Water area storage volume investigation platform based on monitoring unmanned aerial vehicle
By using a monitoring platform for water storage surveys of drones, threshold ranges are calculated using RGB values and flight path planning is performed. This platform identifies water images and calculates storage volume, solving the accuracy and efficiency problems of drone-based water storage surveys and achieving efficient and accurate water storage surveys.
Patent Information
- Application Number
- CN202410977152.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-11-12
AI Technical Summary
Existing UAV-based water storage survey technologies suffer from low accuracy, complex operation, and a lack of systematic approach, failing to meet the demand for efficient and accurate water storage surveys.
The water storage survey platform based on monitoring drones includes a sample training unit, a self-identification unit, a self-planning unit, an acquisition and processing unit, and a measurement and calculation unit. It calculates the storage volume by acquiring RGB values, calculating threshold ranges, planning flight paths, identifying water images, and acquiring and processing image data.
It has enabled efficient, accurate, and comprehensive water storage surveys, providing precise data support and reliable technical support for water management and planning.
Smart Images

Figure CN118941989A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of water conservancy engineering, and in particular to a water area storage volume survey platform based on a monitoring unmanned aerial vehicle. Background Art
[0002] As the importance of water resources becomes increasingly prominent, accurate survey of water storage has become a key task in water resources management and planning. Traditional methods of measuring water storage often rely on manual field measurements, which are not only inefficient and costly, but also difficult to fully and accurately cover large areas of water.
[0003] Driven by modern science and technology, drone technology has developed rapidly and has been widely used in many fields. However, the application of drones in water storage surveys still faces many challenges, such as how to accurately identify water areas and non-water areas from images taken by drones, how to plan the flight collection path of drones to ensure the integrity and validity of data, and how to process and analyze large amounts of image data to accurately calculate water storage.
[0004] At present, although there are some water area monitoring technologies based on drones, most of them have problems such as low accuracy, complex operation, and lack of systematicity, which cannot meet the needs of accurate and efficient investigation of water area storage. Therefore, building a comprehensive, accurate and efficient water area storage investigation platform based on monitoring drones has important practical significance and application value. Summary of the invention
[0005] The purpose of the present invention is to provide a water area storage volume survey platform based on monitoring drones, which solves the technical problems raised in the background technology.
[0006] The purpose of the present invention can be achieved by the following technical solutions:
[0007] The water storage volume survey platform based on monitoring drones includes:
[0008] The sample training unit is used to input multiple water image samples containing water bodies in the target water area, obtain the RGB values of the samples, and perform standard deviation calculation, sorting, and adjacent difference calculation on them, and obtain the threshold intervals corresponding to the R value, G value, and B value respectively, and use them as comparative features of the water image samples;
[0009] The self-identification unit is used to obtain the R value, G value, and B value of the pixel points in the image taken by the drone, and compare them with the corresponding threshold intervals, highlight the pixels that meet the requirements, and then determine whether the image is a water area image or a non-water area image by calculating the ratio of the number of pixels with the highlight mark to the total number of pixels and comparing it with a preset ratio coefficient;
[0010] The self-planning unit is used to pre-set the flight altitude of the UAV, obtain the actual side length of the shooting picture at the corresponding position of the target water area at a fixed flight altitude, determine the moving distance of the UAV based on this, and set the initial collection node on the relevant map containing the target water area. The collection point array diagram is obtained by making a rectangular array according to the moving distance through CAD technology;
[0011] The acquisition processing unit is used to allow the drone to shoot and acquire the acquisition image according to the acquisition node position in the acquisition point array diagram of the relevant map, and the self-identification unit distinguishes the acquired image as a water area image or a non-water area image, retains and acquires the water depth parameter for the water area image, and does not retain the non-water area image;
[0012] The measurement and calculation unit is used to calculate the storage capacity of the target area according to the actual side length of the collected image and the water depth parameters of the corresponding collection nodes of each retained collection image.
