A high-precision real-time crack identification method and system based on drone

Abstract

The present invention discloses a high-precision real-time crack identification method and system based on unmanned aerial vehicles, which relates to the field of crack identification technology, determines the management scope, sets multiple anchor points based on the management scope, builds an unmanned aerial vehicle flying network based on the anchor points and the management scope; sets target crack information, wherein the target crack information includes target crack basic information and target crack spatial information, and builds a crack identification library based on the target crack information; flies the unmanned aerial vehicle based on the unmanned aerial vehicle flying network and collects images within the management scope, obtains crack information in the image, and determines the difference information between the crack information and the corresponding target crack information. The present invention ensures the accuracy of crack identification through the positioning of the crack by the unmanned aerial vehicle flying network and the corresponding relationship between the tangents on the edge line of the crack image, ensures that the current and past crack studies are targeted at the same crack, and has a good identification effect.

Description

Technical Field

[0001] The present invention relates to the technical field of crack identification, and in particular to a high-precision real-time crack identification method and system based on an unmanned aerial vehicle. Background Art

[0002] In the UAV crack identification system, the UAV is equipped with a high-precision camera to shoot the target area through the aerial flight network to obtain high-definition ground images. These images are then transmitted to the ground station or cloud server, and the cracks are automatically identified using advanced image processing algorithms. At present, the use of UAVs for crack identification can usually only obtain the crack state at a certain moment, and it is difficult to achieve long-term continuous tracking of the crack development; this may cause some subtle changes in the crack development process to be ignored, affecting the accurate judgment of the crack development trend; when comparing the current cracks with the past cracks, it is difficult to ensure that the cracks correspond. In reality, due to changes in the UAV shooting environment, the identification of cracks may result in the current cracks and past cracks not being the same cracks, which increases the difficulty of data comparison. Summary of the invention

[0003] The purpose of the present invention is to provide a high-precision real-time crack identification method and system based on unmanned aerial vehicle to solve the shortcomings of the background technology.

[0004] In order to achieve the above object, the present invention provides the following technical solution: a high-precision real-time crack identification method based on drone, comprising the following steps:

[0005] Determine the management scope, set multiple anchor points based on the management scope, and build a UAV aerial flight network based on the anchor points and the management scope;

[0006] Setting target crack information, wherein the target crack information includes target crack basic information and target crack spatial information, and building a crack identification library based on the target crack information;

[0007] The UAV is flown based on the UAV aerial flight network to collect images within the management range, and the crack information in the image is obtained, wherein the crack information includes crack basic information and crack spatial information, and the corresponding target crack is obtained by comparing and identifying the crack information in the crack identification library;

[0008] Determine the difference information between the crack information and the corresponding target crack information.

[0009] In a preferred embodiment, the steps of determining the management scope, setting a plurality of anchor points based on the management scope, and constructing a UAV aerial flight network based on the anchor points and the management scope include:

[0010] Determine the scope where crack identification is required as the management scope;

[0011] Set multiple anchor points on the edge line of the management scope, and connect the multiple anchor points according to the edge line of the management scope;

[0012] Set the flying height of the UAV, set up a flight network at the flying height of the UAV within the management range, and associate the flight network with the anchor point to obtain the UAV flying network.

[0013] In a preferred embodiment, the step of associating the positions of the flight network and the anchor points to obtain the aerial flight network of the drone includes:

[0014] The best altitude for drone shooting is obtained as the drone's flight altitude, and the altitude surface is set at the drone's flight altitude within the management range;

[0015] Divide the height surface into multiple sub-grids by gridding, assign points to the multiple sub-grids as grid points, and set codes for the sub-grids and corresponding grid points respectively;

[0016] Set corresponding first service points for each of the multiple network points, the service points corresponding to the multiple network points are loaded in the first server, set corresponding second service points for each of the multiple anchor points, the second service points of the multiple anchor points are loaded in the second server, and connect the first server to the second server;

[0017] Based on the positions of multiple anchor points and the coding of the network points, a connection relationship between the anchor points and the mapped network points is established in the first server and the second server, and the position association between the flight network and the anchor points is obtained as the drone aerial flight network.

