Unmanned aerial vehicle inspection image reconstruction method and device, electronic equipment and storage medium

Abstract

The invention discloses an unmanned aerial vehicle inspection image reconstruction method and device, electronic equipment and a storage medium, and the method comprises the steps: determining a target inspection region of a target unmanned aerial vehicle; controlling the target unmanned aerial vehicle to execute inspection operation on the target inspection area to obtain a target inspection image; acquiring a plurality of camera images of a plurality of cameras; and performing reconstruction operation on the target inspection image based on the plurality of camera images to obtain a target reconstruction image. According to the method, the technical problem of how to efficiently and precisely reconstruct the inspection image of the unmanned aerial vehicle is solved.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a method, device, electronic device and storage medium for reconstructing unmanned aerial vehicle inspection images. Background Art

[0002] As the level of intelligence and automation of power systems continues to improve, the inspection method of substations has gradually shifted from manual inspections to drone inspections to improve inspection efficiency and reduce labor costs. Drone inspections, especially indoor drone inspections, show great potential in power facility maintenance and fault detection. However, during the inspection process, the indoor environment of the substation is complex, and the accuracy of the camera carried by the drone is often difficult to meet the requirements of high-precision image reconstruction due to weight and size limitations. In addition, the viewing angle and field of view of the drone camera are limited, which makes it difficult to fully capture the on-site information during the image reconstruction process, especially when the drone needs to focus on a specific inspection object. If the image reconstruction only relies on the data of the drone camera, it may cause information loss in the non-inspection area, thereby affecting the reconstruction quality and comprehensiveness of the entire site.

[0003] Existing drone inspection systems usually use drones to perform multiple flight inspections of the inspection area to ensure that the drone takes pictures of every corner of the inspection area during the flight, thereby improving the quality of the reconstructed image. However, the above method will reduce the efficiency of drone inspections.

[0004] There is currently no corresponding solution to the above problems. Summary of the invention

[0005] The embodiments of the present invention provide a method, device, electronic device and storage medium for reconstructing drone inspection images, so as to at least solve the technical problem of how to efficiently and accurately reconstruct drone inspection images.

[0006] According to one aspect of an embodiment of the present invention, a method for reconstructing an unmanned aerial vehicle inspection image is provided, comprising: determining a target inspection area of ​​a target unmanned aerial vehicle; controlling the target unmanned aerial vehicle to perform an inspection operation on the target inspection area to obtain a target inspection image; acquiring multiple camera images of multiple cameras; and performing a reconstruction operation on the target inspection image based on the multiple camera images to obtain a target reconstructed image.

[0007] Optionally, determining the target inspection area of ​​the target UAV includes: acquiring multiple inspection objects of the target UAV; and determining the target inspection area according to the multiple inspection objects.

[0008] Optionally, controlling the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image includes: controlling the target UAV to perform an inspection operation on the target inspection area to obtain an initial inspection image; and performing a reconstruction operation on the initial inspection image to obtain a target inspection image.

[0009] Optionally, performing a reconstruction operation on the target inspection image based on multiple camera images to obtain a target reconstructed image includes: performing a fusion operation on the target inspection image based on multiple camera images to obtain a fused image; performing a reconstruction operation on the fused image based on a preset deep learning algorithm to obtain a target reconstructed image, wherein the target reconstructed image includes: a first resolution image.

[0010] Optionally, the method for reconstructing the drone inspection image also includes: acquiring multiple position information of multiple cameras; and determining the target modeling area according to the multiple position information.

[0011] Optionally, the method for reconstructing the UAV inspection image also includes: determining a target reconstruction area based on a target modeling area and a target inspection area, wherein the target reconstruction area is an area within the target modeling area excluding the target inspection area; obtaining a preset reconstructed image corresponding to the target reconstruction area; performing a reconstruction operation on the preset reconstructed image based on a preset deep learning algorithm to obtain a second resolution image, wherein the image resolution of the first resolution image is higher than that of the second resolution image.

