Ground fixed target identification method, device, equipment, medium and product
By calculating the identification information and image feature matching of the aircraft relative to the target, the problem that pilots find it difficult to accurately identify fixed targets on the ground in complex ground scenes is solved, and the accuracy of automatic accurate identification and material deployment is improved.
Patent Information
- Application Number
- CN202411767032.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-04
- Publication Date
- 2025-03-25
AI Technical Summary
In complex ground scenarios, it is difficult for pilots to accurately identify fixed targets on the ground, resulting in inaccurate material delivery.
By obtaining the position information of the aircraft and the real-time image of the target area, the identification information of the aircraft relative to the target is calculated, and the matching target reference template feature map is obtained from the reference template feature library, image features are extracted and matched, and the target position is determined.
It realizes automatic and accurate identification of ground fixed targets in complex backgrounds, improving the accuracy and reliability of material deployment.
Smart Images
Figure HDA0005169225090000011 
Figure HDA0005169225090000012 
Figure HDA0005169225090000021
Abstract
Description
Technical Field
[0001] The present application relates to the field of target identification, and in particular to a method, device, equipment, medium and product for identifying ground fixed targets. Background Art
[0002] Material delivery refers to the process of placing, transporting and distributing various items (such as materials, goods, etc.) to specific locations or target groups in a planned manner. These materials can include daily necessities (such as food, drinking water, clothing, etc.), disaster relief materials (tents, medical supplies, etc.), commercial products, and many other types.
[0003] Aerial material delivery is a method of material delivery, which mainly uses aircraft and other aircraft to transport materials from the air and deliver them to designated areas.
[0004] In actual cities, when delivering supplies to a fixed target location, the pilot often manually flies the aircraft to the top of the supply delivery location and then delivers the supplies.
[0005] However, since the fixed ground targets to be delivered are often located in complex ground scenes, pilots at high altitudes are often unable to accurately identify the fixed ground targets in complex ground scenes to achieve precise delivery. Summary of the invention
[0006] The purpose of this application is to provide a method, device, equipment, medium and product for identifying ground fixed targets, which can automatically and accurately identify ground fixed targets in complex backgrounds.
[0007] To achieve the above objectives, this application provides the following solutions:
[0008] In a first aspect, the present application provides a method for identifying a ground fixed target, the method for identifying a ground fixed target comprising:
[0009] Obtain the position information of the aircraft at the time of identification and the real-time image of the target area where the target to be identified is located;
[0010] Calculating target recognition information of the aircraft relative to the target to be recognized based on the position information at the recognition time and the position information of the target to be recognized in the target area, wherein the target recognition information includes: the direction of the aircraft relative to the target to be recognized, the distance between the aircraft and the target to be recognized, and the flight altitude of the aircraft;
[0011] Obtaining a matching target reference template feature map from a reference template feature map library according to the target recognition information; the reference template feature library stores the corresponding relationship between different recognition information of each target to be recognized in the target area and the reference template feature map;
[0012] Performing image feature extraction on the real-time image to obtain a real-time feature map of the target area;
[0013] Obtain a correlation matrix between the target area real-time feature map and the target reference template feature map; determine the position in the target area real-time feature map corresponding to the maximum value in each of the correlation matrixes as the position of the target to be identified.
[0014] In a second aspect, the present application provides a ground fixed target identification device, comprising:
[0015] The first acquisition module is used to acquire the position information of the aircraft at the time of identification and the real-time image of the target area where the target to be identified is located;
[0016] a calculation module, configured to calculate target recognition information of the aircraft relative to the target to be recognized based on the position information at the recognition time and the position information of the target to be recognized in the target area, wherein the target recognition information includes: the direction of the aircraft relative to the target to be recognized, the distance between the aircraft and the target to be recognized, and the flight altitude of the aircraft;
[0017] A second acquisition module is used to acquire a matching target reference template feature map from a reference template feature map library according to the target recognition information; the reference template feature library stores the corresponding relationship between different recognition information of each target to be recognized in the target area and the reference template feature map;
[0018] An extraction module, used to extract image features from the real-time image to obtain a real-time feature map of the target area;
[0019] A third acquisition module is used to acquire a correlation matrix between the target area real-time feature map and the target reference template feature map;
[0020] The determination module is used to determine that the position in the target area real-time feature map corresponding to the maximum value in each of the correlation arrays is the position where the target to be identified is located.
[0021] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for identifying ground fixed targets described in any one of the above-mentioned methods.
[0022] In a fourth aspect, the present application provides a 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 identifying ground fixed targets.
