Geographic information detection system based on multi-information graph remote sensing data

Abstract

The invention relates to the technical field of electric digital data processing, and particularly discloses a geographic information detection system based on multi-information graph remote sensing data, which is characterized in that a data acquisition equipment optimization module, a remote sensing data preprocessing module and a target detection and feedback module are arranged, and operation data of data acquisition equipment is acquired according to the data acquisition equipment connected with a remote sensing platform; secondly, the remote sensing platform receives information graph data transmitted by each device as target area remote sensing data, data fusion, feature extraction and feature fusion are performed to obtain anchor frame detection geographic data, the data are matched with actual data of the target area, and the target area remote sensing data is obtained; according to the method, the matching accuracy index is evaluated and verified with the preset value, the geographic information detection of the multi-information graph remote sensing data is completed according to the verification result, and the process aims at improving the accuracy and efficiency of the geographic information detection by optimizing the equipment quality and fusing the multi-source information.

Description

Technical Field

[0001] The invention relates to the technical field of electronic digital data processing, in particular to a geographic information detection system based on multi-information graph remote sensing data. Background Art

[0002] With the continuous advancement of aerospace remote sensing technology, the spatial resolution of high-resolution remote sensing images can gradually reach the decimeter and centimeter levels. Such high-resolution remote sensing images provide a more sophisticated data source for geographic information detection, enabling automatic detection and identification of fine ground objects in remote sensing images, providing an increasingly rich and diverse source of information for resource, environment, and disaster investigation and monitoring.

[0003] For example, the invention patent with announcement number CN109165272B announces a geographic information visualization analysis system for high-resolution remote sensing images, including overall system construction, remote sensing data processing module, and target detection module; establishing a functional visualization sharing system; and providing a target detection algorithm that accurately captures key target location information in response to the problems of massive information processing and key target location information extraction from remote sensing data caused by the big data of remote sensing satellite images.

[0004] For example, the invention patent with announcement number CN103488801B announces an airport target detection method based on a geographic information space library, which involves the field of target detection technology in UAV computer vision. Offline, the direction, feature lines, feature line distance and other attribute information of the airport target are extracted based on satellite remote sensing images and geographic information; online, the line features in the UAV remote sensing image are extracted based on the direction of the current suspected airport in the geographic information space of the remote sensing image acquired in real time, with reference to the accuracy of image correction in the UAV remote sensing image processing system. Finally, the airport target in the remote sensing image is located by comprehensively judging the degree of matching between the feature lines on the UAV remote sensing image and the feature lines of the suspected airport target in the geographic information space.

[0005] Combining the above technical solutions, it is found that there are geographic information detection technical solutions that only perform target detection on single remote sensing information, so that the obtained geographic information has fewer dimensions and its data accuracy needs to be considered. Therefore, it cannot achieve the purpose of accurately locating the target, and ultimately affects the detection of specific target geographic information. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a geographic information detection system based on multi-information graph remote sensing data, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a geographic information detection system based on multi-information graph remote sensing data, including a data acquisition device optimization module, which is used to obtain the operating data of each data acquisition device connected to the remote sensing platform, determine the operating quality evaluation value of each data acquisition device, and compare it with a predefined operating quality evaluation threshold to determine whether to optimize the operating quality of each data acquisition device; a remote sensing data preprocessing module, which is used to obtain the information graphs transmitted by each data acquisition device received by the remote sensing platform, recorded as each information graph remote sensing data of the target area, and sequentially perform information graph fusion operations, feature extraction operations and feature fusion operations, and the final results are recorded as each anchor frame detection geographic data of the target area; a target detection and feedback module, which is used to obtain the actual anchor frame detection geographic data of the target area, match it with the anchor frame detection geographic data of the target area, determine the matching accuracy index of each anchor frame detection geographic data, and verify it with the predefined matching accuracy index preset value, and obtain and complete the geographic information detection of the multi-information graph remote sensing data based on the verification result.

[0008] As a further solution, the operation quality evaluation value of each data acquisition device is determined, and the specific determination process is: the data processing amount of each data acquisition device in the operation quality evaluation period is ratioed with the corresponding duration of the operation quality evaluation period to obtain the data processing speed of each data acquisition device in the operation quality evaluation period; the power value of each data acquisition device at the start time of the operation quality evaluation is differenced with the power value of each data acquisition device at the end time of the operation quality evaluation to obtain the power consumption of each data acquisition device in the operation quality evaluation period, and the power value of each data acquisition device at the start time of the operation quality evaluation is ratioed to obtain the energy consumption proportion of each data acquisition device in the operation quality evaluation period; the network transmission quality data between each data acquisition device and the remote sensing platform is obtained. Specifically, it includes the network transmission bandwidth, network transmission packet loss rate and network delay time between each data acquisition device and the remote sensing platform within the operation quality assessment period, and the network transmission quality impact coefficient between each data acquisition device and the remote sensing platform is obtained by comprehensive analysis; the operation response definition time, data processing reference speed and operation temperature reference value are extracted from the remote sensing platform information database; according to the operation response time of each data acquisition device within the operation quality assessment period, the average operation temperature of each data acquisition device within the operation quality assessment period, the data processing speed of each data acquisition device within the operation quality assessment period, the energy consumption proportion of each data acquisition device within the operation quality assessment period and the network transmission quality impact coefficient between each data acquisition device and the remote sensing platform, the operation quality assessment value of each data acquisition device is obtained by comprehensive analysis.

[0009] As a further solution, the determination of whether to optimize the operating quality of each data acquisition device is as follows: comparing the operating quality evaluation value of each data acquisition device with the operating quality evaluation threshold predefined in the remote sensing platform information library; if the operating quality evaluation value of a data acquisition device is greater than or equal to the operating quality evaluation threshold, there is no need to optimize the operating quality of the data acquisition device; if the operating quality evaluation value of a data acquisition device is less than the operating quality evaluation threshold, it is necessary to optimize the operating quality of the data acquisition device.

[0010] As a further solution, the information graph remote sensing data of the target area are sequentially subjected to information graph fusion operation, feature extraction operation and feature fusion operation, and the specific operation process is as follows: the information graph fusion operation is specifically to preprocess the information graph remote sensing data of the target area, and to perform superposition and fusion operation on the preprocessed information graph remote sensing data according to a preset channel sequence, and the final result is recorded as the information graph data of each channel; the feature extraction operation is specifically to use a multi-layer convolutional neural network structure for the information graph data of each channel, and to extract the spatial features of the information graph data of each channel in sequence through a preset channel sequence to obtain the feature data of each channel information graph; the feature fusion operation is specifically to use a cross-layer connection strategy to perform feature fusion on the feature data of each channel information graph, and the final result is recorded as the anchor frame detection geographic data of the target area.

