Onboard flood data processing method, device and system based on aviation emergency mode

Abstract

The present invention provides an onboard flood data processing method, device and system based on the aviation emergency mode, the method comprising: based on the ground observation image, performing image mosaicking to obtain the target image; inputting the target image into the water body recognition model to obtain the water body recognition result; based on the water body recognition result and the geographic information data, performing spatial analysis to obtain the comprehensive flood analysis information; when the aircraft lands, the comprehensive flood analysis information is sent to the remote terminal, so that the remote terminal receives the comprehensive flood analysis information, makes a preliminary judgment on the disaster situation, and guides the generation of rescue command decisions. The onboard flood data processing method, device and system based on the aviation emergency mode provided by the present invention can realize the splicing, correction, information extraction and the like of flood data in real time and quasi-real time on board during the flight, and transmit the information to the ground command system after the aircraft lands, so as to improve the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

Description

Technical Field

[0001] The present invention relates to the field of aviation remote sensing technology, and in particular to an onboard flood data processing method, device and system based on an aviation emergency mode. Background Art

[0002] At present, flood disasters, as one of the major natural disasters, cause huge casualties and economic losses every year. For flood emergency rescue work, the first thing to do is to master the basic information of the disaster situation in the disaster-stricken areas, complete the general assessment of the disaster situation, and provide necessary guidance information for rescue.

[0003] However, floods are often accompanied by heavy rain or cloudy weather. Optical satellite remote sensing is limited by weather and cannot continuously collect ground image data during the rescue period. Although microwave remote sensing satellites have a certain ability to penetrate clouds and rain, their data processing technology is difficult. Therefore, aerial remote sensing uses a flying platform equipped with various ground observation equipment to obtain information about the disaster area, which is less or less restricted by clouds and rain. It is undoubtedly one of the best means of obtaining information in this situation.

[0004] In the existing technology, the flight platform of aerial remote sensing, whether it is a drone or a manned aircraft, obtains ground image data of the disaster area through flight, and completes the splicing of the disaster area image map through traditional photogrammetry methods after the aircraft lands. On this basis, disaster information extraction is carried out. It usually takes more than several hours from the landing of the aircraft to the splicing of the image map and the extraction of the flood range (affected by factors such as the flight area and the relative altitude of the flight), resulting in a relatively delayed processing time of remote sensing data. In the emergency rescue of large-scale flood disasters, the efficiency of remote sensing data processing on flood inundation is low, and it is impossible to quickly assist emergency rescue decision-making. Summary of the invention

[0005] The present invention provides an onboard flood data processing method, device and system based on aviation emergency mode, which are used to solve the defects of low efficiency and lack of timeliness in the prior art of remote sensing data processing of flood inundation, realize the processing of comprehensive flood analysis information on board, reduce the time difference between data acquisition and data use, and ensure high efficiency and timeliness of data.

[0006] The present invention provides an onboard flood data processing method based on an aviation emergency mode, comprising:

[0007] Based on the earth observation images, image mosaicking is performed to obtain the target image;

[0008] Inputting the target image into a water body recognition model to obtain a water body recognition result;

[0009] Based on the water body identification results and geographic information data, spatial analysis is performed to obtain comprehensive flood analysis information;

[0010] When the aircraft lands, the comprehensive flood analysis information is sent to a remote terminal, so that the remote terminal receives the comprehensive flood analysis information, makes a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions;

[0011] The earth observation image includes an observation image taken by an aerial camera carried by the aircraft and GPS location information corresponding to the observation image, and the water body recognition model is obtained after training based on training samples and predetermined water body labels.

[0012] According to a method for processing on-board flood data based on an aviation emergency mode provided by the present invention, after performing spatial analysis based on the water body identification results and target thematic geographic data to obtain comprehensive flood analysis information, the method also includes: during the flight of the aircraft, sending the comprehensive flood analysis information to a remote terminal, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions.

[0013] According to an onboard flood data processing method based on an aviation emergency mode provided by the present invention, the image mosaicking is performed based on the earth observation image to obtain the target image, including:

[0014] extracting a target observation image from the earth observation image based on the target parameters;

[0015] Based on the target observation image, image mosaicking is performed to obtain a target image.

[0016] According to an onboard flood data processing method based on an aviation emergency mode provided by the present invention, the target observation image is extracted from the ground observation image based on the target parameter, comprising:

[0017] Based on the target parameters, acquiring the target observation image and recording the target exposure time;

[0018] Based on the target exposure time, matching the target GPS location information with the target observation image;

[0019] The target exposure time is the exposure time of the target observation image.

[0020] According to an onboard flood data processing method based on an aviation emergency mode provided by the present invention, the image mosaicking is performed based on the target observation image to obtain the target image, including:

[0021] Based on the target observation image, image registration is performed to extract the same-name point pairs;

[0022] Based on the same-name point pairs, an affine transformation is performed using the least squares method to obtain a spliced ​​image;

[0023] Based on the stitched image, the spatial geographic information is associated using the least squares principle to generate the target image.

[0024] According to an onboard flood data processing method based on an aviation emergency mode provided by the present invention, the spatial analysis is performed based on the water body identification result and the geographic information data to obtain comprehensive flood analysis information, including:

[0025] Based on the water body identification results and target geographic information data, spatial analysis is performed to obtain thematic information;

[0026] Based on the thematic information, statistics are performed to obtain comprehensive flood analysis information.

[0027] According to an onboard flood data processing method based on an aviation emergency mode provided by the present invention, the target image is input into a water body recognition model to obtain a water body recognition result, including:

[0028] Based on the target image, acquiring water body characteristic information;

[0029] Based on the water body feature information, a water body recognition result corresponding to the target image is obtained and output in a vector format.

[0030] The present invention also provides an onboard data processing device for an aviation remote sensing emergency mode, comprising:

[0031] An image acquisition module is used to perform image mosaicking based on earth observation images to acquire target images;

[0032] A water body recognition module, used to input the target image into a water body recognition model to obtain a water body recognition result;

[0033] An onboard data generation module, used to perform spatial analysis based on the water body identification results and geographic information data to obtain comprehensive flood analysis information;

[0034] A landing data sending module is used to send the comprehensive flood analysis information to a remote terminal when the aircraft lands, so that the remote terminal receives the comprehensive flood analysis information, makes a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions;

[0035] The earth observation image includes an observation image taken by an aerial camera carried by the aircraft and GPS location information corresponding to the observation image, and the water body recognition model is obtained after training based on training samples and predetermined water body labels.

[0036] According to an onboard flood data processing device based on an aviation emergency mode provided by the present invention, the device further includes:

[0037] The onboard data sending module is used to send the comprehensive flood analysis information to a remote terminal during the flight of the aircraft, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions.