[0018] In a preferred embodiment, the step of constructing a crack identification library based on target crack information includes:

[0019] Determine the target cracks within the management scope, and obtain the target crack information, wherein the target crack information includes the target crack basic information and the target crack spatial information;

[0020] Constructing a plane storage interval based on the target crack space information, and storing the target crack basic information in the plane storage interval corresponding to the target crack space information;

[0021] The plane storage intervals corresponding to multiple target cracks are combined to obtain a crack identification library.

[0022] In a preferred embodiment, the step of determining the target cracks within the management scope and obtaining the target crack information includes:

[0023] The corresponding UAV aerial flight network plans the flight route, and the UAV collects target images within the management range according to the flight route;

[0024] Extract target features for crack identification in the target image, divide and mark the edges of the graphics in the target features to obtain multiple graphics, enlarge the multiple graphics, obtain the colors inside the graphics, and use the graphics that meet the preset conditions as target cracks;

[0025] A base surface is set, and the target crack pattern is laid on the base surface. The tangents of the target crack edge line are calibrated based on the base surface. The intersection points of the tangents of adjacent tangents whose tangent angles are less than the preset angles are obtained as calibration points. The calibration points are connected to obtain a calibration relationship network.

[0026] Binding the calibration relationship network, the target crack and the corresponding target features as the target crack basic information;

[0027] The location of the target crack basic information is obtained and mapped upward to the sub-grid on the UAV aerial flight network and the corresponding network point coding as the target crack spatial information, and the target crack spatial information and the target crack basic information are used as the target crack information.

[0028] In a preferred embodiment, the step of performing comparative identification in a crack identification library based on crack information to obtain corresponding target cracks includes:

[0029] Set a collection cycle, collect images of the management area through the drone according to the flight route based on the collection cycle, and obtain the current cracks in the image based on the target features in the image;

[0030] The current crack is spread on the base surface to obtain a calibration relationship network of the current crack, and the calibration relationship network of the current crack, the current crack and the corresponding target features are used as the basic information of the crack;

[0031] Obtain the crack basic information and map it upward to the sub-grid on the aerial flight network and the corresponding network point codes as the crack space information;

[0032] The crack space information and the crack basic information are used as crack information, and the corresponding target crack space information is obtained by matching in the crack identification library based on the crack space information, and the corresponding target crack basic information is obtained based on the target crack space information to obtain the target crack.

[0033] In a preferred embodiment, the step of determining the difference information between the crack information and the corresponding target crack information includes:

[0034] The calibration network in the crack information is matched with the calibration network in the target crack information by tangent recognition, and the tangent of the calibration network in the crack information is compared with the tangent of the calibration network in the corresponding target crack information to obtain the difference displacement and the difference angle;

[0035] Based on the difference displacement and the difference angle, the target crack is simulated to move to obtain the current crack, and the movement process information of the current crack is obtained by simulating the movement of the target crack as the difference information, wherein the difference information includes the movement distance and the movement angle.

[0036] The present invention also provides a high-precision real-time crack identification system based on an unmanned aerial vehicle, comprising:

[0037] A construction module is used to determine the management scope, set multiple anchor points based on the management scope, and build a UAV aerial flight network based on the anchor points and the management scope;

[0038] A setting module connected to the building module, used to set target crack information, wherein the target crack information includes target crack basic information and target crack spatial information, and build a crack identification library based on the target crack information;

[0039] The identification module is connected to the setting module and is used to fly the UAV based on the UAV aerial flight network and collect images within the management range to obtain crack information in the image, wherein the crack information includes crack basic information and crack spatial information, and compare and identify the corresponding target crack in the crack identification library based on the crack information;

[0040] The identification and comparison module is connected to the identification module and is used to determine the difference information between the crack information and the corresponding target crack information.

