Method for unmanned aerial vehicle intelligent identification of river and lake water area changes and river and lake abnormal problem early warning

Abstract

The application discloses a method for intelligent identification of changes in river and lake water areas by a UAV and early warning of abnormal problems in rivers and lakes, comprising the following steps: constructing a digital twin bottom plate data of rivers and lakes; manually editing a basic flight route of the rivers and lakes, intelligently generating a shooting flight route based on the basic flight route and the digital twin bottom plate data of the rivers and lakes, splicing the basic flight route and the shooting flight route, and obtaining a dynamic flight route of the UAV; performing dynamic flight route flight according to the dynamic flight route, collecting data information during flight, and returning the data information to a UAV base station through a network and transmitting the data information to an edge computing device; generating river and lake orthographic images and river and lake spatial terrain data by combining the data information collected by the UAV with water area management lines; and identifying and comparing the orthographic images before and after the two times by using a water area surface identification model and an image difference method, so as to obtain a water area surface change area and abnormal problems in the rivers and lakes. The method solves the problems of low identification accuracy, low image spatial resolution and unintelligent flight route scheduling in the existing UAV patrol of the rivers and lakes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image recognition, and in particular to a method for intelligent identification of river and lake water area changes and early warning of river and lake abnormal problems by a UAV. BACKGROUND

[0002] The river and lake system is a key carrier for ecological environment and water resource management, and its health status is directly related to people's livelihood safety and sustainable development. At present, the management of rivers and lakes in China mainly manages two aspects of problems, i.e., water resource occupation and water environment pollution. The water resource occupation mainly considers the carrying capacity and flood discharge capacity of the river and lake, and timely discovers the occupation problems of the river and lake to take necessary measures to ensure that the river and lake has normal water storage capacity and flood discharge function. The water environment pollution mainly considers the water quality problem of the river and lake. Due to the illegal discharge of sewage, blue-green algae bloom and human activities, the water quality becomes poor. Timely monitoring and taking corresponding measures can ensure normal water quality. The traditional supervision methods mainly rely on the following two methods:

[0003] The first method relies on on-site investigation by patrol personnel, which has inherent defects such as low efficiency, limited coverage, and untimely discovery of problems. For example, in the river and lake patrol, the artificial method can only check less than 20 kilometers per day. For water environment pollution problems, there is no data support, and it is impossible to comprehensively analyze the pollution source to discover the pollution source in time.

[0004] The second method realizes fixed-point monitoring by deploying cameras at dam, bridge and other positions, which can partially replace manpower, but has problems such as many blind spots, high deployment cost and poor flexibility. Taking the Yellow River Basin as an example, in order to realize continuous shoreline coverage, equipment needs to be densely deployed, the cost of a single kilometer is more than 200,000 yuan, and it cannot adapt to the mobile monitoring demand of emergency events.

[0005] In recent years, the UAV has gradually been applied to the river and lake supervision field due to its advantages such as flexible maneuvering, high-resolution imaging and multi-sensor integration, but there are still problems such as low identification accuracy, low image spatial resolution and unintelligent flight route scheduling in the existing river and lake UAV patrol. SUMMARY

[0006] Therefore, the present application provides a method for intelligent identification of river and lake water area changes and early warning of river and lake abnormal problems by a UAV to solve the problems of low identification accuracy, low image spatial resolution and unintelligent flight route scheduling in the existing river and lake UAV patrol.

[0007] The present application provides a method for intelligent identification of river and lake water area changes and early warning of river and lake abnormal problems by a UAV, which comprises the following steps:

[0008] Collecting hydrographic data, environmental quality data and engineering operation data by an edge computing device to construct a river and lake digital twin bottom plate data;

[0009] The basic route of the artificial edited river and lake is set, a shooting route is intelligently generated based on the basic route and the river and lake digital twin bottom plate data, a dynamic route generation rule is set, the basic route and the shooting route are spliced, and a dynamic route of the unmanned aerial vehicle is obtained;

[0010] Dynamic route flight is performed according to the dynamic route, data information during flight is collected, and the data information is returned to the unmanned aerial vehicle base station through a network and transmitted to an edge computing device;

[0011] The data information collected by the unmanned aerial vehicle is combined with the water area management line to generate river and lake orthographic images and river and lake spatial terrain data, a water area surface recognition model and an image difference method are used to recognize and compare the front and rear two orthographic images, and then the water area surface change area and the early warning result of the river and lake abnormal problem are obtained.