[0038] The present invention also provides an onboard flood data processing system based on aviation emergency mode, comprising one or more onboard flood data processing devices based on aviation emergency mode as described above, and further comprising: a remote terminal;

[0039] The remote terminal is communicatively connected with the landing data sending module or the onboard data sending module;

[0040] The remote terminal is used to receive the comprehensive flood analysis information, conduct a preliminary assessment of the disaster situation, and guide the generation of rescue command decisions.

[0041] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-mentioned methods for processing on-board flood data based on the aviation emergency mode are implemented.

[0042] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described methods for processing onboard flood data based on an aviation emergency mode.

[0043] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any of the above-mentioned methods for processing on-board flood data based on the aviation emergency mode are implemented.

[0044] The on-board flood data processing method, device and system based on the aviation emergency mode provided by the present invention are based on the large-scale collection of ground observation images in the air, image mosaicking by a simplified method, obtaining a full-frame target image, identifying the target image by a simplified water body recognition model to obtain a water body recognition result, combining geographic information data for spatial analysis to obtain comprehensive flood analysis information, and sending the comprehensive flood analysis information to a remote terminal when the aircraft lands. It is possible to achieve real-time and quasi-real-time completion of flood data splicing, correction, information extraction, etc. on board during flight, and transmit data to the ground command system and front-line rescue units after the aircraft lands, thereby improving information processing efficiency and timeliness, which is of obvious significance to emergency rescue. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 It is a flow chart of an onboard flood data processing method based on an aviation emergency mode provided by the present invention;

[0047] Figure 2 It is one of the sub-flow diagrams of the onboard flood data processing method based on the aviation emergency mode provided by the present invention;

[0048] Figure 3 This is the second sub-flow diagram of the onboard flood data processing method based on the aviation emergency mode provided by the present invention;

[0049] Figure 4 This is the third sub-flow diagram of the onboard flood data processing method based on the aviation emergency mode provided by the present invention;

[0050] Figure 5 This is the fourth sub-flow diagram of the onboard flood data processing method based on the aviation emergency mode provided by the present invention;

[0051] Figure 6 It is a structural schematic diagram of an onboard flood data processing device based on an aviation emergency mode provided by the present invention;

[0052] Figure 7 It is a structural schematic diagram of an onboard flood data processing system based on an aviation emergency mode provided by the present invention;

[0053] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] The terms "first", "second", etc. in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more.

[0056] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0057] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0058] Figure 1 FIG. 1 is a flow chart of an onboard flood data processing method based on an aviation emergency mode provided by the present invention. Figure 1 As shown, the onboard flood data processing method based on the aviation emergency mode provided by the embodiment of the present invention includes: step 101, performing image mosaicking based on the ground observation image to obtain the target image.

[0059] The earth observation images include observation images taken by the aerial camera carried by the aircraft, and the GPS location information corresponding to the observation images.

[0060] It should be noted that the execution subject of the onboard flood data processing method based on the aviation emergency mode provided in the embodiment of the present invention is an onboard flood data processing device based on the aviation emergency mode.

[0061] The onboard flood data processing device based on the aviation emergency mode is a device installed on an aircraft. The application scenario of the onboard flood data processing device based on the aviation emergency mode is that in disaster relief, for the purpose of emergency, by sacrificing a certain accuracy in exchange for a certain processing time, the real-time or quasi-real-time processing of data is completed on board, thereby obtaining highly current information results.

[0062] The application scenario of the onboard flood data processing device based on the aviation emergency mode in the embodiment of the present invention does not affect the work of producing aviation remote sensing data and extracting information according to normal procedures.

[0063] The embodiment of the present invention does not specifically limit the type and layout position of the aerial camera. By way of example, the aerial camera may be an array optical instrument, a linear array optical instrument, a spectral instrument, or a radar instrument.

[0064] Preferably, the aerial camera is installed on a stable aircraft to ensure that the camera lens is facing vertically downward. By abandoning the attitude correction of the image, the onboard flood data processing device based on the aviation emergency mode only processes the deformation caused by different resolutions (terrain scaling), rotation and translation caused by the change of flight altitude, so as to achieve geometric correction of the image in a short time.

[0065] It should be noted that, before step 101, it is necessary to receive the ground observation image captured by the aerial camera carried by the aircraft.

[0066] Earth observation images refer to images collected by the aircraft's onboard camera after takeoff to monitor ground conditions, and the Global Positioning System (GPS) location information is recorded at the time of image collection. Earth observation images are used to reflect the ground conditions of each geographical location that the aircraft passes through during flight.

[0067] The embodiment of the present invention does not specifically limit the number of earth observation images.

[0068] Earth observation images can be all images collected by an aircraft during flight.

[0069] Preferably, the earth observation image may be an image extracted from all collected images according to a specified rule.

[0070] Specifically, in step 101, the onboard flood data processing device based on the aviation emergency mode combines the ground observation image with the corresponding GPS position information into a data pair, and selects an image with relatively uniform brightness and color as a reference image for mosaicking. The other images are geometrically aligned based on the reference image, so that all images to be mosaicked are mosaicked from near to far in the same spatial coordinate system to form a target image.

[0071] The target image refers to a number of ground observation images collected by the aircraft during its travel, which are mosaicked with overlapping areas to generate a larger image. The target image includes each ground observation image and the corresponding GPS location information. The target image is used to represent the ground conditions along the aircraft's route and the summary information of the corresponding geographical location.

[0072] Step 102: Input the target image into the water body recognition model to obtain the water body recognition result.

[0073] Among them, the water body recognition model is obtained after training based on training samples and predetermined water body labels.

[0074] It should be noted that the water body recognition model may be an artificial intelligence model, and the embodiment of the present invention does not specifically limit the model type.

[0075] Preferably, the water body identification model is a supervised learning-based model, which is trained from a training data set and then predicts the test data.

[0076] The training data includes various water sample data and corresponding water labels. Through training, the machine can find the connection between samples and labels, and can determine the labels when faced with test data.

[0077] The supervised algorithms used in the embodiments of the present invention include, but are not limited to, linear regression, neural network, decision tree, support vector machine, k-nearest neighbor classification algorithm (k-NearestNeighbo, kNN), etc.

[0078] Moreover, the finer the feature division in the supervised algorithm, the longer the processing time and the better the classification effect.

[0079] Preferably, in order to obtain highly current data, the supervised algorithm is used to complete classification and extraction based on the huge differences between ground-based water bodies and ground-based spectra (spectral reflectance values ​​and curve shapes), combined with the differences in spatial distribution and texture between flood water bodies and other ground objects. Since the algorithm only focuses on the water body itself, the supervised classification method degenerates into target recognition, which can improve the efficiency of water body recognition and extraction.

