An engineering acceptance method and system based on cooperation of unmanned aerial vehicles and unmanned vehicles

Abstract

The application provides an engineering acceptance method and system based on cooperation of a UAV and an unmanned vehicle, and comprises the following steps: acquiring a first cruising track of the UAV on target area land, and simultaneously acquiring a second cruising track of the unmanned vehicle on the target area land; controlling the UAV to perform first cruising shooting on the target area land according to the first cruising track, and obtaining a first image sequence collected by the UAV; simultaneously, controlling the unmanned vehicle to perform second cruising shooting on the target area land according to the second cruising track, and obtaining a second image sequence collected by the unmanned vehicle; associating and screening the first image sequence and the second image sequence, and obtaining an effective image sequence set; and analyzing the effective image sequence set, and obtaining a land reclamation acceptance result of the target area land. The accuracy of land reclamation acceptance is improved, the intelligence of image collection is ensured, and the efficiency of land reclamation acceptance is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to an engineering acceptance method and system based on cooperation of unmanned aerial vehicles and unmanned vehicles. BACKGROUND

[0002] At present, with the rapid development of industrialization and urbanization, a large amount of land is mined, excavated or destroyed, such as open-pit coal mining, construction engineering soil taking and other activities. Land reclamation aims to restore these damaged lands to a usable state, which is of great significance to the protection of land resources, the maintenance of ecological balance, the protection of agricultural production and the meeting of land demand for social development.

[0003] However, traditional land reclamation acceptance mainly relies on manual site investigation. This method has many limitations. First, manual investigation is low in efficiency and requires a large amount of manpower, material resources and time. For a large area of reclamation land, it is very time-consuming to measure and evaluate one by one. Second, the accuracy of manual investigation is limited by the subjective factors and technical level of the investigators. Different investigators may have different understandings and implementations of the acceptance standards, which affects the reliability of the acceptance results. In addition, some complex terrains and environmental conditions, such as steep slopes and muddy wetlands, may bring difficulties or even dangers to manual investigation.

[0004] Therefore, in order to overcome the above technical problems, the present application provides an engineering acceptance method and system based on cooperation of unmanned aerial vehicles and unmanned vehicles SUMMARY

[0005] The present application provides an engineering acceptance method and system based on cooperation of unmanned aerial vehicles and unmanned vehicles, which can effectively guarantee the integrity and comprehensiveness of land image collection by unmanned aerial vehicles based on a first cruising trajectory and unmanned vehicles based on a second cruising trajectory, realize the multi-angle analysis goal of land reclamation acceptance, help to conduct land reclamation acceptance of the target area from different angles by correlating the first image sequence and the second image sequence, effectively guarantee the accuracy and quality of the effective image sequence obtained by screening, obtain the land reclamation acceptance result of the target area land by analyzing the effective image sequence set, effectively realize multi-source data acceptance of land reclamation acceptance, thereby improving the accuracy of land reclamation acceptance. At the same time, by controlling the unmanned aerial vehicles and unmanned vehicles to take pictures, the intelligence of image collection is effectively guaranteed, thereby improving the efficiency of land reclamation acceptance.

[0006] The present application provides an engineering acceptance method based on cooperation of unmanned aerial vehicles and unmanned vehicles, comprising:

[0007] Step 1: Obtain a first cruising track of the unmanned aerial vehicle in the target area land, and simultaneously obtain a second cruising track of the unmanned vehicle in the target area land;

[0008] Step 2: Control the unmanned aerial vehicle to perform first cruising shooting in the target area land according to the first cruising track, and obtain a first image sequence collected by the unmanned aerial vehicle, and simultaneously control the unmanned vehicle to perform second cruising shooting in the target area land according to the second cruising track, and obtain a second image sequence collected by the unmanned vehicle;

[0009] Step 3: Correlate and screen the first image sequence and the second image sequence to obtain an effective image sequence set;

[0010] Step 4: Analyze the effective image sequence set to obtain a land reclamation acceptance result of the target area land.

[0011] Preferably, in the step 1 of the engineering acceptance method based on the cooperation of the unmanned aerial vehicle and the unmanned vehicle, the first cruising track of the unmanned aerial vehicle in the target area land is obtained, including:

[0012] Obtain the target area land for land reclamation acceptance, and obtain the position distribution characteristics of the target area land;

[0013] Position the boundary contour of the target area land according to the position distribution characteristics of the target area land, and obtain the boundary contour position points of the target area land;

[0014] Obtain the topographic characteristics of the target area land;

[0015] Determine the geometric parameters of the target area land according to the boundary contour position points of the target area land and the topographic characteristics of the target area land, and determine the first cruising track of the unmanned aerial vehicle in the target area land according to the geometric parameters.

[0016] Preferably, in the step 1 of the engineering acceptance method based on the cooperation of the unmanned aerial vehicle and the unmanned vehicle, the second cruising track of the unmanned vehicle in the target area land is obtained, including:

[0017] Collect the land characteristics of the target area land, and determine the static obstacles of the target area land;

[0018] Obtain a set of land boundary points of the target area land, and take any land boundary point in the set of land boundary points as a coordinate origin to construct a two-dimensional coordinate system of the target area land;

[0019] Obtain a first relative position relationship between the remaining land boundary points in the set of land boundary points and the coordinate origin, and determine a plurality of first coordinate points of the remaining land boundary points in the two-dimensional coordinate system according to the first relative position relationship, and perform first labeling on the plurality of first coordinate points in the two-dimensional coordinate system to obtain a first region plane;

[0020] Obtain the obstacle boundary point set of static obstacles, and obtain the second relative positional relationship between each obstacle boundary point in the obstacle boundary point set and the origin of coordinates. Based on the second relative positional relationship, determine the second coordinate point of the obstacle boundary point set in the two-dimensional coordinate system. At the same time, perform a second annotation on multiple second coordinate points in the two-dimensional coordinate system to obtain the second region plane.

[0021] The second cruising trajectory of the unmanned vehicle in the target area is constructed based on the positional distribution of the second regional plane in the first regional plane.