[0012] The hydrological geographical data includes river and lake water area management lines, river and lake underwater terrain data, water level monitoring data, the environmental quality data includes water quality monitoring data and meteorological data, and the engineering operation data includes ground camera monitoring positions, water quality monitoring equipment positions, water level monitoring equipment positions, sewage outlet position data and gate station opening scheduling data.

[0013] The flight time range is calculated according to the basic route, and the basic route determines the path, flight height, shooting angle, shooting focal length and shooting mode of the unmanned aerial vehicle flight.

[0014] The water quality monitoring equipment position, the sewage outlet position and the ground camera monitoring position are set as the flight focus points, and two shooting modes of fixed position multi-angle shooting and fixed area scanning shooting are set for the flight focus points, wherein the sewage outlet position and the water quality monitoring equipment position are shot by fixed position multi-angle shooting, and the ground camera monitoring position is shot by fixed area scanning shooting.

[0015] The fixed position multi-angle shooting focuses on and enlarges the shooting of the fixed position by three different angles to obtain images of different angles;

[0016] The fixed area scanning shooting draws a square area with the fixed position where the abnormal early warning occurs as a center point, focuses on and enlarges the shooting of the square area by scanning in steps.

[0017] The dynamic route generation rule includes,

[0018] If the water quality level monitored by the water quality monitoring equipment data is abnormal, the fixed position multi-angle shooting is performed on the sewage outlet position and the water quality monitoring equipment position; if the ground camera monitoring early warning data is found, the fixed area scanning shooting is performed on the ground camera monitoring position.

[0019] The intelligent generation of the dynamic route is based on an existing dynamic route generation rule, combined with river and lake spatial terrain data and flight attention point positions, and an intelligent space calculation of the route is performed to obtain an intelligent shooting point, intelligently generate a shooting route, splice a basic route and the shooting route, and generate the dynamic route.

[0020] The intelligent shooting point is obtained by,

[0021] The shortest distance position point of the monitoring point from the route is the intelligent shooting point.

[0022] The intelligent generation of the dynamic route includes,

[0023] The dynamic route generation rule is set, a shooting route is generated for the intelligent shooting point, if the shooting mode of the intelligent shooting point is fixed position multi-angle shooting, an angle A formed by a connecting line of a UAV shooting angle and a fixed position and a perpendicular line of a UAV plumb direction is greater than 30 degrees, three UAV shooting points need to be automatically set for the shooting route, and a shooting route is formed through the three UAV shooting points; wherein, the three UAV shooting points need to satisfy the following two conditions: an interval distance B between the three UAV shooting points is greater than 20 meters; a non-obstructing object in a line-of-sight corridor buffer C formed by a connecting line of a shooting point and a fixed position is calculated in combination with river and lake spatial terrain data, to ensure that the fixed position can be shot, and the line-of-sight corridor buffer C is a conical line-of-sight corridor buffer C with a half vertex angle of 5 degrees, with the UAV shooting point as a vertex and the fixed position as an axis end point.

[0024] The water area surface change region is obtained by,

[0025] River and lake orthographic images and river and lake spatial terrain data are generated by combining data collected by a UAV with a water area management line, a water area surface difference value of an entire river channel is extracted by using a water area surface recognition model to recognize and compare the front and rear two orthographic images; an elevation terrain surface is obtained by elevation calculation of river and lake underwater terrain data and water level monitoring data; and the water area surface change region is obtained by comparison and calculation of the water area surface difference value and the elevation terrain surface.

[0026] The water area surface difference value is calculated by a water area surface recognition model to recognize river and lake orthographic images obtained by two continuous flights, to obtain a water area surface obtained by the first flight through recognition, and a water area surface obtained by the second flight through recognition, to obtain a water area surface difference value W through difference calculation.

[0027] ;

[0028] The water area surface difference W is subtracted from the elevation terrain surface T, in the method of corrosion first and expansion later, the false change water area surface is filtered out, the accurate water area surface change area is obtained, and the area size of the water area surface change area is calculated through the conversion of the spatial resolution.

[0029] The acquisition of the early warning result of the river and lake abnormal problem comprises,

[0030] The image difference method is used to compare the front and rear orthographic images, the change object space range in the water area management line is extracted, the pixel position of the original unmanned aerial vehicle image and the change object in the image is obtained through reverse calculation of the change object space range, and thus the early warning result of the river and lake abnormal problem is obtained.

