band values ​​are: C R =13*λ*C / 1000+λ;C G =13*λ*C / 1000+λ;C B =13*λ*C / 1000+λ; Based on the artificial neural network algorithm, the artificial neural network was trained using the band values ​​of seawater suspended sediment concentration images and the measured suspended sediment concentration to establish a low-altitude inversion model for seawater suspended sediment concentration. The artificial neural network is a BP neural network model with an input layer, two hidden layers and an output layer. A neural network model consisting of an input layer, a hidden layer, and an output layer was constructed using Matlab. The input layer consisted of RGB band data extracted from camera images, and the output layer consisted of suspended sediment concentration data. The band data for each pixel served as input, and the suspended sediment concentration served as the corresponding output. The neural network algorithm training data consisted of two parts: input data and output data. These two parts of data were combined into data pairs. 70% of the measured seawater suspended sediment concentration data and processed and corrected image band RGB data pairs were used as training data sets, and the remaining data pairs served as validation test data sets. The input and output data are normalized to the range [0, 1] using the mapminmax function. A neural network is constructed using the feedforwardnet function. The hidden layer uses the tansig activation function. The number of neurons in the hidden layer is set through experimental comparison and analysis. The number of neurons in the first hidden layer is 12, and the number of neurons in the second hidden layer is 7. The output layer uses the linear activation function purelin. The training algorithm selects trainlm to train the neural network and sets the training parameters, including the maximum number of training times 1000, the learning rate 0.01, and the training target error 10. -6 ,After optimization, 5-fold cross validation and L2 regularization are used to avoid overfitting and optimize the neural network parameters; In step S4, the low-altitude inversion model of seawater suspended sediment concentration is verified by comparing it with the measured suspended sediment concentration and the suspended sediment concentration inverted by remote sensing. The accuracy of the suspended sediment concentration inversion model is evaluated using three indicators: mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAPE). The formula is as follows: Among them, y i is the measured value, is the model prediction value, and n is the number of samples.

2. The method for low-altitude observation of suspended sediment concentration in seawater according to claim 1, characterized in that: In step S1, the low-altitude observation platform is a platform with a certain height capable of performing low-altitude photography and obtaining digital images of suspended sediment concentration in seawater, including but not limited to drones, airships, coastal heights, structures, offshore platforms or ships.

3. The method for low-altitude observation of suspended sediment concentration in seawater according to claim 1, characterized in that: In step S1, digital photography includes a smart phone, a camera, and a high-resolution visible light digital camera.

4. The method for low-altitude observation of suspended sediment concentration in seawater according to claim 1, characterized in that: In step S1, when obtaining an image of suspended sediment concentration in seawater by digital photography, a fixed angle is used or an angle is corrected to an orthographic angle.

Images