Unmanned aerial vehicle remote sensing image real-time quality inspection and automatic mapping control method, device, equipment and medium

Abstract

This application relates to a method, apparatus, equipment, and medium for real-time quality inspection and automatic mapping control of UAV remote sensing images. The method includes: a ground processing unit performing multiple quality checks on archived remote sensing images and generating quality assessment indicators based on the results of each quality check; the ground processing unit determining whether preset mapping trigger conditions are met based on the quality assessment indicators, and automatically generating a mapping task when met; for the generated mapping task, the ground processing unit adaptively generating mapping processing parameters based on the task feature information corresponding to the image data; calling the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters, outputting the mapping results, and feeding back the relevant information of the mapping results to the user terminal. This application effectively improves the management efficiency and overall operational efficiency of UAV remote sensing mapping tasks.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method, device, electronic equipment and computer-readable storage medium for real-time quality inspection and automatic mapping control of UAV remote sensing images. Background Technology

[0002] UAV remote sensing mapping technology typically uses a preset flight path to acquire multiple remote sensing images with overlapping areas. After the flight mission is completed, the original images are imported into photogrammetry software to complete post-processing steps such as feature matching, aerial triangulation, point cloud generation, and orthophoto stitching mapping.

[0003] Currently, mainstream software still primarily operates on the model of batch importing original photos for high-precision reconstruction after image acquisition. Real-time mapping results are mostly used for quick previews or low-precision output, which is insufficient to support the demand for high-efficiency and high-quality mapping applications.

[0004] However, existing technical solutions generally suffer from the following technical defects:

[0005] Firstly, the data processing chain is long and the mapping timeliness is insufficient. Imagery and image attribute data are usually centrally processed and imported after the flight mission. There is a long time delay between data acquisition and output, which makes it difficult to meet the application scenarios that have high requirements for rapid mapping, such as emergency mapping and agricultural monitoring.

[0006] Secondly, the delayed detection of image quality issues easily leads to repetitive work. Due to the lack of a real-time quality detection mechanism, problems such as insufficient image clarity, abnormal exposure, missed shots, or incomplete coverage are usually only discovered in the post-processing stage. By this time, the drone has already left the work area, requiring re-flying or re-collection, which prolongs the work cycle and increases costs.

[0007] Third, the triggering of mapping tasks and parameter configuration rely on human experience, resulting in a low degree of automation. The timing of map-building task initiation, image selection strategies, and mosaic parameter settings still largely depend on the professional judgment and manual intervention of operators, lacking a standardized automatic decision-making mechanism, which is not conducive to the promotion of large-scale routine operations.

[0008] Fourth, it is difficult to balance real-time performance and mapping accuracy. In existing technologies, rapid preview mapping usually comes at the cost of accuracy, while high-precision mapping requires a long post-processing time. There is a clear contradiction between the two, making it difficult to meet the dual requirements of high timeliness and high accuracy in the same system.

[0009] In summary, given the existing technologies suffer from problems such as the separation of data acquisition and mapping processing, lagging quality control, low automation, and difficulty in balancing real-time performance and accuracy, the applicant has made corresponding explorations to address these issues. Summary of the Invention

[0010] The purpose of this application is to solve the above-mentioned problems by providing a method, device, electronic equipment and computer-readable storage medium for real-time quality inspection and automatic mapping control of UAV remote sensing images.

[0011] To achieve the various objectives of this application, the following technical solution is adopted:

[0012] A method for real-time quality inspection and automatic mapping control of UAV remote sensing images, proposed to meet one of the purposes of this application, includes:

[0013] The drone will transmit the collected remote sensing images and their corresponding image attribute data to the ground processing terminal in real time. During the transmission process, the remote sensing images will be divided into multiple data packets according to preset units.

[0014] The ground processing terminal establishes a buffer area according to the task identifier and image identifier, verifies each data packet, and reassembles the original image data according to the fragment sequence number after the verification is passed. The remote sensing image is then matched and associated with the image attribute data according to the unified identifier.

[0015] The ground processing terminal renames the remote sensing images according to a unified naming rule. Based on the image identifier, it establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data, generates an associated record containing the image identifier, task identifier, acquisition time, spatial location, attitude information and storage path, and archives and stores it according to the task identifier.

[0016] The ground processing unit performs multiple quality checks on the archived remote sensing images and generates quality assessment indicators based on the results of each quality check. The ground processing unit then determines whether the preset mapping trigger conditions are met based on the quality assessment indicators, and automatically generates a mapping task when the conditions are met.

[0017] For the generated mapping task, the ground processing end adaptively generates mapping processing parameters based on the task feature information corresponding to the image data; it calls the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters, outputs the mapping results, and feeds back the relevant information of the mapping results to the user end.

[0018] Optionally, the UAV transmits the collected remote sensing images and their corresponding image attribute data to the ground processing terminal in real time. During transmission, the remote sensing images are divided into multiple data packets according to preset units, including:

[0019] The remote sensing image is divided into multiple data slices according to a preset slice size. Each data slice is encapsulated into a data packet. The image attribute data is associated with the remote sensing image through the image frame number. The image attribute data includes a task identifier, image frame number, acquisition timestamp, geographic coordinates, attitude angle, camera intrinsic parameter matrix, extrinsic parameter matrix, and flight altitude. Each data packet contains a task identifier, image frame number, slice sequence number, and verification field.

[0020] The data packets are managed through a sequence numbering, buffer queue, timeout retransmission, and cyclic redundancy check mechanism. When a fragment sequence number is found to be discontinuous or the cyclic redundancy check fails, a retransmission request for the corresponding fragment is triggered to ensure the integrity and correct order of the data received by the ground processing end.

[0021] Optionally, the ground processing end establishes a buffer based on the task identifier and image identifier, verifies each data packet, reassembles the original image data according to the fragment sequence number after successful verification, and matches and associates the remote sensing image with the image attribute data according to the unified identifier. This includes the following steps:

[0022] After receiving each data packet, the ground processing unit establishes a corresponding receiving buffer based on the task identifier and image frame number in each data packet;

[0023] The ground processing unit performs integrity verification on each data packet. If any data packet fails verification, an exception flag is generated and logged. The integrity verification is implemented based on any one or more of the following methods: Cyclic Redundancy Check (CRC), MD5 checksum, or hash digest.

[0024] After each data packet passes verification, the ground processing unit sorts and combines all data packets belonging to the same image frame number according to the fragment sequence number of each data packet to restore the original remote sensing image.

[0025] The ground processing unit matches and associates the restored original remote sensing image with the corresponding image attribute data by using the image frame number or the acquisition timestamp. When matching by acquisition timestamp, if the difference between the acquisition time of the remote sensing image and the recording time of the image attribute data is not greater than the preset time deviation threshold, it is determined that the two belong to the same acquisition frame.

[0026] If a match cannot be established using either the image frame number or the acquisition timestamp, an anomaly marker is generated and logged.

[0027] Optionally, the ground processing unit renames the remote sensing images according to a unified naming rule, establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data based on the image identifier, generates an associated record containing the image identifier, task identifier, acquisition time, spatial location, attitude information, and storage path, and archives and stores it according to the task identifier. This includes the following steps:

[0028] The ground processing unit renames the restored remote sensing images according to a unified naming rule;

[0029] The ground processing terminal establishes a unique mapping and binding relationship between the renamed remote sensing image and the corresponding image attribute data based on the image identifier, and generates an association record. The association record includes at least the image identifier, task identifier, acquisition time, spatial location, attitude information and storage path fields.

[0030] The ground processing unit creates an archive directory according to the task identifier, and archives and stores the renamed remote sensing images, corresponding image attribute data, and associated records in the archive directory.

