A method for measuring water flow and velocity of an unmanned aerial vehicle

Abstract

The present application discloses a method for measuring flow and velocity in hydrology, which relates to the technical field of hydrological measurement. The method comprises the following steps: selecting a dynamic anchor point in a target flow measurement area, and anchoring a drone according to the dynamic anchor point; the drone collects images of its own surroundings, obtains obstacle data, and generates a motion route in real time; the drone collects video of the target flow measurement area, generates a fused water surface video, and analyzes the fused water surface video to generate an initial velocity distribution map; the drone collects riverbed data and water surface fluctuation data of the target flow measurement area, obtains riverbed topography based on the riverbed data, obtains water depth data and cross-sectional average velocity; constructs a three-dimensional water flow model, collects the motion trajectory of a buoy group and gyroscope data, corrects the three-dimensional water flow model, and outputs the hydrological flow and velocity measurement results. The present application has the effect of improving the measurement accuracy of flow and velocity measurement and the flexibility and universality when dealing with complex waters.

Description

Technical Field

[0001] The present application relates to the technical field of hydrological measurement, and in particular to a method for hydrological flow and velocity measurement using an unmanned aerial vehicle (UAV). Background Art

[0002] my country is a country with many river basins, especially in the southern region, where the water network is densely distributed and there are many large rivers. As a result, the water flow detection and monitoring of these rivers has become a very important work content, especially the hydrological flow rate monitoring of rivers, which is of great significance for river flood prevention and early warning.

[0003] In the existing technology, when detecting the flow velocity of a river, the traditional method is still to adopt a fixed-point monitoring method, setting up a detection device on the bank of the river, or placing a flow meter in the river, etc., which consumes a lot of time and manpower, and takes a certain amount of time to collect data, and the measurement time is long. At the same time, during the flow measurement process, it is affected by various factors such as wind speed, waves, impurities in the water flow, etc., resulting in low accuracy of the measurement results. In addition, the above method is only applicable to rivers with a certain width and depth. When facing waters with more complex environments, it is difficult to install equipment and measure equipment, and it is not universal. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for measuring hydrological flow and velocity using an unmanned aerial vehicle (UAV) to solve the problems raised in the above-mentioned background technology.

[0005] This application provides a method for measuring hydrological flow and velocity using an unmanned aerial vehicle, the method comprising:

[0006] Obtaining a target flow measurement area of the water area to be inspected, calling a drone to go to the target flow measurement area, selecting a dynamic anchor point in the target flow measurement area, and anchoring and hovering the drone based on the dynamic anchor point;

[0007] The drone collects images of its surroundings, obtains obstacle data based on the images, and generates a motion route in real time based on the obstacle data and the dynamic anchor points;

[0008] The drone collects video of the target flow measurement area, obtains visible light water surface video, infrared light water surface video and polarized light water surface video, fuses them to generate a fused water surface video, and analyzes the fused water surface video to generate an initial flow velocity distribution map;

[0009] The drone collects riverbed data and water surface fluctuation data of the target flow measurement area, obtains riverbed topography based on the riverbed data, and obtains water depth data and cross-sectional average flow velocity based on the riverbed topography and the water surface fluctuation data;

[0010] Based on the riverbed topography, the water depth data and the initial flow velocity distribution map, a three-dimensional water flow model is constructed, and a drone is called to release a group of buoys to collect the motion trajectory and gyroscope data of the buoy group. The three-dimensional water flow model is corrected according to the motion trajectory, the gyroscope data and the average flow velocity of the section, and the hydrological flow and velocity measurement results are output.

[0011] Preferably, the steps of selecting a dynamic anchor point in the target flow measurement area and anchoring the drone according to the dynamic anchor point are specifically as follows:

[0012] Invoking the drone to collect images of the target flow measurement area to obtain a watershed image, extracting water flow information from the watershed image to obtain a water flow image, and particle-izing the water flow image to obtain water flow image particles;

[0013] Extracting multiple particle features from the water flow image particles based on the water flow image particles, and recording a conspicuousness value of each particle feature;

[0014] Selecting the particle feature with the highest conspicuity value as the target particle feature, and performing anchor point selection on the water flow image particles according to the target particle feature to obtain a dynamic anchor point;

[0015] Taking photos of the dynamic anchor point at a preset time interval, recording the displacement of the dynamic anchor point in the two photos, and obtaining the hovering speed according to the displacement and the time interval;

[0016] The propeller power distribution of the UAV is adjusted according to the hovering speed so that the UAV and the dynamic anchor point remain relatively stationary.

[0017] Preferably, the steps of collecting the surrounding images of the drone, obtaining obstacle data based on the surrounding images, and generating a motion route in real time based on the obstacle data and the dynamic anchor points are specifically as follows:

[0018] The drone collects images of its surrounding environment to obtain a surrounding image, and performs content identification on the surrounding image to obtain an obstacle image;

[0019] Extracting visual distance features and external shape features of the obstacle image, obtaining the obstacle distance based on the visual distance features, obtaining the obstacle spatial position based on the external shape features, and generating initial obstacle data by combining the obstacle distance and the obstacle spatial position;

[0020] Predicting obstacles outside or behind the surrounding image based on the initial obstacle data and the external shape features to obtain predicted obstacle data, and adding the predicted obstacle data to the initial obstacle data to generate obstacle data;

[0021] The current drone spatial position of the drone and the current anchor spatial position of the dynamic anchor point are obtained, and the movement path of the drone is generated in real time by combining the obstacle data, the drone spatial position and the anchor spatial position.