[0031] The generation of the orthographic image comprises,

[0032] The water area mask texture fusion method is used to generate the orthographic image, a model for recognizing the water area surface is trained by using an image segmentation model segformer, the water area surface is segmented from all original unmanned aerial vehicle images, a black and white binary mask file Mask is automatically generated, an initial orthographic image is generated by using an orthographic image generation tool opendronemap, the corresponding camera rotation angle and unmanned aerial vehicle position information during shooting of the original unmanned aerial vehicle image are combined, a water area mask is generated through aerial triangulation calculation, the water area mask is imported into the orthographic image, and finally, one water area surface texture is selected in the spatial overlapping area to replace the water area problem image in the orthographic image through a texture optimization method, wherein the conditions for texture optimization are that the angle is the most positive, the reflection is the least, the quality is the best, and the BRISQUE score is the smallest.

[0033] Beneficial effects: the water area surface is recognized by comparing the front and rear orthographic images, the water area surface change area is accurately recognized in combination with the river and lake underwater terrain data, the water area occupation misrecognition problem is solved, the accuracy is improved, and quantitative monitoring of the river water area surface change is realized, which is more intuitive;

[0034] Based on the river and lake digital twin bottom plate data, in combination with water quality and camera early warning information, the unmanned aerial vehicle is intelligently scheduled in a targeted manner, the fixed position multi-angle shooting and fixed area scanning shooting are used to solve the problem that the sun glare and water surface reflection cause the image with obvious features to be unable to be collected, the reliability of image quality collection is improved, the image spatial resolution of the key supervision area is improved, and the river patrol efficiency is not reduced.

[0035] The river and lake problem is recognized in the manner of orthographic image comparison, without considering the type of the compared transformation object, the spatial position of the original unmanned aerial vehicle image and the transformation object is obtained through reverse calculation, early warning recognition is realized, and the generalization and usability of recognition are improved.

[0036] The water area mask texture fusion method can significantly improve the generation quality of the water area surface in the orthographic image.

[0037] It should be understood that the matters described in this section are not intended to identify key or important features of the embodiments of the present application, nor are they used to limit the scope of the present application. Other features of the present application will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0038] The accompanying drawings are used to better understand the present application, and do not constitute a limitation of the present application. Among them:

[0039] Figure 1 is a flowchart provided according to the present application;

[0040] Figure 2 is a schematic diagram of an intelligent shooting point provided according to the present application;

[0041] Figure 3 is a schematic diagram of multi-angle shooting at a fixed position provided according to the present application;

[0042] Figure 4 is a schematic diagram of fixed area scanning shooting provided according to the present application

[0043] Figure 5 is a schematic diagram of equipment provided according to the present application;

[0044] Figure 6 is a schematic diagram of a water area surface change area calculation process provided according to the present application;

[0045] Figure 7 is a schematic diagram of a conventional orthographic image provided according to the present application;

[0046] Figure 8 is a schematic diagram of an orthographic image provided according to the water area mask texture fusion method according to the present application;

[0047] Figure 9 is a schematic diagram of a change object mask in an orthographic image provided according to the present application;

[0048] Figure 10 is a schematic diagram of a change object mask in an original unmanned aerial vehicle image provided according to the present application;

[0049] Figure 11 is a schematic diagram of a data flow provided according to the present application. DETAILED DESCRIPTION

[0050] Exemplary embodiments of the present application are described herein below with reference to the accompanying drawings, in which various specific details are set forth to assist in a thorough understanding of these embodiments. It should be understood, however, that various changes and modifications can be made to the embodiments described herein, without departing from the scope and spirit of the application. Also, in the interest of clarity, not all features of an actual implementation are described in this description. It will be apparent to one skilled in the art, however, that various additional implementations can be employed apart from the examples noted herein. Additionally, one skilled in the art will recognize that the features described herein can apply to any social networking system.

[0051] As shown in Figure 1 The present application provides a method for intelligent identification of river and lake water area changes and early warning of river and lake abnormal problems by a UAV, comprising:

[0052] S1: Collecting hydrographic data, environmental quality data and engineering operation data by an edge computing device to construct a river and lake digital twin bottom plate data. It should be noted that:

[0053] The hydrographic data includes river and lake water area management lines, river and lake underwater topographic data, water level monitoring data, the environmental quality data includes water quality monitoring data and meteorological data, and the engineering operation data includes ground camera monitoring positions, water quality monitoring equipment positions, water level monitoring equipment positions, sewage outlet position data and gate station opening scheduling data.