[0011] As a further solution, the matching accuracy index of the geographic data detected by each anchor frame is determined. The specific determination process is: according to the length deviation value of each anchor frame, the width deviation value of each anchor frame, the height deviation value of each anchor frame, the horizontal coordinate deviation value of the center point of each anchor frame, and the vertical coordinate deviation value of the center point of each anchor frame, the matching accuracy index of the geographic data detected by each anchor frame is obtained by comprehensive processing. The specific method is as follows: ; In the formula, is the matching accuracy index of the mth anchor frame to detect geographic data, m is the number of each anchor frame, , M is the total amount of anchor boxes, is the length deviation value of the mth anchor box, is the width deviation value of the mth anchor box, is the height deviation value of the mth anchor box, is the horizontal coordinate deviation value of the center point of the mth anchor box, is the vertical coordinate deviation value of the center point of the mth anchor box, is the operating quality evaluation value of the bth data acquisition device, b is the serial number of each data acquisition device, , B is the total amount of data acquisition equipment, The matching accurate influencing parameter corresponding to the predefined length deviation value in the remote sensing platform information database, The matching accurate influencing parameter corresponding to the predefined width deviation value in the remote sensing platform information library, The accurate influencing parameters corresponding to the predefined height deviation values ​​in the remote sensing platform information database are matched. It is the matching accurate influencing parameter corresponding to the horizontal coordinate deviation value of the center point predefined in the remote sensing platform information database. It is the matching accurate influencing parameter corresponding to the ordinate deviation value of the center point predefined in the remote sensing platform information database. It is the matching accurate influencing parameter corresponding to the predefined operation quality assessment value in the remote sensing platform information database, and e is a natural constant.

[0012] As a further solution, the geographic information detection of multi-information map remote sensing data is completed based on the verification result, and the specific analysis process is: the verification result is specifically the first verification result or the second verification result; the first verification result is that the matching accuracy index of a certain anchor frame detection geographic data is equal to the matching accuracy index preset value; the second verification result is that the matching accuracy index of a certain anchor frame detection geographic data is not equal to the matching accuracy index preset value; if the verification result is displayed as the first verification result, the anchor frame detection geographic data corresponding to the matching accuracy index of the anchor frame detection geographic data is used, and the anchor frame corresponding to the anchor frame detection geographic data is obtained through a preset anchor frame design method, and is visualized on the remote sensing platform; if the verification result is displayed as the second verification result, the matching accuracy index of the anchor frame detection geographic data is used to feedback the anchor frame detection geographic data.

[0013] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention provides a geographic information detection system based on multi-information graph remote sensing data. According to the data acquisition device connected to the remote sensing platform, its operation data is obtained, the operation quality is evaluated and compared with the preset threshold value to decide whether to optimize. Then, the remote sensing platform receives the information graph data transmitted by each device as the remote sensing data of the target area, performs data fusion, feature extraction and feature fusion, obtains the geographic data of each anchor frame detection, and then matches this data with the actual data of the target area, evaluates the matching accuracy index and verifies it with the preset value, and completes the geographic information detection of the multi-information graph remote sensing data according to the verification result. This process aims to improve the accuracy and efficiency of geographic information detection by optimizing equipment quality and fusing multi-source information.

[0014] (2) The present invention obtains the operating data of each data acquisition device connected to the remote sensing platform and determines the operating quality evaluation value of each data acquisition device. By evaluating the operating quality of the equipment, equipment failure or performance degradation can be discovered in a timely manner, thereby ensuring the quality and accuracy of the remote sensing data. By continuously monitoring and optimizing the operating quality of the equipment, it can be ensured that the remote sensing platform can operate stably and efficiently.

[0015] (3) The present invention obtains the actual anchor frame detection geographic data of the target area, matches it with the various anchor frame detection geographic data of the target area, and determines the matching accuracy index of each anchor frame detection geographic data. By matching with the actual anchor frame detection geographic data, the accuracy and reliability of the target detection anchor frame can be verified. If the matching accuracy index is high, it means that the anchor frame can accurately identify and locate the geographic features in the target area, thereby improving the practicality and credibility of the anchor frame. By screening out the anchor frame detection geographic data with better performance, the processing of redundant or erroneous data is reduced, thereby improving the detection efficiency.

[0016] (4) The present invention efficiently fuses multiple remote sensing information graphs into multi-channel data, fully utilizes the multi-dimensional information of remote sensing data, improves the comprehensiveness and accuracy of target detection, and designs an adaptive anchor frame generation method based on the scale and shape of specific targets in remote sensing data, thereby significantly improving the accuracy of target positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0018] Figure 1 It is a schematic diagram of system module connection of the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0020] See also Figure 1 As shown, an embodiment of the present invention provides a technical solution: a geographic information detection system based on multi-information graph remote sensing data, including a data acquisition device optimization module, a remote sensing data preprocessing module and a target detection and feedback module.

[0021] An embodiment of the present invention provides a technical solution: a geographic information detection system based on multi-information map remote sensing data, which also includes: a remote sensing platform information library for storing operation response definition time, data processing reference speed, operation temperature reference value and preset values ​​of various parameters.

[0022] The data acquisition device optimization module is connected to the remote sensing data preprocessing module, the remote sensing data preprocessing module is connected to the target detection and feedback module, and the data acquisition device optimization module and the target detection and feedback module are both connected to the remote sensing platform information library.

[0023] The data acquisition device optimization module is used to obtain the operating data of each data acquisition device connected to the remote sensing platform, determine the operating quality evaluation value of each data acquisition device, and compare it with a predefined operating quality evaluation threshold to determine whether to optimize the operating quality of each data acquisition device.

[0024] The above-mentioned data acquisition devices can specifically be panchromatic cameras, multispectral cameras, multispectral scanners, three-dimensional laser scanners, etc. Panchromatic cameras can capture high-resolution images of ground targets and generate topographic maps through image processing technology; multispectral cameras can calculate vegetation indices such as NDVI (normalized difference vegetation index) by capturing the reflection or radiation information of ground targets in different spectral bands, and then generate vegetation index maps; three-dimensional laser scanners can perform high-precision data collection on the ground to supplement the deficiencies of aerial remote sensing data and improve the accuracy and integrity of remote sensing information.

[0025] The above-mentioned remote sensing platform is usually composed of remote sensing sensors, data recording devices, communication systems, power systems, thermal control systems, etc. These components work together to ensure that the remote sensing platform can stably and accurately obtain remote sensing data of ground targets. The remote sensing platform can record the exact position of the sensor to ensure that the acquired remote sensing data is of high precision and reliability. By carrying different types of remote sensing sensors, the remote sensing platform can obtain various types of remote sensing data, such as the above-mentioned optical images, radar images, multispectral images, etc.; the remote sensing platform can transmit the acquired remote sensing data to a ground station or other designated locations for subsequent processing and analysis.