[0031] Optionally, the ground processing unit performs multiple quality checks on the archived remote sensing images and generates quality assessment indicators based on the results of each quality check, including:

[0032] The ground processing unit performs quality inspection on the archived remote sensing images and calculates the sharpness value of the remote sensing images. When the sharpness value is less than the preset sharpness threshold, it is determined that the image has a blur or out-of-focus problem.

[0033] In addition, the pixel brightness distribution of the remote sensing image is statistically analyzed. When the proportion of bright pixels is greater than the first preset threshold or the proportion of dark pixels is greater than the second preset threshold, the image is determined to have overexposure or underexposure problems.

[0034] And calculate the similarity value between different remote sensing images. When the similarity value is greater than a preset similarity threshold and the distance between the corresponding shooting locations is less than a preset distance threshold, it is determined that there are duplicate images.

[0035] And compare the theoretical number of images in the preset flight route shooting plan with the actual number of received images. When the distance between adjacent acquisition points is greater than the preset distance threshold or the actual number of received images is less than the theoretical number of images, it is determined that there are missing images.

[0036] And based on the geographic coordinates of each remote sensing image, the field of view of each image is projected onto the ground to form a coverage polygon. The ratio of the union of each coverage polygon to the area of ​​the target area is calculated as the coverage rate. When the coverage rate is less than a preset coverage rate threshold, it is determined that the coverage is insufficient.

[0037] And calculate the overlap ratio of adjacent remote sensing images in the ground coverage area as the overlap degree. When the average forward overlap degree is not less than the third preset ratio threshold and the average lateral overlap degree is not less than the fourth preset ratio threshold, it is determined that the overlap degree meets the mapping requirements; otherwise, it is determined that the overlap degree does not meet the mapping requirements.

[0038] Quality assessment indicators are generated based on the results of various quality tests.

[0039] Optionally, the step of the ground processing terminal determining whether the preset mapping triggering conditions are met based on the quality assessment indicators, and automatically generating a mapping task when the conditions are met, includes:

[0040] The ground processing terminal determines whether the preset mapping triggering conditions are met based on the quality assessment indicators. The preset mapping triggering conditions include: the ratio of the actual number of received images to the theoretical number of images is not less than a first preset ratio threshold, the coverage is not less than a second preset ratio threshold, the average forward overlap is not less than a third preset ratio threshold, the average lateral overlap is not less than a fourth preset ratio threshold, or the ratio of the number of abnormal images to the total number of images is not greater than a fifth preset ratio threshold.

[0041] When any of the above mapping triggering conditions are met, the ground processing terminal automatically generates a mapping task.

[0042] Optionally, the ground processing end adaptively generates mapping processing parameters based on the task feature information corresponding to the image data for the generated mapping task; calls the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters, and outputs the mapping result, including the following steps:

[0043] The ground processing unit extracts the task feature vector corresponding to the generated mapping task, wherein the task feature vector includes ground resolution, image quality index, area, terrain complexity and mapping result type;

[0044] The ground processing unit adaptively generates stitching and mapping parameters based on the task feature vector through a parameter mapping table or rule base. The stitching and mapping parameters include image scaling level, number of feature points extracted, reprojection error threshold, output resolution, and coordinate projection parameters.

[0045] The ground processing unit calls the mapping processing engine to perform automatic mosaicking and mapping processing on the image data based on the mosaicking and mapping parameters, and outputs the mapping results, wherein the mapping results include at least one of orthophoto map, digital surface model or three-dimensional model.

[0046] A real-time quality inspection and automatic mapping control device for UAV remote sensing images, provided for another purpose of this application, includes:

[0047] The data transmission module is configured to transmit the collected remote sensing images and their corresponding image attribute data from the UAV to the ground processing terminal in real time. During the transmission process, the remote sensing images are divided into multiple data packets according to preset units.

[0048] The data association and reassembly module is configured to establish a buffer area on the ground processing end according to the task identifier and image identifier, verify each data packet, and reassemble the original image data according to the fragment sequence number after the verification is passed, and match and associate the remote sensing image with the image attribute data according to the unified identifier.

[0049] The data archiving and storage module is configured to rename remote sensing images according to a unified naming rule on the ground processing end, establish a unique mapping and binding relationship between the renamed image data and the corresponding attribute data based on the image identifier, generate an associated record containing image identifier, task identifier, acquisition time, spatial location, attitude information and storage path, and archive and store it according to the task identifier.

[0050] The mapping trigger module is configured to perform multiple quality checks on the archived remote sensing images by the ground processing end, and generate quality assessment indicators based on the results of each quality check; the ground processing end determines whether the preset mapping trigger conditions are met based on the quality assessment indicators, and automatically generates a mapping task when the conditions are met.

[0051] The automatic mapping module is configured to, on the ground processing end, adaptively generate mapping processing parameters based on the task feature information corresponding to the image data for the generated mapping task; call the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters, output the mapping results, and feed back the relevant information of the mapping results to the user end.

[0052] An electronic device provided for another purpose of this application includes a central processing unit and a memory, wherein the central processing unit is used to invoke and run a computer program stored in the memory to perform the steps of the real-time quality inspection and automatic mapping control method for UAV remote sensing images described in this application.

[0053] A computer-readable storage medium is provided for another purpose of this application, which stores, in the form of computer-readable instructions, a computer program implemented according to the UAV remote sensing image real-time quality inspection and automatic mapping control method, which, when called by a computer, executes the steps included in the corresponding method.

[0054] Compared to existing technologies, this application addresses the problems of existing technologies, such as the separation of data acquisition and mapping processing, lagging quality control, low automation, and difficulty in balancing real-time performance and accuracy. This application offers the following beneficial effects, including but not limited to:

[0055] Firstly, this application utilizes a UAV to simultaneously record image attribute data such as timestamps, geographical location, attitude angles, flight altitude, and camera parameters while acquiring remote sensing images. The image data and attribute data are then transmitted in real-time to a ground processing unit via a wireless link, eliminating the need to wait for data processing after the flight mission. The ground processing unit performs real-time parsing, integrity verification, reconstruction and restoration, and automatic binding and archiving of the received data, achieving fully automated processing from image acquisition to archiving and storage. Compared to the traditional "post-processing via software after flight" model, this application can simultaneously complete data reception, processing, and mapping trigger condition determination during flight, fundamentally solving the problems of long processing links and insufficient mapping timeliness in existing technologies. This meets the application scenarios with high requirements for rapid mapping, such as emergency mapping and agricultural monitoring.

[0056] Secondly, this application performs multiple quality checks on the archived remote sensing images in real time during flight, including sharpness detection, exposure anomaly detection, duplicate image detection, missing image detection, coverage integrity detection, and overlap detection. Each check is based on clear mathematical or physical judgment standards. Image quality issues can be detected and anomaly markers recorded in real time during flight, eliminating the need to wait until after the flight to check in post-processing software. When issues such as missing images or insufficient coverage are detected, the UAV can promptly perform supplementary acquisitions or flights, avoiding the repetitive work costs and time delays caused by discovering data defects hours or even days after the flight in the traditional mode, which necessitates reorganizing supplementary flights. This significantly improves the reliability and economy of the operation.

[0057] Thirdly, this application automatically determines whether the preset mapping triggering conditions are met based on the quality inspection results through the ground processing terminal. When all conditions are met, it automatically generates a mapping task and adaptively generates stitching mapping parameters based on task feature information, then calls the mapping processing engine to execute automatic mapping processing. From data reception, quality inspection, condition judgment, task triggering to parameter generation and mapping execution, the entire process is completed automatically. There is no need for manual data processing, manual data quality checking, manual judgment of mapping timing, or manual setting of mapping parameters, which effectively reduces the dependence of mapping operations on the professional experience of operators and is conducive to the promotion and implementation of large-scale, routine UAV remote sensing mapping operations.