[0022] Preferably, after the step of generating the movement route of the drone in real time by combining the obstacle data, the spatial position of the drone, and the spatial position of the anchor point, the method further includes:

[0023] When the UAV moves along the movement route, the surrounding image is updated in real time, and the obstacle image is updated in real time to obtain an updated obstacle image;

[0024] obtaining updated obstacle data based on the updated obstacle image, and determining whether the movement route is blocked by the obstacle based on the updated obstacle data;

[0025] If it is determined that the movement route is blocked by the obstacle, extracting the blocked route segment from the movement route;

[0026] Based on the route segment and the updated obstacle data, the route segment is reconstructed to obtain an updated route segment, and the updated route segment is added to the motion route.

[0027] Preferably, before the step of fusing the visible light water surface video, the infrared light water surface video and the polarized light water surface video to generate a fused water surface video, the step further includes:

[0028] extracting a light reflection area from the visible light water surface video, extracting a light quantity from the light reflection area, and performing exposure reduction processing on the light reflection area according to the light quantity to obtain a target light reflection area;

[0029] Adding the target light reflection area to the visible light water surface video to obtain a target visible light water surface video;

[0030] Extracting the temperature distribution area and infrared light scattering amount of the water surface based on the infrared light water surface video, and performing clarity processing on the infrared light water surface video based on the temperature distribution area and the infrared light scattering amount to obtain a target infrared light water surface video;

[0031] Based on the polarized water surface video, the polarization state change of the water surface is extracted, and polarization compensation parameters are generated according to the polarization state change. The polarized water surface video is parameter compensated according to the polarization compensation parameters to obtain the target polarized water surface video.

[0032] Preferably, the steps of fusing the visible light water surface video, the infrared light water surface video and the polarized light water surface video to generate a fused water surface video, and analyzing the fused water surface video to generate the initial flow velocity distribution map are specifically as follows:

[0033] fusing the target visible light water surface video, the target infrared light water surface video, and the target polarized light water surface video to generate a fused water surface video;

[0034] Eliminating the water mist effect and temperature effect appearing in the fused water surface video based on the target infrared light water surface video;

[0035] Based on the target polarized light water surface video, enhancing the content clarity of the underwater image in the fused water surface video;

[0036] Performing water surface frame recognition on the fused water surface video to obtain water surface video frames, and recording visible light water surface differences and infrared light water surface differences between the water surface video frames;

[0037] Mutual verification and identification of the visible light water surface difference and the infrared light water surface difference to obtain an initial water surface velocity distribution;

[0038] Performing underwater frame recognition on the fused surface video to obtain underwater video frames, and recording the differences in water ripple shapes and water ripple colors between the underwater video frames;

[0039] Based on the differences in the shapes and colors of the water ripples, underwater water flow variation characteristics are obtained, and according to the underwater water flow variation characteristics, an initial underwater flow velocity distribution is obtained;

[0040] An initial flow velocity distribution map of a target flow measurement area is generated according to the initial water surface flow velocity distribution and the initial underwater flow velocity distribution.

[0041] Preferably, based on the target infrared water surface video, the step of eliminating the water mist effect and temperature effect appearing in the fused water surface video is specifically:

[0042] Extracting infrared frames from the target infrared light water surface video, and obtaining infrared temperature distribution and water surface infrared characteristics based on the infrared frames;

[0043] Locating the water surface boundary based on the water surface infrared characteristics to obtain water surface boundary data;

[0044] Performing layout segmentation on the infrared temperature distribution according to the water surface boundary data to obtain a water surface temperature distribution and an above-water temperature distribution;

[0045] Based on the water surface temperature distribution, the water mist influence in the target flow measurement area is obtained; based on the water surface temperature distribution, the temperature influence in the target flow measurement area is obtained;

[0046] A water mist image compensation is generated based on the water mist influence, and a temperature image compensation is generated based on the temperature influence. The water mist influence and the temperature influence in the fused water surface video are eliminated according to the water mist image compensation and the temperature image compensation.

[0047] Preferably, the steps of obtaining riverbed topography according to the riverbed data, and obtaining water depth data and cross-sectional average flow velocity according to the riverbed topography and the water surface fluctuation data are specifically as follows:

[0048] Modeling the riverbed of the target flow measurement area according to the riverbed data to obtain the riverbed topography;

[0049] Obtaining an average water surface spatial position of the target flow measurement area based on the water surface fluctuation data, and obtaining water depth data based on the riverbed topography and the average water surface spatial position;

[0050] Obtaining a cross-sectional spatial curve of the riverbed according to the riverbed topography, and constructing a watershed cross-sectional image of the target flow measurement area based on the cross-sectional spatial curve and the water depth data;

[0051] Based on the water surface fluctuation data and the water depth data, a plurality of deep flow velocity relationship diagrams are obtained by dividing the watershed cross-sectional image;

[0052] According to the initial flow velocity distribution diagram and the deep flow velocity relationship diagram, cross-sectional flow velocity data on the watershed cross-sectional image is constructed, and the cross-sectional average flow velocity is obtained according to the cross-sectional flow velocity data.