[0012] According to another aspect of an embodiment of the present invention, a device for reconstructing an unmanned aerial vehicle inspection image is also provided, including: a first determination module, used to determine a target inspection area of ​​a target unmanned aerial vehicle; a control module, used to control the target unmanned aerial vehicle to perform an inspection operation on the target inspection area to obtain a target inspection image; a second acquisition module, used to acquire multiple camera images of multiple cameras; and a first execution module, used to perform a reconstruction operation on the target inspection image based on the multiple camera images to obtain a target reconstructed image.

[0013] According to another aspect of an embodiment of the present invention, an electronic device is also provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any of the above-mentioned methods for reconstructing drone inspection images.

[0014] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, wherein the computer program is configured to execute any of the above-mentioned methods for reconstructing drone inspection images when running on a computer or a processor.

[0015] According to another aspect of an embodiment of the present invention, a computer program product is also provided, including a computer program, which implements the method for reconstructing an unmanned aerial vehicle inspection image in any of the above items when executed by a processor.

[0016] In an embodiment of the present invention, a target inspection area of ​​a target drone is determined; the target drone is controlled to perform an inspection operation on the target inspection area to obtain a target inspection image; multiple camera images of multiple cameras are acquired; and a reconstruction operation is performed on the target inspection image based on the multiple camera images to obtain a target reconstructed image. The present invention optimizes the target inspection image by combining the camera images acquired by multiple cameras, thereby achieving the purpose of avoiding the drone from performing multiple flight inspections on the inspection area to improve the quality of the reconstructed image, thereby achieving the technical effect of efficiently and accurately reconstructing the drone inspection image, and further solving the technical problem of how to efficiently and accurately reconstruct the drone inspection image. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1 is a flow chart of a method for reconstructing an unmanned aerial vehicle inspection image according to one embodiment of the present invention;

[0019] Figure 2 is a schematic diagram of a method for reconstructing an unmanned aerial vehicle inspection image according to one embodiment of the present invention;

[0020] Figure 3 4 is a structural block diagram of a device for reconstructing UAV inspection images according to one embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] According to an embodiment of the present invention, an embodiment of a method for reconstructing an unmanned aerial vehicle inspection image is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system comprising at least one set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0024] The method embodiment may also be performed in an electronic device, a similar control device or an electronic device including a memory and a processor. Taking an electronic device as an example, the electronic device may include one or more processors and a memory for storing data. Optionally, the electronic device may also include a communication device and a display device for communication functions. It will be understood by those skilled in the art that the above structural description is only illustrative and does not limit the structure of the above electronic device. For example, the electronic device may also include more or fewer components than the above structural description, or have a configuration different from the above structural description.

[0025] The processor may include one or more processing units. For example, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processing (DSP) chip, a microcontroller unit (MCU), a field-programmable gate array (FPGA), a neural-network processing unit (NPU), a tensor processing unit (TPU), an artificial intelligence (AI) type processor, and other processing devices. Among them, different processing units may be independent components or integrated into one or more processors. In some instances, the electronic device may also include one or more processors.

[0026] The memory can be used to store computer programs, such as storing a computer program corresponding to the method for reconstructing the inspection image of the unmanned aerial vehicle in the embodiment of the present invention. The processor implements the above-mentioned method for reconstructing the inspection image of the unmanned aerial vehicle by running the computer program stored in the memory. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the electronic device via a grid. Examples of the above-mentioned grid include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0027] The communication device is used to receive or send data via a grid. The specific examples of the above-mentioned grid may include a wireless grid provided by a communication provider of the mobile terminal. In one example, the communication device includes a grid adapter (network interface controller, NIC), which can be connected to other grid devices through a base station so as to communicate with the Internet. In one example, the communication device can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly. In some embodiments of the present solution, the communication device is used to connect to mobile devices such as mobile phones and tablets, and instructions can be sent to the vehicle-mounted terminal through the mobile device.