[0080] Specifically, in step 102, the onboard flood data processing device based on the aviation emergency mode uses a pre-trained simplified water body recognition model to quickly recognize the water body of the target image in step 101, and can obtain a water body recognition result corresponding to the target image.

[0081] The water body recognition result may be a probability value or a label result. The embodiment of the present invention does not specifically limit the form of the water body recognition result.

[0082] Optionally, the water body recognition result may be a probability value, and the probability that any pixel point in the target image is a water body or a non-water body is described by the probability value.

[0083] Optionally, the water body identification result may be a label result, and an intermediate numerical result is obtained through the model. If the numerical result meets a preset target condition, a corresponding label result is assigned to a target user corresponding to the numerical result.

[0084] For example, an intermediate numerical result is obtained through the water body recognition model to characterize the probability that a certain pixel in the target image is a water body. According to the pre-set target condition, when the probability is greater than the preset threshold, a label is generated to express that the pixel is a water body, and the label is assigned to the user.

[0085] Step 103: Based on the water body identification results and geographic information data, spatial analysis is performed to obtain comprehensive flood analysis information.

[0086] It should be noted that, before step 103 , the onboard flood data processing device based on the aviation emergency mode will need to obtain the geographic information data of the corresponding position according to the GPS position information contained in the target image obtained in step 101 .

[0087] Geographic information data refers to the basic geographic information of the area involved in the target image before it suffers from flood disasters. This information can be provided by various densely distributed information acquisition and monitoring devices, or various administrative units.

[0088] The embodiment of the present invention does not specifically limit the geographic information data. For example, the geographic information data may be related information of roads, residential areas, reservoirs, rivers, mining areas, tailings ponds, factories and mines, or ecological protection areas.

[0089] Specifically, in step 103, during the flight of the aircraft, the onboard flood data processing device based on the aviation emergency mode incorporates the water body identification result obtained in step 102 into a unified spatial geographic framework by aligning it with the existing geographic information data, and obtains comprehensive flood analysis information.

[0090] Comprehensive flood analysis information refers to data that associates the pixel coordinate system of the water body identification result with the actual spatial geographic coordinate system. Comprehensive flood analysis information is used to characterize whether a specific actual geographic location has a water body, that is, whether it is flooded.

[0091] This process is not specifically limited in the embodiment of the present invention.

[0092] For example, during the flight of the aircraft, the target image generated by combining several earth observation images only has the geographical location corresponding to each earth observation image, but does not have spatial geographic information in the conventional sense. Therefore, after the water body is identified in step 102, only the approximate geographical location of the water body area can be determined, and the location where the flood occurs in the target image cannot be specifically located.

[0093] Therefore, in step 103, the water body identification result is superimposed with basic geographic information data such as roads, buildings or regions in the actual geographic coordinate space to obtain comprehensive flood analysis information for Huqiu. The comprehensive flood analysis information can indicate whether a road, a building or a region is flooded.

[0094] The embodiment of the present invention does not specifically describe the presentation form of the comprehensive flood analysis information.

[0095] Optionally, the comprehensive flood analysis information may include whether a road, a building or an area is flooded, and is saved in the original data format.

[0096] Optionally, the comprehensive flood analysis information can further extract the area submerged by flooding, the partition area, the buffer zone analysis results, etc. according to whether the actual geographic space is submerged by flooding.

[0097] Optionally, the comprehensive flood analysis information may be statistical information generated based on whether the actual geographic space is flooded.

[0098] Exemplarily, the statistical information can use a GIS software (geographic information software, such as ArcGIS) mapping template to automatically insert comprehensive flood analysis information to complete the view, and automatically insert the view into a designed formatted disaster report document (word document template). Statistical analysis information is also automatically inserted into the report for use in quantitative decision-making.

[0099] Step 104: When the aircraft lands, the comprehensive flood analysis information is sent to the remote terminal, so that the remote terminal receives the comprehensive flood analysis information, makes a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions.

[0100] Specifically, in step 104, after the processing of the comprehensive flood analysis information is completed on board in step 103, when the aircraft lands, the onboard flood data processing device based on the aviation emergency mode sends the comprehensive flood analysis information to the remote terminal through wireless communication technology.

[0101] Among them, wireless communication technology includes but is not limited to WIFI wireless cellular signals (2G, 3G, 4G, 5G).

[0102] In the prior art, the use of aerial remote sensing to observe the flood disaster in a certain area generally involves using an aircraft equipped with an aerial camera to fly from the starting point to the end point to collect image data. After returning to the ground, the collected image data is transmitted to a remote terminal so that the remote terminal can further process and analyze the original image data. Since the round trip of the aircraft takes time, the data processed by the remote terminal only retains the disaster information at the time of image collection. However, since the disaster situation changes rapidly, the time interval between the image collection time and the data landing processing time is large, which may cause the disaster information at the time of image collection to guide the current disaster information in a non-real-time and unreliable manner, and the information is less current.

[0103] In an embodiment of the present invention, an onboard flood data processing device based on an aviation emergency mode processes and analyzes the original image data onboard the aircraft during its return journey, and uploads the data to a remote terminal after landing. The time interval between the image acquisition moment and the data processing moment is small, and after landing, the remote terminal can directly provide disaster relief guidance based on the processed data, and the information is highly current.

[0104] The embodiment of the present invention is based on collecting large-scale ground observation images in the air, performing image mosaicking through a simplified method to obtain a full-frame target image, identifying the target image through a simplified water body recognition model to obtain a water body recognition result, combining geographic information data for spatial analysis to obtain comprehensive flood analysis information, and sending the comprehensive flood analysis information to a remote terminal when the aircraft lands. It is possible to complete the splicing, correction, and information extraction of flood data in real time and quasi-real time on board during the flight, and transmit the data to the ground command system and front-line rescue units after the aircraft lands, thereby improving the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

[0105] On the basis of any of the above embodiments, after performing spatial analysis based on the water body identification results and target thematic geographic data to obtain comprehensive flood analysis information, it also includes: during the flight of the aircraft, the comprehensive flood analysis information is sent to a remote terminal, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions.

[0106] Specifically, after step 103, when the aircraft is in flight, the onboard flood data processing device based on the aviation emergency mode sends the comprehensive flood analysis information to the remote terminal through wireless communication technology.

[0107] Among them, wireless communication technology includes but is not limited to WIFI wireless cellular signals (2G, 3G, 4G, 5G), or instant messaging applications (such as email).