[0058] Fourth, this application sets multi-dimensional quality inspection thresholds in the quality inspection stage, including sharpness, exposure, coverage integrity, and overlap. In the mapping triggering stage, it sets multiple combined threshold conditions for comprehensive judgment; the mapping task is only triggered when all conditions are met. This multi-dimensional combined guarantee mechanism ensures that the image data participating in mapping meets the mapping requirements in terms of quantity, spatial coverage, directional geometric constraints, lateral geometric constraints, and data validity, thereby guaranteeing the accuracy and reliability of the output mapping results. It can adaptively adjust the mapping strategy according to different task requirements, while taking into account both processing timeliness and result accuracy, effectively supporting diverse application scenarios such as emergency mapping, agricultural monitoring, and rapid mapping.

[0059] Fifth, this application feeds back the file path, task status, processing time, and abnormal alarm information of the mapping results to the user terminal, clearly indicating whether re-collection, re-flight, or re-mapping is required. From image acquisition, real-time transmission, quality inspection, automatic mapping to result feedback, a complete data processing closed loop is formed. Users can monitor the execution status and result quality of the mapping task in real time through the user terminal, making subsequent operational decisions accordingly, effectively improving the management efficiency and overall operational efficiency of UAV remote sensing mapping tasks. Attached Figure Description

[0060] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0061] Figure 1 This is a flowchart illustrating the real-time quality inspection and automatic mapping control method for UAV remote sensing images in this embodiment of the application.

[0062] Figure 2 This is a schematic diagram illustrating the process of matching and associating remote sensing images with image attribute data according to a unified identifier in the embodiments of this application;

[0063] Figure 3 This is a schematic diagram of the process of archiving and storing data according to task identifiers in an embodiment of this application;

[0064] Figure 4 This is a schematic diagram of the process of performing automatic mapping processing on image data based on the mapping processing parameters and outputting mapping results in an embodiment of this application;

[0065] Figure 5 This is a schematic diagram of the real-time quality inspection and automatic mapping control device for UAV remote sensing images in the embodiments of this application;

[0066] Figure 6 This is a schematic diagram of the structure of the computer device in the embodiments of this application. Detailed Implementation

[0067] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0068] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this application means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0069] Those skilled in the art will understand that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0070] Those skilled in the art will understand that the terms "client," "terminal," and "terminal device" as used herein include both devices that receive wireless signals, devices that only possess wireless signal receiver capabilities without transmission capabilities, and devices with receiving and transmitting hardware, devices that have receiving and transmitting hardware capable of bidirectional communication over a bidirectional communication link. Such devices may include: cellular or other communication devices such as personal computers or tablets, having single-line displays, multi-line displays, or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service) that can combine voice, data processing, fax, and / or data communication capabilities; PDAs (Personal Digital Assistants) that may include radio frequency receivers, pagers, internet / intranet access, web browsers, notebooks, calendars, and / or GPS (Global Positioning System) receivers; and conventional laptops and / or handheld computers or other devices that have and / or include radio frequency receivers. As used herein, "client," "terminal," and "terminal device" can be portable, transportable, installed in a means of transportation (air, sea, and / or land), or suitable and / or configured to operate locally and / or in a distributed manner, operating in any other location on Earth and / or in space. "Client," "terminal," and "terminal device" as used herein can also be a communication terminal, an internet access terminal, or a music / video playback terminal, such as a PDA, a MID (Mobile Internet Device), and / or a mobile phone with music / video playback capabilities, or a smart TV, set-top box, etc.

[0071] The hardware referred to by the names "server," "client," and "service node" in this application is essentially an electronic device with the equivalent capabilities of a personal computer. It is a hardware device with the necessary components revealed by the von Neumann architecture, such as a central processing unit (including an arithmetic logic unit and a control unit), memory, input devices, and output devices. The computer program is stored in its memory, and the central processing unit loads the program stored in the secondary storage into the main memory to run it, execute the instructions in the program, and interact with the input and output devices to complete specific functions.

[0072] It should be noted that the concept of "server" used in this application can also be extended to the case of server clusters. Based on the network deployment principles understood by those skilled in the art, the servers should be logically divided. Physically, these servers can be independent of each other but accessible through interfaces, or they can be integrated into a single physical computer or a computer cluster. Those skilled in the art should understand this flexibility and should not use it to constrain the implementation of the network deployment method in this application.

[0073] Unless otherwise expressly specified, one or more technical features of this application may be deployed on a server and accessed by a client through remote invocation of the online service interface provided by the server, or they may be directly deployed and run on a client for access.

[0074] Unless otherwise specified, the neural network models referenced or potentially referenced in this application may be deployed on a remote server and invoked remotely on the client, or deployed on a client with the capability to invoke directly. In some embodiments, when running on the client, the corresponding intelligence may be acquired through transfer learning in order to reduce the requirements on the client's hardware resources and avoid excessive consumption of the client's hardware resources.

[0075] Unless otherwise specified, all data involved in this application may be stored remotely on a server or on a local terminal device, as long as it is suitable for use by the technical solution of this application.

[0076] Those skilled in the art will understand that although the various methods in this application are described based on the same concept and thus present commonality among them, they can be performed independently unless otherwise specified. Similarly, the various embodiments disclosed in this application are all based on the same inventive concept; therefore, concepts expressed in the same way, as well as concepts that are appropriately changed for convenience but are expressed differently, should be understood equivalently.

[0077] Unless otherwise expressly stated, the various embodiments disclosed in this application can be combined in a cross-cutting manner to flexibly construct new embodiments, as long as such combination does not depart from the inventive spirit of this application and can meet the needs of the prior art or solve a certain deficiency in the prior art. Those skilled in the art should be aware of such modifications.

[0078] Please see Figure 1 In one embodiment of the UAV remote sensing image real-time quality inspection and automatic mapping control method of this application, the method includes:

[0079] Step S10: The UAV will transmit the collected remote sensing images and their corresponding image attribute data to the ground processing terminal in real time. During the transmission process, the remote sensing images will be divided into multiple data packets according to preset units.

[0080] The UAV collects remote sensing images of the target area according to a preset flight path and generates corresponding image attribute data simultaneously with each collection. The image attribute data includes image frame number, collection timestamp, geographic coordinates, attitude angle, camera intrinsic parameter matrix, extrinsic parameter matrix, flight altitude, and mission identifier. A one-to-one correspondence is established between the remote sensing images and the image attribute data according to the same frame number.

[0081] In this embodiment, the flight path spacing and shooting interval can be set according to flight altitude, camera focal length, pixel size, and target overlap, wherein the ground resolution is based on... Calculate the ground coverage length of a single frame image. and width The forward shooting interval is determined by the image pixel size and ground resolution, respectively. Set the lateral route spacing according to Settings, where Forward overlap, This refers to lateral overlap. Using the above method, a synchronization relationship between remote sensing imagery and flight status parameters can be established during the acquisition phase.

[0082] In some embodiments, Indicates ground resolution; Indicates flight altitude; Indicates the pixel size;

[0083] Indicates the camera's focal length; Indicates the length of the image's ground coverage; Indicates the width of the image's ground coverage; Indicates the forward shooting interval; Indicates the spacing between lateral routes; Indicates forward overlap; Indicates the degree of lateral overlap.

[0084] In some embodiments, the UAV transmits the collected remote sensing images and their corresponding image attribute data to a ground processing terminal in real time. The step of dividing the remote sensing images into multiple data packets according to preset units during transmission includes:

[0085] Step S101: Divide the remote sensing image into multiple data slices according to a preset slice size. Each data slice is encapsulated into a data packet. The image attribute data is associated with the remote sensing image through the image frame number. The image attribute data includes a task identifier, image frame number, acquisition timestamp, geographic coordinates, attitude angle, camera intrinsic parameter matrix, extrinsic parameter matrix, and flight altitude. Each data packet contains a task identifier, image frame number, slice sequence number, and verification field.