[0028] The display device may be a touch screen type liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or "touch display screen"). The liquid crystal display may enable a user to interact with a user interface of the vehicle-mounted terminal. In some embodiments, the vehicle-mounted terminal has a graphical user interface (GUI), and a user may interact with the GUI by finger contact and / or gestures on a touch-sensitive surface. The human-computer interaction function here may include a vehicle gear shifting function, and executable instructions for executing the human-computer interaction function are configured / stored in a computer program product or a readable storage medium executable by one or more processors.

[0029] Figure 1 is a flow chart of a method for reconstructing an unmanned aerial vehicle inspection image according to one embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0030] Step S101, determining a target inspection area of ​​a target UAV.

[0031] This application takes the use of target drones to inspect substations as an example to elaborate on the method of reconstructing drone inspection images.

[0032] Substation inspection refers to the regular inspection, testing, maintenance and other operations on substation equipment and facilities to ensure the safe operation of power equipment and reduce the possibility of accidents.

[0033] For example, key information can be extracted from the work ticket or operation ticket of the power system, including but not limited to the scope of the inspection task, the location of key equipment, the purpose and sequence of the inspection. By extracting the above key information, the target inspection area for the drone to perform the inspection task can be clarified.

[0034] Step S102, controlling the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image.

[0035] Specifically, based on the target inspection area of ​​the target drone determined in step S101, the flight path of the drone is planned. The path planning needs to consider the flight safety of the drone, the sampling angle of the camera, and how to maximize the image acquisition efficiency. In addition, the path planning also needs to ensure that the drone can approach the target device from the optimal angle to obtain high-quality image data.

[0036] Based on the planned flight path, control commands are sent to the target drone, including takeoff instructions, flight speed, flight altitude, detailed coordinates of the flight path, and preset angles and resolution settings of the camera. After receiving the command, the drone takes off and inspects the target inspection area according to the preset path.

[0037] During the flight of the target UAV, the camera carried by the target UAV collects image data of the target inspection area according to the instructions, obtains a preliminary inspection image, and reconstructs the preliminary inspection image to obtain the target inspection image. Among them, the parameters of the camera (such as resolution, focal length, exposure time, etc.) are dynamically adjusted according to the task requirements to ensure that high-precision images are collected in key areas and low-precision images are collected in auxiliary areas to balance image quality and data transmission volume.

[0038] Optionally, the obtained preliminary inspection image can be further processed, including detail enhancement, contrast adjustment, noise removal, etc., to improve the clarity and recognition rate of the image, thereby improving the accuracy of subsequent image reconstruction.

[0039] Step S103, acquiring multiple camera images from multiple cameras.

[0040] Specifically, all available fixed cameras in the substation are identified, including their location, field of view, resolution and other technical parameters. According to the inspection tasks and path planning of the drone, fixed cameras that can cover the drone inspection area are screened out, and multiple camera images captured by each camera during the drone inspection are obtained.

[0041] Optionally, the parameter settings of the fixed camera, such as resolution, focal length, exposure time, etc., can be adaptively adjusted according to the needs of the drone inspection to optimize the image quality and data transmission efficiency.

[0042] Step S104: Reconstruct the target inspection image based on the multiple camera images to obtain a target reconstructed image.

[0043] Specifically, multiple camera images and target inspection images are fused to generate a fused image, and then deep learning techniques such as convolutional neural networks or generative adversarial networks are used to deeply analyze and reconstruct the fused image. The deep learning model can learn and understand complex spatial structures and texture features to generate high-precision target reconstruction images.

[0044] Optionally, the target reconstructed image is further analyzed to automatically identify and mark potential abnormal conditions, such as equipment damage, foreign body intrusion, temperature abnormality, etc., to provide guidance for subsequent detailed inspections by maintenance personnel.

[0045] In an embodiment of the present invention, a target inspection area of ​​a target drone is determined; the target drone is controlled to perform an inspection operation on the target inspection area to obtain a target inspection image; multiple camera images of multiple cameras are acquired; and a reconstruction operation is performed on the target inspection image based on the multiple camera images to obtain a target reconstructed image. The present invention optimizes the target inspection image by combining the camera images acquired by multiple cameras, thereby achieving the purpose of avoiding the drone from performing multiple flight inspections on the inspection area to improve the quality of the reconstructed image, thereby achieving the technical effect of efficiently and accurately reconstructing the drone inspection image, and further solving the technical problem of how to efficiently and accurately reconstruct the drone inspection image.