[0108] The embodiment of the present invention is based on collecting large-scale ground observation images in the air, performing image mosaicking through a simplified method, obtaining a full-frame target image, identifying the target image through a simplified water body recognition model to obtain a water body recognition result, combining geographic information data for spatial analysis to obtain comprehensive flood analysis information, and sending the comprehensive flood analysis information to a remote terminal during the flight of the aircraft. After the data is spliced, corrected, and information extracted in real time or quasi-real time on board during the flight, the data is directly transmitted to the ground command system and front-line rescue units, improving the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

[0109] On the basis of any of the above embodiments, image mosaicking is performed based on the earth observation image to obtain the target image, including: extracting the target observation image from the earth observation image based on the target parameters.

[0110] It should be noted that for earth observation images, usually exposure is timed and the exposure time is recorded, or exposure is fixed at a distance and the exposure time is recorded. Finally, the image position (usually the center point of the image) is obtained by matching the exposure time with the airborne GPS data.

[0111] In normal production, in order to ensure the mapping accuracy of image products, a large proportion of (heading, lateral) overlap needs to be set, which leads to a large overlap of images, but correspondingly leads to a large increase in the amount of data. In the embodiment of the present invention, it is necessary to reduce the amount of data, in order to sacrifice a certain observation accuracy in exchange for data processing time.

[0112] The target parameter refers to the parameter corresponding to the boundary condition of the extraction rule which is guided by reducing the overlap between images.

[0113] The embodiment of the present invention does not specifically limit the degree of overlap between images.

[0114] Preferably, the overlap between images is set to 5%. The corresponding target parameters include but are not limited to one or more of exposure mode, flight speed, overlap and meteorological conditions.

[0115] Specifically, in step 101, the onboard flood data processing device based on the aviation emergency mode extracts a number of target observation images from the ground observation image according to different target parameters, so that the overlap between the target observation images is 5%.

[0116] The embodiment of the present invention does not specifically limit the acquisition of the target observation image.

[0117] Optionally, by programming, the aerial camera storage unit is accessed, and the original earth observation images acquired in the memory are extracted at intervals according to the designed exposure mode, flight speed, overlap, and meteorological conditions.

[0118] At this time, the image is original spatial resolution data, and the stitching effect is the best, but the corresponding data volume is large and the processing speed is slow.

[0119] Optionally, by programming, the aerial camera operation management system or memory is accessed, and thumbnails of the ground observation images are obtained by interval extraction in the memory according to the designed exposure mode, flight speed, overlap, and meteorological conditions.

[0120] Image thumbnails are down-scaled versions of the original images, covering the same area as the original images, but with lower spatial resolution and smaller data volumes. The use of either method depends on the capabilities of the onboard computer and the urgency of the disaster.

[0121] Based on the target observation image, image mosaicking is performed to obtain the target image.

[0122] Specifically, the onboard flood data processing device based on the aviation emergency mode selects an image with relatively uniform brightness and color from the target observation image as a reference image, and performs registration and geometric correction on the remaining target observation images to mosaic the parts overlapping with the reference image to obtain the target image.

[0123] The embodiment of the present invention extracts the target observation image from the ground observation image based on the target parameters, performs image mosaicking by a simplified method, and obtains the full-frame target image. It is possible to complete data splicing, correction, and information extraction in real time and quasi-real time on board during flight. Furthermore, after the aircraft lands, the data is transmitted to the ground command system and front-line rescue units, which improves the efficiency and timeliness of information processing, and has obvious significance for emergency rescue.

[0124] On the basis of any of the above embodiments, extracting a target observation image from the ground observation image based on the target parameters includes: acquiring the target observation image based on the target parameters, and recording the target exposure time.

[0125] The target exposure time is the exposure time of the target observation image.

[0126] Specifically, based on the setting of target parameters such as exposure mode, flight speed, overlap, and meteorological conditions in the comprehensive flood analysis information processing device of aerial remote sensing, the target exposure time is recorded while the target observation image is extracted.

[0127] The target exposure time refers to the exposure time corresponding to the target observation image. The number of data of the target exposure time is equal to the number of target observation images.

[0128] Based on the target exposure time, the target GPS location information is matched with the target observation image.

[0129] Specifically, the onboard flood data processing device based on the aviation emergency mode retrieves the GPS record of the aircraft according to the target exposure time to obtain the target GPS position information, and uses the target exposure time to use the target GPS position information as the position information corresponding to the exposure center point of the target observation image, that is, each target observation image and the target GPS position information form a data pair.

[0130] The embodiment of the present invention does not specifically limit the target GPS location information.

[0131] Optionally, the target GPS position information may be a coarse coordinate position directly obtained from a GPS device carried by an aircraft by an onboard flood data processing device based on an aviation emergency mode.

[0132] Optionally, the target GPS position information may be a precise coordinate position calculated by differentially calculating the coarse coordinate position and ground base station observation data or precise ephemeris.

[0133] This process is not specifically limited in the embodiment of the present invention.

[0134] For example, Figure 2 This is one of the sub-flow diagrams of the onboard flood data processing method based on the aviation emergency mode provided by the present invention. Figure 2 As shown, a specific implementation method for obtaining a target observation image is given below:

[0135] Step 201: extract the target observation image or the thumbnail of the target observation image according to the capability level of the onboard computer and the urgency of the disaster.

[0136] Step 202: Extract and match target GPS location information.

[0137] Step 203: The target observation image and the target GPS position are combined into a data pair through the target exposure time to complete the extraction of the optical aerial camera image.

[0138] Based on the target parameters, the embodiment of the present invention records the target exposure time while extracting the target observation image, and associates the target GPS location information with the target observation image through the target exposure time. Under the condition of meeting a certain accuracy, it is possible to reduce the overlapping area between the images and improve the image processing efficiency. Furthermore, it improves the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

[0139] On the basis of any of the above embodiments, image mosaicking is performed based on the target observation image to obtain the target image, including: performing image registration based on the target observation image to extract the same-name point pairs.

[0140] Specifically, the onboard flood data processing device based on the aviation emergency mode extracts feature points with the same name in the overlapping area between the target observation image as the reference image and each target observation image to be mosaicked.

[0141] The embodiment of the present invention does not specifically limit the method for extracting feature points with the same name.

[0142] Preferably, a splicing method based on Harris corner detector is used to extract the same-name point pairs, and its basic principle is as follows:

[0143] A local window is set in the image to be detected, and the intensity is changed by changing a small amount of the window in different directions.