[0053] Preferably, the steps of calling a drone to release a group of buoys, collecting motion trajectories and gyroscope data of the buoy group, and correcting the three-dimensional water flow model according to the motion trajectories, the gyroscope data, and the average flow velocity of the cross section are specifically as follows:

[0054] Invoking a drone to release a group of buoys, the group of buoys moving on the water surface of the target flow measurement area, and collecting the movement trajectory and gyroscope data of the group of buoys;

[0055] Based on the motion trajectory, a basic water flow direction of the target flow measurement area is obtained, and based on the gyroscope data, water flow resistance data of the target flow measurement area is obtained;

[0056] Constructing an underwater water flow model in the three-dimensional water flow model based on the average flow velocity of the cross section;

[0057] Based on the basic water flow direction and the water flow confrontation data, the surface water flow in the three-dimensional water flow model is refined and corrected.

[0058] In summary, this application includes at least one of the following beneficial technical effects:

[0059] The target flow measurement area is inspected using a drone. First, a dynamic anchor point is selected within the water flow in the target flow measurement area. The drone is then positioned perpendicular to the anchor point, maintaining relative stillness on the surface. As the drone follows the dynamic anchor point, it captures images of its surroundings. Based on this data, it obtains obstacle data and sets its movement path in real time. The drone then captures visible light, infrared light, and polarized light video of the water surface in the target flow measurement area. These three videos are then pre-processed and fused to generate a fused water surface video. The fused water surface video eliminates the effects of mist and temperature on flow velocity analysis, yielding an initial flow velocity distribution map for the water surface. Riverbed data and water surface fluctuation data for the target flow measurement area are then collected to obtain water depth data. The average flow velocity across the section is then determined based on this water depth data and the initial flow velocity distribution. A three-dimensional flow model is constructed based on riverbed topography, water depth data, and an initial flow velocity distribution map. A drone is then deployed to release a swarm of buoys, which collect their motion trajectories and gyroscope data. This data is used to determine basic flow direction and current resistance data. The surface flow in the three-dimensional flow model is then refined or corrected based on these two data points. The underwater flow model within the three-dimensional flow model is then constructed based on the average cross-sectional flow velocity. Finally, a hydrological flow and velocity measurement report is generated based on the corrected three-dimensional flow model. This improves the accuracy of flow and velocity measurement, as well as its flexibility and applicability in complex waters. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a flowchart of the steps of a drone hydrological flow and velocity measurement method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] The following combination Figure 1 This application will be described in further detail, but the actual aspects of the present invention are not limited thereto.

[0062] The embodiment of the present application discloses a method for measuring hydrological flow and velocity using an unmanned aerial vehicle (UAV).

[0063] In this embodiment, a method for measuring hydrological flow and velocity using an unmanned aerial vehicle (UAV) is provided, the method comprising:

[0064] S100: Obtain a target flow measurement area of the water area to be inspected, call a drone to the target flow measurement area, select a dynamic anchor point in the target flow measurement area, and anchor the drone to hover according to the dynamic anchor point;

[0065] S200: The drone collects images of its surroundings, obtains obstacle data based on the images, and generates a motion route in real time based on the obstacle data and dynamic anchor points;

[0066] S300: The UAV collects video of the target flow measurement area, obtains visible light water surface video, infrared light water surface video, and polarized light water surface video, fuses them to generate a fused water surface video, and analyzes the fused water surface video to generate an initial flow velocity distribution map;

[0067] S400: The drone collects riverbed data and water surface fluctuation data in the target flow measurement area, obtains riverbed topography based on the riverbed data, and obtains water depth data and cross-section average flow velocity based on the riverbed topography and water surface fluctuation data;

[0068] S500: Based on the riverbed topography, water depth data and initial flow velocity distribution map, a three-dimensional water flow model is constructed, and a drone is called to release a group of buoys to collect the motion trajectory and gyroscope data of the buoy group. The three-dimensional water flow model is corrected according to the motion trajectory, gyroscope data and average flow velocity of the cross section, and the hydrological flow and velocity measurement results are output.

[0069] It should be noted that the above process is only the basic steps of this embodiment. During the specific implementation process, some steps can be appropriately added, reduced or modified without affecting the overall implementation effect.

[0070] The steps to select a dynamic anchor point in the target flow measurement area and hover the drone based on the dynamic anchor point are as follows:

[0071] Calling a UAV to collect images of the target flow measurement area to obtain a watershed image, extracting water flow information from the watershed image to obtain a water flow image, and particle-izing the water flow image to obtain water flow image particles;

[0072] Based on the water flow image particles, multiple particle features in the water flow image particles are extracted, and the obvious degree value of each particle feature is recorded;

[0073] The particle feature with the highest apparent value is selected as the target particle feature, and anchor points are selected for the water flow image particles according to the target particle feature to obtain dynamic anchor points.