[0046] Optionally, in step S101, determining the target inspection area of ​​the target drone may include the following steps:

[0047] S1011, obtaining multiple inspection objects of the target UAV;

[0048] S1012: Determine a target inspection area according to multiple inspection objects.

[0049] Specifically, after receiving the work ticket or operation ticket of the power system, natural language processing and pattern recognition technology are used to automatically parse the specific requirements of the inspection task, including the list of equipment that needs to be inspected, the location information of the equipment, the inspection priority and special requirements (such as equipment status monitoring, fault detection, etc.), and determine the multiple inspection objects of the target drone from the information obtained.

[0050] Based on the determined multiple inspection objects, a final target inspection area is determined.

[0051] Optionally, based on the priority of the inspection equipment, the target inspection area can be divided into multiple sub-target inspection areas. For sub-target inspection areas that contain key equipment and high-priority inspection objects, the UAV needs to fly near these areas and collect high-resolution images to ensure the details and quality of the reconstructed image. For sub-target inspection areas that contain low-priority inspection objects, images with lower image resolution can be collected to reduce the amount of data processing and improve overall efficiency.

[0052] Optionally, in step S102, controlling the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image may include the following steps:

[0053] S1021, controlling the target UAV to perform an inspection operation on the target inspection area to obtain an initial inspection image;

[0054] S1022, performing a reconstruction operation on the initial inspection image to obtain a target inspection image.

[0055] Specifically, after receiving the work ticket / operation ticket information, the specific inspection requirements are parsed, including the equipment list and priority to be inspected, the location information of the equipment, and the required image reconstruction accuracy. The UAV system starts the inspection task based on this information. Based on the parsed equipment information and substation map, the flight path of the UAV is planned to ensure that the path can cover the entire target inspection area. Flight path planning needs to consider the safe flight of the UAV, including avoiding obstacles and high-voltage equipment.

[0056] Specifically, when the drone enters the target inspection area according to the planned path, the drone is controlled to perform specific inspection operations, and the camera on the drone begins to shoot the inspection object and its surrounding environment to obtain a series of initial inspection images. During the flight, the drone can adaptively adjust parameters such as flight altitude, camera angle, and exposure time to obtain high-quality image data. These image data will be transmitted to the adaptive image reconstruction server in real time for processing.

[0057] Optionally, the received initial inspection images are preprocessed, including image correction, denoising and color adjustment, to eliminate interference factors such as vibration and light changes during the flight of the UAV and improve the accuracy of image reconstruction.

[0058] The preprocessed initial inspection image is reconstructed. Exemplarily, based on a deep learning algorithm, such as a convolutional neural network or a generative adversarial network, the acquired initial inspection image is analyzed and reconstructed to generate a preliminary target inspection image.

[0059] It should be noted that the generated preliminary target inspection image does not necessarily meet the preset accuracy requirements, so further optimization is required on this basis.

[0060] Optionally, in step S104, performing a reconstruction operation on the target inspection image based on the multiple camera images to obtain the target reconstructed image may include the following steps:

[0061] S1041, performing a fusion operation on the target inspection image based on multiple camera images to obtain a fused image;

[0062] S1042, performing a reconstruction operation on the fused image based on a preset deep learning algorithm to obtain a target reconstructed image, wherein the target reconstructed image includes: a first resolution image.

[0063] Specifically, the image data received from drones and fixed cameras are first synchronized in time to ensure the effective fusion of image data, so that all images reflect the scene at the same moment. Next, all images are preprocessed, including color correction, contrast adjustment, brightness standardization, and image cropping, to eliminate the image quality differences caused by different camera characteristics, and remove parts of the image that are not related to the inspection task.