[0026] Specifically, the operating data of each data acquisition device includes the operating response time of each data acquisition device during the operation quality assessment cycle, the average operating temperature of each data acquisition device during the operation quality assessment cycle, the data processing volume of each data acquisition device during the operation quality assessment cycle, the power value of each data acquisition device at the start time point of the operation quality assessment, and the power value of each data acquisition device at the end time point of the operation quality assessment.

[0027] The above-mentioned operation quality assessment cycle refers to a period of time used to conduct operation quality assessment on data acquisition equipment, wherein the start time point of the operation quality assessment is the first time point of the operation quality assessment cycle, and the end time point of the operation quality assessment is the last time point of the operation quality assessment cycle. The operation quality assessment cycle is determined by the geographic information detection personnel based on a comprehensive analysis of factors such as the equipment status, detection environment, and actual detection needs.

[0028] The above-mentioned operation response time, operation temperature, data processing amount and power value can all be extracted from the operation report of each data acquisition device.

[0029] Furthermore, the operation quality evaluation value of each data acquisition device is determined, and the specific determination process is as follows: The data processing amount of each data acquisition device in the operation quality assessment cycle is ratio-processed with the corresponding duration of the operation quality assessment cycle to obtain the data processing speed of each data acquisition device in the operation quality assessment cycle.

[0030] The power consumption of each data acquisition device during the operation quality assessment period is obtained by performing a difference processing on the power value of each data acquisition device at the start time point of the operation quality assessment and the power value of each data acquisition device at the end time point of the operation quality assessment. The power consumption of each data acquisition device during the operation quality assessment period is obtained by performing a ratio processing on the power value of each data acquisition device at the start time point of the operation quality assessment, and the energy consumption proportion of each data acquisition device during the operation quality assessment period is obtained.

[0031] The network transmission quality data between each data acquisition device and the remote sensing platform is obtained, including the network transmission bandwidth, network transmission packet loss rate and network delay duration between each data acquisition device and the remote sensing platform during the operation quality assessment cycle. A comprehensive analysis is performed to obtain the network transmission quality influence coefficient between each data acquisition device and the remote sensing platform.

[0032] The above network transmission bandwidth, network transmission packet loss rate and network delay duration can all be extracted from the network transmission report between each data acquisition device and the remote sensing platform.

[0033] According to the network transmission bandwidth between each data acquisition device and the remote sensing platform during the operation quality assessment cycle, the network transmission packet loss rate between each data acquisition device and the remote sensing platform during the operation quality assessment cycle, and the network delay duration between each data acquisition device and the remote sensing platform during the operation quality assessment cycle, a comprehensive analysis is performed to obtain the network transmission quality influence coefficient between each data acquisition device and the remote sensing platform. The specific method is as follows: ; In the formula, is the network transmission quality impact coefficient between the bth data acquisition device and the remote sensing platform. In this embodiment, it is a quantitative indicator used to measure the impact of the transmission quality between different devices and platforms on the operation quality of each data acquisition device during the network transmission process. It reflects the performance of different devices and platforms in network transmission. The larger the network transmission quality impact coefficient, the poorer the network transmission quality of the device, and the more serious the negative impact on the device operation quality.

[0034] b is the number of each data acquisition device, , B is the total number of data acquisition devices.

[0035] It is the network transmission bandwidth between the bth data acquisition device and the remote sensing platform during the operation quality assessment cycle. It refers to the maximum amount of data that can be transmitted by the network transmission channel between a data acquisition device and the remote sensing platform during the operation quality assessment cycle, usually measured in bits per second.

[0036] is the network transmission packet loss rate between the bth data acquisition device and the remote sensing platform within the operation quality assessment cycle, which refers to the proportion of data packets lost during the network transmission between a data acquisition device and the remote sensing platform within the operation quality assessment cycle, specifically, the ratio of the lost data packets to the total number of transmitted data packets. The larger the network transmission packet loss rate, the worse the data transmission quality.

[0037] is the network delay duration between the bth data acquisition device and the remote sensing platform during the operation quality assessment cycle. It refers to the total time required for a data acquisition device and the remote sensing platform to receive the request and respond during the operation quality assessment cycle. The longer the network delay duration, the worse the data transmission quality.

[0038] is the network quality weight parameter corresponding to the network transmission bandwidth predefined in the remote sensing platform information database. is the network quality weight parameter corresponding to the network transmission packet loss rate predefined in the remote sensing platform information database, It is the network quality weight parameter corresponding to the network delay duration predefined in the remote sensing platform information database.

[0039] The network quality weight parameters corresponding to the network transmission bandwidth, the network quality weight parameters corresponding to the network transmission packet loss rate, and the network quality weight parameters corresponding to the network delay duration are all obtained in the remote sensing platform information library. For example, the network transmission bandwidth and the network quality weight parameters corresponding to the network transmission bandwidth preset in the remote sensing platform information library form a mapping set, and the real-time network transmission bandwidth is brought into the mapping set to obtain the network quality weight parameters corresponding to the network transmission bandwidth, wherein the mapping relationship may be a one-to-one correspondence or a many-to-one relationship; the network transmission packet loss rate and the network quality weight parameters corresponding to the network transmission packet loss rate preset in the remote sensing platform information library form a mapping set, and the real-time network transmission bandwidth is brought into the mapping set to obtain the network quality weight parameters corresponding to the network transmission bandwidth. The input packet loss rate is brought into the mapping set to obtain the network quality weight parameter corresponding to the network transmission packet loss rate, where the mapping relationship can be a one-to-one correspondence or a many-to-one relationship; the network delay duration and the network quality weight parameter corresponding to the preset network delay duration in the remote sensing platform information library form a mapping set, and the real-time network delay duration is brought into the mapping set to obtain the network quality weight parameter corresponding to the network delay duration, where the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. In this embodiment, the value range of the network quality weight parameter corresponding to the network transmission bandwidth, the network quality weight parameter corresponding to the network transmission packet loss rate, and the network quality weight parameter corresponding to the network delay duration is (0, 1).

[0040] In this embodiment, when the network transmission bandwidth is larger, the data transmission speed in the network is faster, because more data can be transmitted at the same time, so the delay caused by network congestion will be greatly reduced, thereby improving the network transmission quality between the device and the platform, and the increase in network transmission bandwidth also helps to reduce packet loss caused by network congestion. When the network transmission bandwidth is large enough, the network can more effectively handle the transmission of large amounts of data, thereby reducing the packet loss rate of network transmission, and can also improve the network transmission quality between the device and the platform; if the network transmission packet loss rate is large, it often leads to network communication interruption, forcing the data packet to be resent, which directly increases the total communication delay, because resending the data packet requires additional time, resulting in an increase in the network delay duration; in summary, the network transmission bandwidth, the network transmission packet loss rate and the network delay duration affect each other. In order to optimize the network transmission performance between the device and the platform, it is necessary to comprehensively consider multiple factors to improve the quality of network transmission and reduce the negative impact on the quality of the device.