[0086] Step S102: The data packets are managed through a sequence numbering, buffer queue, timeout retransmission, and cyclic redundancy check mechanism. When a fragment sequence number is found to be discontinuous or the cyclic redundancy check fails, a retransmission request for the corresponding fragment is triggered to ensure the integrity and correct order of the data received by the ground processing end.

[0087] Specifically, the UAV transmits the collected remote sensing images and their corresponding image attribute data to the ground processing terminal in real time via a wireless communication link. During transmission, the remote sensing images are divided into multiple data packets according to a preset segment size. Each data packet contains a task identifier, image frame number, segment sequence number, segment length, timestamp, and verification field. The image attribute data is sent synchronously with the remote sensing images in JSON format, binary structure format, or key-value pair format, or sent independently before or after the remote sensing images and associated with the corresponding remote sensing images through the image frame number. The wireless communication link can be any one of a data link, image transmission link, fourth-generation mobile communication network, fifth-generation mobile communication network, or private network communication link. The data packets are managed using sequential numbering, buffer queues, timeout retransmission, and cyclic redundancy check mechanisms. When a segment sequence number is detected to be discontinuous or a cyclic redundancy check fails, a retransmission request for the corresponding segment is triggered to ensure the integrity and correct order of the data received by the ground processing terminal.

[0088] Step S20: The ground processing terminal establishes a buffer area according to the task identifier and image identifier, verifies each data packet, and reassembles the original image data according to the fragment sequence number after the verification is passed. The remote sensing image is then matched and associated with the image attribute data according to the unified identifier.

[0089] The UAV transmits the collected remote sensing images and their corresponding image attribute data to the ground processing terminal in real time. During the transmission, the remote sensing images are divided into multiple data packets according to preset units. The ground processing terminal establishes a buffer area according to the task identifier and image identifier, verifies each data packet, and reassembles the original image data according to the fragment sequence number after the verification is passed. The remote sensing images are then matched and associated with the image attribute data according to the unified identifier.

[0090] In some embodiments, please refer to Figure 2The ground processing unit establishes a buffer based on the task identifier and image identifier, verifies each data packet, reassembles the original image data according to the fragment sequence number after successful verification, and matches and associates the remote sensing image with the image attribute data according to the unified identifier. This includes the following steps:

[0091] Step S201: After receiving each data packet, the ground processing terminal establishes a corresponding receiving buffer based on the task identifier and image frame number in each data packet.

[0092] Step S202: The ground processing terminal performs integrity verification on each data packet. If any data packet fails verification, an anomaly mark is generated and logged. The integrity verification is implemented based on any one or more of Cyclic Redundancy Check (CRC) codes, MD5 checksums, or hash digests.

[0093] Step S203: After each data packet passes verification, the ground processing terminal sorts and combines all data packets belonging to the same image frame number according to the fragment sequence number of each data packet to restore the original remote sensing image.

[0094] Step S204: The ground processing end matches and associates the restored original remote sensing image with the corresponding image attribute data by using the image frame number or the acquisition timestamp. When matching by acquisition timestamp, if the difference between the acquisition time of the remote sensing image and the recording time of the image attribute data is not greater than the preset time deviation threshold, then the two are determined to belong to the same acquisition frame.

[0095] Step S205: If a matching association cannot be completed using the image frame number and the acquisition timestamp, an anomaly marker is generated and logged.

[0096] Specifically, after receiving image data slices and image attribute data, the ground processing terminal first establishes a receiving buffer based on the task identifier and image frame number, and performs integrity verification, sequence verification, and data reassembly on each data slice. The integrity verification is implemented using one or more of the following methods: Cyclic Redundancy Check (CRC), MD5 checksum, or hash digest. When the slice set corresponding to a certain remote sensing image is complete and each slice passes verification, reassembly is performed according to the slice sequence number to restore the original remote sensing image. The ground processing terminal matches and associates the restored original remote sensing image with the corresponding image attribute data using the image frame number or acquisition timestamp. When matching by acquisition timestamp, if the difference between the acquisition time of the remote sensing image and the recording time of the image attribute data is not greater than a preset time deviation threshold, they are determined to belong to the same acquisition frame. If slice loss, disordered order, missing fields, or mismatch between image and attribute data occurs, an anomaly marker is generated and logged.

[0097] Step S30: The ground processing terminal renames the remote sensing images according to a unified naming rule. Based on the image identifier, it establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data, generates an associated record containing the image identifier, task identifier, acquisition time, spatial location, attitude information and storage path, and archives and stores it according to the task identifier.

[0098] The ground processing terminal establishes a buffer area based on the task identifier and image identifier, verifies each data packet, and reassembles the original image data according to the fragment sequence number after the verification is passed. After matching and associating the remote sensing image with the image attribute data according to the unified identifier, the ground processing terminal renames the remote sensing image according to the unified naming rules. Based on the image identifier, it establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data, generates an association record containing the image identifier, task identifier, acquisition time, spatial location, attitude information and storage path, and archives and stores it according to the task identifier.

[0099] In some embodiments, please refer to Figure 3 The ground processing unit renames remote sensing images according to a unified naming rule, establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data based on the image identifier, generates an associated record containing the image identifier, task identifier, acquisition time, spatial location, attitude information, and storage path, and archives and stores it according to the task identifier. This process includes:

[0100] Step S301: The ground processing terminal renames the restored remote sensing images according to a unified naming rule;

[0101] Step S302: The ground processing terminal establishes a unique mapping and binding relationship between the renamed remote sensing image and the corresponding image attribute data according to the image identifier, and generates an association record. The association record includes at least the image identifier, task identifier, acquisition time, spatial location, attitude information and storage path fields.

[0102] Step S303: The ground processing terminal creates an archive directory according to the task identifier, and archives and stores the renamed remote sensing images, corresponding image attribute data and the associated records into the archive directory.

[0103] Specifically, the ground processing unit performs image naming standardization on the reconstructed remote sensing images, automatically binds image data and image attribute data, and archives and stores them according to task identifiers. The image naming standardization process renames the received remote sensing images according to a unified naming rule in the format "task identifier_acquisition date_acquisition time_image frame number_longitude_latitude". The naming rule for image attribute data is set to a file format with the same name as the corresponding remote sensing image, or it is stored in association using a database index. The automatic binding process between image data and image attribute data is used to establish a unique mapping relationship between each frame of remote sensing image and its corresponding image attribute data. Specifically, the ground processing terminal matches the remote sensing image and image attribute data based on the image identifier and generates a corresponding association record. The association record includes at least the fields of image identifier, task identifier, acquisition time, spatial location, attitude information, and storage path, so that each frame of remote sensing image and its acquisition status, spatial location, and task attributes form a one-to-one correspondence. In some embodiments, the binding relationship can be implemented through a relational database, key-value index table, lightweight embedded database, or task cache mapping table. When the ground processing terminal receives new remote sensing image or image attribute data, it automatically queries the corresponding primary key field and completes the binding update.

[0104] For cases involving inconsistent acquisition times, missing image identifiers, incomplete attribute data fields, or duplicate reception, anomaly markers are generated and written to the verification log for subsequent manual review or automatic error correction. After completing naming standardization and automatic binding, the data is categorized and stored according to preset archiving rules. The ground processing terminal creates an archiving directory based on the task identifier, uniformly archiving and storing the renamed remote sensing images, corresponding image attribute data, associated records, and verification logs in the archiving directory, providing a structured data foundation for subsequent quality inspection, coverage analysis, overlap calculation, and automatic mosaicking.