[0022] Preferably, in an engineering acceptance method based on the collaboration of unmanned aerial vehicles (UAVs) and unmanned vehicles, step 2 involves controlling the UAV to conduct a first cruise and take pictures in the target area according to the first cruise trajectory, obtaining a first image sequence collected by the UAV, including:

[0023] The first cruise trajectory is read to determine the first cruise trajectory data, and at the same time, the command reading format of the UAV is obtained;

[0024] The first cruise trajectory data is converted into the first instruction element according to the instruction reading format of the UAV. At the same time, the preset UAV image acquisition frequency is obtained, and the second instruction element is generated according to the preset UAV image acquisition frequency.

[0025] A first shooting command is generated based on the first and second command elements. At the same time, the drone is controlled to conduct a first cruise shooting in the target area based on the first shooting command.

[0026] The first images captured are sorted and stored according to the order in which they were captured, thus obtaining the first image sequence acquired by the drone.

[0027] Preferably, in an engineering acceptance method based on the collaboration of unmanned aerial vehicles (UAVs) and unmanned vehicles (UAVs), step 2 involves controlling the UAV to conduct a second cruise and take pictures in the target area according to the second cruise trajectory, thereby obtaining a second image sequence collected by the UAV, including:

[0028] The command reading format of the autonomous vehicle is obtained, and the second cruise trajectory is converted according to the command reading format of the autonomous vehicle to obtain the third command element;

[0029] Read the preset unmanned vehicle image acquisition frequency and generate the fourth instruction element based on the preset unmanned vehicle image acquisition frequency;

[0030] The second shooting command is generated based on the third and fourth instruction units, and the unmanned vehicle is controlled to conduct a second cruise shooting in the target area according to the second shooting command. The second image sequence collected by the unmanned vehicle is obtained according to the shooting sequence.

[0031] Preferably, in step 3, the first image sequence and the second image sequence are associated and screened to obtain an effective image sequence set, which includes:

[0032] The obtained first image sequence and second image sequence are sequentially frame-divided to obtain an image frame set;

[0033] Edge detection is performed on each image frame in the image frame set to determine the object edge features recorded in each image frame. Meanwhile, the image acquisition angles of the unmanned aerial vehicle and the unmanned vehicle are determined based on the object edge features, and different manifestation features of the same object are determined based on the first cruise trajectory and the second cruise trajectory.

[0034] Based on the different manifestation features of the same object, the image frame sets of the first image sequence and the second image sequence are traversed, and the first image sequence and the second image sequence are associated according to the same object index based on the traversal result to obtain an image pair sequence.

[0035] Based on the image pair sequence, the first image sequence and the second image sequence are screened to obtain an effective image sequence set.

[0036] Preferably, in step 3, the first image sequence and the second image sequence are associated and screened to obtain an effective image sequence set, which includes:

[0037] An image pair of the first image sequence and the second image sequence is obtained, and the mapping relationship of the image frames of the same object in the first image sequence and the second image sequence is determined based on the image pair.

[0038] Based on the mapping relationship, it is determined whether there is a many-to-one relationship in the image pair of the same object. When there is a many-to-one relationship, it is determined that there is a duplicate image frame, and the duplicate image frame is selected and reserved.

[0039] Based on the selected and reserved result, an effective image pair is obtained, and the effective image pair is integrated and associated to obtain an effective image sequence set.

[0040] Preferably, in step 4, the effective image sequence set is analyzed to obtain a land reclamation acceptance result of the target area land, which includes:

[0041] An acceptance dimension for land reclamation acceptance of the target area land is obtained, and the effective image sequence set is analyzed according to the acceptance dimension. The land features of the target area land corresponding to each acceptance dimension are determined according to the analysis result.

[0042] Obtain the acceptance standard of each acceptance dimension, and match the land features corresponding to each acceptance dimension with the acceptance standard of each acceptance dimension to obtain the matching degree corresponding to each acceptance dimension, and take the matching degree corresponding to each acceptance dimension as the target score of each acceptance dimension;

[0043] Obtain the target weight of each acceptance dimension in the land reclamation acceptance result, and calculate according to the target weight of each acceptance dimension in the land reclamation acceptance result and the target score corresponding to each acceptance dimension to obtain the land reclamation acceptance comprehensive score of the target region land, wherein the land reclamation acceptance comprehensive score is the land reclamation acceptance result of the target region land.

[0044] Preferably, an engineering acceptance method based on cooperation of unmanned aerial vehicles and unmanned vehicles, according to geometric parameters, a first cruising track of the unmanned aerial vehicle in the target region land is formulated, comprising:

[0045] Read the geometric parameters to determine the geometric shape of the target region land and the longitudinal height distribution state of the target region land;

[0046] According to the geometric shape of the target region land and the longitudinal height distribution state of the target region land, a virtual target region land is simulated in the computer;

[0047] According to the longitudinal height distribution state of the target region land, different longitudinal height closed regions in the virtual target region land are divided to obtain a plurality of sub-virtual regions, wherein the longitudinal height of each sub-virtual region is consistent;

[0048] Obtain a preset flight starting point and a flight direction, and sequentially add a serial number label to each sub-virtual region according to the flight starting point and the flight direction;

[0049] Determine the first track element of the unmanned aerial vehicle in the virtual target region land according to the serial number label;

[0050] Map the plurality of sub-virtual regions in the geometric shape of the virtual target region land, and divide the geometric shape of the virtual target region land into a plurality of sub-geometric shapes according to the matching result, wherein the sub-geometric shapes correspond one-to-one to the sub-virtual regions;

[0051] Obtain a preset track flight management library;

[0052] Input the sub-geometric shape into the preset track flight management library for matching, and output the second track element of each sub-virtual region corresponding to each sub-geometric shape;

[0053] Determine the connection point of each adjacent two sub-virtual regions according to the serial number label, and determine the flight starting point and the flight ending point in each sub-virtual region according to the serial number value of each adjacent two sub-virtual regions and the connection point;

[0054] According to the flight starting point and the flight ending point in each sub-virtual area, the second track element of each sub-virtual area is adaptively adjusted to obtain a third track element;

[0055] The longitudinal height of each sub-virtual area is read, and the longitudinal height of each sub-virtual area is marked as a target label in the third track element to obtain a target third track element;

[0056] The first track element is associated with the plurality of target third track elements to obtain a first cruise track of the unmanned aerial vehicle in the target area land.