[0013] As a further solution of the present invention: the specific method of the measurement and calculation unit is as follows:
[0014] By formula: Calculate the storage volume V of the target area;
[0015] Wherein, a is the actual side length of the acquired image, hj represents the water depth parameter of each retained acquired image corresponding to the corresponding acquisition node, j=1, 2, ... m, and m represents the number of the retained acquired images corresponding to the corresponding acquisition nodes.
[0016] As a further solution of the present invention: the specific method of the sample training unit is as follows:
[0017] SA1. Enter water area image samples
[0018] In the target water area, multiple water area image samples containing water bodies are recorded;
[0019] SA2, get RGB value
[0020] Obtain the RGB value of the water area image sample, and then the corresponding R value, G value, and B value of all pixel points of the multiple water area image samples;
[0021] SA3. Calculate the standard deviation of R value
[0022] Select the R value corresponding to all pixels;
[0023] The standard deviation of the R values corresponding to all pixels is calculated using the standard deviation calculation formula;
[0024] SA4, R value sorting and calculation of adjacent differences
[0025] Sort all R values from small to large, and calculate the difference between all two adjacent R values among all the sorted R values;
[0026] SA5. Determine the R-value threshold
[0027] From all the sorted R values, select the middle value, then take the difference between adjacent R values on both sides based on the middle value, and compare the corresponding difference with the standard deviation one by one:
[0028] If the difference between a group of adjacent R values is less than or equal to the standard deviation, then the difference between the next group of adjacent R values is obtained and compared with the standard deviation;
[0029] If the difference between a group of adjacent R values is greater than or equal to the standard deviation, then in the group of adjacent R values, the R value with the smallest absolute difference calculated from the middle value is selected as the maximum threshold or minimum threshold;
[0030] SA6. Generate R value threshold interval
[0031] After the SA5 comparison is completed, the corresponding maximum threshold and minimum threshold are obtained, and then the maximum threshold and the minimum threshold are combined into a threshold interval corresponding to the R value;
[0032] SA7. Get the G and B value threshold ranges
[0033] According to the method of SA3-SA6, the threshold interval corresponding to the G value and the threshold interval corresponding to the B value are obtained respectively;
[0034] SA8. Determine the comparative characteristics of water area image samples
[0035] Then, the threshold interval corresponding to the R value, the threshold interval corresponding to the G value, and the threshold interval corresponding to the B value are used as comparative features of the water area image samples.
[0036] As a further solution of the present invention: the specific method of the self-identification unit is as follows:
[0037] SB1. Obtain the corresponding R value, G value, and B value of all pixels in the image taken by the drone;
[0038] SB2, the R value, G value, and B value corresponding to each pixel point are respectively compared with the threshold interval corresponding to the R value, the threshold interval corresponding to the G value, and the threshold interval corresponding to the B value;
[0039] If the R value, G value, and B value corresponding to a pixel point are respectively within the corresponding threshold intervals, the pixel point is highlighted, indicating that the pixel point is a pixel point of the image corresponding to the water area;
[0040] Otherwise, no highlight mark will be made;
[0041] SB3. Obtain the number u of all pixels with prominent marks added to the image, then compare u / n with a preset proportionality coefficient β, and determine whether the image is a water area image based on the comparison result:
[0042] Among them, n represents the number of all pixels in the image.
[0043] As a further solution of the present invention: in SB3, if u / n≥β, it means that the image is a water area image, which means that the image contains water;
[0044] If u / n<β, it means that the image is a non-water area image, which means that the image does not contain water.
[0045] As a further solution of the present invention: the specific method of the self-planning unit is as follows:
[0046] SS1, pre-set the flight altitude of the drone;
[0047] SS2. When the UAV is at a fixed flight altitude, obtain the actual side length of the UAV’s shooting image corresponding to the corresponding position in the target waters;
[0048] SS3. Determine the moving distance of the drone based on the actual side length;
[0049] SS4. On the relevant map containing the target waters, set an initial collection node, and then use CAD technology to make a rectangular array at the initial collection node according to the moving distance, and obtain a collection point array diagram;
[0050] Wherein, the collection point array diagram includes a number of collection nodes;
[0051] As a further solution of the present invention: the photographed picture that no one has obtained is in a square shape.