[0064] Specifically, on the preprocessed image, computer vision algorithms, such as scale-invariant feature transform, accelerated robust features or oriented gradient histogram, are used to extract the key feature points of the image. Then, feature matching algorithms, such as brute force matching, FLANN matching or a matching network based on deep learning, are used to find corresponding points in different images, establish the correlation between camera views, and obtain feature point matching results.

[0065] Specifically, the feature point matching results obtained above are used to apply the multi-view geometry principle to calculate the relative pose (position and posture) between the cameras and the 3D coordinates of the scene points.

[0066] On the basis of the above, image fusion algorithms such as weighted averaging, multi-resolution pyramid fusion or deep learning-based fusion network are used to merge image information from different cameras into a single fused image. The generation of the fused image needs to take into account the effects of perspective, illumination and occlusion to produce an image that is closest to the real scene.

[0067] Based on the fused image, the preset deep learning algorithms, such as fully convolutional networks, generative adversarial networks or variational autoencoders, are used to enhance and optimize the details of the fused image to generate a highly realistic target reconstructed image.

[0068] It should be noted that the target reconstructed image is an image that meets the preset high-precision requirements.

[0069] In addition, by making full use of the viewing angle advantages of multiple cameras and the intelligent processing capabilities of deep learning algorithms, high-quality, adaptive resolution target reconstruction images are generated, significantly improving the reconstruction efficiency and accuracy of drone inspection images.

[0070] Optionally, the method for reconstructing the UAV inspection image further includes:

[0071] S1043, obtaining multiple position information of multiple cameras;

[0072] S1044: Determine a target modeling area according to the multiple location information.

[0073] Specifically, based on the fixed cameras that can cover the drone inspection area selected in step S103, the installation position and viewing angle range of the camera are obtained, and the coverage area of ​​each camera is calculated. The coverage area of ​​the fixed camera is calculated based on its installation position and fixed viewing angle. The coverage area of ​​each camera is analyzed to identify which areas are covered by multiple cameras and which areas are covered by only a part of the cameras, and the area that can be covered by multiple cameras is determined as the target modeling area.

[0074] Optionally, the method for reconstructing the UAV inspection image further includes:

[0075] S1045, determining a target reconstruction area according to the target modeling area and the target inspection area, wherein the target reconstruction area is an area within the target modeling area excluding the target inspection area;

[0076] S1046, obtaining a preset reconstructed image corresponding to the target reconstruction area;

[0077] S1047, performing a reconstruction operation on the preset reconstructed image based on a preset deep learning algorithm to obtain a second resolution image, wherein the image resolution of the first resolution image is higher than that of the second resolution image.

[0078] Specifically, after determining the target modeling area (the area covered by multiple cameras) and the target inspection area (the equipment area that needs to be inspected), a comparison analysis is performed between the two. Through the comparison analysis, the area outside the target inspection area in the target modeling area is identified and defined as the "target reconstruction area". This area may contain auxiliary facilities, passages, walls, etc., which are not the focus of the inspection task, but still need to be reconstructed to provide complete environmental background information.

[0079] For the target reconstruction area, a preset reconstruction image with a lower resolution is obtained. These preset reconstruction images are generated based on the image data of the fixed camera. The use of the preset reconstruction images reduces unnecessary data transmission and processing burden.

[0080] Furthermore, a reconstructed operation is performed on the preset reconstructed image based on a deep learning algorithm to obtain a second resolution image. The deep learning algorithm here can be exemplarily a deep residual network, a densely connected network or an efficient feature extraction algorithm, which can recover more details from the low-resolution image and improve the image clarity. However, compared with the first resolution image, the image resolution of the second resolution image is lower because the target reconstruction area is not the focus of the inspection and its demand on image details is relatively low.

[0081] It should be noted that for the target inspection area, high-precision image reconstruction is used to generate an image of the first resolution; while for the target reconstruction area, a low-precision reconstruction strategy is used to generate an image of the second resolution. This adaptive adjustment can ensure the monitoring quality of key areas while optimizing the resource consumption of the entire system.