[0057] The application provides an engineering acceptance system based on cooperation of an unmanned aerial vehicle and an unmanned vehicle, comprising:

[0058] A cruise track determination module is configured to obtain a first cruise track of the unmanned aerial vehicle in the target area land, and simultaneously obtain a second cruise track of the unmanned vehicle in the target area land;

[0059] An image acquisition module is configured to control the unmanned aerial vehicle to perform first cruise shooting in the target area land according to the first cruise track, and obtain a first image sequence collected by the unmanned aerial vehicle, and simultaneously control the unmanned vehicle to perform second cruise shooting in the target area land according to the second cruise track, and obtain a second image sequence collected by the unmanned vehicle;

[0060] An image processing module is configured to associate and filter the first image sequence and the second image sequence to obtain an effective image sequence set;

[0061] An acceptance module is configured to analyze the effective image sequence set to obtain a land reclamation acceptance result of the target area land.

[0062] Compared with the prior art, the application has the following advantages:

[0063] The unmanned aerial vehicle performs shooting based on the first cruise track to obtain the first image sequence, and the unmanned vehicle performs shooting based on the second cruise track to obtain the second image sequence, which can effectively guarantee the integrity and comprehensiveness of land image collection, realize multi-angle analysis of land reclamation acceptance, help reclamation acceptance of the target area land from different angles by associating the first image sequence and the second image sequence, effectively guarantee the accuracy and quality of the effective image sequence by filtering, obtain the land reclamation acceptance result of the target area land by analyzing the effective image sequence set, effectively realize multi-source data acceptance of land reclamation acceptance, thereby improving the accuracy of land reclamation acceptance, and by controlling the unmanned aerial vehicle and the unmanned vehicle to perform shooting, effectively guarantee the intelligence of image collection, thereby improving the efficiency of land reclamation acceptance.

[0064] Other features and advantages of the present application will be set forth in the description that follows, and in part will be apparent from the description, or can be learned by practice of the application. The purposes and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims hereof.

[0065] The technical solutions of the present application are described in further detail below with the aid of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0066] The accompanying drawings are included to provide a further understanding of the present application and are incorporated in and constitute a part of the specification, illustrate embodiments of the present application and are used to explain the present application, but do not constitute a limitation of the present application. In the drawings:

[0067] Figure 1 A flowchart of an engineering acceptance method based on cooperation of unmanned aerial vehicles and unmanned vehicles in an embodiment of the present application;

[0068] Figure 2 A flowchart of step 1 in an engineering acceptance method based on cooperation of unmanned aerial vehicles and unmanned vehicles in an embodiment of the present application;

[0069] Figure 3 A structural diagram in an engineering acceptance system based on cooperation of unmanned aerial vehicles and unmanned vehicles in an embodiment of the present application. DETAILED DESCRIPTION

[0070] The preferred embodiments of the present application are described below with reference to the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and do not constitute a limitation of the present application.

[0071] Embodiment 1:

[0072] This embodiment provides an engineering acceptance method based on cooperation of unmanned aerial vehicles and unmanned vehicles, as shown in Figure 1 , comprising:

[0073] Step 1: Obtain a first cruising trajectory of the unmanned aerial vehicle on the target area land, and simultaneously, obtain a second cruising trajectory of the unmanned vehicle on the target area land;

[0074] Step 2: Control the unmanned aerial vehicle to perform first cruising shooting in the target area land according to the first cruising trajectory, and obtain a first image sequence collected by the unmanned aerial vehicle, and simultaneously, control the unmanned vehicle to perform second cruising shooting in the target area land according to the second cruising trajectory, and obtain a second image sequence collected by the unmanned vehicle;

[0075] Step 3: Associate and filter the first image sequence and the second image sequence to obtain an effective image sequence set;

[0076] Step 4: analyzing the set of effective image sequences to obtain the land reclamation acceptance result of the target area land.

[0077] In this embodiment, the first cruise trajectory can be a trajectory used to represent the operation of the UAV in the target area land for image collection, and the collected images constitute the first image sequence.

[0078] In this embodiment, the second cruise trajectory can be a trajectory used to represent the operation of the unmanned vehicle in the target area land for image collection, and the collected images constitute the second image sequence.

[0079] In this embodiment, the target area land can be a land area that needs to be subjected to land reclamation acceptance (i.e., project acceptance).

[0080] In this embodiment, the first image sequence and the second image sequence are associated and filtered, wherein the association means that the first image sequence and the second image sequence are associated according to the same area, i.e., the purpose of being able to simultaneously read the first image collected by the UAV and the second image collected by the unmanned vehicle in the same area is achieved; the filtering means that the repeated images existing in the first image sequence and the repeated images existing in the second image sequence are removed, thereby obtaining the effective image sequence.

[0081] In this embodiment, the set of effective image sequences is a set of images that can represent the target area land after removing the repetitions.

[0082] In this embodiment, the acceptance result is a comprehensive score of land reclamation acceptance obtained by analyzing the effective image sequence.

[0083] The working principle and beneficial effects of the above technical solution are as follows: the first image sequence obtained by the UAV based on the first cruise trajectory and the second image sequence obtained by the unmanned vehicle based on the second cruise trajectory can effectively guarantee the completeness and comprehensiveness of the land image collection, achieve the multi-angle analysis goal of land reclamation acceptance, the association of the first image sequence and the second image sequence helps to perform reclamation acceptance of the target area land from different angles, the filtering can effectively guarantee the accuracy and quality of the obtained effective image sequence, the analysis of the set of effective image sequences to obtain the land reclamation acceptance result of the target area land can effectively realize multi-source data acceptance of land reclamation acceptance, thereby improving the accuracy of land reclamation acceptance, at the same time, by controlling the UAV and the unmanned vehicle to perform image collection, the intelligence of image collection is effectively guaranteed, thereby improving the efficiency of land reclamation acceptance.

[0084] Embodiment 2:

[0085] On the basis of embodiment 1, the present embodiment provides a project acceptance method based on cooperation of a UAV and an unmanned vehicle, like Figure 2As shown, in step 1, a first cruising track of the unmanned aerial vehicle in the target area land is acquired, including:

[0086] The target area land is acquired for land reclamation acceptance, and the position distribution characteristics of the target area land are acquired;

[0087] The boundary contour of the target area land is positioned according to the position distribution characteristics of the target area land, and the boundary contour position point of the target area land is acquired;

[0088] The topographic characteristics of the target area land are acquired;

[0089] The geometric parameters of the target area land are determined according to the boundary contour position point of the target area land and the topographic characteristics of the target area land, and the first cruising track of the unmanned aerial vehicle in the target area land is formulated according to the geometric parameters.