[0041] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0042] The present invention ensures the accuracy of crack recognition through the positioning of cracks by the UAV aerial flight network and the correspondence between the tangents on the edge lines of the crack image, ensures that the current and past crack studies are aimed at the same crack, and has a good recognition effect. Afterwards, the target crack needs to be simulated in motion to obtain the motion process information of the current crack as distinguishing information, so as to obtain a development situation of the crack. After the corresponding crack is identified, the development situation of the crack can be understood. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0044] Figure 1 The present invention is a flow chart of the method.

[0045] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Example 1, please refer to Figure 1 As shown, the high-precision real-time crack identification method based on drone described in this embodiment includes the following steps:

[0048] S1. Determine the management scope, set multiple anchor points based on the management scope, and build a UAV air flight network based on the anchor points and the management scope;

[0049] S2. Setting target crack information, wherein the target crack information includes target crack basic information and target crack spatial information, and building a crack identification library based on the target crack information;

[0050] S3, flying the drone based on the drone aerial flight network and collecting images within the management range to obtain crack information in the image, wherein the crack information includes crack basic information and crack spatial information, and comparing and identifying the crack information in the crack identification library to obtain the corresponding target crack;

[0051] S4, determining difference information between the crack information and the corresponding target crack information;

[0052] As described in the above steps S1-S4, the target area is photographed through the aerial flight network to obtain high-definition ground images. These images are then transmitted to the ground station or cloud server, and the cracks are automatically identified using advanced image processing algorithms. At present, the use of drones for crack identification can usually only obtain the crack state at a certain moment, and it is difficult to achieve long-term continuous tracking of the development of cracks; this may cause some subtle changes to be ignored during the development of cracks, affecting the accurate judgment of the development trend of cracks; when comparing the current cracks with the past cracks, it is difficult to ensure whether the cracks correspond. In reality, due to changes in the drone shooting environment, the identification of cracks may cause the current cracks to be different from the past cracks, which increases the difficulty of data comparison. The present application uses the drone aerial flight network to locate the cracks and the corresponding relationship between the tangents on the edge line of the crack image to ensure the accuracy of crack identification, ensure that the current and past cracks are studied for the same crack, and have a good identification effect. After that, the target crack needs to be simulated to obtain the movement process information of the current crack as the distinguishing information, so that a development of the crack can be obtained. After the corresponding crack is identified, the development of a crack can be understood.

[0053] In one embodiment, the step S1 of determining the management scope, setting a plurality of anchor points based on the management scope, and constructing a UAV aerial flight network based on the anchor points and the management scope includes:

[0054] S11. Determine the scope where crack identification is required as the management scope;

[0055] S12, setting a plurality of anchor points on the edge line of the management scope, and connecting the plurality of anchor points according to the edge line of the management scope;

[0056] S13, setting the flight altitude of the drone, setting a flight network at the flight altitude of the drone within the management range, and associating the flight network with the anchor point to obtain the flight network of the drone;

[0057] As described in the above steps S11-S13, the identification of cracks is determined within a range, or a section of road or a piece of ground, as the management range, and then multiple anchor points are set on the edge line of the management range. Multiple anchor points are set on the edge line, and the edge lines between multiple anchor points are all drawn by anchor points. Therefore, adjacent anchor points are connected by edge lines, and then the aerial flight height of the drone is set. In order to collect crack images with higher accuracy, it is necessary to shoot at the optimal height for drone shooting. Then, the flight network is set at the aerial flight height position of the drone within the management range, and the flight network is associated with the anchor points to obtain the aerial flight network of the drone. The flight network here is a plane network set at the same height (the aerial flight height of the drone). Here, it is a virtual setting, which can facilitate the subsequent better collection and identification of cracks.