[0074] Take photos of the dynamic anchor point at a preset time interval, record the displacement of the dynamic anchor point in the two photos, and calculate the hovering speed based on the displacement and time interval;

[0075] Adjust the propeller power distribution of the drone according to the hovering speed so that the drone remains relatively stationary with the dynamic anchor point.

[0076] In practice, a drone was used to capture images of a specific river flow measurement area, producing a watershed image. The image showed multiple vortices and waves on the water surface. Flow information was extracted from the image to produce a flow image, which showed a predominant southeasterly flow. The flow image was then particle-processed, displaying each water molecule as an independent particle. Particle features were extracted, including particle size, velocity, and color depth. The significance of each feature was calculated, with the velocity feature having the highest significance (95%). The particle region corresponding to this feature was selected as a dynamic anchor point, located in the center of the river. Images were taken of the dynamic anchor point every 5 seconds, revealing a 2-meter displacement between the two images. Based on the 2-meter displacement and the 5-second interval, the hovering speed was calculated to be 0.4 m / s. The power distribution of the drone's four propellers was adjusted: the power of the two front propellers was reduced to 70%, while the power of the two rear propellers was increased to 120%, ensuring that the drone remained stationary relative to the dynamic anchor point.

[0077] The steps for the drone to collect its own surrounding images, obtain obstacle data based on the surrounding images, and generate a motion route in real time based on the obstacle data and dynamic anchor points are as follows:

[0078] The drone collects images of its surrounding environment to obtain surrounding images, and then identifies the content of the surrounding images to obtain obstacle images;

[0079] Extract visual distance features and external shape features of the obstacle image, obtain obstacle distance based on visual distance features, obtain obstacle spatial position based on external shape features, and generate initial obstacle data by combining obstacle distance and obstacle spatial position;

[0080] Based on the initial obstacle data and external shape features, obstacles outside or behind the surrounding image are predicted to obtain predicted obstacle data, and the predicted obstacle data is added to the initial obstacle data to generate obstacle data;

[0081] Obtain the current UAV spatial position and the current anchor spatial position of the dynamic anchor point, and combine the obstacle data, UAV spatial position and anchor spatial position to generate the UAV's motion route in real time.

[0082] In this application, a drone captures images of the target flow measurement area of a river. The image shows a tree branch obstacle on the left and a rock on the right. The visual distance of the tree branch obstacle (calculated using the lens focal length) is 15 meters, and its shape is characterized by a long, thin strip. The visual distance of the rock is 20 meters, and its shape is characterized by an irregular block. Initial obstacle data is generated: the tree branch (15 meters, coordinates X=10, Y=5) and the rock (20 meters, coordinates X=30, Y=8). Based on the blocky shape of the rock, a hidden obstacle (such as a submerged rock) is predicted behind it. The predicted obstacle data is added with coordinates X=32, Y=10. The current position of the drone (X=50, Y=0, Z=100) and the dynamic anchor point position (X=55, Y=5, Z=0) are obtained. A movement route is generated: from the current position, heading southeast to avoid the tree branch and rock, with the path points (X=52, Y=3) → (X=58, Y=7) → the dynamic anchor point.

[0083] After the step of combining obstacle data, the spatial position of the drone, and the spatial position of the anchor point to generate the drone's motion path in real time, the following steps are also included:

[0084] When the UAV moves on the movement route, the surrounding images are updated in real time, and the obstacle image is updated in real time to obtain an updated obstacle image;

[0085] Obtaining updated obstacle data based on the updated obstacle image, and determining whether the movement route is blocked by an obstacle based on the updated obstacle data;

[0086] If it is determined that the movement route is blocked by an obstacle, the blocked route segment in the movement route is extracted;

[0087] Based on the route segment and the updated obstacle data, the route segment is reconstructed to obtain an updated route segment, and the updated route segment is added to the motion route.

[0088] In practice, for example, a drone flying along a river's target flow measurement area discovered a new obstacle while following its initially planned route. The drone continuously captured surrounding footage and discovered a floating tree trunk near the originally planned pathpoint (X=58, Y=7). The trunk was approximately 6 meters long, 15 meters from the drone, and moving downstream at a speed of 0.3 meters per second. Simultaneously, an infrared sensor detected an underwater reef to the right (X=60, Y=5), just 0.8 meters below the water surface. The system immediately updated the obstacle data, marking the coordinates of the tree trunk and reef. The drone's current location was (X=52, Y=3), with a dynamic anchor point at (X=55, Y=5). The system recalculated the route and discovered that the original path segment from (X=58, Y-7) to (X=55, Y=5) was blocked by the tree trunk. The system generated three new routes: the first route detoured to the right, passing through (X=60, Y=10) and (X=62, Y=6), increasing the flight distance by 13 meters; the second route climbed to the left to an altitude of 120 meters, but would consume 18% more battery power; the third route passed 2 meters below the tree trunk, but there was a risk of collision. Based on the remaining battery power of 75% and the urgency of the mission, the first detour was selected. The drone adjusted the propeller power, increasing the left propeller power to 130% and the right propeller power to 80%, and began to turn to the new path point. During the flight, the position of the tree trunk was updated every second. After the trunk drifted to X=59, Y=8, the system fine-tuned the path again, and finally arrived safely above the dynamic anchor point after 1 minute and 20 seconds.