[0090] In this embodiment, the topographic characteristics are the height distribution state of the target area land.

[0091] In this embodiment, the position distribution characteristics are the contour distribution of the target area land.

[0092] In this embodiment, the boundary contour position point refers to the boundary distribution point of the target area land.

[0093] In this embodiment, the first cruising track refers to the flight track of the unmanned aerial vehicle in the target area land.

[0094] The working principle and beneficial effects of the above technical solution are: by determining the position distribution characteristics of the target area land, the boundary contour point of the target area land is effectively determined, and the geometric parameters are effectively obtained, and the first cruising track of the unmanned aerial vehicle in the target area land is formulated through the geometric parameters, which effectively realizes the effectiveness and accuracy of the unmanned aerial vehicle flying in the target area land, and improves the working efficiency of the unmanned aerial vehicle.

[0095] Embodiment 3:

[0096] Based on embodiment 1, the embodiment provides an engineering acceptance method based on cooperation of unmanned aerial vehicles and unmanned vehicles, in step 1, a second cruising track of the unmanned vehicle in the target area land is acquired, including:

[0097] The land characteristics of the target area land are collected, and the static obstacles of the target area land are determined;

[0098] The land boundary point set of the target area land is acquired, and any land boundary point in the land boundary point set is taken as a coordinate origin to construct a two-dimensional coordinate system of the target area land;

[0099] acquire a first relative position relationship between the rest of the land boundary points in the land boundary point set and the coordinate origin, and determine a plurality of first coordinate points of the rest of the land boundary points in the two-dimensional coordinate system according to the first relative position relationship, and perform first labeling on the plurality of first coordinate points in the two-dimensional coordinate system to obtain a first regional plane;

[0100] acquire a static obstacle boundary point set, and acquire a second relative position relationship between each obstacle boundary point in the obstacle boundary point set and the coordinate origin, and determine a plurality of second coordinate points of the obstacle boundary point set in the two-dimensional coordinate system according to the second relative position relationship, and perform second labeling on the plurality of second coordinate points in the two-dimensional coordinate system to obtain a second regional plane;

[0101] construct a second cruise trajectory of the unmanned vehicle in the target region land according to the position distribution of the second regional plane in the first regional plane.

[0102] In this embodiment, the static obstacle is an object that cannot move and blocks the movement of the unmanned vehicle, such as a stone.

[0103] In this embodiment, the land feature is the distribution of obstacles and buildings in the target region land.

[0104] In this embodiment, the first regional plane is a plane formed by the land boundary points in the two-dimensional coordinate system (two-dimensional rectangular coordinate system).

[0105] In this embodiment, the second regional plane is a plane formed by the obstacle boundary point set of the static obstacle in the two-dimensional coordinate system, and the second regional plane belongs to the first regional plane.

[0106] In this embodiment, the second cruise trajectory is the walking path of the unmanned vehicle in the target region land.

[0107] The working principle and beneficial effects of the above technical solution are: by determining the first regional plane and the second regional plane, the second cruise trajectory of the unmanned vehicle in the target region land is effectively constructed, thereby effectively realizing the effectiveness and accuracy of the walking path of the unmanned vehicle, and improving the operation efficiency of the unmanned vehicle in the target region land.

[0108] Embodiment 4:

[0109] Based on the embodiment 1, the embodiment provides an engineering acceptance method based on cooperation of unmanned aerial vehicle and unmanned vehicle, in step 2, according to the first cruise trajectory, the unmanned aerial vehicle is controlled to perform first cruise shooting in the target region land, and a first image sequence collected by the unmanned aerial vehicle is obtained, including:

[0110] read the first cruise trajectory, determine the first cruise trajectory data, and acquire the instruction reading format of the unmanned aerial vehicle;

[0111] The first cruise trajectory data is converted into a first instruction element according to the instruction reading format of the unmanned aerial vehicle, meanwhile, a preset unmanned aerial vehicle image acquisition frequency is obtained, and a second instruction element is generated according to the preset unmanned aerial vehicle image acquisition frequency;

[0112] The first shooting instruction is generated according to the first instruction element and the second instruction element, and the unmanned aerial vehicle is controlled to perform first cruise shooting in the target area land according to the first shooting instruction.

[0113] The first image after shooting is sorted and stored according to the shooting sequence, and a first image sequence collected by the unmanned aerial vehicle is obtained.

[0114] In this embodiment, the instruction reading format is set in advance to represent the inherent format of the instruction reading of the unmanned aerial vehicle.

[0115] In this embodiment, the first shooting instruction is used to realize the instruction of the first cruise shooting of the unmanned aerial vehicle in the target area land, which is composed of the first instruction element and the second instruction element.

[0116] In this embodiment, the first image sequence refers to the image set collected by the unmanned aerial vehicle.

[0117] The working principle and beneficial effects of the above technical solution are as follows: by performing format conversion on the instruction reading format of the unmanned aerial vehicle, the first instruction element is effectively obtained, the second instruction element is obtained through the image acquisition frequency, and then the first shooting instruction is effectively obtained based on the first instruction element and the second instruction element, which is beneficial to guarantee the accuracy and intelligence of the flight control of the unmanned aerial vehicle in the target area land, and the correlation and order of image acquisition are effectively guaranteed through the acquisition of the first image sequence.

[0118] Embodiment 5:

[0119] Based on the embodiment 1, the embodiment provides an engineering acceptance method based on cooperation of unmanned aerial vehicle and unmanned vehicle, in step 2, the unmanned vehicle is controlled to perform second cruise shooting in the target area land according to the second cruise trajectory, and a second image sequence collected by the unmanned vehicle is obtained, including:

[0120] The instruction reading format of the unmanned vehicle is obtained, and the second cruise trajectory is format-converted according to the instruction reading format of the unmanned vehicle to obtain a third instruction element;

[0121] A preset unmanned vehicle image acquisition frequency is read, and a fourth instruction element is generated according to the preset unmanned vehicle image acquisition frequency;

[0122] The second shooting instruction is generated according to the third instruction element and the fourth instruction element, the unmanned vehicle is controlled to perform second cruise shooting in the target area land according to the second shooting instruction, and the second image sequence collected by the unmanned vehicle is obtained according to the shooting sequence.