[0105] Step S40: The ground processing terminal performs multiple quality checks on the archived remote sensing images and generates quality assessment indicators based on the results of each quality check; the ground processing terminal determines whether the preset mapping trigger conditions are met based on the quality assessment indicators, and automatically generates a mapping task when the conditions are met.

[0106] The ground processing unit renames the remote sensing images according to a unified naming rule. Based on the image identifier, it establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data, generating an association record containing the image identifier, task identifier, acquisition time, spatial location, attitude information, and storage path. After archiving and storing the image according to the task identifier, the ground processing unit performs multiple quality checks on the archived remote sensing images and generates quality evaluation indicators based on the results of each quality check. The ground processing unit then determines whether the preset mapping trigger conditions are met based on the quality evaluation indicators. If the conditions are met, a mapping task is automatically generated.

[0107] In some embodiments, the steps of performing multiple quality checks on archived remote sensing images at the ground processing end and generating quality assessment indicators based on the results of each quality check include:

[0108] Step S401: The ground processing terminal performs quality inspection on the archived remote sensing images and calculates the sharpness value of the remote sensing images. When the sharpness value is less than the preset sharpness threshold, it is determined that the image has a blur or out-of-focus problem.

[0109] The sharpness value of the remote sensing image was calculated using the Laplace variance evaluation index, and the remote sensing image was recorded. The resolution value is When the resolution value If the image is less than the preset sharpness threshold, it is determined that the image is blurry or out of focus.

[0110] Step S402, and statistical analysis of the pixel brightness distribution of the remote sensing image, when the proportion of bright pixels is greater than the first preset threshold or the proportion of low-brightness pixels is greater than the second preset threshold, it is determined that the image has an overexposure or underexposure problem.

[0111] Based on the grayscale histogram, the pixel brightness distribution of remote sensing images is statistically analyzed, and the proportion of bright pixels is calculated. and low brightness pixel ratio When the proportion of bright pixels Greater than the first preset threshold or low brightness pixel ratio Greater than the second preset threshold If the image is overexposed or underexposed, it is determined that the image has an overexposure or underexposure problem. Before mapping, images with abnormal exposure are removed to ensure that the images participating in the mapping have a suitable dynamic range of brightness and recognizable texture information, so as to avoid feature extraction and matching failure due to exposure problems.

[0112] in, High-brightness pixel ratio represents the proportion of pixels in an image whose brightness value exceeds the high-brightness threshold; Low brightness pixel percentage represents the proportion of pixels in an image whose brightness value is lower than the low brightness threshold. Pixel brightness value, representing the grayscale value or brightness value of each pixel in the image; This is the overexposure threshold, also known as the first preset threshold. This is the underexposure threshold, also known as the second preset threshold. This is the brightness threshold, used to distinguish the brightness of highlighted pixels from normal pixels; The low-brightness threshold is a brightness threshold used to distinguish low-brightness pixels from normal pixels; This refers to the number of highlighted pixels, with a brightness value greater than [missing information]. The number of pixels; The number of low-brightness pixels, with a brightness value less than The number of pixels; This indicates the total number of pixels in the image.

[0113] Step S403 involves calculating the similarity value between different remote sensing images. When the similarity value is greater than a preset similarity threshold and the distance between the corresponding shooting locations is less than a preset distance threshold, it is determined that there are duplicate images.

[0114] Duplicate image detection can employ perceptual hashing, feature descriptor similarity, or joint temporal and spatial constraints. Perceptual hashing scales the image to a fixed size and calculates a hash value, measuring similarity by comparing the Hamming distance (i.e., the number of different bits) between the hash values ​​of two images. Feature descriptor methods calculate similarity by extracting and matching local feature points from the images. Joint temporal and spatial constraints combine the acquisition time interval and the distance between the shooting locations for a comprehensive judgment. For example, when the perceptual hash Hamming distance between two remote sensing images... The similarity is less than the preset similarity threshold, and the distance to the corresponding shooting location is... If the distance is less than a preset distance threshold, it is determined that there is a duplicate image.

[0115] Step S404 involves comparing the theoretical number of images in the preset flight path shooting plan with the actual number of received images. If the distance between adjacent acquisition points is greater than the preset distance threshold or the actual number of received images is less than the theoretical number of images, it is determined that there are missing images.

[0116] Specifically, missing image detection is used to identify missed or lost images due to reasons such as packet loss during transmission, camera failure to trigger, or storage anomalies. The theoretical number of images is calculated based on a preset flight path shooting plan (theoretical shooting positions determined by parameters such as flight path length, flight altitude, and overlap), and compared with the number of images actually received and successfully reconstructed. Simultaneously, spatial continuity can be determined by the distance between adjacent acquisition points. When the number of actually received images is less than the theoretical number of images, it indicates that images are missing or missed. Identifying incomplete data before mapping provides a basis for supplementary flight decisions. Missing images can lead to data gaps or discontinuous coverage in the mapping results.

[0117] Step S405: Project the field of view of each remote sensing image onto the ground to form a coverage polygon based on the geographic coordinates of each image, calculate the ratio of the union of each coverage polygon to the area of ​​the target area as the coverage rate, and determine that the coverage is insufficient when the coverage rate is less than a preset coverage rate threshold.

[0118] Coverage integrity detection can be achieved by projecting the field of view of each image onto the ground to form a coverage polygon. And calculate the target area. coverage When coverage Below the preset coverage threshold At that time, it was determined that there was insufficient coverage. This represents the ratio of the area covered by the union of all images to the area of ​​the target region, i.e., the coverage rate; Indicates the first The overlay polygon of the frame image; Indicates the target area; Represents the area function; This indicates the preset coverage threshold.

[0119] Step S406 involves calculating the overlap ratio of adjacent remote sensing images in the ground coverage area as the overlap degree. If the average forward overlap degree is not less than the third preset ratio threshold and the average lateral overlap degree is not less than the fourth preset ratio threshold, the overlap degree is determined to meet the mapping requirements; otherwise, the overlap degree is determined not to meet the mapping requirements. Forward overlap degree refers to the degree of overlap between two adjacent frames of images along the UAV's flight direction within the same flight strip in the ground coverage area. Lateral overlap degree refers to the degree of overlap between two corresponding frames of images on the left and right sides of adjacent flight strips in the ground coverage area.

[0120] The overlap ratio of adjacent remote sensing images in the ground coverage area is calculated as the overlap degree, which is expressed as: If the average forward overlap is not less than a third preset threshold and the average lateral overlap is not less than a fourth preset threshold, the overlap is determined to meet the mapping requirements; otherwise, the overlap is determined not to meet the mapping requirements. For the degree of overlap, the first Frame and the The overlap ratio of the ground coverage area of ​​the frame image; For the first The overlay polygon of the frame image; For the first The overlay polygon of the frame image; To cover the intersection of polygons, let represent the th Frame and the The common portion of the frame image in the ground coverage area; The intersection area represents the area of ​​the overlapping region; The smaller coverage area indicates the smaller of the two coverage areas; Represents the average forward overlap, which is the sum of all forward-adjacent image pairs. The average value; This represents the average lateral overlap, which is the sum of all lateral adjacent image pairs. The average value.

[0121] Step S407: Generate quality assessment indicators based on the results of various quality tests.

[0122] The results of image sharpness detection (whether it is blurry or out of focus), exposure detection (whether it is overexposed or underexposed), duplication detection (whether it is duplicated), missing detection (whether there are missed shots), coverage integrity detection, and overlap detection are uniformly organized to generate structured data that can be used for subsequent mapping trigger condition judgment. The quality detection results of each dimension are aggregated into a unified data structure, providing a data foundation for mapping trigger condition judgment.