[0144] E(u,v)=∑ i w(x i ,y i )[I(x i +u,y i +v)-I(x i ,y i )]

[0145] Among them, w(x i ,y i ) is the detection window (x i ,y i ). I(x i ,y i ) is in (x i ,y i ) position, I(x i +u,y i +v) is the displacement strength to (u,v). Local texture around pixel (local texture around pixel), (x i ,y i ) can be expressed as the following autocorrelation matrix:

[0146]

[0147] in, and They are I(x i ,y i ) In the gradient in the x direction and the gradient in the y direction, the maximum eigenvalues ​​corresponding to the two matrices C correspond to a corner point. To make the algorithm more robust, the "cornerness" R value can be used to eliminate edge points:

[0148] R=Det(C)-αTr 2 (C)

[0149] Where Tr(C) is the tracking function of C, and the range of α is 0.04≤α≤0.06. If the local maximum of R is higher than the threshold T, then this point is the corner point. After all the corner points of the two images are detected, their correspondence can be determined by the Normalization Cross-Correlation (NCC) or Sum of Squared Differences (SSD) method.

[0150] Based on the same-name point pairs, the least squares method is used to perform affine transformation to obtain the stitched image.

[0151] Specifically, the onboard flood data processing device based on the aviation emergency mode solves the affine transformation parameters through the same-name point pairs, and uses the geometric correspondence corresponding to the affine transformation parameters to unify the target observation images to be mosaicked into the global reference system to obtain the stitched image.

[0152] A stitched image refers to an image that is stitched together based on the image's own features and does not contain spatial geographic information in the conventional sense.

[0153] The embodiment of the present invention does not specifically limit the method for obtaining the stitched image.

[0154] Preferably, the least squares method is used to perform affine transformation to obtain a spliced ​​image. The specific process is as follows:

[0155] Affine Transformation refers to the process of performing a linear transformation (multiplying by a matrix) and a translation (adding a vector) in a vector space to transform it to another vector space.

[0156] Affine transformation represents the mapping relationship between two images, which can be expressed as:

[0157]

[0158] Among them, the affine transformation matrix M is a 2*3 matrix, the matrix B plays the role of translation, and the diagonal in the matrix A determines the scaling, and the anti-diagonal determines the rotation or shearing.

[0159] The original pixel coordinates (x, y), the coordinates of the point after affine transformation are T, so affine transformation is a linear transformation from two-dimensional coordinates (x, y) to two-dimensional coordinates (u, v), and its mathematical expression is as follows:

[0160]

[0161] A and B in the equation need to be solved through pairs of points with the same name. The basic process is:

[0162] (1) Taking the first input image as a benchmark, complete the selection of feature point pairs with the same name as its neighboring images.

[0163] (2) Using a large number of feature point pairs with the same name, the least squares method is used to solve the affine transformation parameters A and B.

[0164] (3) Apply the solved parameters to the affine transformation of the stitched image.

[0165] (4) Stitching the transformed image with the reference image.

[0166] Based on the stitched images, the least squares method is used to associate spatial geographic information and generate the target image.

[0167] Specifically, the onboard flood data processing device based on the aviation emergency mode associates the stitched image with the existing spatial geographic information, and outputs the stitched image incorporated into a unified spatial geographic framework as a target image.

[0168] The embodiment of the present invention does not specifically limit the method for acquiring the target image.

[0169] Preferably, the center point of each target observation image involved in the stitching is used as the ground control point (GCP), and the least squares method is used to solve the correction equation parameters through the associated target GPS position information, thereby completing the stitching image correction and incorporating it into a unified spatial geographic framework to obtain the target image.

[0170] This process is not specifically limited in the embodiment of the present invention.

[0171] For example, Figure 3 This is the second sub-flow diagram of the onboard flood data processing method based on the aviation emergency mode provided by the present invention. Figure 3 As shown, a specific implementation method for obtaining a target observation image is given below:

[0172] Step 301: perform image registration on the target observation image and extract the same-name point pairs.

[0173] Step 302: Calculate affine transformation parameters.

[0174] Step 303: Perform affine transformation on the image to be registered.

[0175] Step 304: Obtain a stitched image.

[0176] Step 305: Add geospatial information coding to the stitched image.

[0177] Step 306: Complete image mosaicking.

[0178] The embodiment of the present invention extracts same-name point pairs based on the fixed-pose target observation image, obtains a spliced ​​image using the same-name point pairs through the principle of least squares method, and associates spatial geographic information through the principle of least squares method to obtain the target image. It can reduce the matching operation of the same-name feature point pairs at the expense of a certain geometric accuracy, which saves the matching time of the same-name feature point pairs on the one hand, and avoids the difficulty of matching the same-name feature point pairs on the other hand, which is an effective compromise between accuracy and efficiency. Furthermore, it improves the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

[0179] Based on any of the above embodiments, spatial analysis is performed based on water body identification results and geographic information data to obtain comprehensive flood analysis information, including: performing spatial analysis based on water body identification results and target geographic information data to obtain thematic information.

[0180] It should be noted that target geographic information data refers to the classification of geographic information data according to the requirements of the rescue mission to obtain directional data.

[0181] Exemplarily, the target geographic information data may be geographic data related to one or more of roads, residential areas, reservoirs, rivers, mining areas, tailings ponds, factories and mines, and ecological protection areas.

[0182] Specifically, in step 103, during the flight of the aircraft, the onboard flood data processing device based on the aviation emergency mode determines the water body identification result obtained in step 102 according to the spatial distribution status and attention level of the regional key targets, aligns it with the existing target geographic information data, incorporates it into a unified spatial geographic framework, and obtains thematic information corresponding to the target geographic information data.

[0183] Thematic information refers to data that associates the pixel coordinate system of the water body recognition result with the actual target geographic coordinate system. Thematic information is used to characterize whether there is a water body in the actual geographic location corresponding to the topic, that is, whether it is flooded. The embodiment of the present invention does not specifically limit the thematic information.

[0184] Optionally, if the target geographic information data is road-related geographic data, the corresponding thematic information is road damage or availability information, which can be used to determine whether the road is flooded to guide rescue road planning, emergency repair, etc.

[0185] Optionally, if the target geographic information data is geographic data related to a residential area, the corresponding thematic information is disaster information of the residential area, which can guide the deployment of rescue forces, etc.

[0186] Optionally, if the target geographic information data is reservoir and river-related geographic data, the corresponding thematic information is the river and surrounding disaster information, which can guide the inspection of river embankments with hidden dangers, disaster development simulation and disaster rescue dispatch.

[0187] Optionally, if the target geographic information data is mining area, tailings pond, factory and mine-related geographic data, the corresponding thematic information is the current status of tailings pond inundation in the mining area, which can guide the current status investigation and simulation of the development trend of secondary disasters.

[0188] Based on the thematic information, statistics are conducted to obtain comprehensive flood analysis information.

[0189] Specifically, the onboard flood data processing device based on the aviation emergency mode performs basic operations such as merging (Union), intersect (Intersect), identifying (Identity), deleting (Erase), updating (Update) and clipping (Clip) on the thematic information to complete the analysis of spatial position, spatial distribution, spatial form, spatial distance, and topological relationship, and to statistically generate comprehensive flood analysis information.