[0123] In this embodiment, the third instruction element is the result of format conversion of the second cruise trajectory based on the instruction reading format of the unmanned vehicle, wherein the instruction reading format of the unmanned vehicle is also set in advance.

[0124] In this embodiment, the second shooting instruction is an instruction for controlling the unmanned aerial vehicle to perform second cruise shooting in the target area land, which is composed of the third instruction element and the fourth instruction element.

[0125] In this embodiment, the second image sequence refers to the image set collected by the unmanned vehicle.

[0126] The working principle and beneficial effects of the above technical solution are: by generating the second shooting instruction, the accuracy and intelligence of the second cruise shooting of the unmanned vehicle are effectively realized, and by acquiring the second image sequence, the relevance and order of image acquisition are effectively guaranteed.

[0127] Embodiment 6:

[0128] Based on embodiment 1, this embodiment provides an engineering acceptance method based on the cooperation of unmanned aerial vehicles and unmanned vehicles. In step 3, the first image sequence and the second image sequence are associated and screened to obtain an effective image sequence set, including:

[0129] The first image sequence and the second image sequence are obtained, and the first image sequence and the second image sequence are sequentially frame-divided to obtain an image frame set;

[0130] Edge detection is performed on each image frame in the image frame set to determine the object edge features recorded in each image frame. At the same time, based on the object edge features, the image acquisition angles of the unmanned aerial vehicle and the unmanned vehicle are determined, and based on the first cruise trajectory and the second cruise trajectory, the different performance characteristics of the same object under the image acquisition angles of the unmanned aerial vehicle and the unmanned vehicle are determined.

[0131] Based on the different performance characteristics of the same object, the image frame sets of the first image sequence and the second image sequence are traversed, and based on the traversal results, the first image sequence and the second image sequence are associated according to the same object index to obtain an image pair sequence.

[0132] Based on the image pair sequence, the first image sequence and the second image sequence are screened to obtain an effective image sequence set.

[0133] In this embodiment, frame division refers to the splitting of the image corresponding to each time in the first image sequence and the second image sequence, i.e., obtaining the static image corresponding to each time.

[0134] In this embodiment, edge detection refers to analyzing the image edges of each image frame, aiming to determine the edge features (i.e. edge shape, etc.) of the objects recorded in each image frame.

[0135] In this embodiment, different performance features refer to different appearance features of the same object under the image acquisition angle of the UAV and the unmanned vehicle.

[0136] In this embodiment, object traversal refers to determining the soil area or object recorded in each image frame in the image frame set of the first image sequence and the second image sequence.

[0137] In this embodiment, the same object indicator refers to a unified recording subject.

[0138] In this embodiment, the image pair sequence refers to associating the images of the unified object recorded in the first image sequence and the second image sequence, that is, each image pair sequence contains a UAV image and an unmanned vehicle image.

[0139] The working principle and beneficial effects of the above technical solution are: by dividing the first image sequence and the second image sequence into frames, the image frame set is determined, at the same time, the edge detection of each image frame in the image frame set is carried out, the image acquisition angle of the UAV and the unmanned vehicle is locked according to the edge detection result, then the different performance features of the same object under the image acquisition angle of the UAV and the unmanned vehicle are determined, finally, the image frames of the unified object in the image frame set of the first image sequence and the second image sequence are associated according to the performance features, and the association result is screened, and finally the effective image sequence set is determined, which provides reliable image support and convenience for land reclamation acceptance.

[0140] Embodiment 7:

[0141] Based on embodiment 6, the embodiment provides an engineering acceptance method based on the cooperation of UAV and unmanned vehicle, the first image sequence and the second image sequence are screened based on the association result, and the effective image sequence set is obtained, including:

[0142] Obtain the image pairs of the first image sequence and the second image sequence, and determine the mapping relationship of the image frames of the same object in the first image sequence and the second image sequence based on the image pairs;

[0143] Determine whether there is a many-to-one relationship in the image pairs of the same object based on the mapping relationship, and when there is a many-to-one relationship, determine that there is a duplicate image frame, and retain one of the duplicate image frames;

[0144] Obtain the effective image pairs based on the one-to-one retention result, and integrate and associate the effective image pairs to obtain the effective image sequence set.

[0145] The working principle and beneficial effects of the above technical solution are: by analyzing the mapping relationship of the image frames of the same object in the first image sequence and the second image sequence in the image pair, effectively determining the repeated image frames when there are multiple one-to-one, and realizing the selection and retention of repeated images, effectively guaranteeing the accuracy of obtaining the effective image sequence set, and providing effective data support for the engineering acceptance of the cooperation of the unmanned aerial vehicle and the unmanned vehicle.

[0146] Embodiment 8:

[0147] Based on the embodiment 1, the embodiment provides an engineering acceptance method based on the cooperation of the unmanned aerial vehicle and the unmanned vehicle, in step 4, the effective image sequence set is analyzed to obtain the land reclamation acceptance result of the target area land, including:

[0148] The acceptance dimensions of the land reclamation acceptance of the target area land are obtained, and the effective image sequence set is analyzed according to the acceptance dimensions, and the land features corresponding to each acceptance dimension of the target area land are determined according to the analysis result;

[0149] The acceptance standards of each acceptance dimension are obtained, and the land features corresponding to each acceptance dimension are matched with the acceptance standards of each acceptance dimension, and the matching degree corresponding to each acceptance dimension is obtained, and the matching degree corresponding to each acceptance dimension is taken as the target score of each acceptance dimension;

[0150] The target weight of each acceptance dimension in the land reclamation acceptance result is obtained, and the target weight of each acceptance dimension in the land reclamation acceptance result and the target score corresponding to each acceptance dimension are calculated to obtain the comprehensive score of the land reclamation acceptance of the target area land, wherein the comprehensive score of the land reclamation acceptance is the land reclamation acceptance result of the target area land.

[0151] In this embodiment, the acceptance dimensions of the land reclamation acceptance of the target area land include: vegetation coverage acceptance, soil quality acceptance (soil color, texture and other features), and topographic reconstruction acceptance.

[0152] In this embodiment, the acceptance standards are the indicators corresponding to the qualified of each acceptance dimension, for example: the standard coverage of vegetation, the soil color, texture and other features corresponding to the good soil features, and the standard topographic features.

[0153] In this embodiment, the target weight is the influence degree of each acceptance dimension on the land reclamation acceptance result, which is helpful to realize the objectivity and effectiveness of obtaining the comprehensive score of the land reclamation acceptance based on the target weight.