[0123] In a further embodiment, the step of the ground processing terminal determining whether the preset mapping triggering conditions are met based on the quality assessment indicators, and automatically generating a mapping task when the conditions are met, includes:

[0124] Step S4001: The ground processing terminal determines whether the preset mapping triggering conditions are met based on the quality assessment indicators. The preset mapping triggering conditions include: the ratio of the actual number of received images to the theoretical number of images is not less than a first preset ratio threshold, the coverage rate is not less than a second preset ratio threshold, the average forward overlap is not less than a third preset ratio threshold, the average lateral overlap is not less than a fourth preset ratio threshold, or the ratio of the number of abnormal images to the total number of images is not greater than a fifth preset ratio threshold.

[0125] Step S4002: When any of the above mapping triggering conditions are met, the ground processing terminal automatically generates a mapping task.

[0126] For different application scenarios, the preset mapping triggering condition package can also adopt single threshold triggering, combined threshold triggering or hierarchical triggering methods to meet the needs of fast preview mapping and high-precision mapping tasks respectively.

[0127] Step S50: For the generated mapping task, the ground processing end adaptively generates mapping processing parameters based on the task feature information corresponding to the image data; calls the mapping processing engine, performs automatic mapping processing on the image data based on the mapping processing parameters, outputs the mapping results, and feeds back the relevant information of the mapping results to the user end.

[0128] The ground processing unit performs multiple quality checks on the archived remote sensing images and generates quality assessment indicators based on the results of each quality check. The ground processing unit determines whether the preset mapping trigger conditions are met based on the quality assessment indicators. If the conditions are met, a mapping task is automatically generated. The ground processing unit then adaptively generates mapping processing parameters for the generated mapping task based on the task feature information corresponding to the image data. The mapping processing engine is invoked to perform automatic mapping processing on the image data based on the mapping processing parameters, outputting the mapping results and feeding back the relevant information of the mapping results to the user terminal.

[0129] In some embodiments, please refer to Figure 4 The ground processing unit adaptively generates mapping processing parameters based on the task feature information corresponding to the image data for the generated mapping task; it then calls the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters and outputs the mapping results. The steps include:

[0130] Step S501: The ground processing end extracts the task feature vector corresponding to the generated mapping task, wherein the task feature vector includes ground resolution, image quality index, area, terrain complexity and mapping result type.

[0131] Step S502: The ground processing end adaptively generates stitching and mapping parameters based on the task feature vector through a parameter mapping table or rule base. The stitching and mapping parameters include image scaling level, number of feature points extracted, reprojection error threshold, output resolution, and coordinate projection parameters.

[0132] Step S503: The ground processing terminal calls the mapping processing engine to perform automatic mosaicking and mapping processing on the image data based on the mosaicking and mapping parameters, and outputs the mapping result, wherein the mapping result includes at least one of orthophoto map, digital surface model or three-dimensional model.

[0133] For the generated mapping task, the ground processing end extracts the task feature vector corresponding to the mapping task. Based on the task feature vector, it adaptively generates stitching mapping parameters through a parameter mapping table or rule base. The task feature vector includes ground resolution, image quality index, area, terrain complexity, and mapping result type. The stitching mapping parameters include image scaling level, number of feature points extracted, reprojection error threshold, output resolution, and coordinate projection parameters.

[0134] The task feature vector is represented as follows: ,in For ground resolution, The average forward overlap, The average lateral overlap, Image quality indicators. For the area, Due to terrain complexity, The mapping result type; the stitching mapping parameters are expressed as follows: ,in For image scaling levels, The number of feature points extracted. The reprojection error threshold, For output resolution, These are the coordinate projection parameters.

[0135] Furthermore, the adaptive generation of stitched mapping parameters can be implemented using a preset parameter mapping table, a rule-based matching method, a model prediction method, or a combination of the above methods. Specifically, the preset parameter mapping table can pre-store corresponding parameter templates based on the ground resolution range and mapping result type; the rule-based matching method can automatically select the corresponding mapping strategy based on data quality, terrain conditions, and task objectives; and the model prediction method can optimize and recommend parameters based on historical task data.

[0136] In some implementations, when the ground resolution is high, the image quality is good, and the overlap is sufficient, high-precision mapping parameters are automatically selected; when the task objective is rapid preview, the feature extraction density and reconstruction accuracy are automatically reduced to improve processing speed.

[0137] The ground processing terminal calls the mapping processing engine to perform automatic mosaicking and mapping processing on the image data based on the mosaicking and mapping parameters, outputs the mapping results, and feeds back the relevant information of the mapping results to the user terminal.

[0138] Preferably, the automatic mosaicking process includes: performing feature extraction and matching on image data, removing mismatched points using a random sampling consensus algorithm, performing aerial triangulation and pose optimization, generating a dense point cloud or surface model, and outputting at least one of an orthophoto map, a digital surface model, or a 3D model. Specifically, the feature extraction algorithm employs any one of the following: Scale Invariant Feature Transform (SIFT), Speed-Up Robust Feature Transform (SURF), or Rotation-Oriented Enhancement (ORB); the matched geometric relationships are estimated using a homography matrix or fundamental matrix, and the reprojection error is minimized using a random sampling consensus algorithm (RANSAC).

[0139] In the optimization stage of aerial triangulation, the objective function is to minimize the reprojection error through bundle adjustment. After mapping is completed, the ground processing terminal sends the mapping result file path, task status, processing time, and abnormal alarm information to the user terminal to indicate whether re-collection, re-flight, or re-mapping is required. It is the square norm; Let be the observed value, representing the th . Pixel coordinates of the k-th feature point on the frame image The reprojection error represents the reprojection error of a point in three-dimensional space. Projected to the Predicted pixel position on frame image vs. actual observed position The difference between them. When the camera pose is accurate, the projected point should coincide with the observation point, and the error should approach zero; For projection functions, points in three-dimensional space are projected. Functions that project onto a two-dimensional image plane, including camera imaging geometric transformations such as perspective projection or distortion models; Let be the camera intrinsic parameter matrix, representing the th The camera's internal parameters for a frame image, including focal length, principal point coordinates, distortion coefficients, etc., determine how three-dimensional points are mapped to the image pixel coordinate system. Let be a rotation matrix, representing the first... The camera attitude rotation component of the frame image describes the orientation of the UAV at that moment (e.g., yaw angle, pitch angle, roll angle, etc.). Let be the translation vector, representing the first... The camera position translation component of the frame image describes the UAV's spatial three-dimensional coordinates (e.g., position in the longitude, latitude, and altitude directions) at that moment. For a three-dimensional point, it represents the spatial coordinates of the k-th three-dimensional point, that is, the true three-dimensional position of the feature point on the surface of the ground object.

[0140] As can be seen from the above embodiments, compared with the prior art, this application addresses the problems of data acquisition and mapping processing separation, lagging quality control, low automation, and difficulty in balancing real-time performance and accuracy in the prior art. This application has, but is not limited to, the following beneficial effects:

[0141] Firstly, this application utilizes a UAV to simultaneously record image attribute data such as timestamps, geographical location, attitude angles, flight altitude, and camera parameters while acquiring remote sensing images. The image data and attribute data are then transmitted in real-time to a ground processing unit via a wireless link, eliminating the need to wait for data processing after the flight mission. The ground processing unit performs real-time parsing, integrity verification, reconstruction and restoration, and automatic binding and archiving of the received data, achieving fully automated processing from image acquisition to archiving and storage. Compared to the traditional "post-processing via software after flight" model, this application can simultaneously complete data reception, processing, and mapping trigger condition determination during flight, fundamentally solving the problems of long processing links and insufficient mapping timeliness in existing technologies. This meets the application scenarios with high requirements for rapid mapping, such as emergency mapping and agricultural monitoring.