[0041] The operation response limit time, data processing reference speed and operation temperature reference value are extracted from the remote sensing platform information database.

[0042] According to the operation response time of each data acquisition device in the operation quality assessment cycle, the average operation temperature of each data acquisition device in the operation quality assessment cycle, the data processing speed of each data acquisition device in the operation quality assessment cycle, the energy consumption proportion of each data acquisition device in the operation quality assessment cycle, and the network transmission quality influence coefficient between each data acquisition device and the remote sensing platform, the operation quality assessment value of each data acquisition device is obtained through comprehensive analysis. The specific method is as follows: ; In the formula, is the operating quality evaluation value of the bth data acquisition device. In this embodiment, it is a quantitative indicator obtained by comprehensively evaluating the performance, stability, reliability and data quality of the data acquisition device during operation. The larger the operating quality evaluation value of the data acquisition device is, the better the performance of the device in detecting geographic information is, and the more accurate the data it acquires is, which has a positive impact on the subsequent generation of anchor frames.

[0043] b is the number of each data acquisition device, , B is the total number of data acquisition devices.

[0044] The operation response time of the bth data acquisition device within the operation quality assessment cycle refers to the time from the time the device receives the data acquisition instruction to the time the device makes an effective response to its instruction, that is, the device starts to acquire data.

[0045] The time defined for an operational response refers to the maximum value of the pre-set operational response time.

[0046] is the average operating temperature of the bth data acquisition device during the operation quality assessment cycle, which refers to the average value of the temperature data of the data acquisition device when it is operating during the operation quality assessment cycle.

[0047] The operating temperature reference value refers to the pre-set standard value for the operating temperature.

[0048] is the data processing speed of the bth data acquisition device within the operation quality assessment cycle, which refers to the ability of the data acquisition device to process data within the operation quality assessment cycle.

[0049] The reference speed for data processing refers to the standard value corresponding to the pre-set data processing speed.

[0050] It is the energy consumption ratio of the bth data acquisition device during the operation quality assessment cycle, which refers to the ratio of the power consumption value of the data acquisition device to the total power value when the data acquisition device is running during the operation quality assessment cycle.

[0051] is the network transmission quality impact coefficient between the bth data acquisition device and the remote sensing platform, which is obtained by comprehensive analysis of the network transmission bandwidth between each data acquisition device and the remote sensing platform during the operation quality assessment period, the network transmission packet loss rate between each data acquisition device and the remote sensing platform during the operation quality assessment period, and the network delay duration between each data acquisition device and the remote sensing platform during the operation quality assessment period.

[0052] It is the operation quality influencing parameter corresponding to the data processing speed predefined in the remote sensing platform information database. It is the operating quality influencing parameter corresponding to the energy consumption ratio predefined in the remote sensing platform information database, and e is a natural constant.

[0053] The operating quality influencing parameters corresponding to the data processing speed and the operating quality influencing parameters corresponding to the energy consumption ratio are both obtained from the remote sensing platform information library. For example, the data processing speed and the operating quality influencing parameters corresponding to the data processing speed preset in the remote sensing platform information library form a mapping set, and the real-time data processing speed is brought into the mapping set to obtain the operating quality influencing parameters corresponding to the data processing speed, wherein the mapping relationship may be a one-to-one correspondence or a many-to-one relationship; the energy consumption ratio and the operating quality influencing parameters corresponding to the energy consumption ratio preset in the remote sensing platform information library form a mapping set, and the real-time energy consumption ratio is brought into the mapping set to obtain the operating quality influencing parameters corresponding to the energy consumption ratio, wherein the mapping relationship may be a one-to-one correspondence or a many-to-one relationship. In this embodiment, the value range of the operating quality influencing parameters corresponding to the data processing speed and the operating quality influencing parameters corresponding to the energy consumption ratio is (0, 1).

[0054] In this embodiment, the improvement of data processing speed helps to reduce its operation response time. If the data acquisition device has a faster data processing speed during the operation quality evaluation cycle, it means that the data acquisition device can effectively respond to the data processing task in a shorter time and quickly complete the data processing task, thereby shortening the operation response time of the data acquisition device. When the data processing speed deviates from its reference value and is too high, more power is consumed by the device to support data processing, increasing the energy consumption ratio of the device, and excessive energy consumption may have a negative impact on the operation of the device; the operating temperature of the data acquisition device has an important impact on its performance. When the device temperature is too high, it may cause problems such as processor frequency reduction and increased memory latency, thereby reducing the data processing speed. When the device temperature is too low, it may mean that the device's operating performance is low, which also leads to a decrease in data processing speed; the operating temperature of the device will also affect its energy consumption ratio. When the device temperature is too high, it may be necessary to increase the power of the heat dissipation system to reduce the temperature, thereby increasing energy consumption and reducing the operation quality of the device.

[0055] In this embodiment, the quality of network transmission is directly related to the communication stability between the device and the platform. If the network transmission quality is poor, such as unstable signal, high packet loss rate, large delay, etc., the communication quality between the device and the platform will be reduced, and communication interruption, data loss and other problems may occur, which will directly affect the operation effect of the device and may even cause the device to fail to work properly. The network transmission quality also directly affects the data transmission speed and operation response time. In a high-quality network transmission environment, that is, when the network transmission quality impact coefficient between each data acquisition device and the remote sensing platform is small, the data transmission speed is fast and the operation response time is short, which can support real-time data transmission and low-latency applications. If the network transmission quality is poor, that is, when the network transmission quality impact coefficient between each data acquisition device and the remote sensing platform is large, the data transmission speed is slow and the operation response time is long, which will seriously affect the real-time performance and response speed of the device and reduce the operation efficiency of the device.

[0056] Specifically, the determination of whether to optimize the operation quality of each data acquisition device is as follows: The operation quality evaluation value of each data acquisition device is compared with the operation quality evaluation threshold predefined in the remote sensing platform information library. If the operation quality evaluation value of a data acquisition device is greater than or equal to the operation quality evaluation threshold, there is no need to optimize the operation quality of the data acquisition device. If the operation quality evaluation value of a data acquisition device is less than the operation quality evaluation threshold, it is necessary to optimize the operation quality of the data acquisition device.

[0057] The above-mentioned optimization of the operating quality of the data acquisition device can specifically include the use of cooling fans, heat pipes and other cooling elements to improve the heat dissipation capacity of the data acquisition device, ensure smooth air circulation, reduce heat accumulation, so that the device can operate at a suitable ambient temperature, and avoid excessively high or low temperatures from affecting the performance of the device; use multi-threading or parallel computing technology to decompose the data acquisition task into multiple subtasks for parallel execution to improve the data processing speed of the device; adjust the network configuration parameters of the data acquisition device, such as bandwidth allocation, routing strategy, etc., to improve network transmission efficiency and reduce the negative impact of network transmission.