[0089] Before the step of fusing the visible light water surface video, the infrared light water surface video, and the polarized light water surface video to generate a fused water surface video, the following steps are further included:

[0090] Extract the light reflection area in the visible light water surface video, extract the light amount in the light reflection area, and reduce the exposure of the light reflection area according to the light amount to obtain the target light reflection area;

[0091] Add the target light reflection area to the visible light water surface video to obtain the target visible light water surface video;

[0092] Based on the infrared water surface video, the temperature distribution area of the water surface and the infrared light scattering amount are extracted, and the infrared water surface video is clarified based on the temperature distribution area and the infrared light scattering amount to obtain the target infrared water surface video;

[0093] Based on the polarized water surface video, the polarization state transformation of the water surface is extracted, and polarization compensation parameters are generated according to the polarization state change. The polarized water surface video is compensated according to the polarization compensation parameters to obtain the target polarized water surface video.

[0094] In practice, for example, a target flow measurement area on a river was captured by a drone. In the visible light video, the area (X=15-20, Y=30-35) showed strong reflections, reaching 2500 lumens. The system reduced the exposure in this area, adjusting the light intensity to 800 lumens, making the ripples on the water surface clearly visible. Infrared video showed the water surface temperature at the center was 25°C, while the edge was 28°C. The infrared scattering in the center was 45% higher than at the edge. The system sharpened the center area, amplifying the temperature difference by three times to enhance the thermal convection. Polarized video detected that the polarization angle of the water surface fluctuated between 30° and 45°, generating a polarization compensation parameter of +5°. The processed fused water surface video showed that details as small as 0.5 mm in diameter bubbles in the original reflective area were discernible, and the outline clarity of rocks 1.2 meters underwater was improved by 85%. The system then analyzed the effects of water mist in the infrared video and found that mist blurred the water surface boundary by approximately 20%. Using temperature distribution data, the system identifies foggy areas (temperatures between 22 and 24°C) in the video and increases the contrast in these areas by 50%. The resulting fused water surface video reduces the error in surface velocity calculation from 12% to 3%.

[0095] The steps of fusing the visible light water surface video, the infrared light water surface video, and the polarized light water surface video to generate a fused water surface video and analyzing the fused water surface video to generate an initial flow velocity distribution map are as follows:

[0096] The target visible light water surface video, the target infrared light water surface video and the target polarized light water surface video are fused to generate a fused water surface video;

[0097] Based on the target infrared water surface video, the water mist and temperature effects appearing in the fused water surface video are eliminated;

[0098] Based on the target polarized light surface video, the clarity of the underwater image in the fused surface video is enhanced;

[0099] Perform water surface frame recognition on the fused water surface video to obtain water surface video frames, and record the visible light water surface difference and infrared light water surface difference between the water surface video frames;

[0100] The visible light water surface difference and the infrared light water surface difference are mutually verified and identified to obtain the initial water surface velocity distribution;

[0101] Perform underwater frame recognition on the fused surface video to obtain underwater video frames, and record the differences in water ripple shape and water ripple color between underwater video frames;

[0102] Based on the differences in water ripple shape and color, the underwater water flow variation characteristics are obtained, and the initial underwater flow velocity distribution is obtained according to the underwater water flow variation characteristics;

[0103] According to the initial water surface velocity distribution and the initial underwater velocity distribution, an initial velocity distribution map of the target flow measurement area is generated.

[0104] In practice, using a target flow measurement area on a river as an example, the fused three-band video was broken down into 30 frames per second for analysis. The surface frame showed that the wave height at position (X=40, Y=50) rose from 0.2 meters to 0.5 meters in 0.3 seconds, and the infrared difference showed that the temperature in this area dropped by 0.6°C per second. By comparing visible light and infrared data, the system confirmed the presence of a vortex and calculated the surface flow velocity to be 1.4 meters per second. In the underwater frame analysis, the water ripple at position (X=38, Y=55) changed from dark blue to light blue over 2 seconds, with a shape stretch of 18%. Based on this data, the system calculated a flow velocity of 0.9 meters per second at a depth of 0.8 meters underwater. The system then overlaid the surface and underwater data to generate an initial flow velocity distribution map. The image shows that the surface layer (0-0.5 meters) has a flow velocity of 1.2-1.6 m / s, the middle layer (0.5-1.5 meters) has a flow velocity of 0.8-1.2 m / s, and the bottom layer (1.5-3 meters) has a flow velocity of 0.5-0.8 m / s. The system also detected an abnormally low velocity zone (0.3 m / s) at (X=50, Y=60). Combined with polarized light data, this area was found to contain a buildup of aquatic plants and marked as a special observation point. After integrating all this data, the system generated an initial distribution map consisting of 12 velocity zones for subsequent 3D modeling.