[0154] The working principle and beneficial effects of the technical solution are as follows: by acquiring the acceptance dimensions for land reclamation acceptance of the target area land and the land characteristics of the target area land corresponding to each acceptance dimension, the acquisition of the matching degree corresponding to each acceptance dimension is effectively realized, and then the matching degree corresponding to each acceptance dimension is taken as the target score of each acceptance dimension. The target score is comprehensively calculated according to the target weight of each acceptance dimension in the land reclamation acceptance result, so as to effectively obtain the land reclamation acceptance result of the target area land, improve the accuracy of obtaining the land reclamation acceptance result of the target area land, and help to ensure the objectivity and effectiveness of obtaining the comprehensive score of land reclamation acceptance, thereby improving the efficiency of land reclamation acceptance.

[0155] Embodiment 9:

[0156] Based on the embodiment 2, the embodiment provides an engineering acceptance method based on cooperation of unmanned aerial vehicle and unmanned vehicle, the first cruising track of the unmanned aerial vehicle in the target area land is formulated according to the geometric parameters, comprising:

[0157] Reading the geometric parameters, determining the geometric shape of the target area land and the longitudinal height distribution state of the target area land;

[0158] According to the geometric shape of the target area land and the longitudinal height distribution state of the target area land, simulation is carried out in the computer to obtain a virtual target area land;

[0159] According to the longitudinal height distribution state of the target area land, the closed regions with different longitudinal heights in the virtual target area land are divided to obtain a plurality of sub-virtual regions, wherein the longitudinal height of each sub-virtual region is consistent;

[0160] Obtaining a preset flight starting point and a flight direction, and sequentially adding a serial number label to each sub-virtual region according to the flight starting point and the flight direction;

[0161] According to the serial number label, the first track element of the unmanned aerial vehicle in the virtual target area land is determined;

[0162] The plurality of sub-virtual regions are matched and mapped in the geometric shape of the virtual target area land, and the geometric shape of the virtual target area land is divided into a plurality of sub-geometric shapes according to the matching result, wherein the sub-geometric shapes correspond one by one to the sub-virtual regions;

[0163] Obtaining a preset track flight management library;

[0164] The sub-geometric shapes are input into the preset track flight management library for matching, and the second track element of each sub-virtual region corresponding to each sub-geometric shape is output;

[0165] determine the connection point of each two adjacent sub-virtual areas according to the serial number label, and determine the flight start point and the flight end point in each sub-virtual area according to the serial number value of each two adjacent sub-virtual areas and the connection point;

[0166] adaptively adjust the second track element of each sub-virtual area according to the flight start point and the flight end point in each sub-virtual area, and obtain a third track element;

[0167] read the longitudinal height of each sub-virtual area, and mark the longitudinal height of each sub-virtual area as a target label in the third track element to obtain a target third track element;

[0168] associate the first track element with the plurality of target third track elements to obtain a first cruising track of the unmanned aerial vehicle in the target area land.

[0169] In this embodiment, the geometric parameter refers to data information capable of representing the shape of the target area.

[0170] In this embodiment, the longitudinal height distribution state refers to the distribution of different horizontal heights in the target area land, for example, the different address heights corresponding to the valleys and peaks of the target area land are the longitudinal height distribution state.

[0171] In this embodiment, the virtual target area land refers to a simulation model of the target area land obtained by simulating the target area land in the computer according to the longitudinal height distribution state of the target area land.

[0172] In this embodiment, the closed area refers to a plane of different longitudinal heights in the virtual target area land, and the plane of one height is a closed area.

[0173] In this embodiment, the sub-virtual area refers to a different height corresponding to the land area obtained by dividing the closed area of different longitudinal heights in the virtual target area land.

[0174] In this embodiment, the serial number label refers to a symbol for marking different sub-virtual areas, and the order of image acquisition of the different sub-virtual areas corresponding to the land area can be determined through the serial number label.

[0175] In this embodiment, the first track element is determined according to the serial number label, and the order is determined as the serial number label increases.

[0176] In this embodiment, the sub-geometric shape refers to the result obtained by mapping the plurality of sub-virtual areas in the geometric shape of the virtual target area land.

[0177] In this embodiment, the preset track flight library pre-stores a plurality of geometric shapes and an optimal cruise track corresponding to each sub-geometric shape.

[0178] In this embodiment, the second track element refers to a flight route in the corresponding land area of each sub-virtual area when image acquisition is performed on the corresponding land area of each sub-virtual area.

[0179] In this embodiment, the connection point refers to a boundary line of two adjacent sub-virtual areas.

[0180] In this embodiment, the flight start point and the flight end point in each sub-virtual area according to the serial number value of each adjacent two sub-virtual areas and the connection point refer to that the flight order of each sub-virtual area is determined according to the increasing order of the serial number label (i.e., the serial number value), and the flight start point and the flight end point of each sub-virtual area are determined according to the connection point. The flight start point and the flight end point of each sub-virtual area can be set as the connection point, the purpose is to ensure that the unmanned aerial vehicle can comprehensively and effectively collect images of the corresponding land area.

[0181] In this embodiment, the third track element refers to a result obtained by adjusting the second track element of each sub-virtual area according to the determined flight start point and the flight end point in each sub-virtual area, that is, the finally corresponding track.

[0182] In this embodiment, the target third track element refers to a result obtained by adding a flight height label to the obtained third track element, the purpose is to guide the flight height of the unmanned aerial vehicle.

[0183] The working principle and beneficial effects of the technical scheme are as follows: by determining the geometric shape and longitudinal height distribution state of the target area land, the target area land is simulated in the computer to obtain an accurate and effective virtual target area land; secondly, different longitudinal height closed areas in the virtual target area land are divided according to the longitudinal height distribution state of the target area land, the sub-virtual areas are determined, and the sequence number labels are added to the plurality of sub-virtual areas according to the preset flight starting point and flight direction, so that the first track element of the unmanned aerial vehicle in the virtual target area land is determined; finally, the plurality of sub-virtual areas are matched and mapped in the geometric shape of the virtual target area land, so that the corresponding sub-geometric shape of each sub-virtual area is determined, and the second track element of each sub-virtual area corresponding to each sub-geometric shape is determined according to the preset track flight management library; meanwhile, the flight starting point and the flight ending point in each sub-virtual area are determined according to the sequence number label and the connection point of each adjacent two sub-virtual areas; finally, the second track element of each sub-virtual area is adaptively adjusted according to the flight starting point and the flight ending point in each sub-virtual area to obtain a third track element, the flight height of the third track element is marked according to the longitudinal height of each sub-virtual area, and finally the first track element is associated with a plurality of target third track elements according to the marking result, so that the first cruising track of the unmanned aerial vehicle on the target area land is locked, and the efficiency and accuracy of the land reclamation acceptance are improved.