[0190] It is understandable that the comprehensive flood analysis information includes the focus of different topics, the corresponding areas under flood inundation, zoning areas, buffer zone analysis results, etc.

[0191] This process is not specifically limited in the embodiment of the present invention.

[0192] For example, Figure 4 This is the third sub-flow diagram of the onboard flood data processing method based on the aviation emergency mode provided by the present invention. Figure 3 As shown, a specific implementation method for obtaining comprehensive flood analysis information is given below:

[0193] Step 401: Perform GIS (geographic information) spatial analysis based on water body identification results and target geographic information data to obtain thematic information.

[0194] Step 402: Statistics are collected on the analyzed special information to obtain comprehensive flood analysis information.

[0195] Step 403: Based on the comprehensive flood analysis information, a thematic map of regional disaster distribution and severity is produced to provide necessary guidance information for rescue (such as road accessibility information, disaster severity classification information, etc.).

[0196] The embodiment of the present invention performs spatial analysis based on water body identification results and target geographic information data, quantitatively extracts various types of thematic information, and completes the establishment of comprehensive flood analysis information through statistics. It can achieve the expansion of the deep processing capability of data on board in real time and quasi-real time while ensuring a certain accuracy, directly transmit data to the ground command system and front-line rescue units, quickly convert high-current data into high-current information, improve information processing efficiency and timeliness, and has obvious significance for emergency rescue.

[0197] On the basis of any of the above embodiments, the target image is input into the water body recognition model to obtain the water body recognition result, including: obtaining water body feature information based on the target image.

[0198] Specifically, in step 102, the onboard flood data processing device based on the aviation emergency mode inputs the target image generated in step 101 into the water body recognition model to perform feature extraction and obtain water body feature information.

[0199] The embodiment of the present invention does not specifically limit the feature extraction process.

[0200] Preferably, considering the distinctiveness and recognition efficiency of the features, the spectral angle can be extracted as the water body feature information.

[0201] Based on the water body feature information, the water body recognition result corresponding to the target image is obtained and output in vector format.

[0202] Specifically, the onboard flood data processing device based on the aviation emergency mode performs target recognition and classification on the extracted features, completes the extraction of water body recognition results, and outputs the binary results of water bodies and non-water bodies.

[0203] For example, when identifying (classifying) water bodies, the simplest spectral angle or correlation coefficient can be used to quickly determine whether the pixel to be classified is consistent with or close to the water body (sample) by comparing with the threshold. If it is below the threshold, it is judged as a water body, otherwise it is a non-water body.

[0204] The embodiment of the present invention does not specifically limit the output format of the water body identification result.

[0205] Preferably, the onboard flood data processing device based on the aviation emergency mode converts the binary raster data (each pixel has only two optional values, such as either 0 or 1) into vector data through a vectorization method.

[0206] In step 103, the water body identification result vector file containing point, line and surface annotations is superimposed with real space geographic information for analysis, so as to be used for positioning and navigation during ground rescue.

[0207] This process is not specifically limited in the embodiment of the present invention.

[0208] For example, Figure 5 This is the fourth sub-flow diagram of the onboard flood data processing method based on the aviation emergency mode provided by the present invention. Figure 5 As shown, a specific implementation method for obtaining water body recognition results is given below:

[0209] Step 501: Acquire a target image for water body identification.

[0210] Step 502: Extract spectral features of the target image.

[0211] Step 503: Perform rapid identification and classification based on the spectral characteristics of the target image

[0212] Step 504: Generate a binary classification result in a raster format.

[0213] Step 505: Convert the result of step 504 to generate a binary classification result in a vector format.

[0214] The embodiment of the present invention extracts water feature information of the target image based on the ground characteristics after the flood disaster occurs, and quickly obtains the water body recognition result by combining classification with target recognition, and outputs it in vector format. It can reduce the algorithm complexity to reduce the processing time consumption while ensuring a certain accuracy, and the data volume of the vectorized water body recognition result is small, which improves the information processing efficiency and timeliness to a certain extent, and has obvious significance for emergency rescue.

[0215] Figure 6 is a schematic diagram of the structure of the onboard flood data processing device based on the aviation emergency mode provided by the present invention. Figure 6 As shown, the device includes: an image acquisition module 610, a water body recognition module 620, an onboard data generation module 630 and a landing data sending module 640, wherein:

[0216] The image acquisition module 610 is used to perform image mosaicking based on the earth observation image to acquire the target image.

[0217] The water body recognition module 620 is used to input the target image into the water body recognition model to obtain the water body recognition result.

[0218] The on-board data generation module 630 is used to perform spatial analysis based on the water body identification results and geographic information data to obtain comprehensive flood analysis information.

[0219] The landing data sending module 640 is used to send the comprehensive flood analysis information to the remote terminal when the aircraft lands, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions.

[0220] Among them, the ground observation images include observation images taken by the aerial camera carried by the aircraft, and the GPS location information corresponding to the observation images. The water body recognition model is obtained after training based on training samples and predetermined water body labels.

[0221] Specifically, the image acquisition module 610, the water body recognition module 620, the onboard data generation module 630 and the landing data sending module 640 are electrically connected in sequence.

[0222] The image acquisition module 610 forms a data pair with the ground observation image and the corresponding GPS position information, and selects an image with relatively uniform brightness and color as the reference image for mosaicking. Other images are geometrically aligned based on the reference image, so that all images to be mosaicked are mosaicked from near to far in the same spatial coordinate system to form a target image.

[0223] The water body recognition module 620 performs rapid water body recognition on the target image in the image acquisition module 610 through a simplified water body recognition model trained in advance, and can obtain a water body recognition result corresponding to the target image.

[0224] The water body recognition result may be a probability value or a label result. The embodiment of the present invention does not specifically limit the form of the water body recognition result.

[0225] During the flight of the aircraft, the onboard data generation module 630 incorporates the water body identification results obtained in the water body identification module 620 into a unified spatial geographic framework by aligning them with the existing geographic information data, and obtains comprehensive flood analysis information.

[0226] Comprehensive flood analysis information refers to data that associates the pixel coordinate system of the water body identification result with the actual spatial geographic coordinate system. Comprehensive flood analysis information is used to characterize whether a specific actual geographic location has a water body, that is, whether it is flooded.

[0227] When the aircraft lands, the landing data sending module 640 sends the comprehensive flood analysis information to the remote terminal through wireless communication technology.

[0228] Among them, wireless communication technology includes but is not limited to WIFI wireless cellular signals (2G, 3G, 4G, 5G).