[0058] Through the above-mentioned optimization measures, the operating quality of the data acquisition equipment is optimized, and the operating quality evaluation value of the data acquisition equipment is improved, so that the data acquisition equipment can collect and transmit data more accurately, reduce errors and noise, improve data reliability and accuracy, and provide good equipment support for subsequent data processing and anchor frame generation.

[0059] The remote sensing data preprocessing module is used to obtain information graphs transmitted by various data acquisition devices received by the remote sensing platform, which are recorded as the information graph remote sensing data of the target area, and perform information graph fusion operations, feature extraction operations and feature fusion operations in sequence. The final results are recorded as the anchor frame detection geographic data of the target area.

[0060] Furthermore, the information graph remote sensing data of the target area are sequentially subjected to information graph fusion operation, feature extraction operation and feature fusion operation. The specific operation process is as follows: The above-mentioned information map remote sensing data of the target area may specifically be information maps such as topographic map, vegetation index map, soil type map, etc. of the target area, wherein the information map may include topographic data of the target area, such as height, slope, direction, terrain undulation, etc., which may be obtained through ground measurement, aerial photography, satellite remote sensing, etc., and expressed through digital terrain model (DTM) or digital elevation model (DEM); geological data of the target area, such as rock type, geological structure, crustal movement, seismic activity and other information; Normalized Difference Vegetation Index (NDVI) of the target area, which quantifies the growth status of vegetation by calculating the difference between the near-infrared band and the red band, reflecting the health of vegetation and the growth trend of vegetation; Enhanced Vegetation Index (EVI) of the target area, which enhances vegetation signals by adding a blue band, corrects the influence of soil background and aerosol scattering, and is suitable for densely vegetated areas; Soil physical and chemical property data of the target area, such as soil texture, organic matter content, pH value, fertility, acidity and alkalinity, permeability, degree of salinization, etc.

[0061] The information graph fusion operation specifically preprocesses the remote sensing data of each information graph of the target area, and performs a superposition and fusion operation on the preprocessed remote sensing data of each information graph according to a preset channel sequence, and the final result is recorded as the information graph data of each channel.

[0062] The above-mentioned preprocessing specifically includes image registration, spatial registration of remote sensing images from different sources to ensure the spatial consistency of each information graph, normalization processing, normalization of each information graph to unify the data scale, eliminate the dimensional differences between different information graphs, and image cropping. According to the detection requirements, large-size remote sensing images are cropped into tiles of appropriate size to improve processing efficiency and detection accuracy.

[0063] The preprocessed multiple information map remote sensing data are superimposed in a preset channel order to form multi-channel input data. For example, topographic map, vegetation index map, soil type map, etc. are respectively input as different channels, where the preset channel order can be topographic map, soil type map, vegetation index map; if the input information map exceeds the preset number of channels (such as 3 channels), key feature selection or dimensionality reduction technology (such as principal component analysis PCA) is used to integrate the multiple information maps into a fixed number of channels to meet the input requirements of subsequent processing operations.

[0064] The feature extraction operation specifically uses a multi-layer convolutional neural network structure to extract the spatial features of each channel information graph data in sequence through a preset channel order to obtain the feature data of each channel information graph.

[0065] A multi-layer convolutional neural network (CNN) structure is used to extract the spatial features of the input data channel by channel, where channel by channel means layer by layer, and each channel corresponds to each layer. Each layer of convolution operations includes convolution operations, batch normalization, and activation function processing to enhance the ability to express features; residual connections are introduced to effectively alleviate the gradient vanishing problem in deep networks and improve the stability and efficiency of feature extraction; a multi-scale feature extraction module is designed to capture target information of different scales through convolution kernel sizes and pooling operations of different scales, thereby improving the detection accuracy of small and large targets. A feature pyramid structure is constructed to fuse shallow and deep features to enhance the recognition ability of the CNN structure for targets of different scales.

[0066] The feature fusion operation specifically involves fusing the feature data of each channel information graph using a cross-layer connection strategy, and the final result is recorded as the geographical data of each anchor box detection in the target area.

[0067] A cross-layer connection strategy is adopted to fuse features at different levels, combining high-level semantic information and low-level spatial information to improve the accuracy of target detection; attention mechanisms (such as channel attention and spatial attention) are introduced to dynamically adjust the importance of different feature channels and enhance the model's ability to focus on key information; weighted summation or concatenation operations are performed on different feature layers to form a comprehensive feature representation for subsequent target detection; a multi-path feature fusion network is designed to extract and fuse different features through parallel paths to improve the diversity and robustness of feature expression.

[0068] After the above-mentioned information graph fusion operation, feature extraction operation and feature fusion operation, the remote sensing image data of the target area is obtained. According to the scale and shape characteristics of the target in the remote sensing image, multiple groups of anchor frames of different sizes and aspect ratios are designed to meet the detection needs of different targets, and the detection geographic data of each anchor frame in the target area are obtained.

[0069] The target detection and feedback module is used to obtain the actual anchor frame detection geographic data of the target area, match it with the anchor frame detection geographic data of the target area, determine the matching accuracy index of each anchor frame detection geographic data, and verify it with the predefined matching accuracy index preset value, and obtain and complete the geographic information detection of the multi-information map remote sensing data based on the verification result.

[0070] Specifically, the detected geographic data of each anchor frame of the target area specifically includes the detected length, detected width, detected height, detected horizontal coordinate of the center point, and detected vertical coordinate of the center point of each anchor frame.

[0071] The above-mentioned anchor frame detection geographic data is specifically obtained by scaling the remote sensing image data.

[0072] In a specific embodiment, the detection length of the anchor frame in the image remote sensing data may be 8 centimeters, and if it is scaled at 1:10, the detection length of the anchor frame in the anchor frame detection geographic data may be 8 meters.

[0073] Furthermore, the actual anchor frame detection geographic data of the target area specifically includes the actual length, actual width, actual height, actual horizontal coordinate of the center point and actual vertical coordinate of the center point of each anchor frame.

[0074] The above-mentioned actual length, actual width, actual height, actual horizontal coordinate of the center point, and actual vertical coordinate of the center point can all be obtained by field measurement using a drone.

[0075] Specifically, the matching accuracy index of each anchor frame detection geographic data is determined by the following specific determination process: The detected length of each anchor frame is subtracted from the actual length of each anchor frame to obtain the length deviation value of each anchor frame.

[0076] The detected width of each anchor box is subtracted from the actual width of each anchor box to obtain the width deviation value of each anchor box.

[0077] The detected height of each anchor frame is subtracted from the actual height of each anchor frame to obtain the height deviation value of each anchor frame.