[0054] The river and lake water area management lines are vector data of the scope of river and lake management.

[0055] The river and lake underwater topographic data is the river and lake underwater topographic data detected by a multi-beam probe of an unmanned ship.

[0056] The water quality monitoring equipment position and data are spatial position data of water quality monitoring equipment and water quality parameter data collected at regular intervals for the current monitoring section to monitor water quality changes.

[0057] The water level monitoring equipment position and data are spatial position data of water level monitoring equipment and current river and lake water level elevation data collected at regular intervals.

[0058] The sewage outlet position data are spatial position data of a sewage outlet.

[0059] The ground camera monitoring position is spatial position data of a ground camera and video recognition early warning result information, and the recognition content includes floating objects, piles, personnel intrusion and other river and lake supervision related recognitions.

[0060] The gate station opening scheduling data is time data of gate station opening scheduling. When the gate station is opened for scheduling, the water level of the river will change in a short time due to the water level difference before and after the gate.

[0061] The meteorological data are short-term weather forecast data connected to a third-party cloud platform.

[0062] S2: manually edit the basic route of the river and lake, intelligently generate a shooting route based on the basic route and the river and lake digital twin bottom plate data, set a dynamic route generation rule, splice the basic route and the shooting route, and obtain a dynamic route of the unmanned aerial vehicle. It should be noted that:

[0063] The flight time range is calculated according to the basic route, and the basic route determines the path, flight height, shooting angle, shooting focal length and shooting mode of the unmanned aerial vehicle.

[0064] The flight time range is to fly according to the route once a day as required, and the time range is from 8am to 5pm, ensuring good visibility, combining meteorological data and lock station scheduling data to generate automatic flight time of the unmanned aerial vehicle on the same day. The automatic flight time range meets the following two conditions:

[0065] 1) The meteorological data in the automatic flight time range meets the flight meteorological requirements.

[0066] 2) Since the lock station will cause abnormal water level for a short time when it is opened and dispatched, there is no lock station opening and dispatching within 1 hour before and after the automatic flight time range.

[0067] Set the water quality monitoring device position, sewage outlet position and ground camera monitoring position as the flight focus point, set the fixed position multi-angle shooting and fixed area scanning shooting as the shooting mode of the flight focus point, and the sewage outlet position and water quality monitoring device position adopt fixed position multi-angle shooting, and the ground camera monitoring position adopts fixed area scanning shooting.

[0068] The flight focus point and the shooting mode set the water quality monitoring device position data, the sewage outlet position data and the ground camera monitoring position data as the flight focus point.

[0069] The fixed position multi-angle shooting focuses on the fixed position and enlarges the shooting to obtain images at different angles to solve the image quality problem caused by solar glare.

[0070] The fixed area scanning shooting draws a square area with the fixed position where the abnormal alarm occurs as the center point, focuses on the square area and enlarges the shooting to scan the area.

[0071] S3: According to the dynamic route, perform dynamic route flight, collect data information during flight, and return the data information to the unmanned aerial vehicle base station through the network and transmit it to the edge computing device. It should be noted that:

[0072] The dynamic route generation rule includes,

[0073] If the water quality level monitored by the water quality monitoring device is abnormal, the position of the multi-angle shooting of the fixed position of the sewage outlet and the position of the water quality monitoring device are fixed; if the ground camera monitoring early warning data is found, the fixed area scanning shooting of the ground camera monitoring position is carried out.

[0074] The intelligent generation of the dynamic flight path is based on the existing dynamic flight path generation rule, combined with the river and lake space terrain data and the flight focus point position, and the intelligent shooting point is obtained through the flight path intelligent space calculation, the shooting flight path is intelligently generated, the basic flight path and the shooting flight path are spliced, and the dynamic flight path is generated;

[0075] The intelligent shooting point is obtained by,

[0076] The shortest distance position point of the monitoring point from the flight path is the intelligent shooting point, such as Figure 2 As shown in the figure, 1, 2 and 3 represent three different flight focus points, and 4 represents an intelligent shooting point.