[0052] As a further solution of the present invention: the specific collection method of the collection processing unit is as follows:
[0053] In the corresponding collection point array diagram on the relevant map, according to the corresponding position of each collection node, the corresponding collection image and the water depth parameters of the corresponding collection node are obtained by drone photography;
[0054] As a further solution of the present invention: wherein, after obtaining a captured image at a position corresponding to a capture node, the self-identification unit distinguishes the captured image as a water area image or a non-water area image;
[0055] When the collected image is a water area image, the collected image is retained, and then the water depth parameters are obtained at the collection node through the laser radar or depth sounder pre-equipped by the drone;
[0056] When the collected image is a non-water area image, the collected image will not be retained.
[0057] Beneficial effects of the present invention:
[0058] Through the sample training unit, the threshold range of RGB values in water image samples can be accurately obtained, laying the foundation for the subsequent accurate identification of water images.
[0059] The self-identification unit can efficiently and accurately determine whether the image taken by the drone is a water image or a non-water image based on the comparison of the threshold range and the RGB value of the pixel point, greatly improving the accuracy and efficiency of image recognition.
[0060] The self-planning unit can pre-set the flight altitude of the drone, determine the actual side length and moving distance of the shooting picture, and generate an array diagram of collection points through CAD technology, ensuring the scientific and rational distribution of the collection points and realizing comprehensive and targeted water area data collection.
[0061] The acquisition and processing unit can effectively screen the images taken by the UAV at the acquisition node, retain the water area images and obtain the water depth parameters, remove the non-water area images, and improve the data validity and processing efficiency.
[0062] The measurement and calculation unit accurately calculates the storage capacity of the target area based on the actual side length and water depth parameters of the collected image, providing accurate and reliable data support for the management and planning of water resources.
[0063] In summary, the present invention can realize efficient, accurate and comprehensive monitoring and storage volume calculation of target water areas, and provide powerful technical support and data guarantee for water area management and related decision-making. BRIEF DESCRIPTION OF THE DRAWINGS
[0064] The present invention will be further described below in conjunction with the accompanying drawings.
[0065] Figure 1 It is a system block diagram of the water area storage capacity survey platform based on monitoring drones of the present invention.
[0066] Figure 2 It is a flow chart of a sample training unit in a water area storage volume survey platform based on monitoring drones of the present invention. DETAILED DESCRIPTION
[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0068] Embodiment 1
[0069] See also Figure 1 and Figure 2 As shown, the present invention is a water area storage volume survey platform based on monitoring drones, comprising:
[0070] The self-planning unit is used to pre-set the flight altitude of the UAV, obtain the actual side length of the shooting picture at the corresponding position of the target water area at a fixed flight altitude, determine the moving distance of the UAV based on this, and set the initial collection node on the relevant map containing the target water area. The collection point array diagram is obtained by making a rectangular array according to the moving distance through CAD technology;
[0071] The specific method is as follows:
[0072] SS1, pre-set the flight altitude of the drone;
[0073] SS2. When the UAV is at a fixed flight altitude, obtain the actual side length of the UAV’s shooting image corresponding to the corresponding position in the target waters;
[0074] In this embodiment, the shooting screen is in a square shape;
[0075] SS3. Determine the moving distance of the drone based on the actual side length;
[0076] SS4. On the relevant map containing the target waters, set an initial collection node, and then use CAD technology to make a rectangular array at the initial collection node according to the moving distance, and obtain a collection point array diagram;
[0077] Among them, the initial collection node is a designated point at the edge of the target water area;
[0078] The acquisition processing unit is used to allow the drone to shoot and acquire acquisition images according to the acquisition node positions in the acquisition point array diagram of the relevant map, and then divide the acquired images into water images or non-water images, and remove the non-water images, and then obtain the water depth parameters at the acquisition nodes corresponding to the retained water images;
[0079] The specific collection methods are as follows:
[0080] In the corresponding collection point array diagram on the relevant map, according to the corresponding position of each collection node, the corresponding collection image is obtained by shooting with a drone;
[0081] After obtaining a collected image at a position corresponding to a collection node, the collected image is divided into a water area image or a non-water area image;
[0082] When the collected image is a water area image, the collected image is retained, and then the water depth parameters are obtained at the collection node by using a UAV equipped with a laser radar or a depth sounder;
[0083] When the collected image is a non-water area image, the collected image will not be retained;
[0084] A measurement and calculation unit, used to calculate the storage volume of the target area according to the actual side length of the collected image and the water depth parameters of the corresponding collection nodes of each retained collected image;
[0085] The specific method is as follows:
[0086] By formula: Calculate the storage volume V of the target area;
[0087] Where a is the actual side length of the acquired image, hj represents the water depth parameter of each retained acquired image corresponding to the corresponding acquisition node, j = 1, 2, ... m, m represents the number of retained acquired images corresponding to the corresponding acquisition nodes;
[0088] This embodiment can ensure comprehensive and regular coverage of the target water area by scientifically planning the collection path of the drone through the self-planning unit. The collection and processing unit effectively screens the collected images, retains only the water area images and obtains the water depth parameters, thereby improving the validity and processing efficiency of the data. The measurement and calculation unit calculates the storage capacity of the target area based on accurate data, providing a reliable quantitative basis for water area management and decision-making. The overall solution realizes efficient and accurate investigation of the storage capacity of the target water area.