[0078] The detected horizontal coordinate of the center point of each anchor frame is subtracted from the actual horizontal coordinate of the center point of each anchor frame to obtain the horizontal coordinate deviation value of the center point of each anchor frame.

[0079] The detected horizontal coordinate of each anchor frame center point refers to the detected geographical longitude corresponding to the anchor frame center point, and the actual horizontal coordinate of each anchor frame center point refers to the actual geographical longitude corresponding to the anchor frame center point.

[0080] The detected ordinate of the center point of each anchor frame is subtracted from the actual ordinate of the center point of each anchor frame to obtain the ordinate deviation value of the center point of each anchor frame.

[0081] The detected ordinate of each anchor frame center point refers to the detected geographic latitude corresponding to the anchor frame center point, and the actual ordinate of each anchor frame center point refers to the actual geographic latitude corresponding to the anchor frame center point.

[0082] According to the length deviation value of each anchor frame, the width deviation value of each anchor frame, the height deviation value of each anchor frame, the horizontal coordinate deviation value of the center point of each anchor frame, and the vertical coordinate deviation value of the center point of each anchor frame, the matching accuracy index of each anchor frame detection geographic data is obtained through comprehensive processing. The specific method is as follows: ; In the formula, It is the matching accuracy index of the mth anchor frame detection geographic data. In this embodiment, it is used to measure the matching degree between the obtained anchor frame data and the actual target area in the target detection task. If the matching accuracy index of the anchor frame detection geographic data is higher, it means that the anchor frame detection geographic data is closer to the true level.

[0083] m is the number of each anchor box, , M is the total number of anchor boxes.

[0084] is the length deviation value of the mth anchor box, which refers to the difference between the preset length of the anchor box and the actual length in the target detection task, where the length refers to a dimension of the anchor box in the image.

[0085] is the width deviation value of the mth anchor box, which refers to the difference between the preset width of the anchor box and the actual width in the target detection task.

[0086] is the height deviation value of the mth anchor box, which refers to the difference between the height of the preset anchor box and the actual height in the target detection task.

[0087] It is the horizontal coordinate deviation value of the center point of the mth anchor frame, which refers to the difference between the longitude of the preset anchor frame and the actual longitude in the target detection task.

[0088] It is the vertical coordinate deviation value of the center point of the mth anchor box, which refers to the difference between the latitude of the preset anchor box and the actual latitude in the target detection task.

[0089] is the operation quality evaluation value of the bth data acquisition device, which means that the operation quality evaluation value of each data acquisition device is obtained through comprehensive analysis of the operation response time of each data acquisition device within the operation quality evaluation period, the average operation temperature of each data acquisition device within the operation quality evaluation period, the data processing speed of each data acquisition device within the operation quality evaluation period, the energy consumption proportion of each data acquisition device within the operation quality evaluation period, and the network transmission quality influence coefficient between each data acquisition device and the remote sensing platform.

[0090] b is the number of each data acquisition device, , B is the total number of data acquisition devices.

[0091] The matching accurate influencing parameter corresponding to the predefined length deviation value in the remote sensing platform information database, The matching accurate influencing parameter corresponding to the predefined width deviation value in the remote sensing platform information library, The accurate influencing parameters corresponding to the predefined height deviation values ​​in the remote sensing platform information database are matched. It is the matching accurate influencing parameter corresponding to the horizontal coordinate deviation value of the center point predefined in the remote sensing platform information database. It is the matching accurate influencing parameter corresponding to the ordinate deviation value of the center point predefined in the remote sensing platform information database. It is the matching accurate influencing parameter corresponding to the predefined operation quality assessment value in the remote sensing platform information database, and e is a natural constant.

[0092] The matching accurate influencing parameters corresponding to the length deviation value, the matching accurate influencing parameters corresponding to the width deviation value, the matching accurate influencing parameters corresponding to the height deviation value, the matching accurate influencing parameters corresponding to the horizontal coordinate deviation value of the center point, the matching accurate influencing parameters corresponding to the vertical coordinate deviation value of the center point, and the matching accurate influencing parameters corresponding to the operation quality assessment value are extracted from the remote sensing platform information library, and the length deviation value and the matching accurate influencing parameters corresponding to the preset length deviation value in the remote sensing platform information library form a mapping set, and the real-time length deviation value is brought into the mapping set to obtain the matching accurate influencing parameters corresponding to the length deviation value, wherein the mapping relationship can be one-to-one or multiple One-to-one relationship; the width deviation value and the matching accurate influencing parameter corresponding to the preset width deviation value in the remote sensing platform information library form a mapping set, and the real-time width deviation value is brought into the mapping set to obtain the matching accurate influencing parameter corresponding to the width deviation value, wherein the mapping relationship can be one-to-one correspondence or many-to-one relationship; the height deviation value and the matching accurate influencing parameter corresponding to the preset height deviation value in the remote sensing platform information library form a mapping set, and the real-time height deviation value is brought into the mapping set to obtain the matching accurate influencing parameter corresponding to the height deviation value, wherein the mapping relationship can be one-to-one correspondence or many-to-one relationship; the horizontal coordinate deviation value of the center point and the horizontal coordinate deviation value of the center point preset in the remote sensing platform information library The accurate influencing parameters corresponding to the horizontal coordinate deviation value of the real-time center point are formed into a mapping set, and the horizontal coordinate deviation value of the real-time center point is brought into the mapping set to obtain the accurate influencing parameters corresponding to the horizontal coordinate deviation value of the center point, wherein the mapping relationship can be one-to-one or many-to-one; the vertical coordinate deviation value of the center point and the accurate influencing parameters corresponding to the vertical coordinate deviation value of the center point preset in the remote sensing platform information library are formed into a mapping set, and the real-time vertical coordinate deviation value of the center point is brought into the mapping set to obtain the accurate influencing parameters corresponding to the vertical coordinate deviation value of the center point, wherein the mapping relationship can be one-to-one or many-to-one; the operation quality assessment value and the remote sensing platform The matching accurate influencing parameters corresponding to the preset operation quality assessment values ​​in the information database form a mapping set, and the real-time operation quality assessment value is brought into the mapping set to obtain the matching accurate influencing parameters corresponding to the operation quality assessment value, wherein the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. In this embodiment, the matching accurate influencing parameters corresponding to the length deviation value, the matching accurate influencing parameters corresponding to the width deviation value, the matching accurate influencing parameters corresponding to the height deviation value, the matching accurate influencing parameters corresponding to the horizontal coordinate deviation value of the center point, the matching accurate influencing parameters corresponding to the vertical coordinate deviation value of the center point, and the matching accurate influencing parameters corresponding to the operation quality assessment value have a value range of (0, 1).