[0077] The intelligent generation of the dynamic flight path includes,

[0078] The dynamic flight path generation rule is set, the shooting flight path is generated for the intelligent shooting point, if the shooting mode of the intelligent shooting point is multi-angle shooting of fixed position, the angle A formed by the connecting line of the shooting angle of the unmanned aerial vehicle and the fixed position and the perpendicular line of the plumb direction of the unmanned aerial vehicle is greater than 30 degrees, three unmanned aerial vehicle shooting points need to be automatically set for the shooting flight path, and the shooting flight path is formed through the three unmanned aerial vehicle shooting points; wherein, the three unmanned aerial vehicle shooting points need to meet the following two conditions: the interval distance B between the three unmanned aerial vehicle shooting points is greater than 20 meters; the unobstructed object in the visual corridor buffer C formed by the connecting line of the shooting point and the fixed position is calculated in combination with the river and lake space terrain data, so as to ensure that the fixed position can be shot, and the visual corridor buffer C is a conical visual corridor buffer C with a half vertex angle of 5 degrees, with the unmanned aerial vehicle shooting point as the vertex and the fixed position as the axis end point (as Figure 3 As shown in the figure, the number 17 is an intelligent shooting point).

[0079] The intelligent shooting point is taken as the starting point and the ending point, the intelligent shooting flight path is formed in combination with the three unmanned aerial vehicle shooting points, the focal length of the intelligent shooting flight path is set to the maximum magnification, and the shooting angle is determined according to the three unmanned aerial vehicle shooting points and the fixed position to be shot.

[0080] If the shooting mode of the intelligent shooting point is fixed area scanning shooting, a square region of D is generated with the intelligent shooting point as the center point, the maximum magnification is used for oblique shooting, the grid coverage shooting route, the shooting angle and the focal length information are obtained through the river and lake space terrain data, the camera shooting angle, the maximum magnification shooting focal length and the flight height information calculation, and the intelligent shooting flight path is generated. The intelligent shooting flight path is also taken as the starting point and the ending point, asFigure 4 As shown, 5 is an intelligent shooting point, and 6 is a UAV.

[0081] The grid coverage shooting route is to perform full coverage shooting on a square region, the shooting method adopts a grid shooting manner, an S-shaped shooting is performed, and an intelligent shooting route is generated.

[0082] The route splicing is to obtain a shooting route according to an intelligent shooting point on the basis of a basic route, the intelligent shooting point position of the basic route is interrupted, then the corresponding shooting route is spliced, a complete dynamic route is generated, and is sent to a UAV airport, and is updated once a day.

[0083] S4: The data information collected by the UAV is combined with the water area management line to generate river and lake orthographic images and river and lake spatial terrain data, a water area surface recognition model and an image difference method are used to recognize and compare the front and rear two orthographic images, and then the water area surface change area and the early warning result of the river and lake abnormal problem are obtained. It should be noted that:

[0084] The water area surface change area includes,

[0085] The data collected by the UAV is combined with the water area management line to generate river and lake orthographic images and river and lake spatial terrain data, a water area surface recognition model is used to recognize and compare the front and rear two orthographic images, and the water area surface difference value of the whole river channel is extracted; the elevation terrain surface is obtained through the river and lake underwater terrain data and the water level monitoring data elevation calculation; the water area surface change area is obtained by comparing and calculating the water area surface difference value and the elevation terrain surface.

[0086] Dynamic route flight data collection: dynamic route flight is performed according to the dynamic route, video data, UAV position information, camera rotation angle, camera focal length, and focusing magnification shooting data during flight are collected, and are returned to the UAV base station through the network, and are transmitted to the edge computing device. Equipment such as Figure 5 As shown, 7 is a UAV, 8 is a UAV nest, 9 is a water level monitoring device, 10 is a ground camera monitoring device, 11 is a water quality monitoring device, 12 is a switch, 13 is an edge computing device, 14 is a 4G router, 15 is a network gateway, and 16 is a third-party cloud platform.

[0087] The generation of the orthographic image includes,

[0088] The water area mask texture fusion method is used to generate the orthographic image. First, a water area surface recognition model is trained using the image segmentation model segformer to segment the water area from all original unmanned aerial vehicle images, extract the water area surface, and automatically generate a black and white binary mask file Mask. Then, the initial orthographic image is generated from the original unmanned aerial vehicle image using the orthographic image generation tool opendronemap. The corresponding camera rotation angle and unmanned aerial vehicle position information during image shooting are combined to calculate the water area mask through aerial triangulation. The water area mask is imported into the orthographic image. Finally, a water area surface texture is selected in the spatial overlap area to replace the water area problem image in the orthographic image through texture optimization, as shown in Figure 7 and Figure 8 The texture optimization conditions are the most positive angle, the least reflection, the best quality, and the lowest BRISQUE score.