[0089] Embodiment 2
[0090] As the second embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the technical solution of the present embodiment is different from that of the first embodiment only in that,
[0091] The sample training unit is used to input multiple water image samples containing water bodies in the target water area, obtain the RGB values of the samples, and perform standard deviation calculation, sorting, and adjacent difference calculation on them, and obtain the threshold intervals corresponding to the R value, G value, and B value respectively, and use them as comparative features of the water image samples;
[0092] The specific method is as follows:
[0093] SA1. In the target water area, record multiple water area image samples containing water bodies;
[0094] SA2, obtaining the RGB value of the water area image sample, and then the corresponding R value, G value, and B value of all pixels of the multiple water area image samples;
[0095] SA3, select the R value corresponding to all pixels;
[0096] The standard deviation of the R values corresponding to all pixels is calculated using the standard deviation calculation formula;
[0097] SA4. Sort all R values in ascending order, and calculate the difference between all two adjacent R values among all the sorted R values;
[0098] SA5. From all the sorted R values, select the middle value, then take the difference between the adjacent R values on both sides based on the middle value, and compare the corresponding difference with the standard deviation one by one:
[0099] If the difference between a group of adjacent R values is less than or equal to the standard deviation, then the difference between the next group of adjacent R values is obtained and compared with the standard deviation;
[0100] If the difference between a group of adjacent R values is greater than or equal to the standard deviation, then in the group of adjacent R values, the R value with the smallest absolute difference calculated from the middle value is selected as the maximum threshold or minimum threshold;
[0101] After the comparison of SA6 and SA5 is completed, the corresponding maximum threshold and minimum threshold are obtained, and then the maximum threshold and the minimum threshold are combined into a threshold interval corresponding to the R value;
[0102] Among them, the threshold interval is [Rmin, Rmax], Rmin represents the minimum threshold corresponding to the comparison result, and Rmax represents the maximum threshold corresponding to the comparison result;
[0103] SA7. According to the method of SA3-SA6, the threshold interval [Gmin, Gmax] corresponding to the G value and the threshold interval [Bmin, Bmax] corresponding to the B value are obtained respectively;
[0104] SA8, then use the threshold interval [Rmin, Rmax] corresponding to the R value, the threshold interval [Gmin, Gmax] corresponding to the G value, and the threshold interval [Bmin, Bmax] corresponding to the B value as comparison features of the water area image samples;
[0105] A self-identification unit, used to distinguish between water images and non-water images corresponding to the images taken by the drone;
[0106] The specific method is as follows:
[0107] SB1. Obtain the corresponding R value, G value, and B value of all pixels in the image taken by the drone;
[0108] SB2. Compare the R value, G value, and B value corresponding to each pixel with the threshold interval [Rmin, Rmax] corresponding to the R value, the threshold interval [Gmin, Gmax] corresponding to the G value, and the threshold interval [Bmin, Bmax] corresponding to the B value;
[0109] If the R value, G value, and B value corresponding to a pixel point are respectively in [Rmin, Rmax], [Gmin, Gmax], and [Bmin, Bmax], the pixel point is highlighted, indicating that the pixel point is a pixel point of the image corresponding to the water area;
[0110] Otherwise, no highlight mark will be made;
[0111] In this embodiment, the highlight mark is used to count the number of corresponding pixels whose R value, G value, and B value are respectively within [Rmin, Rmax], [Gmin, Gmax], and [Bmin, Bmax];
[0112] SB3. Obtain the number u of all pixels with highlight marks added to the image, and then compare u / n with the preset proportionality coefficient β:
[0113] Where n represents the number of all pixels in the image;
[0114] If u / n≥β, it means that the image is a water area image, which means that the image contains water.