[0142] Secondly, this application performs multiple quality checks on the archived remote sensing images in real time during flight, including sharpness detection, exposure anomaly detection, duplicate image detection, missing image detection, coverage integrity detection, and overlap detection. Each check is based on clear mathematical or physical judgment standards. Image quality issues can be detected and anomaly markers recorded in real time during flight, eliminating the need to wait until after the flight to check in post-processing software. When issues such as missing images or insufficient coverage are detected, the UAV can promptly perform supplementary acquisitions or flights, avoiding the repetitive work costs and time delays caused by discovering data defects hours or even days after the flight in the traditional mode, which necessitates reorganizing supplementary flights. This significantly improves the reliability and economy of the operation.

[0143] Thirdly, this application automatically determines whether the preset mapping triggering conditions are met based on the quality inspection results through the ground processing terminal. When all conditions are met, it automatically generates a mapping task and adaptively generates stitching mapping parameters based on task feature information, then calls the mapping processing engine to execute automatic mapping processing. From data reception, quality inspection, condition judgment, task triggering to parameter generation and mapping execution, the entire process is completed automatically. There is no need for manual data processing, manual data quality checking, manual judgment of mapping timing, or manual setting of mapping parameters, which effectively reduces the dependence of mapping operations on the professional experience of operators and is conducive to the promotion and implementation of large-scale, routine UAV remote sensing mapping operations.

[0144] Fourth, this application sets multi-dimensional quality inspection thresholds in the quality inspection stage, including sharpness, exposure, coverage integrity, and overlap. In the mapping triggering stage, it sets multiple combined threshold conditions for comprehensive judgment; the mapping task is only triggered when all conditions are met. This multi-dimensional combined guarantee mechanism ensures that the image data participating in mapping meets the mapping requirements in terms of quantity, spatial coverage, directional geometric constraints, lateral geometric constraints, and data validity, thereby guaranteeing the accuracy and reliability of the output mapping results. It can adaptively adjust the mapping strategy according to different task requirements, while taking into account both processing timeliness and result accuracy, effectively supporting diverse application scenarios such as emergency mapping, agricultural monitoring, and rapid mapping.

[0145] Fifth, this application feeds back the file path, task status, processing time, and abnormal alarm information of the mapping results to the user terminal, clearly indicating whether re-collection, re-flight, or re-mapping is required. From image acquisition, real-time transmission, quality inspection, automatic mapping to result feedback, a complete data processing closed loop is formed. Users can monitor the execution status and result quality of the mapping task in real time through the user terminal, making subsequent operational decisions accordingly, effectively improving the management efficiency and overall operational efficiency of UAV remote sensing mapping tasks.

[0146] Please see Figure 5This application provides a real-time quality inspection and automatic mapping control device for UAV remote sensing images, comprising a data transmission module 1100, a data association and reassembly module 1200, a data archiving and storage module 1300, a mapping triggering module 1400, and an automatic mapping module 1500. The data transmission module 1100 is configured to transmit the acquired remote sensing images and their corresponding image attribute data from the UAV to the ground processing terminal in real time, dividing the remote sensing images into multiple data packets according to preset units during transmission. The data association and reassembly module 1200 is configured to allow the ground processing terminal to establish a buffer based on task identifiers and image identifiers, verify each data packet, and reassemble the original image data according to the segment sequence number after verification, and match and associate the remote sensing images with image attribute data according to a unified identifier. The data archiving and storage module 1300 is configured to allow the ground processing terminal to rename the remote sensing images according to a unified naming rule, establish a unique mapping and binding relationship between the renamed image data and the corresponding attribute data based on the image identifier, and generate a data archiving and storage module containing the image identifier, task identifier, and image attribute data. The system collects associated records of time, spatial location, attitude information, and storage path, and archives them according to task identifiers. A mapping trigger module 1400 is configured to perform multiple quality checks on the archived remote sensing images on the ground processing end, and generate quality assessment indicators based on the results of each quality check. The ground processing end determines whether preset mapping trigger conditions are met based on the quality assessment indicators, and automatically generates a mapping task when met. An automatic mapping module 1500 is configured to adaptively generate mapping processing parameters based on the task feature information corresponding to the image data for the generated mapping task on the ground processing end; it calls the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters, outputs the mapping results, and feeds back the relevant information of the mapping results to the user end.

[0147] Based on any embodiment of this application, please refer to Figure 6 Another embodiment of this application also provides an electronic device, which can be implemented by a computer device, such as... Figure 6The diagram shows the internal structure of a computer device. This computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. The computer-readable storage medium stores an operating system, a database, and computer-readable instructions. The database may store control information sequences. When the processor executes the computer-readable instructions, it enables the processor to implement a real-time quality inspection and automatic mapping control method for UAV remote sensing images. The processor provides computing and control capabilities, supporting the operation of the entire computer device. The memory stores computer-readable instructions, which, when executed by the processor, enable the processor to execute the UAV remote sensing image real-time quality inspection and automatic mapping control method of this application. The network interface of the computer device is used for communication with a terminal. Those skilled in the art will understand that… Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0148] In this embodiment, the processor is used to execute... Figure 5 The specific functions of each module are defined within the device, and the memory stores the program code and various data required to execute these modules. The network interface is used for data transmission between the user terminal and the server. In this embodiment, the memory stores the program code and data required to execute all modules in the UAV remote sensing image real-time quality inspection and automatic mapping control device of this application. The server can call the server's program code and data to execute the functions of all modules.

[0149] This application also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the real-time quality inspection and automatic mapping control method for UAV remote sensing images described in any embodiment of this application.

[0150] This application also provides a computer program product, including a computer program / instruction, which, when executed by one or more processors, implements the steps of the real-time quality inspection and automatic mapping control method for UAV remote sensing images described in any embodiment of this application.

[0151] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. This computer program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0152] The above description is only a partial embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for real-time quality inspection and automatic mapping control of UAV remote sensing images, characterized in that, include: The drone will transmit the collected remote sensing images and their corresponding image attribute data to the ground processing terminal in real time. During the transmission process, the remote sensing images will be divided into multiple data packets according to preset units. The ground processing terminal establishes a buffer area according to the task identifier and image identifier, verifies each data packet, and reassembles the original image data according to the fragment sequence number after the verification is passed. The remote sensing image is then matched and associated with the image attribute data according to the unified identifier. The ground processing terminal renames the remote sensing images according to a unified naming rule. Based on the image identifier, it establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data, generates an associated record containing the image identifier, task identifier, acquisition time, spatial location, attitude information and storage path, and archives and stores it according to the task identifier. The ground processing unit performs multiple quality checks on the archived remote sensing images and generates quality assessment indicators based on the results of each quality check. The ground processing unit then determines whether the preset mapping trigger conditions are met based on the quality assessment indicators, and automatically generates a mapping task when the conditions are met. For the generated mapping task, the ground processing end adaptively generates mapping processing parameters based on the task feature information corresponding to the image data; it calls the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters, outputs the mapping results, and feeds back the relevant information of the mapping results to the user end.

2. The method for real-time quality inspection and automatic mapping control of UAV remote sensing images according to claim 1, characterized in that, The UAV transmits the collected remote sensing images and their corresponding image attribute data to the ground processing terminal in real time. During transmission, the remote sensing images are divided into multiple data packets according to preset units, including: The remote sensing image is divided into multiple data slices according to a preset slice size. Each data slice is encapsulated into a data packet. The image attribute data is associated with the remote sensing image through the image frame number. The image attribute data includes a task identifier, image frame number, acquisition timestamp, geographic coordinates, attitude angle, camera intrinsic parameter matrix, extrinsic parameter matrix, and flight altitude. Each data packet contains a task identifier, image frame number, slice sequence number, and verification field. The data packets are managed through a sequence numbering, buffer queue, timeout retransmission, and cyclic redundancy check mechanism. When a fragment sequence number is found to be discontinuous or the cyclic redundancy check fails, a retransmission request for the corresponding fragment is triggered to ensure the integrity and correct order of the data received by the ground processing end.