[0105] Based on the target infrared water surface video, the steps for eliminating the water mist and temperature effects appearing in the fused water surface video are as follows:

[0106] Extract infrared frames from target infrared light water surface video, and obtain infrared temperature distribution and water surface infrared characteristics based on the infrared frames;

[0107] Locate the water surface boundary based on the water surface infrared characteristics to obtain water surface boundary data;

[0108] The infrared temperature distribution is segmented according to the water surface boundary data to obtain the water surface temperature distribution and the above-water temperature distribution respectively;

[0109] Based on the water surface temperature distribution, the water mist influence in the target flow measurement area is obtained; based on the water surface temperature distribution, the temperature influence in the target flow measurement area is obtained;

[0110] Water mist image compensation is generated based on the water mist influence, and temperature image compensation is generated based on the temperature influence. The water mist influence and temperature influence in the fused water surface video are eliminated according to the water mist image compensation and the temperature image compensation.

[0111] In practice, using a target flow measurement area on a river as an example, the system extracted a frame from infrared video and analyzed it to determine the water surface boundary within the (Y=50-70) coordinate zone. Based on this boundary, the system segmented the temperature distribution into a water surface area (25-28°C) and an airborne mist area (22-24°C). The system detected that the mist increased the blurriness of the water surface video by 25%. The system applied a contrast enhancement algorithm to this area, increasing the edge sharpening parameter by 60%. Furthermore, the water surface temperature distribution revealed an abnormally high temperature point (29°C) in the (X=45, Y=55) area. This was later determined to be a miscalculation caused by direct sunlight. The system generated a temperature compensation formula to correct the flow velocity data in this area to 85% of its original value. In the processed video, the flow velocity at (X=45, Y=55) was corrected from 1.5 m / s to 1.28 m / s. In addition, the system detected that the infrared speed measurement deviation in the edge area of the water surface (Y=50-55) was caused by low temperature, and the weight coefficient of this area was increased by 0.7 and recalculated, ultimately controlling the overall speed measurement error within ±2%.

[0112] The steps for obtaining riverbed topography based on riverbed data, and obtaining water depth data and cross-section average flow velocity based on riverbed topography and water surface fluctuation data are as follows:

[0113] Model the riverbed in the target flow measurement area based on the riverbed data to obtain the riverbed topography;

[0114] The average water surface spatial position of the target flow measurement area is obtained based on the water surface fluctuation data, and the water depth data is obtained based on the riverbed topography and the average water surface spatial position;

[0115] According to the riverbed topography, the cross-sectional space curve of the riverbed is obtained. Based on the cross-sectional space curve and water depth data, the cross-sectional image of the watershed in the target flow measurement area is constructed;

[0116] Based on the water surface fluctuation data and water depth data, multiple deep-layer flow velocity relationship diagrams are obtained on the basin cross-sectional image;

[0117] According to the initial velocity distribution map and the deep velocity relationship map, the cross-sectional velocity data on the basin cross-sectional image is constructed, and the cross-sectional average velocity is obtained based on the cross-sectional velocity data.

[0118] In practice, for example, a drone acquired riverbed data using sonar at a target flow measurement area on a particular river. The data showed the deepest point at (X=40, Y=60), with a depth of 3.2 meters. Analysis of water surface fluctuation data revealed an average wave height of 0.25 meters, allowing the system to calculate an average water surface elevation of 100.3 meters above sea level. Incorporating the riverbed topography, a water depth distribution map was generated: the area (X=30-50) had a water depth of 2.1-3.2 meters, and the area (X=50-70) had a water depth of 1.5-2.0 meters. The system then plotted a riverbed cross-section curve based on the parabolic formula y=0.02x², dividing the flow measurement area into six sections. Each section was then layered at 0.5-meter intervals, generating a total of 12 velocity measurement layers. Combining the water surface fluctuation data with the initial velocity distribution map yielded a calculated average velocity of 0.87 meters per second for the section. For example, at the third section (X=45-50), the surface velocity is 1.2 m / s, the middle layer is 0.9 m / s, and the bottom layer is 0.6 m / s. The weighted average velocity for this section is 0.92 m / s. After integrating the data from all sections, the system outputs a basin-wide average velocity of 0.84 m / s and marks (X=55, Y=58) as a velocity anomaly (1.8 m / s), indicating the presence of a possible undercurrent.

[0119] The steps for calling a drone to release a group of buoys, collecting the motion trajectory and gyroscope data of the buoy group, and correcting the three-dimensional water flow model based on the motion trajectory, gyroscope data, and average flow velocity of the cross section are as follows:

[0120] Call the drone to release the buoy group, which moves on the water surface of the target flow measurement area, and collects the movement trajectory and gyroscope data of the buoy group;

[0121] Based on the motion trajectory, the basic water flow direction of the target flow measurement area is obtained, and the water flow confrontation data of the target flow measurement area is obtained according to the gyroscope data;

[0122] Construct an underwater flow model in a three-dimensional flow model based on the average flow velocity of the cross section;

[0123] Based on the basic water flow direction and water flow confrontation data, the surface water flow in the three-dimensional water flow model is refined and corrected.