[0089] The reverse calculation is achieved by matching the original unmanned aerial vehicle image and calculating the pixel position of the changed object in the original unmanned aerial vehicle image within the changed object spatial range in the orthographic image. As shown in Figure 9 , 10

[0090] The matching of the original unmanned aerial vehicle image is achieved by obtaining the minimum circumscribed rectangle Z from the changed object spatial range, calculating the ground coverage range for each original unmanned aerial vehicle image, and constructing the image space index U. All original unmanned aerial vehicle images containing the minimum circumscribed rectangle Z are searched, and the original unmanned aerial vehicle image with the highest spatial resolution of the minimum circumscribed rectangle Z is extracted to obtain the original unmanned aerial vehicle image.

[0091] The calculation of the pixel position of the changed object in the original unmanned aerial vehicle image is achieved by using the camera 3D coordinate system conversion to obtain the camera 3D coordinates of the changed object based on the original unmanned aerial vehicle image and the corresponding camera rotation angle and unmanned aerial vehicle position information during image shooting. The image pixel coordinates are obtained by calculating the image plane 2D coordinates based on the camera focal length, and the image pixel coordinates are obtained by converting the image plane coordinates.

[0092] The orthographic image and the river and lake spatial terrain data are generated by inputting video data and flight data using opendronemap, and then the generated data is cropped using the water area management line to only retain the data within the water area management line.

[0093] The water area surface recognition model is trained by manually annotating a large number of orthographic images of water area surfaces.

[0094] The water area surface difference calculation is achieved by identifying the river and lake orthographic images obtained by the water area surface recognition model in the first flight and the second flight. ​and the second flight through the obtained water area surface , the water area surface difference W is obtained by difference calculation,

[0095] ;

[0096] The water area surface difference W is subtracted from the elevation terrain surface T, in the method of first corrosion and then expansion, the false changed water area surface caused by errors is filtered out, the accurate water area surface change area is obtained, and the size of the water area surface change area is calculated through the conversion of the spatial resolution.

[0097] The data collected by the unmanned aerial vehicle is combined with the water area management line to generate the river and lake orthographic image and the river and lake spatial terrain data, the water area surface recognition model is used to identify and compare the front and rear two orthographic images, and the water area surface difference of the whole river channel is extracted; the elevation terrain surface is obtained through the elevation calculation of the river and lake underwater terrain data and the water level monitoring data; the water area surface difference and the elevation terrain surface are compared and calculated, the water area surface change area is obtained, and the calculation method of the water area surface change area is that the original water area surface change area K is obtained by subtracting the water level elevation terrain surface T from the water area surface difference W, then the false changed water area surface caused by errors is filtered out by using the method of first corrosion and then expansion, the accurate water area surface change area L is obtained, and the size of the water area surface change area is calculated through the conversion of the spatial resolution, as shown in Figure 6 At the same time, the image difference method is used to compare and extract the change object spatial range in the water area management line from the front and rear two orthographic images, the pixel position of the change object in the original unmanned aerial vehicle image and the image is obtained through reverse calculation of the change object spatial range, so as to obtain the identification early warning result and the early warning result of the river and lake abnormal problem.

[0098] The acquisition of the early warning result of the river and lake abnormal problem includes,

[0099] The image difference method is used to compare the front and rear two orthographic images, the change object spatial range in the water area management line is extracted, the pixel position of the change object in the original unmanned aerial vehicle image and the image is obtained through reverse calculation of the change object spatial range, so as to obtain the early warning result of the river and lake abnormal problem. As shown in Figure 11 The water area surface change area, the identification early warning result, the generated three-dimensional orthographic image and the river and lake spatial terrain data are uploaded to the third party cloud platform, and the edge end calculation is completed.