3. The method for real-time quality inspection and automatic mapping control of UAV remote sensing images according to claim 1, characterized in that, The ground processing unit establishes a buffer based on task identifiers and image identifiers, verifies each data packet, reassembles the original image data according to the fragment sequence number after successful verification, and matches and associates the remote sensing image with the image attribute data according to a unified identifier. This process includes: After receiving each data packet, the ground processing unit establishes a corresponding receiving buffer based on the task identifier and image frame number in each data packet; The ground processing unit performs integrity verification on each data packet. If any data packet fails verification, an exception flag is generated and logged. The integrity verification is implemented based on any one or more of the following methods: Cyclic Redundancy Check (CRC), MD5 checksum, or hash digest. After each data packet passes verification, the ground processing unit sorts and combines all data packets belonging to the same image frame number according to the fragment sequence number of each data packet to restore the original remote sensing image. The ground processing unit matches and associates the restored original remote sensing image with the corresponding image attribute data by using the image frame number or the acquisition timestamp. When matching by acquisition timestamp, if the difference between the acquisition time of the remote sensing image and the recording time of the image attribute data is not greater than the preset time deviation threshold, it is determined that the two belong to the same acquisition frame. If a match cannot be established using either the image frame number or the acquisition timestamp, an anomaly marker is generated and logged.

4. The method for real-time quality inspection and automatic mapping control of UAV remote sensing images according to claim 1, characterized in that, The ground processing unit renames remote sensing images according to a unified naming rule, establishes a unique mapping and binding relationship between the renamed image data and the corresponding attribute data based on the image identifier, generates an associated record containing the image identifier, task identifier, acquisition time, spatial location, attitude information, and storage path, and archives and stores it according to the task identifier. This process includes: The ground processing unit renames the restored remote sensing images according to a unified naming rule; The ground processing terminal establishes a unique mapping and binding relationship between the renamed remote sensing image and the corresponding image attribute data based on the image identifier, and generates an association record. The association record includes at least the image identifier, task identifier, acquisition time, spatial location, attitude information and storage path fields. The ground processing unit creates an archive directory according to the task identifier, and archives and stores the renamed remote sensing images, corresponding image attribute data, and associated records in the archive directory.

5. The method for real-time quality inspection and automatic mapping control of UAV remote sensing images according to claim 1, characterized in that, The steps involved in the ground processing unit performing multiple quality checks on the archived remote sensing images and generating quality assessment indicators based on the results of each quality check include: The ground processing unit performs quality inspection on the archived remote sensing images and calculates the sharpness value of the remote sensing images. When the sharpness value is less than the preset sharpness threshold, it is determined that the image has a blur or out-of-focus problem. In addition, the pixel brightness distribution of the remote sensing image is statistically analyzed. When the proportion of bright pixels is greater than the first preset threshold or the proportion of dark pixels is greater than the second preset threshold, the image is determined to have overexposure or underexposure problems. And calculate the similarity value between different remote sensing images. When the similarity value is greater than a preset similarity threshold and the distance between the corresponding shooting locations is less than a preset distance threshold, it is determined that there are duplicate images. And compare the theoretical number of images in the preset flight route shooting plan with the actual number of received images. When the distance between adjacent acquisition points is greater than the preset distance threshold or the actual number of received images is less than the theoretical number of images, it is determined that there are missing images. And based on the geographic coordinates of each remote sensing image, the field of view of each image is projected onto the ground to form a coverage polygon. The ratio of the union of each coverage polygon to the area of ​​the target area is calculated as the coverage rate. When the coverage rate is less than a preset coverage rate threshold, it is determined that the coverage is insufficient. And calculate the overlap ratio of adjacent remote sensing images in the ground coverage area as the overlap degree. When the average forward overlap degree is not less than the third preset ratio threshold and the average lateral overlap degree is not less than the fourth preset ratio threshold, it is determined that the overlap degree meets the mapping requirements; otherwise, it is determined that the overlap degree does not meet the mapping requirements. Quality assessment indicators are generated based on the results of various quality tests.

6. The method for real-time quality inspection and automatic mapping control of UAV remote sensing images according to claim 5, characterized in that, The ground processing terminal determines whether the preset mapping triggering conditions are met based on the quality assessment indicators, and automatically generates a mapping task when the conditions are met, including: The ground processing terminal determines whether the preset mapping triggering conditions are met based on the quality assessment indicators. The preset mapping triggering conditions include: the ratio of the actual number of received images to the theoretical number of images is not less than a first preset ratio threshold, the coverage is not less than a second preset ratio threshold, the average forward overlap is not less than a third preset ratio threshold, the average lateral overlap is not less than a fourth preset ratio threshold, or the ratio of the number of abnormal images to the total number of images is not greater than a fifth preset ratio threshold. When any of the above mapping triggering conditions are met, the ground processing terminal automatically generates a mapping task.

7. The method for real-time quality inspection and automatic mapping control of UAV remote sensing images according to any one of claims 1 to 6, characterized in that, The ground processing unit adaptively generates mapping processing parameters based on the task feature information corresponding to the image data for the generated mapping task; it then calls the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters and outputs the mapping results. The steps include: The ground processing unit extracts the task feature vector corresponding to the generated mapping task, wherein the task feature vector includes ground resolution, image quality index, area, terrain complexity and mapping result type; The ground processing unit adaptively generates stitching and mapping parameters based on the task feature vector through a parameter mapping table or rule base. The stitching and mapping parameters include image scaling level, number of feature points extracted, reprojection error threshold, output resolution, and coordinate projection parameters. The ground processing unit calls the mapping processing engine to perform automatic mosaicking and mapping processing on the image data based on the mosaicking and mapping parameters, and outputs the mapping results, wherein the mapping results include at least one of orthophoto map, digital surface model or three-dimensional model.

8. A real-time quality inspection and automatic mapping control device for UAV remote sensing images, characterized in that, include: The data transmission module is configured to transmit the collected remote sensing images and their corresponding image attribute data from the UAV to the ground processing terminal in real time. During the transmission process, the remote sensing images are divided into multiple data packets according to preset units. The data association and reassembly module is configured to establish a buffer area on the ground processing end according to the task identifier and image identifier, verify each data packet, and reassemble the original image data according to the fragment sequence number after the verification is passed, and match and associate the remote sensing image with the image attribute data according to the unified identifier. The data archiving and storage module is configured to rename remote sensing images according to a unified naming rule on the ground processing end, establish a unique mapping and binding relationship between the renamed image data and the corresponding attribute data based on the image identifier, generate an associated record containing image identifier, task identifier, acquisition time, spatial location, attitude information and storage path, and archive and store it according to the task identifier. The mapping trigger module is configured to perform multiple quality checks on the archived remote sensing images by the ground processing end, and generate quality assessment indicators based on the results of each quality check; the ground processing end determines whether the preset mapping trigger conditions are met based on the quality assessment indicators, and automatically generates a mapping task when the conditions are met. The automatic mapping module is configured to, on the ground processing end, adaptively generate mapping processing parameters based on the task feature information corresponding to the image data for the generated mapping task; call the mapping processing engine to perform automatic mapping processing on the image data based on the mapping processing parameters, output the mapping results, and feed back the relevant information of the mapping results to the user end.

9. An electronic device comprising a central processing unit and a memory, characterized in that, The central processing unit is used to invoke and run a computer program stored in the memory to perform the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores, in the form of computer-readable instructions, a computer program implemented according to any one of claims 1 to 7, which, when invoked by a computer, executes the steps included in the corresponding method.

Images