[0058] In one embodiment, the step S13 of obtaining the aerial flight network of the drone by associating the positions of the flight network and the anchor points includes:

[0059] S131, obtaining the best altitude for drone photography as the aerial flight altitude of the drone, and setting an altitude plane at the aerial flight altitude position of the drone within the management range;

[0060] S132, gridding the height surface to obtain a plurality of sub-grids, assigning points to the plurality of sub-grids as grid points, and setting codes for the sub-grids and corresponding grid points respectively;

[0061] S133, respectively setting corresponding first service points for the plurality of network points, the service points corresponding to the plurality of network points are loaded in the first server, setting corresponding second service points for the plurality of anchor points, the second service points of the plurality of anchor points are loaded in the second server, and connecting the first server to the second server;

[0062] S134, establishing a connection relationship between the anchor points and the mapped network points in the first server and the second server based on the positions of the plurality of anchor points and the codes of the network points, and obtaining a position association between the flight network and the anchor points as the aerial flight network of the drone;

[0063] As described in the above steps S131-S134, the optimal height for drone photography is obtained as the drone's aerial flight altitude, and a height plane is set at the drone's aerial flight altitude position within the management range. The height plane is the aerial plane at the drone's optimal altitude. Then, grid division is performed in the height plane to obtain multiple sub-grids. By dividing the sub-grids, the flight position of the drone can be obtained when the drone performs subsequent flight photography to collect crack images. After obtaining the sub-grids, since the grids have an internal range, it is necessary to set multiple grid points in the sub-grids, and then encode the sub-grids and the corresponding grid points respectively. The positions of the sub-grids and the grid points can be determined according to the encoding. Then, in order to ensure the stability of the virtual setting between the height plane and the management range, it is necessary to set multiple grid points. The network points are respectively provided with corresponding first service points, and the service points corresponding to the multiple network points are loaded in the first server. The multiple anchor points are correspondingly provided with corresponding second service points, and the second service points of the multiple anchor points are loaded in the second server. The first server is connected with the second server, and the connection relationship between the anchor points and the mapped network points is established in the first server and the second server based on the positions of the multiple anchor points and the encoding of the network points, so as to obtain the position association between the flight network and the anchor points as the aerial flight network of the unmanned aerial vehicle, which can determine the corresponding relationship between the flight network and the jurisdiction, and determine the position of the flight network through the connection relationship between the anchor points and the network points in the flight network, wherein the connection relationship is the corresponding relationship between the anchor points and the upper network points, so as to facilitate the recognition ability of the cracks during the subsequent flight of the unmanned aerial vehicle.

[0064] In one embodiment, the step S2 of building a crack identification library based on target crack information includes:

[0065] S21, determining a target crack within the management scope, and obtaining target crack information, wherein the target crack information includes target crack basic information and target crack spatial information;

[0066] S22, constructing a plane storage interval based on the target crack space information, and storing the target crack basic information in the plane storage interval corresponding to the target crack space information;

[0067] S23, combining the plane storage intervals corresponding to the multiple target cracks to obtain a crack identification library;

[0068] As described in the above steps S21-S23, target cracks within the management scope are collected, and basic information of target cracks and spatial information of target cracks are obtained as target crack information. The target crack information here is collected by the drone in the drone air flight network. Then, the basic information of the target cracks is stored corresponding to the spatial information of the target cracks. One subgrid represents a plane storage interval, that is, a storage location. Then, the plane storage intervals of multiple basic information of target cracks corresponding to the spatial information of the target cracks are combined according to the relationship of the subunit grids to obtain a crack identification library. The crack identification library is used to store the basic information of the target cracks and the spatial information of the target cracks, as well as the corresponding relationship between the basic information of the target cracks and the spatial information of the target cracks, which can facilitate the subsequent daily monitoring and identification of cracks and have a good corresponding relationship.

[0069] In one embodiment, the step S21 of determining target cracks within the management scope and obtaining target crack information includes:

[0070] S211, the corresponding UAV aerial flight network plans a flight route, and the UAV collects target images within the management range according to the flight route;

[0071] S212, extracting target features for crack identification in the target image, dividing and marking the edges of graphics in the target features to obtain multiple graphics, amplifying the multiple graphics, obtaining the colors inside the graphics, and taking the graphics that meet the preset conditions as target cracks;

[0072] S213, setting a base surface, laying the target crack pattern on the base surface, calibrating the tangent of the target crack edge line based on the base surface, obtaining the tangent intersection points of adjacent tangents whose tangent angles are less than a preset angle as calibration points, and connecting the calibration points to obtain a calibration relationship network;

[0073] S214, binding the calibration relationship network, the target crack and the corresponding target features as target crack basic information;