[0093] In this embodiment, the horizontal coordinate deviation value and the vertical coordinate deviation value jointly describe the difference between the position of the anchor frame on the image plane and the position of the actual target area. These two deviation values ​​determine the relative position of the anchor frame and the actual target area in space. If the horizontal coordinate deviation value and the vertical coordinate deviation value are both zero, it means that the center point of the anchor frame completely coincides with the center point of the actual target area; the length deviation value, the width deviation value and the height deviation value jointly describe the difference between the size of the anchor frame and the size of the actual target area. These three deviation values ​​determine the similarity in size between the anchor frame and the actual target area. If the length deviation value, the width deviation value and the height deviation value are all zero, it means that the size of the anchor frame is exactly the same as the size of the actual target area; in summary, these deviation values ​​jointly determine the degree of matching between the anchor frame and the actual target area. In the target detection task, these deviation values ​​are calculated and their sizes are used to determine whether the anchor frame accurately matches the actual target area. The smaller the deviation value, the higher the degree of matching between the anchor frame and the actual target area; the larger the deviation value, the lower the degree of matching.

[0094] In this embodiment, the operating quality of the data acquisition device directly affects the accuracy of the collected data. If the device has a fault or error, the collected data may deviate from the true value, resulting in inaccurate results of subsequent processing and analysis. In the anchor frame matching process, if the input data is inaccurate, the calculated anchor frame parameters, such as length, width, height, horizontal and vertical coordinates of the center point, etc., will also have deviations, thereby affecting the degree of matching between the anchor frame and the actual target area; the operating quality of the data acquisition device also affects the integrity of the data. If the device fails or is interrupted when collecting data, it may cause data loss or incompleteness. In the anchor frame matching process, if key data is missing, it may not be possible to accurately calculate the parameters of the anchor frame, thereby affecting the matching results.

[0095] Furthermore, the geographic information detection of the multi-information map remote sensing data is completed according to the verification result, and the specific analysis process is as follows: The verification result is specifically the first verification result or the second verification result.

[0096] The first verification result is that the matching accuracy index of the geographic data detected by a certain anchor frame is equal to the matching accuracy index preset value predefined in the remote sensing platform information library.

[0097] The second verification result is that a matching accuracy index of geographic data detected by a certain anchor frame is not equal to a preset matching accuracy index value.

[0098] If the verification result is displayed as the first verification result, the anchor frame corresponding to the matching accuracy index of the anchor frame detection geographic data is used to detect the geographic data, and the anchor frame corresponding to the anchor frame detection geographic data is obtained through the preset anchor frame design method, and is visualized on the remote sensing platform.

[0099] The anchor frame is designed according to the anchor frame detection geographic data. Specifically, a preset anchor frame is formed by presetting the length, width, height, horizontal coordinate of the center point and vertical coordinate of the center point, wherein the preset anchor frame will be displayed on the display interface of the remote sensing platform.

[0100] If the verification result is displayed as the second verification result, feedback is provided to the anchor frame detection geographic data according to the matching accuracy index of the anchor frame detection geographic data.

[0101] Specifically, the feeding back of the anchor frame detected geographic data specifically involves performing feature extraction operations and feature fusion operations on the anchor frame detected geographic data again.

[0102] The specific process is to perform difference processing on the matching accuracy index of the anchor frame detection geographic data and the preset value of the matching accuracy index to obtain the matching accuracy index deviation value of the anchor frame detection geographic data, and match it with the second operation mode corresponding to each matching accuracy index deviation value interval predefined in the remote sensing platform information library. The specific matching process is: obtain the mapping set of the matching accuracy index deviation value of the anchor frame detection geographic data and the second operation mode from the remote sensing platform information library, determine the specific interval of the matching accuracy index deviation value of the anchor frame detection geographic data, and obtain the second operation mode corresponding to the interval, that is, perform a second feature extraction operation and a feature fusion operation on the anchor frame detection geographic data to obtain the second detection geographic data of the anchor frame. At this time, the matching accuracy index corresponding to the second detection geographic data of the anchor frame meets the above-mentioned first verification result.

[0103] In a specific embodiment, if the matching accuracy index deviation value of the anchor frame detection geographic data is 10.5, it is determined that the matching accuracy index deviation value of the anchor frame detection geographic data belongs to the second matching accuracy index deviation value interval, and the second operation mode corresponding to the interval is the second feature extraction operation and the feature fusion operation. Specifically, the anchor frame detection geographic data is selected according to the detection target and data characteristics. The corresponding features are extracted. The features can be based on pixel-based statistical features, such as spectral features, texture features, etc., or based on geometric features such as shape and size. A convolutional neural network algorithm is selected to extract features from the geographic data, and the extracted features are quantized and encoded for subsequent processing and analysis; different features are corrected and aligned to ensure that they are on the same spatial and temporal scales. Appropriate transformation or interpolation methods can be used to achieve feature correction and alignment. A fusion algorithm based on sparse representation is used to fuse the corrected and aligned features to generate the second detection geographic data of the anchor frame.

[0104] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A geographic information detection system based on multi-information graph remote sensing data, characterized in that: include: A data acquisition device optimization module is used to obtain the operation data of each data acquisition device connected to the remote sensing platform, determine the operation quality evaluation value of each data acquisition device, and compare it with a predefined operation quality evaluation threshold to determine whether to optimize the operation quality of each data acquisition device; The remote sensing data preprocessing module is used to obtain the information graphs transmitted by the data acquisition devices received by the remote sensing platform, record them as the remote sensing data of each information graph of the target area, and perform information graph fusion operation, feature extraction operation and feature fusion operation in sequence, and the final result is recorded as the geographical data of each anchor frame detection of the target area; The target detection and feedback module is used to obtain the actual anchor frame detection geographic data of the target area, match it with the anchor frame detection geographic data of the target area, determine the matching accuracy index of each anchor frame detection geographic data, and verify it with the predefined matching accuracy index preset value, and obtain and complete the geographic information detection of multi-information map remote sensing data based on the verification result.

2. The geographic information detection system based on multi-information graph remote sensing data according to claim 1, characterized in that: The operating data of each data acquisition device specifically includes the operating response time of each data acquisition device during the operation quality assessment period, the average operating temperature of each data acquisition device during the operation quality assessment period, the data processing volume of each data acquisition device during the operation quality assessment period, the power value of each data acquisition device at the start time point of the operation quality assessment, and the power value of each data acquisition device at the end time point of the operation quality assessment.