[0229] Optionally, the device further includes: an onboard data sending module, wherein:

[0230] The onboard data sending module is used to send comprehensive flood analysis information to a remote terminal during the flight of the aircraft, so that the remote terminal can receive the comprehensive flood analysis information, conduct a preliminary assessment of the disaster situation, and guide the generation of rescue command decisions.

[0231] Optionally, the image acquisition module 610 includes an extraction unit and a mosaic unit, wherein:

[0232] The extraction unit is used to extract a target observation image from the ground observation image based on the target parameters.

[0233] The mosaic unit is used to perform image mosaicking based on the target observation image to obtain the target image.

[0234] Optionally, the extraction unit includes a recording time subunit and a matching subunit, wherein:

[0235] The recording time subunit is used to obtain a target observation image based on target parameters and record the target exposure time.

[0236] The matching subunit is used to match the target GPS position information with the target observation image based on the target exposure time.

[0237] The target exposure time is the exposure time of the target observation image.

[0238] Optionally, the mosaic unit includes a registration subunit, an affine subunit and an association subunit, wherein:

[0239] The registration subunit is used to perform image registration and extract the same-name point pairs based on the target observation image.

[0240] The affine subunit is used to perform affine transformation based on the same-name point pairs using the least squares principle to obtain a stitched image.

[0241] The association subunit is used to associate spatial geographic information based on the spliced ​​images and adopt the least squares principle to generate a target image.

[0242] Optionally, the on-board data generation module 630 includes a thematic information acquisition unit and a statistical unit, wherein:

[0243] The thematic information acquisition unit is used to perform spatial analysis and acquire thematic information based on the water body identification results and target geographic information data.

[0244] The statistical unit is used to perform statistics based on thematic information and obtain comprehensive flood analysis information.

[0245] Optionally, the water body identification module 620 includes a feature extraction unit and an identification unit, wherein:

[0246] The feature extraction unit is used to obtain water body feature information based on the target image.

[0247] The recognition unit is used to obtain a water body recognition result corresponding to the target image based on the water body feature information and output it in a vector format.

[0248] The onboard flood data processing device based on the aviation emergency mode provided in an embodiment of the present invention is used to execute the onboard flood data processing method based on the aviation emergency mode of the present invention. Its implementation method is consistent with the implementation method of the onboard flood data processing method based on the aviation emergency mode provided by the present invention, and can achieve the same beneficial effects, which will not be repeated here.

[0249] The embodiment of the present invention is based on collecting large-scale ground observation images in the air, performing image mosaicking through a simplified method to obtain a full-frame target image, identifying the target image through a simplified water body recognition model to obtain a water body recognition result, combining geographic information data for spatial analysis to obtain comprehensive flood analysis information, and sending the comprehensive flood analysis information to a remote terminal when the aircraft lands. It is possible to complete the splicing, correction, and information extraction of flood data in real time and quasi-real time on board during the flight, and transmit the data to the ground command system and front-line rescue units after the aircraft lands, thereby improving the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

[0250] Based on any of the above embodiments, the device further includes: an onboard data sending module, wherein:

[0251] The onboard data sending module is used to send comprehensive flood analysis information to a remote terminal during the flight of the aircraft, so that the remote terminal can receive the comprehensive flood analysis information, conduct a preliminary assessment of the disaster situation, and guide the generation of rescue command decisions.

[0252] Specifically, the onboard data sending module sends the comprehensive flood analysis information to the remote terminal through wireless communication technology during the flight of the aircraft.

[0253] Among them, wireless communication technology includes but is not limited to WIFI wireless cellular signals (2G, 3G, 4G, 5G), or instant messaging applications (such as email).

[0254] The embodiment of the present invention is based on collecting large-scale ground observation images in the air, performing image mosaicking through a simplified method to obtain a full-frame target image, identifying the target image through a simplified water body recognition model to obtain a water body recognition result, combining geographic information data for spatial analysis to obtain comprehensive flood analysis information, and sending the comprehensive flood analysis information to a remote terminal during the flight of the aircraft. After the data is spliced, corrected, and information extracted in real time or quasi-real time on board during the flight, the data is directly transmitted to the ground command system and front-line rescue units, improving the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

[0255] Figure 7 FIG. 1 is a schematic diagram of the structure of the onboard flood data processing system based on the aviation emergency mode provided by the present invention. Figure 7 As shown, the system includes one or more onboard flood data processing devices 710 based on aviation emergency mode, and also includes: a remote terminal 720.

[0256] The remote terminal is communicatively connected with the landing data sending module or the onboard data sending module.

[0257] It should be noted that a ground command system can be generated around the remote terminal 720 on the ground, so that the ground command system can manage and collect data from different onboard flood data processing devices 710 based on aviation emergency modes, and can command disaster relief in an all-round manner.

[0258] Specifically, each onboard flood data processing device 710 based on the aviation emergency mode establishes a corresponding wireless communication connection with the remote terminal 720 according to the setting of the transceiver device on the aircraft.

[0259] Optionally, when there is no satellite communication system on the aircraft, the landing data sending module can directly communicate with the remote terminal 720. The communication method can be wireless communication technology (Wi-Fi), Bluetooth or serial port, etc., which is not specifically limited in the embodiment of the present invention.

[0260] Optionally, when a satellite communication system is provided on the aircraft, the onboard data transmission module can directly communicate with the remote terminal 720. The communication method can be wireless communication technology (Wi-Fi), instant messaging software, etc., which is not specifically limited in the embodiment of the present invention.

[0261] The remote terminal 720 is used to receive comprehensive flood analysis information, conduct preliminary assessment of the disaster situation, and guide the generation of rescue command decisions.

[0262] Specifically, the remote terminal 720 receives comprehensive flood analysis information sent from each onboard flood data processing device 710 based on the aviation emergency mode, and conducts a first-stage study and judgment on the flood disaster situation at the corresponding moment based on the comprehensive flood analysis information, and provides a rescue command plan to deal with the disaster.

[0263] The embodiment of the present invention does not specifically limit the processing process of the remote terminal 720.

[0264] Optionally, after receiving the comprehensive flood analysis information, the remote terminal 720 displays it, and the user of the remote terminal 720 conducts a subjective analysis of the disaster situation and formulates a corresponding rescue plan.

[0265] Optionally, after receiving the comprehensive flood analysis information, the remote terminal 720 executes a corresponding processing algorithm, objectively analyzes the disaster situation, and formulates a corresponding rescue plan for the user of the remote terminal 720 to select or optimize the plan. It is understandable that the generation time of the comprehensive flood analysis information can be adjusted according to the amount of data and the urgency of information acquisition. Through program control, the order of data processing and flight data acquisition can be adjusted. Similarly, the comprehensive flood analysis information can also be adjusted to a dynamic update state, but considering actual use, the dynamic update interval can be slightly larger, such as an hour interval.