[0124] In practice, using a target flow measurement area on a river as an example, a drone released 20 buoys, 15 of which drifted southeast, while five were trapped by eddies. The buoy trajectories indicated a main current direction of 12 degrees south-east, with an average drift velocity of 1.1 m / s. Gyroscope data indicated that the buoys experienced the greatest lateral force (6.2 N) at the angle (X=50, Y=60), causing a 20-degree deviation in their trajectory. The system input this data into a three-dimensional flow model, first correcting the surface flow direction to 12 degrees south-east and adjusting the velocity in that area from the model's predicted 1.3 m / s to 1.15 m / s. Next, based on the average cross-sectional velocity of 0.84 m / s, velocity gradients were constructed for each underwater layer: 0.9 m / s for the 0-1 m layer, 0.7 m / s for the 1-2 m layer, and 0.5 m / s for the 2-3 m layer. Finally, for the abnormally high-speed area at (X=55, Y=58), the model added a local turbulence coefficient, correcting the velocity at that point to 1.6 meters per second. The revised three-dimensional flow model showed a maximum flow rate of 920 cubic meters per second for the entire basin, an 8% increase over the initial model. All data was wirelessly transmitted to the control center, generating a final flow measurement report that included a flow velocity thermogram and hazardous area markers.

[0125] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for measuring hydrological flow and velocity using an unmanned aerial vehicle, characterized in that: The following steps are involved: Obtaining a target flow measurement area of the water area to be inspected, calling a drone to go to the target flow measurement area, selecting a dynamic anchor point in the target flow measurement area, and anchoring and hovering the drone based on the dynamic anchor point; The drone collects images of its surroundings, obtains obstacle data based on the images, and generates a motion route in real time based on the obstacle data and the dynamic anchor points; The drone collects video of the target flow measurement area, obtains visible light water surface video, infrared light water surface video and polarized light water surface video, fuses them to generate a fused water surface video, and analyzes the fused water surface video to generate an initial flow velocity distribution map; The steps of fusing the visible light water surface video, the infrared light water surface video, and the polarized light water surface video to generate a fused water surface video, and analyzing the fused water surface video to generate an initial flow velocity distribution map are specifically as follows: fusing the visible light water surface video, the infrared light water surface video, and the polarized light water surface video to generate a fused water surface video; Eliminating the effects of water mist and temperature that appear in the fused water surface video based on the infrared water surface video; Based on the polarized water surface video, enhancing the content clarity of the underwater image in the fused water surface video; Performing water surface frame recognition on the fused water surface video to obtain water surface video frames, and recording visible light water surface differences and infrared light water surface differences between the water surface video frames; Mutual verification and identification of the visible light water surface difference and the infrared light water surface difference to obtain an initial water surface velocity distribution; Performing underwater frame recognition on the fused surface video to obtain underwater video frames, and recording the differences in water ripple shapes and water ripple colors between the underwater video frames; Based on the differences in the shapes and colors of the water ripples, underwater water flow variation characteristics are obtained, and according to the underwater water flow variation characteristics, an initial underwater flow velocity distribution is obtained; generating an initial flow velocity distribution map of a target flow measurement area according to the initial water surface flow velocity distribution and the initial underwater flow velocity distribution; The drone collects riverbed data and water surface fluctuation data of the target flow measurement area, obtains riverbed topography based on the riverbed data, and obtains water depth data and cross-sectional average flow velocity based on the riverbed topography and the water surface fluctuation data; Based on the riverbed topography, the water depth data and the initial flow velocity distribution map, a three-dimensional water flow model is constructed, and a drone is called to release a group of buoys to collect the motion trajectory and gyroscope data of the buoy group. The three-dimensional water flow model is corrected according to the motion trajectory, the gyroscope data and the average flow velocity of the section, and the hydrological flow and velocity measurement results are output.

2. The method for measuring flow and velocity using an unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The steps of selecting a dynamic anchor point in the target flow measurement area and anchoring the UAV according to the dynamic anchor point are specifically as follows: Invoking the drone to collect images of the target flow measurement area to obtain a watershed image, extracting water flow information from the watershed image to obtain a water flow image, and particle-izing the water flow image to obtain water flow image particles; Extracting multiple particle features from the water flow image particles based on the water flow image particles, and recording a conspicuousness value of each particle feature; Selecting the particle feature with the highest conspicuity value as the target particle feature, and performing anchor point selection on the water flow image particles according to the target particle feature to obtain a dynamic anchor point; Taking photos of the dynamic anchor point at a preset time interval, recording the displacement of the dynamic anchor point in the two photos, and obtaining the hovering speed according to the displacement and the time interval; The propeller power distribution of the UAV is adjusted according to the hovering speed so that the UAV and the dynamic anchor point remain relatively stationary.

3. The method for measuring flow and velocity using an unmanned aerial vehicle according to claim 2, characterized in that: The steps of collecting the drone's own surrounding images, obtaining obstacle data based on the surrounding images, and generating a motion route in real time based on the obstacle data and the dynamic anchor points are specifically as follows: The drone collects images of its surrounding environment to obtain a surrounding image, and performs content identification on the surrounding image to obtain an obstacle image; Extracting visual distance features and external shape features of the obstacle image, obtaining the obstacle distance based on the visual distance features, obtaining the obstacle spatial position based on the external shape features, and generating initial obstacle data by combining the obstacle distance and the obstacle spatial position; Predicting obstacles outside or behind the surrounding image based on the initial obstacle data and the external shape features to obtain predicted obstacle data, and adding the predicted obstacle data to the initial obstacle data to generate obstacle data; The current drone spatial position of the drone and the current anchor spatial position of the dynamic anchor point are obtained, and the movement path of the drone is generated in real time by combining the obstacle data, the drone spatial position and the anchor spatial position.