[0115] If u / n<β, it means that the image is a non-water image, which means that the image does not contain water.
[0116] In this embodiment, the combination of the sample training unit and the self-identification unit is used to automatically classify the captured image into a water area image or a non-water area image.
[0117] In this embodiment, the sample training unit obtains the threshold interval of the RGB value through complex calculation and analysis, which provides an accurate standard for the subsequent accurate identification of water images. The self-identification unit accurately judges water images and non-water images based on the comparison of the threshold interval and the RGB value of the pixel point, which greatly improves the accuracy of image recognition. The combination of the two can automatically, efficiently and accurately classify the collected images, providing a high-quality data source for subsequent processing and analysis.
[0118] Embodiment 3
[0119] As the third embodiment of the present invention, when the present application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments mentioned above for implementation.
[0120] This embodiment combines the solutions of embodiment 1 and embodiment 2, and realizes the optimization of the whole process from scientific planning of drone collection paths, to accurate screening and classification of collected images, and then to accurate calculation of storage volume in the target area. It can not only ensure comprehensive and regular data collection, but also accurately distinguish between water area and non-water area images, and finally obtain accurate storage volume results, providing comprehensive, accurate and efficient technical support for the effective management and reasonable planning of water resources, with significant comprehensive benefits.
[0121] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.
[0122] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. The water storage volume survey platform based on monitoring drones is characterized by: include: The sample training unit is used to input multiple water image samples containing water bodies in the target water area, obtain the RGB values of the samples, and perform standard deviation calculation, sorting, and adjacent difference calculation on them, and obtain the threshold intervals corresponding to the R value, G value, and B value respectively, and use them as comparative features of the water image samples; The self-identification unit is used to obtain the R value, G value, and B value of the pixel points in the image taken by the drone, and compare them with the corresponding threshold intervals, highlight the pixels that meet the requirements, and then determine whether the image is a water area image or a non-water area image by calculating the ratio of the number of pixels with the highlight mark to the total number of pixels and comparing it with a preset ratio coefficient; The self-planning unit is used to pre-set the flight altitude of the UAV, obtain the actual side length of the shooting picture at the corresponding position of the target water area at a fixed flight altitude, determine the moving distance of the UAV based on this, and set the initial collection node on the relevant map containing the target water area. The collection point array diagram is obtained by making a rectangular array according to the moving distance through CAD technology; The acquisition processing unit is used to allow the drone to shoot and acquire the acquisition image according to the acquisition node position in the acquisition point array diagram of the relevant map, and the self-identification unit distinguishes the acquired image as a water area image or a non-water area image, retains and acquires the water depth parameter for the water area image, and does not retain the non-water area image; The measurement and calculation unit is used to calculate the storage capacity of the target area according to the actual side length of the collected image and the water depth parameters of the corresponding collection nodes of each retained collected image.
2. The water area storage volume survey platform based on monitoring drone according to claim 1 is characterized in that: The specific method of the measurement calculation unit is as follows: By formula: Calculate the storage volume V of the target area; Wherein, a is the actual side length of the acquired image, hj represents the water depth parameter of each retained acquired image corresponding to the corresponding acquisition node, j=1, 2, ... m, and m represents the number of the retained acquired images corresponding to the corresponding acquisition nodes.