[0082] Figure 2 is a schematic diagram of a method for reconstructing an unmanned aerial vehicle inspection image according to one embodiment of the present invention.

[0083] like Figure 2 As shown in the figure, the specific steps of the reconstruction method of the UAV inspection image are as follows:

[0084] Step 1: After the UAV inspection work ticket / operation ticket is issued, the adaptive image reconstruction server divides the flight phase according to the inspection object and determines the high-precision modeling area according to the inspection object;

[0085] Step 2: The drone takes off;

[0086] Step 3: According to the flight phases divided in step 1, call n fixed cameras to cover the area to be inspected.

[0087] Step 4: The UAV performs the inspection task of stage k and uses a deep learning algorithm to reconstruct the image to obtain the reconstructed image G1.

[0088] Step 5: For the inspection object area, the data from the fixed camera is integrated to further optimize G1, and the high-resolution image G2 is further optimized through the deep learning algorithm.

[0089] Step 6: For images covered by fixed cameras but not within the G1 area, obtain image G3 through a deep learning algorithm.

[0090] Step 7: After the current flight phase ends, if there is a next flight phase, enter the next flight phase, k=k+1, and return to step 3; if there is no next flight phase, enter the next phase.

[0091] Step 8: The drone lands and the flight ends.

[0092] In an embodiment of the present invention, a device for reconstructing an unmanned aerial vehicle inspection image is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" may be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0093] Figure 3is a structural block diagram of a device 300 for reconstructing an unmanned aerial vehicle inspection image according to one embodiment of the present invention. Figure 3 As shown, the device includes: a first determination module 301, used to determine the target inspection area of ​​the target UAV; a control module 302, used to control the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image; a second acquisition module 303, used to obtain multiple camera images of multiple cameras; a first execution module 304, used to perform a reconstruction operation on the target inspection image based on the multiple camera images to obtain a target reconstructed image.

[0094] Optionally, the first determination module 301 is further used to: obtain multiple inspection objects of the target UAV; and determine the target inspection area according to the multiple inspection objects.

[0095] Optionally, the control module 302 is further used to: control the target UAV to perform an inspection operation on the target inspection area to obtain an initial inspection image; and perform a reconstruction operation on the initial inspection image to obtain a target inspection image.

[0096] Optionally, the first execution module 304 is also used to: perform a fusion operation on the target inspection image based on multiple camera images to obtain a fused image; perform a reconstruction operation on the fused image based on a preset deep learning algorithm to obtain a target reconstructed image, wherein the target reconstructed image includes: a first resolution image.

[0097] Optionally, the first execution module 304 is further used to: obtain multiple position information of multiple cameras; and determine the target modeling area according to the multiple position information.

[0098] Optionally, the first execution module 304 is also used to: determine a target reconstruction area based on a target modeling area and a target inspection area, wherein the target reconstruction area is an area within the target modeling area excluding the target inspection area; obtain a preset reconstructed image corresponding to the target reconstruction area; perform a reconstruction operation on the preset reconstructed image based on a preset deep learning algorithm to obtain a second resolution image, wherein the image resolution of the first resolution image is higher than that of the second resolution image.

[0099] An embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the method for reconstructing an unmanned aerial vehicle inspection image as described in any of the above embodiments.

[0100] Optionally, in this embodiment, the processor in the above electronic device may be configured to run a computer program to perform the following steps:

[0101] Step S101, determining a target inspection area of ​​a target UAV.

[0102] Step S102, controlling the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image.

[0103] Step S103, acquiring multiple camera images from multiple cameras.

[0104] Step S104: Reconstruct the target inspection image based on the multiple camera images to obtain a target reconstructed image.

[0105] Optionally, the processor in the above electronic device can be configured to run a computer program to perform the following steps: obtaining multiple inspection objects of the target drone; and determining a target inspection area based on the multiple inspection objects.

[0106] Optionally, the processor in the above electronic device can be configured to run a computer program to perform the following steps: control the target UAV to perform an inspection operation on the target inspection area to obtain an initial inspection image; perform a reconstruction operation on the initial inspection image to obtain a target inspection image.