[0266] The embodiment of the present invention is based on collecting large-scale ground observation images in the air, performing image mosaicking through a simplified method to obtain a full-frame target image, identifying the target image through a simplified water body recognition model to obtain a water body recognition result, combining geographic information data for spatial analysis to obtain comprehensive flood analysis information, and sending the comprehensive flood analysis information to a remote terminal when the aircraft lands. It is possible to complete the splicing, correction, and information extraction of flood data in real time and quasi-real time on board during the flight, and transmit the data to the ground command system and front-line rescue units after the aircraft lands, thereby improving the efficiency and timeliness of information processing, which is of obvious significance to emergency rescue.

[0267] Figure 8 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 8As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the on-board flood data processing method based on the aviation emergency mode, the method comprising: based on the ground observation image, performing image mosaic to obtain the target image; inputting the target image into the water body recognition model to obtain the water body recognition result; performing spatial analysis based on the water body recognition result and the geographic information data to obtain the comprehensive flood analysis information; when the aircraft lands, the comprehensive flood analysis information is sent to the remote terminal, so that the remote terminal receives the comprehensive flood analysis information, makes a preliminary judgment on the disaster situation, and guides the generation of rescue command decisions; wherein the ground observation image includes the observation image taken by the aerial camera carried by the aircraft, and the GPS location information corresponding to the observation image, and the water body recognition model is obtained after training based on the training sample and the predetermined water body label.

[0268] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0269] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the on-board flood data processing method based on the aviation emergency mode provided by the above-mentioned methods, the method including: based on ground observation images, image mosaicking is performed to obtain a target image; the target image is input into a water body recognition model to obtain a water body recognition result; based on the water body recognition result and geographic information data, spatial analysis is performed to obtain comprehensive flood analysis information; when the aircraft lands, the comprehensive flood analysis information is sent to a remote terminal, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions; wherein the ground observation image includes an observation image taken by an aerial camera carried by the aircraft, and GPS location information corresponding to the observation image, and the water body recognition model is obtained after training based on training samples and predetermined water body labels.

[0270] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the onboard flood data processing method based on the aviation emergency mode provided by the above-mentioned methods, the method comprising: based on ground observation images, performing image mosaicking to obtain a target image; inputting the target image into a water body recognition model to obtain a water body recognition result; based on the water body recognition result and geographic information data, performing spatial analysis to obtain comprehensive flood analysis information; when the aircraft lands, sending the comprehensive flood analysis information to a remote terminal, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions; wherein the ground observation image includes an observation image taken by an aerial camera carried by the aircraft, and GPS location information corresponding to the observation image, and the water body recognition model is obtained after training based on training samples and predetermined water body labels.

[0271] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0272] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0273] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing on-board flood data based on aviation emergency mode, characterized in that: include: Based on the target parameters, the target observation image is extracted from the ground observation image, and the target exposure time is recorded; Based on the target exposure time, the target GPS location information is matched with the target observation image; wherein the target exposure time is the exposure time of the target observation image; based on the target observation image, image registration is performed to extract the same-name point pairs; based on the same-name point pairs, affine transformation is performed using the least squares principle to obtain a spliced ​​image; based on the spliced ​​image, spatial geographic information is associated using the least squares principle to generate a target image; Inputting the target image into a water body recognition model to obtain a water body recognition result; Based on the water body identification results and the target geographic information data, spatial analysis is performed to obtain thematic information; based on the thematic information, statistics are performed to obtain comprehensive flood analysis information; When the aircraft lands, the comprehensive flood analysis information is sent to a remote terminal, so that the remote terminal receives the comprehensive flood analysis information, makes a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions; The earth observation image includes an observation image taken by an aerial camera carried by the aircraft and GPS location information corresponding to the observation image, and the water body recognition model is obtained after training based on training samples and predetermined water body labels.

2. The method for processing flood data on board based on aviation emergency mode according to claim 1, characterized in that: After performing spatial analysis based on the water body identification results and target thematic geographic data to obtain comprehensive flood analysis information, the method also includes: during the flight of the aircraft, sending the comprehensive flood analysis information to a remote terminal, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions.

3. The method for processing flood data on board based on aviation emergency mode according to claim 1 or 2, characterized in that: The step of inputting the target image into a water body recognition model to obtain a water body recognition result includes: Based on the target image, acquiring water body characteristic information; Based on the water body feature information, a water body recognition result corresponding to the target image is obtained and output in a vector format.

4. An onboard flood data processing device based on aviation emergency mode, characterized in that: include: An image acquisition module is used to extract a target observation image from the earth observation image based on target parameters and record the target exposure time; Based on the target exposure time, the target GPS location information is matched with the target observation image; wherein the target exposure time is the exposure time of the target observation image; based on the target observation image, image registration is performed to extract the same-name point pairs; based on the same-name point pairs, affine transformation is performed using the least squares principle to obtain a spliced ​​image; based on the spliced ​​image, spatial geographic information is associated using the least squares principle to generate a target image; A water body recognition module, used to input the target image into a water body recognition model to obtain a water body recognition result; An onboard data generation module is used to perform spatial analysis based on the water body identification results and target geographic information data to obtain thematic information; based on the thematic information, perform statistics to obtain comprehensive flood analysis information; A landing data sending module is used to send the comprehensive flood analysis information to a remote terminal when the aircraft lands, so that the remote terminal receives the comprehensive flood analysis information, makes a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions; The earth observation image includes an observation image taken by an aerial camera carried by the aircraft and GPS location information corresponding to the observation image, and the water body recognition model is obtained after training based on training samples and predetermined water body labels.

5. The onboard flood data processing device based on aviation emergency mode according to claim 4 is characterized in that: The device also includes: The onboard data sending module is used to send the comprehensive flood analysis information to a remote terminal during the flight of the aircraft, so that the remote terminal receives the comprehensive flood analysis information, conducts a preliminary assessment of the disaster situation, and guides the generation of rescue command decisions.

6. An onboard flood data processing system based on aviation emergency mode, comprising one or more onboard flood data processing devices based on aviation emergency mode as claimed in claim 5, characterized in that: Also includes: a remote terminal; The remote terminal is communicatively connected with the landing data sending module or the onboard data sending module; The remote terminal is used to receive the comprehensive flood analysis information, conduct a preliminary assessment of the disaster situation, and guide the generation of rescue command decisions.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the onboard flood data processing method based on the aviation emergency mode as described in any one of claims 1 to 3 are implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the onboard flood data processing method based on the aviation emergency mode as claimed in any one of claims 1 to 3 are implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the onboard flood data processing method based on the aviation emergency mode as claimed in any one of claims 1 to 3 are implemented.

Images