[0074] S215, obtaining the location of the target crack basic information and mapping it upward to the sub-grid on the UAV aerial flight network and the code of the corresponding grid point as the target crack spatial information, and taking the target crack spatial information and the target crack basic information as the target crack information;

[0075] As described in the above steps S211-S215, the flight route is planned on the drone's aerial flight network. The flight route can capture target features within the management range. After the drone collects the target image within the management range according to the flight route, the features in the target image are extracted, and then the target features for crack identification in the target image are obtained, the edges of the graphics in the target features are divided and marked to obtain multiple graphics, the multiple graphics are enlarged, the colors inside the graphics are obtained, and the graphics that meet the preset conditions are used as target cracks. The preset conditions are preset colors and the approximate shape of the image, for example, the color is black or dark gray, and the approximate shape of the image is long, straight or curved. Then, in order to understand the cracks, Features, set a base surface, which is a surface for tiling images, depict the graphic edge line of the target crack on the base surface and calibrate the tangent of the target crack edge line, obtain the tangent intersection point of adjacent tangents whose tangent angle is less than the preset angle as the calibration point, connect the calibration points to obtain the calibration relationship network, where the tangent angle is the inner angle between adjacent tangents, and the preset angle is a set inner angle, bind the calibration relationship network, the target crack and the corresponding target feature as the target crack basic information, obtain the location of the target crack basic information and map it upward to the sub-grid on the aerial flight network and the code of the corresponding grid point as the target crack space information, which can facilitate the subsequent identification and determination of the crack;

[0076] In one embodiment, the step S3 of performing comparative identification in a crack identification library based on crack information to obtain corresponding target cracks includes:

[0077] S31, setting a collection cycle, collecting images of the management area by using a drone according to a flight route based on the collection cycle, and obtaining current cracks in the image based on target features in the image;

[0078] S32, laying the current crack on the base surface to obtain a calibration relationship network of the current crack, and using the calibration relationship network of the current crack, the current crack and the corresponding target features as basic crack information;

[0079] S33, obtaining the sub-grids mapped upward on the aerial flight network by the basic information of the crack and the codes of the corresponding grid points as the spatial information of the crack;

[0080] S34, taking the fracture spatial information and the fracture basic information as fracture information, matching in a fracture identification library based on the fracture spatial information to obtain corresponding target fracture spatial information, and obtaining corresponding target fracture basic information based on the target fracture spatial information to obtain the target fracture;

[0081] As described in the above steps S31-S34, the collection cycle is to set the cycle for the drone to collect the current cracks, for example, once a month, once a week, etc., and then the drone is used to collect images of the management range according to the flight route according to the collection cycle, and the current crack in the image is obtained based on the target features in the image. The current crack is the crack collected at that time, and the current crack is flattened on the base surface to obtain the calibration relationship network of the current crack. The calibration relationship network of the current crack, the crack and the corresponding target features are used as the basic information of the crack. The calibration relationship network of the current crack is the same as the operation of constructing the calibration relationship network of the target crack in step S21. Then the calibration relationship of the current crack is obtained. The network, the current crack and the corresponding target features are used as the basic information of the crack, and the sub-grids and the corresponding network points on the aerial flying network mapped upward by the basic information of the crack are obtained as the coding of the crack space information, and the crack space information and the basic information of the crack are used as the crack information. Based on the crack space information, the corresponding target crack space information is matched in the crack identification library, and the corresponding target crack basic information is obtained based on the target crack space information to obtain the target crack. The target crack can be matched in the crack identification library through the crack space information, and the corresponding relationship between the current crack and the crack identification library can be quickly identified, which has a good crack management function and is convenient for periodic monitoring of cracks.