[0107] Optionally, the processor in the above-mentioned electronic device can be configured to run a computer program to perform the following steps: perform a fusion operation on the target inspection image based on multiple camera images to obtain a fused image; perform a reconstruction operation on the fused image based on a preset deep learning algorithm to obtain a target reconstructed image, wherein the target reconstructed image includes: a first resolution image.

[0108] Optionally, the processor in the above electronic device can be configured to run a computer program to perform the following steps: obtaining multiple position information of multiple cameras; and determining a target modeling area based on the multiple position information.

[0109] Optionally, the processor in the above-mentioned electronic device can be configured to run a computer program to perform the following steps: determine a target reconstruction area based on a target modeling area and a target inspection area, wherein the target reconstruction area is an area within the target modeling area excluding the target inspection area; obtain a preset reconstructed image corresponding to the target reconstruction area; perform a reconstruction operation on the preset reconstructed image based on a preset deep learning algorithm to obtain a second resolution image, wherein the image resolution of the first resolution image is higher than that of the second resolution image.

[0110] An embodiment of the present invention also provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the method for reconstructing the drone inspection image described in any of the above embodiments when running on a computer or a processor.

[0111] Optionally, in this embodiment, the above-mentioned computer program may be configured to store a computer program for performing the following steps:

[0112] Step S101, determining a target inspection area of ​​a target UAV.

[0113] Step S102, controlling the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image.

[0114] Step S103, acquiring multiple camera images from multiple cameras.

[0115] Step S104: Reconstruct the target inspection image based on the multiple camera images to obtain a target reconstructed image.

[0116] Optionally, the above-mentioned computer program can be configured to store a computer program for executing the following steps: obtaining multiple inspection objects of the target drone; and determining a target inspection area according to the multiple inspection objects.

[0117] Optionally, the above-mentioned computer program can be configured to store a computer program for executing the following steps: controlling the target UAV to perform an inspection operation on the target inspection area to obtain an initial inspection image; performing a reconstruction operation on the initial inspection image to obtain a target inspection image.

[0118] Optionally, the above-mentioned computer program can be configured to store a computer program for executing the following steps: performing a fusion operation on the target inspection image based on multiple camera images to obtain a fused image; performing a reconstruction operation on the fused image based on a preset deep learning algorithm to obtain a target reconstructed image, wherein the target reconstructed image includes: a first resolution image.

[0119] Optionally, the above-mentioned computer program may be configured to store a computer program for executing the following steps: obtaining multiple position information of multiple cameras; and determining a target modeling area according to the multiple position information.

[0120] Optionally, the above-mentioned computer program can be configured to store a computer program for executing the following steps: determining a target reconstruction area based on a target modeling area and a target inspection area, wherein the target reconstruction area is an area within the target modeling area excluding the target inspection area; obtaining a preset reconstructed image corresponding to the target reconstruction area; performing a reconstruction operation on the preset reconstructed image based on a preset deep learning algorithm to obtain a second resolution image, wherein the image resolution of the first resolution image is higher than that of the second resolution image.

[0121] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the method for reconstructing the drone inspection image described in any of the above embodiments.

[0122] Optionally, in this embodiment, the above computer program implements the following steps when executed by a processor:

[0123] Step S101, determining a target inspection area of ​​a target UAV.

[0124] Step S102, controlling the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image.

[0125] Step S103, acquiring multiple camera images from multiple cameras.

[0126] Step S104: Reconstruct the target inspection image based on the multiple camera images to obtain a target reconstructed image.

[0127] Optionally, the above-mentioned computer program implements the following steps when executed by a processor: obtaining multiple inspection objects of the target drone; and determining a target inspection area based on the multiple inspection objects.

[0128] Optionally, the above-mentioned computer program implements the following steps when executed by the processor: controlling the target UAV to perform an inspection operation on the target inspection area to obtain an initial inspection image; performing a reconstruction operation on the initial inspection image to obtain a target inspection image.