[0184] Embodiment 10:

[0185] The embodiment provides an engineering acceptance system based on cooperation of an unmanned aerial vehicle and an unmanned vehicle, as shown in the figure, which comprises: Figure 3

[0186] A cruising track determination module is configured to acquire a first cruising track of the unmanned aerial vehicle on the target area land, and simultaneously acquire a second cruising track of the unmanned vehicle on the target area land.

[0187] An image acquisition module is configured to control the unmanned aerial vehicle to perform first cruising shooting in the target area land according to the first cruising track, and obtain a first image sequence collected by the unmanned aerial vehicle; and simultaneously control the unmanned vehicle to perform second cruising shooting in the target area land according to the second cruising track, and obtain a second image sequence collected by the unmanned vehicle.

[0188] An image processing module is configured to associate and screen the first image sequence and the second image sequence, and obtain an effective image sequence set.

[0189] An acceptance module is configured to analyze the effective image sequence set, and obtain a land reclamation acceptance result of the target area land.

[0190] ​The working principle and beneficial effects of the above technical solution are: the unmanned aerial vehicle obtains the first image sequence by shooting based on the first cruise trajectory, and the unmanned vehicle obtains the second image sequence by shooting based on the second cruise trajectory, which can effectively guarantee the integrity and comprehensiveness of the land picture collection, realize the multi-angle analysis target of land reclamation acceptance, and help to conduct land reclamation acceptance of the target area land from different angles by correlating the first image sequence and the second image sequence. The accuracy and quality of the effective image sequence obtained can be effectively guaranteed by screening, and the land reclamation acceptance result of the target area land can be obtained by analyzing the effective image sequence set, which can effectively realize multi-source data acceptance of land reclamation acceptance, thereby improving the accuracy of land reclamation acceptance. At the same time, by controlling the unmanned aerial vehicle and the unmanned vehicle to shoot, the intelligence of image collection is effectively guaranteed, thereby improving the efficiency of land reclamation acceptance.

[0191] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. An engineering acceptance method based on cooperation of a UAV and an unmanned vehicle, characterized in that, The method comprises the following steps: Step 1: Obtain a first cruising track of a UAV in a target area of land, and simultaneously obtain a second cruising track of an unmanned vehicle in the target area of land; Step 2: Control the UAV to perform first cruising shooting in the target area of land according to the first cruising track, and obtain a first image sequence collected by the UAV; simultaneously, control the unmanned vehicle to perform second cruising shooting in the target area of land according to the second cruising track, and obtain a second image sequence collected by the unmanned vehicle; Step 3: Associate and screen the first image sequence and the second image sequence to obtain an effective image sequence set; Step 4: Analyze the effective image sequence set to obtain a land reclamation acceptance result of the target area of land; In step 3, the first image sequence and the second image sequence are associated and screened to obtain an effective image sequence set, which comprises the following steps: Obtain the first image sequence and the second image sequence, and sequentially perform frame division on the first image sequence and the second image sequence to obtain an image frame set; Perform edge detection on each image frame in the image frame set to determine object edge features recorded in each image frame, simultaneously determine image collection viewing angles of the UAV and the unmanned vehicle based on the object edge features, and determine different performance features of a same object under the image collection viewing angles of the UAV and the unmanned vehicle based on the first cruising track and the second cruising track respectively; Perform object traversal on the image frame set of the first image sequence and the second image sequence based on the different performance features of the same object, and associate the first image sequence and the second image sequence according to the same object index based on the object traversal result to obtain an image pair sequence; Screen the first image sequence and the second image sequence based on the image pair sequence to obtain the effective image sequence set.

2. The method of claim 1, wherein, In step 1, the first cruising track of the UAV in the target area of land is obtained, which comprises the following steps: Obtain a target area of land for land reclamation acceptance, and obtain position distribution characteristics of the target area of land; Position a boundary contour of the target area of land according to the position distribution characteristics of the target area of land, and determine a boundary contour position point of the target area of land; Obtain terrain characteristics of the target area of land; Determine geometric parameters of the target area of land according to the boundary contour position point of the target area of land and the terrain characteristics of the target area of land, and formulate the first cruising track of the UAV in the target area of land according to the geometric parameters. 3.The method of claim 1, wherein, In step 1, the second cruising track of the unmanned vehicle in the target area of land is obtained, which comprises the following steps: Collect land characteristics of the target area of land to determine static obstacles of the target area of land; Obtain a land boundary point set of the target area of land, and take any land boundary point in the land boundary point set as a coordinate origin to construct a two-dimensional coordinate system of the target area of land; Obtain a first relative position relationship between the remaining land boundary points in the land boundary point set and the coordinate origin, and determine a plurality of first coordinate points of the remaining land boundary points in the two-dimensional coordinate system according to the first relative position relationship, and perform first labeling on the plurality of first coordinate points in the two-dimensional coordinate system to obtain a first area plane; Obtain a static obstacle boundary point set, and obtain a second relative position relationship between each obstacle boundary point in the obstacle boundary point set and the coordinate origin, and determine a second coordinate point of the obstacle boundary point set in a two-dimensional coordinate system according to the second relative position relationship, and meanwhile, secondly mark a plurality of second coordinate points in the two-dimensional coordinate system to obtain a second region plane; According to the position distribution of the second region plane in the first region plane, a second cruise track of the unmanned vehicle in the target region land is constructed.

4. The method of claim 1, wherein, In step 2, the first cruise track is controlled to control the unmanned aerial vehicle to perform first cruise shooting in the target region land, and a first image sequence collected by the unmanned aerial vehicle is obtained, including: The first cruise track is read to determine the first cruise track data, and meanwhile, the instruction reading format of the unmanned aerial vehicle is obtained; The first cruise track data is converted into first instruction elements according to the instruction reading format of the unmanned aerial vehicle, and meanwhile, a preset unmanned aerial vehicle image collection frequency is obtained, and second instruction elements are generated according to the preset unmanned aerial vehicle image collection frequency; The first shooting instruction is generated according to the first instruction elements and the second instruction elements, and meanwhile, the unmanned aerial vehicle is controlled to perform first cruise shooting in the target region land according to the first shooting instruction; The first images after shooting are sorted and stored according to the shooting sequence to obtain the first image sequence collected by the unmanned aerial vehicle.