3. The geographic information detection system based on multi-information graph remote sensing data according to claim 2, characterized in that: The specific determination process of determining the operation quality evaluation value of each data acquisition device is as follows: The data processing amount of each data acquisition device in the operation quality assessment cycle is processed by ratio with the corresponding duration of the operation quality assessment cycle to obtain the data processing speed of each data acquisition device in the operation quality assessment cycle; Perform difference processing on the power value of each data acquisition device at the start time of the operation quality assessment and the power value of each data acquisition device at the end time of the operation quality assessment to obtain the power consumption of each data acquisition device during the operation quality assessment period, and perform ratio processing on the power value of each data acquisition device at the start time of the operation quality assessment to obtain the energy consumption ratio of each data acquisition device during the operation quality assessment period; Obtain the network transmission quality data between each data acquisition device and the remote sensing platform, including the network transmission bandwidth, network transmission packet loss rate and network delay duration between each data acquisition device and the remote sensing platform during the operation quality assessment period, and obtain the network transmission quality influence coefficient between each data acquisition device and the remote sensing platform through comprehensive analysis; Extract the operation response limit time, data processing reference speed and operation temperature reference value from the remote sensing platform information database; Based on the operation response time of each data acquisition device within the operation quality assessment period, the average operating temperature of each data acquisition device within the operation quality assessment period, the data processing speed of each data acquisition device within the operation quality assessment period, the energy consumption proportion of each data acquisition device within the operation quality assessment period, and the network transmission quality influence coefficient between each data acquisition device and the remote sensing platform, a comprehensive analysis is performed to obtain the operation quality assessment value of each data acquisition device.

4. The geographic information detection system based on multi-information graph remote sensing data according to claim 1, characterized in that: The specific determination process of whether to optimize the operation quality of each data acquisition device is as follows: The operation quality evaluation value of each data acquisition device is compared with the operation quality evaluation threshold predefined in the remote sensing platform information library. If the operation quality evaluation value of a data acquisition device is greater than or equal to the operation quality evaluation threshold, there is no need to optimize the operation quality of the data acquisition device. If the operation quality evaluation value of a data acquisition device is less than the operation quality evaluation threshold, it is necessary to optimize the operation quality of the data acquisition device.

5. The geographic information detection system based on multi-information graph remote sensing data according to claim 1, characterized in that: The information graph remote sensing data of the target area are sequentially subjected to information graph fusion operation, feature extraction operation and feature fusion operation. The specific operation process is as follows: The information graph fusion operation specifically pre-processes the remote sensing data of each information graph of the target area, and performs a superposition fusion operation on the pre-processed remote sensing data of each information graph according to a preset channel sequence, and the final result is recorded as the information graph data of each channel; The feature extraction operation specifically uses a multi-layer convolutional neural network structure to extract the spatial features of each channel information graph data in sequence through a preset channel order to obtain feature data of each channel information graph; The feature fusion operation specifically involves fusing the feature data of each channel information graph using a cross-layer connection strategy, and the final result is recorded as the geographical data of each anchor box detection in the target area.

6. The geographic information detection system based on multi-information graph remote sensing data according to claim 1, characterized in that: The detected geographic data of each anchor frame of the target area specifically includes the detected length, detected width, detected height, detected horizontal coordinate of the center point and detected vertical coordinate of the center point of each anchor frame; The actual anchor frame detection geographic data of the target area specifically includes the actual length, actual width, actual height, actual horizontal coordinate of the center point and actual vertical coordinate of the center point of each anchor frame.

7. According to the geographic information detection system based on multi-information graph remote sensing data of claim 1, the matching accuracy index of each anchor frame detection geographic data is determined by the specific determination process: The detected length of each anchor frame is subtracted from the actual length of each anchor frame to obtain the length deviation value of each anchor frame; Perform difference processing on the detected width of each anchor frame and the actual width of each anchor frame to obtain the width deviation value of each anchor frame; Perform difference processing on the detected height of each anchor frame and the actual height of each anchor frame to obtain the height deviation value of each anchor frame; Perform difference processing on the detected horizontal coordinate of the center point of each anchor frame and the actual horizontal coordinate of the center point of each anchor frame to obtain the horizontal coordinate deviation value of the center point of each anchor frame; The detected ordinate of the center point of each anchor frame is subtracted from the actual ordinate of the center point of each anchor frame to obtain the ordinate deviation value of the center point of each anchor frame.

8. The geographic information detection system based on multi-information graph remote sensing data according to claim 1, characterized in that: The specific determination process of determining the matching accuracy index of each anchor frame detection geographic data is as follows: According to the length deviation value of each anchor frame, the width deviation value of each anchor frame, the height deviation value of each anchor frame, the horizontal coordinate deviation value of the center point of each anchor frame, and the vertical coordinate deviation value of the center point of each anchor frame, the matching accuracy index of each anchor frame detection geographic data is obtained through comprehensive processing. The specific method is as follows: ； In the formula, is the matching accuracy index of the mth anchor frame to detect geographic data, m is the number of each anchor frame, , M is the total amount of anchor boxes, is the length deviation value of the mth anchor box, is the width deviation value of the mth anchor box, is the height deviation value of the mth anchor box, is the horizontal coordinate deviation value of the center point of the mth anchor box, is the vertical coordinate deviation value of the center point of the mth anchor box, is the operating quality evaluation value of the bth data acquisition device, b is the serial number of each data acquisition device, , B is the total amount of data acquisition equipment, The matching accurate influencing parameter corresponding to the predefined length deviation value in the remote sensing platform information database, The matching accurate influencing parameter corresponding to the predefined width deviation value in the remote sensing platform information library, The accurate influencing parameters corresponding to the predefined height deviation values ​​in the remote sensing platform information database are matched. It is the matching accurate influencing parameter corresponding to the horizontal coordinate deviation value of the center point predefined in the remote sensing platform information database. It is the matching accurate influencing parameter corresponding to the ordinate deviation value of the center point predefined in the remote sensing platform information database. It is the matching accurate influencing parameter corresponding to the predefined operation quality assessment value in the remote sensing platform information database, and e is a natural constant.

9. The geographic information detection system based on multi-information graph remote sensing data according to claim 1, characterized in that: The above-mentioned verification results are used to complete the geographic information detection of multi-information map remote sensing data. The specific analysis process is as follows: The verification result is specifically the first verification result or the second verification result; The first verification result is that the matching accuracy index of the geographic data detected by a certain anchor frame is equal to the preset matching accuracy index value; The second verification result is that the matching accuracy index of the geographic data detected by a certain anchor frame is not equal to the matching accuracy index preset value; If the verification result is displayed as the first verification result, the anchor frame corresponding to the matching accuracy index of the anchor frame detection geographic data is detected, and the anchor frame corresponding to the anchor frame detection geographic data is obtained by a preset anchor frame design method, and is visualized on the remote sensing platform; If the verification result is displayed as the second verification result, feedback is provided to the anchor frame detection geographic data according to the matching accuracy index of the anchor frame detection geographic data.

10. The geographic information detection system based on multi-information graph remote sensing data according to claim 9, characterized in that: The feeding back of the anchor frame detected geographic data specifically involves performing feature extraction and feature fusion operations on the anchor frame detected geographic data again.

Images