[0082] In one embodiment, the step S4 of determining the difference information between the crack information and the corresponding target crack information includes:

[0083] S41, performing tangent recognition and correspondence between the calibration relationship network in the crack information and the calibration relationship network in the target crack information, and comparing the tangent of the calibration relationship network in the crack information with the tangent of the calibration relationship network in the corresponding target crack information to obtain a difference displacement and a difference angle;

[0084] S42, simulating the movement of the target crack to obtain the current crack based on the difference displacement and the difference angle, and obtaining the movement process information of the current crack by simulating the movement of the target crack as the distinguishing information, wherein the distinguishing information includes the movement distance and the movement angle;

[0085] As described in the above steps S41 and S42, the calibration relationship network in the crack information is matched with the calibration relationship network in the target crack information for tangent identification in order to perform a feature identification function on the current crack and the corresponding target crack. The feature here is the tangent. The tangent is used to determine whether they are the same crack. Then, the tangent of the calibration relationship network in the crack information is compared with the tangent of the calibration relationship network in the corresponding target crack information to obtain the difference displacement and the difference angle. In order to obtain the change of the current crack compared with the previous target crack, it is necessary to simulate the movement of the target crack to obtain the movement process information of the current crack as the distinguishing information. The distinguishing information includes the movement distance and the movement angle, which can obtain the development of the crack. After the corresponding crack is identified, the development of the crack can be understood. The crack is positioned by the drone aerial flight network and the correspondence between the tangents on the edge line of the crack image to ensure the accuracy of crack identification and to ensure that the current and past crack studies are for the same crack, which has a good identification effect.

[0086] Example 2, please refer to Figure 2 As shown, the high-precision real-time crack identification system based on drone described in this embodiment includes:

[0087] A construction module is used to determine the management scope, set multiple anchor points based on the management scope, and build a UAV aerial flight network based on the anchor points and the management scope;

[0088] A setting module connected to the building module, used to set target crack information, wherein the target crack information includes target crack basic information and target crack spatial information, and build a crack identification library based on the target crack information;

[0089] The identification module is connected to the setting module and is used to fly the UAV based on the UAV aerial flight network and collect images within the management range to obtain crack information in the image, wherein the crack information includes crack basic information and crack spatial information, and compare and identify the corresponding target crack in the crack identification library based on the crack information;

[0090] The identification and comparison module is connected to the identification module and is used to determine the difference information between the crack information and the corresponding target crack information.

[0091] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A high-precision real-time crack identification method based on drone, characterized in that: The following steps are involved: Determine the management scope, set multiple anchor points based on the management scope, and build a UAV aerial flight network based on the anchor points and the management scope; Setting target crack information, wherein the target crack information includes target crack basic information and target crack spatial information, and building a crack identification library based on the target crack information; The step of building a crack identification library based on target crack information includes: Determine the target cracks within the management scope, and obtain the target crack information, wherein the target crack information includes the target crack basic information and the target crack spatial information; The step of determining the target cracks within the management scope and obtaining the target crack information comprises: The corresponding UAV aerial flight network plans the flight route, and the UAV collects target images within the management range according to the flight route; Extract target features for crack identification in the target image, divide and mark the edges of the graphics in the target features to obtain multiple graphics, enlarge the multiple graphics, obtain the colors inside the graphics, and use the graphics that meet the preset conditions as target cracks; A base surface is set, and the target crack pattern is laid on the base surface. The tangents of the target crack edge line are calibrated based on the base surface. The intersection points of the tangents of adjacent tangents whose tangent angles are less than the preset angles are obtained as calibration points. The calibration points are connected to obtain a calibration relationship network. Binding the calibration relationship network, the target crack and the corresponding target features as the target crack basic information; The location of the target crack basic information is obtained and mapped upward to the sub-grid on the UAV aerial flight network and the code of the corresponding grid point as the target crack spatial information, and the target crack spatial information and the target crack basic information are used as the target crack information; Constructing a plane storage interval based on the target crack space information, and storing the target crack basic information in the plane storage interval corresponding to the target crack space information; The plane storage intervals corresponding to multiple target cracks are combined to obtain a crack identification library; The UAV is flown based on the UAV aerial flight network to collect images within the management range, and the crack information in the image is obtained, wherein the crack information includes crack basic information and crack spatial information, and the corresponding target crack is obtained by comparing and identifying the crack information in the crack identification library; Determine difference information between the crack information and the corresponding target crack information; The step of determining difference information between the crack information and the corresponding target crack information comprises: The calibration network in the crack information is matched with the calibration network in the target crack information by tangent recognition, and the tangent of the calibration network in the crack information is compared with the tangent of the calibration network in the corresponding target crack information to obtain the difference displacement and the difference angle; Based on the difference displacement and the difference angle, the target crack is simulated to move to obtain the current crack, and the movement process information of the current crack is obtained by simulating the movement of the target crack as the difference information, wherein the difference information includes the movement distance and the movement angle.