5. The method of claim 1, wherein, In step 2, the second cruise track is controlled to control the unmanned vehicle to perform second cruise shooting in the target region land, and a second image sequence collected by the unmanned vehicle is obtained, including: The instruction reading format of the unmanned vehicle is obtained, and the second cruise track is converted into third instruction elements according to the instruction reading format of the unmanned vehicle; The preset unmanned vehicle image collection frequency is read, and fourth instruction elements are generated according to the preset unmanned vehicle image collection frequency; The second shooting instruction is generated according to the third instruction elements and the fourth instruction elements, and the unmanned vehicle is controlled to perform second cruise shooting in the target region land according to the second shooting instruction, and the second image sequence collected by the unmanned vehicle is obtained according to the shooting sequence.

6. The method of claim 1, wherein, The first image sequence and the second image sequence are screened based on the correlation result to obtain an effective image sequence set, including: An image pair of the first image sequence and the second image sequence is obtained, and a mapping relationship of image frames of the same object in the first image sequence and the second image sequence is determined based on the image pair; Based on the mapping relationship, it is determined whether there is many-to-one in the image pair of the same object, and when there is many-to-one, it is determined that there is a duplicate image frame, and the duplicate image frame is selected and reserved; Based on the selected and reserved result, an effective image pair is obtained, and the effective image pair is integrated and correlated to obtain the effective image sequence set.

7. The method of claim 1, wherein, In step 4, the effective image sequence set is analyzed to obtain a land reclamation acceptance result of the target region land, including: An acceptance dimension for land reclamation acceptance of the target region land is obtained, and the effective image sequence set is analyzed according to the acceptance dimension, and the land features of the target region land corresponding to each acceptance dimension are determined according to the analysis result; The acceptance standards of each acceptance dimension are obtained, and the land features corresponding to each acceptance dimension are matched with the acceptance standards of each acceptance dimension to obtain a matching degree corresponding to each acceptance dimension, and the matching degree corresponding to each acceptance dimension is taken as a target score of each acceptance dimension; The target weight of each acceptance dimension in the land reclamation acceptance result is obtained, and the target weight of each acceptance dimension in the land reclamation acceptance result and the target score corresponding to each acceptance dimension are calculated to obtain a land reclamation acceptance comprehensive score of the target region land, wherein the land reclamation acceptance comprehensive score is the land reclamation acceptance result of the target region land.

8. The method of claim 2, wherein, According to the geometric parameters, a first cruising track of the unmanned aerial vehicle in the target region land is formulated, comprising: reading the geometric parameters to determine the geometric shape of the target region land and the longitudinal height distribution state of the target region land; According to the geometric shape of the target region land and the longitudinal height distribution state of the target region land, a virtual target region land is simulated in the computer; According to the longitudinal height distribution state of the target region land, the closed regions of different longitudinal heights in the virtual target region land are divided to obtain a plurality of sub-virtual regions, wherein the longitudinal height of each sub-virtual region is consistent; a preset flight starting point and a flight direction are obtained, and each sub-virtual region is sequentially labeled with a serial number according to the flight starting point and the flight direction; According to the serial number label, a first track element of the unmanned aerial vehicle in the virtual target region land is determined; The plurality of sub-virtual regions are matched and mapped in the geometric shape of the virtual target region land, and the geometric shape of the virtual target region land is divided into a plurality of sub-geometric shapes according to the matching result, wherein the sub-geometric shapes correspond to the sub-virtual regions one by one; a preset track flight management library is obtained; The sub-geometric shapes are input into the preset track flight management library for matching, and the second track element of each sub-virtual region corresponding to each sub-geometric shape is output; According to the serial number label, the connection points of each adjacent two sub-virtual regions are determined, and the flight starting point and the flight termination point in each sub-virtual region are determined according to the serial number value of each adjacent two sub-virtual regions and the connection points; According to the flight starting point and the flight termination point in each sub-virtual region, the second track element of each sub-virtual region is adaptively adjusted to obtain a third track element; The longitudinal height of each sub-virtual region is read, and the longitudinal height of each sub-virtual region is taken as a target label in the third track element to obtain a target third track element; The first track element is associated with the plurality of target third track elements to obtain the first cruising track of the unmanned aerial vehicle in the target region land.

9. An engineering acceptance system based on cooperation of a UAV and an unmanned vehicle, characterized in that, Comprising: a cruising track determination module for obtaining a first cruising track of the unmanned aerial vehicle in the target region land, and simultaneously obtaining a second cruising track of the unmanned vehicle in the target region land; an image acquisition module for controlling the unmanned aerial vehicle to perform first cruising shooting in the target region land according to the first cruising track to obtain a first image sequence acquired by the unmanned aerial vehicle, and simultaneously controlling the unmanned vehicle to perform second cruising shooting in the target region land according to the second cruising track to obtain a second image sequence acquired by the unmanned vehicle; An image processing module is configured to associate and filter the first image sequence and the second image sequence to obtain an effective image sequence set; An acceptance module is configured to analyze the effective image sequence set to obtain a land reclamation acceptance result of the target area land; In the image processing module, the first image sequence and the second image sequence are associated and filtered to obtain the effective image sequence set, including: The obtained first image sequence and second image sequence are subjected to frame division in sequence to obtain an image frame set; Edge detection is performed on each image frame in the image frame set to determine object edge features recorded in each image frame. Meanwhile, the image collection visual angle of the unmanned aerial vehicle and the unmanned vehicle is determined based on the object edge features, and different performance features of the same object under the image collection visual angle of the unmanned aerial vehicle and the unmanned vehicle are determined based on the first cruise track and the second cruise track, respectively; The image frame set of the first image sequence and the second image sequence is subjected to object traversal based on the different performance features of the same object, and the first image sequence and the second image sequence are associated according to the same object index based on the object traversal result to obtain an image pair sequence; The first image sequence and the second image sequence are filtered based on the image pair sequence to obtain the effective image sequence set.

Images