3. The water area storage volume survey platform based on monitoring drone according to claim 1 is characterized in that: The specific method of the sample training unit is as follows: SA1. In the target water area, record multiple water area image samples containing water bodies; SA2, obtaining the RGB value of the water area image sample, and then the corresponding R value, G value, and B value of all pixels of the multiple water area image samples; SA3, select the R value corresponding to all pixels; The standard deviation of the R values corresponding to all pixels is calculated using the standard deviation calculation formula; SA4. Sort all R values in ascending order, and calculate the difference between all two adjacent R values among all the sorted R values; SA5. From all the sorted R values, select the middle value, then take the difference between the adjacent R values on both sides based on the middle value, and compare the corresponding difference with the standard deviation one by one: If the difference between a group of adjacent R values is less than or equal to the standard deviation, then the difference between the next group of adjacent R values is obtained and compared with the standard deviation; If the difference between a group of adjacent R values is greater than or equal to the standard deviation, then in the group of adjacent R values, the R value with the smallest absolute difference calculated from the middle value is selected as the maximum threshold or minimum threshold; After the comparison of SA6 and SA5 is completed, the corresponding maximum threshold and minimum threshold are obtained, and then the maximum threshold and the minimum threshold are combined into a threshold interval corresponding to the R value; SA7. According to the method of SA3-SA6, the threshold interval corresponding to the G value and the threshold interval corresponding to the B value are obtained respectively; SA8. Then, the threshold interval corresponding to the R value, the threshold interval corresponding to the G value, and the threshold interval corresponding to the B value are used as comparative features of the water area image samples.
4. The water area storage volume survey platform based on monitoring drone according to claim 1 is characterized in that: The specific method of the self-identification unit is as follows: SB1. Obtain the corresponding R value, G value, and B value of all pixels in the image taken by the drone; SB2, the R value, G value, and B value corresponding to each pixel point are respectively compared with the threshold interval corresponding to the R value, the threshold interval corresponding to the G value, and the threshold interval corresponding to the B value; If the R value, G value, and B value corresponding to a pixel point are respectively within the corresponding threshold intervals, the pixel point is highlighted, indicating that the pixel point is a pixel point of the image corresponding to the water area; Otherwise, no highlight mark will be made; SB3. Obtain the number u of all pixels with prominent marks added to the image, then compare u / n with a preset proportionality coefficient β, and determine whether the image is a water area image based on the comparison result: Among them, n represents the number of all pixels in the image.
5. The water area storage volume survey platform based on monitoring drone according to claim 4 is characterized in that: In SB3, if u / n≥β, it means that the image is a water area image, which means that the image contains water. If u / n<β, it means that the image is a non-water area image, which means that the image does not contain water.
6. The water area storage volume survey platform based on monitoring drone according to claim 1 is characterized in that: The specific method of the self-planning unit is as follows: SS1. Pre-set the flight altitude of the drone; SS2. When the UAV is at a fixed flight altitude, obtain the actual side length of the UAV’s shooting image corresponding to the corresponding position in the target waters; SS3. Determine the moving distance of the drone based on the actual side length; SS4. On the relevant map containing the target waters, set an initial collection node, and then use CAD technology to make a rectangular array at the initial collection node according to the moving distance, and obtain a collection point array diagram; The collection point array diagram includes several collection nodes.
7. The water area storage volume survey platform based on monitoring drone according to claim 6 is characterized in that: The shot that no one got was in a square shape.
8. The water area storage volume survey platform based on monitoring drone according to claim 1 is characterized in that: The specific collection method of the collection processing unit is as follows: In the corresponding collection point array diagram on the relevant map, according to the corresponding position of each collection node, the corresponding collection image and the water depth parameters of the corresponding collection node are obtained by drone photography.
9. The water area storage volume survey platform based on monitoring drone according to claim 8 is characterized in that: in, After obtaining a captured image at a position corresponding to a capture node, the self-identification unit distinguishes the captured image as a water area image or a non-water area image; When the collected image is a water area image, the collected image is retained, and then the water depth parameters are obtained at the collection node through the laser radar or depth sounder pre-equipped by the drone; When the collected image is a non-water area image, the collected image will not be retained.