2. The high-precision real-time crack identification method based on an unmanned aerial vehicle according to claim 1, characterized in that: The steps of determining the management scope, setting a plurality of anchor points based on the management scope, and constructing a UAV aerial flight network based on the anchor points and the management scope include: Determine the scope where crack identification is required as the management scope; Set multiple anchor points on the edge line of the management scope, and connect the multiple anchor points according to the edge line of the management scope; Set the flying height of the UAV, set up a flight network at the flying height of the UAV within the management range, and associate the flight network with the anchor point to obtain the UAV flying network.

3. The high-precision real-time crack identification method based on an unmanned aerial vehicle according to claim 2 is characterized in that: The step of associating the positions of the flight network and the anchor points to obtain the flight network of the drone in the air comprises: The best altitude for drone shooting is obtained as the drone's flight altitude, and the altitude surface is set at the drone's flight altitude within the management range; Divide the height surface into multiple sub-grids by gridding, assign points to the multiple sub-grids as grid points, and set codes for the sub-grids and corresponding grid points respectively; Set corresponding first service points for each of the multiple network points, the service points corresponding to the multiple network points are loaded in the first server, set corresponding second service points for each of the multiple anchor points, the second service points of the multiple anchor points are loaded in the second server, and connect the first server to the second server; Based on the positions of multiple anchor points and the coding of the network points, a connection relationship between the anchor points and the mapped network points is established in the first server and the second server, and the position association between the flight network and the anchor points is obtained as the drone aerial flight network.

4. The high-precision real-time crack identification method based on an unmanned aerial vehicle according to claim 1 is characterized in that: The step of performing comparative identification in a crack identification library based on crack information to obtain corresponding target cracks includes: Set a collection cycle, collect images of the management area through the drone according to the flight route based on the collection cycle, and obtain the current cracks in the image based on the target features in the image; The current crack is spread on the base surface to obtain a calibration relationship network of the current crack, and the calibration relationship network of the current crack, the current crack and the corresponding target features are used as the basic information of the crack; Obtain the crack basic information and map it upward to the sub-grid on the aerial flight network and the corresponding network point codes as the crack space information; The crack space information and the crack basic information are used as crack information, and the corresponding target crack space information is obtained by matching in the crack identification library based on the crack space information, and the corresponding target crack basic information is obtained based on the target crack space information to obtain the target crack.

5. A high-precision real-time crack identification system based on a drone, used to implement a high-precision real-time crack identification method based on a drone as claimed in any one of claims 1 to 4, characterized in that: include: A construction module is used to determine the management scope, set multiple anchor points based on the management scope, and build a UAV aerial flight network based on the anchor points and the management scope; A setting module connected to the building module, used to set target crack information, wherein the target crack information includes target crack basic information and target crack spatial information, and build a crack identification library based on the target crack information; The setting module includes: Determine the target cracks within the management scope, and obtain the target crack information, wherein the target crack information includes the target crack basic information and the target crack spatial information; Constructing a plane storage interval based on the target crack space information, and storing the target crack basic information in the plane storage interval corresponding to the target crack space information; The plane storage intervals corresponding to multiple target cracks are combined to obtain a crack identification library; The identification module is connected to the setting module and is used to fly the UAV based on the UAV aerial flight network and collect images within the management range to obtain crack information in the image, wherein the crack information includes crack basic information and crack spatial information, and compare and identify the corresponding target crack in the crack identification library based on the crack information; The identification and comparison module is connected to the identification module and is used to determine the difference information between the crack information and the corresponding target crack information.

Images