[0129] Optionally, the above-mentioned computer program implements the following steps when executed by the processor: performing a fusion operation on the target inspection image based on multiple camera images to obtain a fused image; performing a reconstruction operation on the fused image based on a preset deep learning algorithm to obtain a target reconstructed image, wherein the target reconstructed image includes: a first resolution image.

[0130] Optionally, the above computer program implements the following steps when executed by a processor: obtaining multiple position information of multiple cameras; and determining a target modeling area according to the multiple position information.

[0131] Optionally, the above-mentioned computer program implements the following steps when executed by the processor: determine the target reconstruction area based on the target modeling area and the target inspection area, wherein the target reconstruction area is the area within the target modeling area excluding the target inspection area; obtain a preset reconstructed image corresponding to the target reconstruction area; perform a reconstruction operation on the preset reconstructed image based on a preset deep learning algorithm to obtain a second resolution image, wherein the image resolution of the first resolution image is higher than that of the second resolution image.

[0132] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0133] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0134] In some embodiments provided in the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0135] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0136] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0137] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a grid device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc., which can store program codes.

[0138] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for reconstructing an unmanned aerial vehicle inspection image, characterized in that: include: Determine the target inspection area of ​​the target drone; Controlling the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image; Get multiple camera images from multiple cameras; A reconstruction operation is performed on the target inspection image based on the multiple camera images to obtain a target reconstructed image.

2. The method for reconstructing an unmanned aerial vehicle inspection image according to claim 1, characterized in that: Determining the target inspection area of ​​the target UAV includes: Acquire multiple inspection objects of the target UAV; The target inspection area is determined according to the multiple inspection objects.

3. The method for reconstructing an unmanned aerial vehicle inspection image according to claim 1, characterized in that: Controlling the target UAV to perform the inspection operation on the target inspection area to obtain the target inspection image includes: Controlling the target UAV to perform the inspection operation on the target inspection area to obtain an initial inspection image; A reconstruction operation is performed on the initial inspection image to obtain the target inspection image.

4. The method for reconstructing an unmanned aerial vehicle inspection image according to claim 3, characterized in that: Performing the reconstruction operation on the target inspection image based on the multiple camera images to obtain the target reconstructed image includes: Performing a fusion operation on the target inspection image based on the multiple camera images to obtain a fused image; The reconstruction operation is performed on the fused image based on a preset deep learning algorithm to obtain the target reconstructed image, wherein the target reconstructed image includes: a first resolution image.

5. The method for reconstructing an unmanned aerial vehicle inspection image according to claim 4, characterized in that: The method further comprises: Acquire multiple position information of the multiple cameras; A target modeling area is determined according to the plurality of position information.

6. The method for reconstructing an unmanned aerial vehicle inspection image according to claim 5, characterized in that: The method further comprises: Determine a target reconstruction area according to the target modeling area and the target inspection area, wherein the target reconstruction area is an area within the target modeling area excluding the target inspection area; Acquire a preset reconstructed image corresponding to the target reconstruction area; The reconstruction operation is performed on the preset reconstructed image based on the preset deep learning algorithm to obtain a second resolution image, wherein the image resolution of the first resolution image is higher than that of the second resolution image.

7. A device for reconstructing drone inspection images, characterized in that: include: The first determination module is used to determine the target inspection area of ​​the target UAV; A control module, used to control the target UAV to perform an inspection operation on the target inspection area to obtain a target inspection image; A second acquisition module is used to acquire multiple camera images from multiple cameras; The first execution module is used to perform a reconstruction operation on the target inspection image based on the multiple camera images to obtain a target reconstructed image.

8. An electronic device, comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to execute the method for reconstructing the drone inspection image as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the method for reconstructing the drone inspection image described in any one of claims 1 to 6 when running on a computer or a processor.

10. A computer program product, characterized in that It includes a computer program, which, when executed by a processor, implements the steps of the method for reconstructing an unmanned aerial vehicle inspection image as described in any one of claims 1 to 6.

Images