4. The method for measuring flow and velocity using an unmanned aerial vehicle according to claim 3, wherein: After the step of generating the movement path of the drone in real time by combining the obstacle data, the spatial position of the drone, and the spatial position of the anchor point, the method further includes: When the UAV moves along the movement route, the surrounding image is updated in real time, and the obstacle image is updated in real time to obtain an updated obstacle image; obtaining updated obstacle data based on the updated obstacle image, and determining whether the movement route is blocked by the obstacle based on the updated obstacle data; If it is determined that the movement route is blocked by the obstacle, extracting the blocked route segment from the movement route; Based on the route segment and the updated obstacle data, the route segment is reconstructed to obtain an updated route segment, and the updated route segment is added to the motion route.

5. The method for measuring hydrological flow and velocity using an unmanned aerial vehicle according to claim 1, characterized in that: Before the step of fusing the visible light water surface video, the infrared light water surface video, and the polarized light water surface video to generate a fused water surface video, the following steps are further included: extracting a light reflection area from the visible light water surface video, extracting a light quantity from the light reflection area, and performing exposure reduction processing on the light reflection area according to the light quantity to obtain a target light reflection area; Adding the target light reflection area to the visible light water surface video to obtain a target visible light water surface video; Extracting the temperature distribution area and infrared light scattering amount of the water surface based on the infrared light water surface video, and performing clarity processing on the infrared light water surface video based on the temperature distribution area and the infrared light scattering amount to obtain a target infrared light water surface video; Based on the polarized water surface video, the polarization state change of the water surface is extracted, and polarization compensation parameters are generated according to the polarization state change. The polarized water surface video is parameter compensated according to the polarization compensation parameters to obtain the target polarized water surface video.

6. The method for measuring hydrological flow and velocity using an unmanned aerial vehicle according to claim 5, characterized in that: The step of eliminating the water mist effect and temperature effect appearing in the fused water surface video based on the target infrared light water surface video is specifically as follows: Extracting infrared frames from the target infrared light water surface video, and obtaining infrared temperature distribution and water surface infrared characteristics based on the infrared frames; Locating the water surface boundary based on the water surface infrared characteristics to obtain water surface boundary data; Performing layout segmentation on the infrared temperature distribution according to the water surface boundary data to obtain a water surface temperature distribution and an above-water temperature distribution; Based on the water surface temperature distribution, the water mist influence in the target flow measurement area is obtained; based on the water surface temperature distribution, the temperature influence in the target flow measurement area is obtained; A water mist image compensation is generated based on the water mist influence, and a temperature image compensation is generated based on the temperature influence. The water mist influence and the temperature influence in the fused water surface video are eliminated according to the water mist image compensation and the temperature image compensation.

7. The method for measuring hydrological flow and velocity using an unmanned aerial vehicle according to claim 6, characterized in that: The steps of obtaining riverbed topography according to the riverbed data, and obtaining water depth data and cross-sectional average flow velocity according to the riverbed topography and the water surface fluctuation data are specifically as follows: Modeling the riverbed of the target flow measurement area according to the riverbed data to obtain the riverbed topography; Obtaining an average water surface spatial position of the target flow measurement area based on the water surface fluctuation data, and obtaining water depth data based on the riverbed topography and the average water surface spatial position; According to the riverbed topography, a cross-sectional space curve of the riverbed is obtained, and based on the cross-sectional space curve and the water depth data, a watershed cross-sectional image of the target flow measurement area is constructed; Based on the water surface fluctuation data and the water depth data, a plurality of deep layer flow velocity relationship diagrams are obtained by dividing the watershed cross-sectional image; According to the initial flow velocity distribution diagram and the deep flow velocity relationship diagram, cross-sectional flow velocity data on the watershed cross-sectional image is constructed, and the cross-sectional average flow velocity is obtained according to the cross-sectional flow velocity data.

8. The method for measuring hydrological flow and velocity using an unmanned aerial vehicle according to claim 7, characterized in that: The steps of calling a drone to release a group of buoys, collecting motion trajectories and gyroscope data of the buoy group, and correcting the three-dimensional water flow model based on the motion trajectories, the gyroscope data, and the average flow velocity of the cross section are specifically as follows: Invoking a drone to release a group of buoys, the group of buoys moving on the water surface of the target flow measurement area, and collecting the movement trajectory and gyroscope data of the group of buoys; Based on the motion trajectory, a basic water flow direction of the target flow measurement area is obtained, and based on the gyroscope data, water flow resistance data of the target flow measurement area is obtained; Constructing an underwater water flow model in the three-dimensional water flow model based on the average flow velocity of the cross section; Based on the basic water flow direction and the water flow confrontation data, the surface water flow in the three-dimensional water flow model is refined and corrected.

Images