[0100] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited to this, any change or replacement within the technical scope disclosed by the present application should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for intelligent identification of changes in river and lake water areas and early warning of abnormal problems in rivers and lakes by using an unmanned aerial vehicle, characterized in that, The method comprises the following steps: Collecting hydrological and geographical data, environmental quality data and engineering operation data by using edge computing equipment to build a river and lake digital twin bottom plate data; Manually editing the basic route of the river and lake, intelligently generating a shooting route based on the basic route and the river and lake digital twin bottom plate data, setting a dynamic route generation rule, splicing the basic route and the shooting route, and obtaining a dynamic route of the unmanned aerial vehicle; According to the dynamic route, dynamic route flight is performed, data information during flight is collected, and the data information is returned to the unmanned aerial vehicle base station through the network and transmitted to the edge computing equipment; The data information collected by the unmanned aerial vehicle is combined with the water area management line to generate river and lake orthographic images and river and lake spatial terrain data, and a water area surface recognition model and an image difference method are used to identify and compare the front and rear two orthographic images, thereby obtaining a warning result of the water area surface change area and the river and lake abnormal problem; The dynamic route generation rule includes, If the water quality level monitored by the water quality monitoring equipment data is abnormal, the position of the water quality monitoring equipment and the position of the sewage outlet are fixed and multi-angle shot; if the ground camera monitoring early warning data is found, the fixed area scanning shooting of the ground camera monitoring position is performed; The intelligent generation of the dynamic route is based on the existing dynamic route generation rule, combined with the river and lake spatial terrain data and the flight focus point position, and through intelligent space calculation of the route, an intelligent shooting point is obtained, a shooting route is intelligently generated, the basic route and the shooting route are spliced, and the dynamic route is generated; The acquisition of the intelligent shooting point includes, The shortest distance position point of the monitoring point from the route is the intelligent shooting point; The intelligent generation of the dynamic route includes, Setting a dynamic route generation rule, generating a shooting route for the intelligent shooting point, if the intelligent shooting point shooting mode is fixed position multi-angle shooting, the angle A formed by the connecting line of the unmanned aerial vehicle shooting angle and the fixed position and the perpendicular line of the unmanned aerial vehicle plumb direction is greater than 30 degrees, three unmanned aerial vehicle shooting points need to be automatically set for the shooting route, and the shooting route is formed through the three unmanned aerial vehicle shooting points; wherein, the three unmanned aerial vehicle shooting points need to satisfy the following two conditions: the interval distance B between the three unmanned aerial vehicle shooting points is greater than 20 meters; combined with the river and lake spatial terrain data calculation, the non-obstructive object in the visual corridor buffer C formed by the connecting line of the shooting point and the fixed position is calculated to ensure that the fixed position can be shot, and the visual corridor buffer C is a conical visual corridor buffer C with a half vertex angle of 5 degrees, with the unmanned aerial vehicle shooting point as the top point and the fixed position as the axis end point.

2. The method for intelligent identification of river and lake water changes and early warning of abnormal river and lake problems by unmanned aerial vehicles according to claim 1, characterized in that: The flight time range is calculated based on the basic route and the river and lake digital twin bottom plate data, and the basic route determines the path, flight height, shooting angle and shooting focal length of the unmanned aerial vehicle flight.

3. The method of claim 2, wherein the method further comprises: determining a water level of the water body based on the image; and determining a water level change of the water body based on the image. The water quality monitoring equipment position, sewage outlet position and ground camera monitoring position are set as flight focus points, and two shooting modes of fixed position multi-angle shooting and fixed area scanning shooting are set for the flight focus points, wherein the sewage outlet position and the water quality monitoring equipment position adopt fixed position multi-angle shooting, and the ground camera monitoring position adopts fixed area scanning shooting.

4. The method of claim 3, wherein the method further comprises: determining a water level of the water body based on the image; and determining a water level change of the water body based on the image. The fixed position multi-angle shooting is focused on the fixed position and enlarged to obtain images of different angles through three different angles; The fixed area scanning shooting draws a square area with the fixed position as the center point when an abnormality warning occurs, and performs focused and enlarged frame-by-frame scanning shooting on the square area.

5. The method of claim 4, wherein the method further comprises: determining a water level of the river or lake based on the image; and determining whether the water level is abnormal based on the water level and the water level threshold. The water area face change area acquisition includes, The water area face change area acquisition includes, 6. The method of claim 1 or 5, wherein the method further comprises: determining a water level of the river or lake based on the image; and determining whether the water level is abnormal based on the water level. The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, 7. The method of claim 6, wherein the method further comprises: determining a water level of the water body based on the image; and determining a water level change of the water body based on the image. W = W1-W2; The water area face change area acquisition includes, 8. The method of claim 7, wherein the method further comprises: determining a water level of the river or lake based on the image; and determining whether the water level is abnormal based on the water level and a water level threshold. The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area acquisition includes, The water area face change area

Images