[0023] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned methods for identifying ground fixed targets.
[0024] According to the specific embodiments provided in this application, this application discloses the following technical effects:
[0025] The present application provides a method, device, equipment, medium and product for identifying ground fixed targets, the method comprising: obtaining position information of an aircraft at the time of identification and a real-time image of a target area where a target to be identified is located; calculating target identification information of the aircraft relative to the target to be identified based on the position information at the time of identification and the position information of the target to be identified in the target area, the target identification information comprising: the direction of the aircraft relative to the target to be identified, the distance between the aircraft and the target to be identified and the flight altitude of the aircraft; obtaining a matching target reference template feature map from a reference template feature map library based on the target identification information; the reference template feature library stores the corresponding relationship between different identification information of each target to be identified in the target area and the reference template feature map; performing image feature extraction on the real-time image to obtain a real-time feature map of the target area; obtaining a correlation matrix between the real-time feature map of the target area and the target reference template feature map; determining that the position in the real-time feature map of the target area corresponding to the maximum value in each correlation matrix is the position where the target to be identified is located. In the present disclosure, since the reference template feature library stores the correspondence between the different identification information of each target to be identified in the target area and the reference template feature map, the accuracy and reliability of identification can be improved when the position of the target to be identified is identified in the real-time image through the reference template feature library. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0027] Figure 1 is a flow chart of a method for identifying a ground fixed target according to an exemplary embodiment;
[0028] Figure 2 is a flow chart of a method for generating a reference template feature library according to an exemplary embodiment;
[0029] Figure 3 A schematic diagram of a spatial position change provided by an embodiment of the present application;
[0030] Figure 4A schematic diagram of a preset window provided in an embodiment of the present application;
[0031] Figure 5 An embodiment of the present application provides Figure 4 Schematic diagram of the result matrix corresponding to the preset window in;
[0032] Figure 6 A schematic diagram of functional modules of a ground fixed target identification device provided in one embodiment of the present application;
[0033] Figure 7 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION
[0034] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0035] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0036] Figure 1 is a flow chart of a method for identifying a ground fixed target according to an exemplary embodiment. Figure 1 As shown, the method includes the following steps S101-S106:
[0037] In step S101, the position information of the aircraft at the time of identification and the real-time image of the target area where the target to be identified is located are obtained.
[0038] The imaging device mounted on the aircraft can be used to capture real-time images of the target area where the target to be identified is located.
[0039] In step S102, target recognition information of the aircraft relative to the target to be recognized is calculated based on the position information at the recognition time and the position information of the target to be recognized in the target area.
[0040] The target identification information includes: the direction of the aircraft relative to the target to be identified, the distance between the aircraft and the target to be identified, and the flight altitude of the aircraft.
[0041] Among them, the location information of the target to be identified in the target area can be preset in advance.
[0042] In step S103, a matching target reference template feature map is obtained from a reference template feature map library according to the target recognition information.
[0043] The reference template feature library stores the corresponding relationship between different identification information of each target to be identified in the target area and the reference template feature map.
[0044] In one embodiment, it is also necessary to prepare a reference template feature library. Figure 2 As shown, preparing a reference template feature library includes the following sub-steps S201-S203:
[0045] S201. In a three-dimensional scene model, a two-dimensional slice corresponding to each preset recognition target in a target area is obtained according to preset recognition information as a basic template image.
[0046] The preset identification information includes: a preset identification direction, a preset identification distance and a preset identification height.
[0047] Based on the existing three-dimensional scene model of the target area, a preset recognition target to be recognized is selected and determined, and then a two-dimensional slice is obtained in the three-dimensional scene model according to the preset recognition direction, preset recognition distance and preset recognition height with this as the viewing center, and saved as the basic template image.
[0048] Among them, the preset identification distance can be the distance between the aircraft and the preset identification target during identification, the preset identification altitude can be the preset flight altitude of the aircraft, and the preset identification direction is the direction of the aircraft relative to the preset identification target during identification.
[0049] S202 , transforming the basic template image in spatial position to obtain reference template images corresponding to the basic template image in different spatial positions.
[0050] Among them, Figure 3 As shown, four preset recognition targets are respectively subjected to spatial position transformation, wherein the spatial position transformation includes: angle transformation, distance transformation and height transformation.
[0051] The basic template image can be transformed in angle, distance and height according to information such as 360° direction, preset recognition distance range of ±500m and preset recognition height range of ±300m to generate multiple reference template images.
[0052] In one embodiment, step S202 includes the following sub-steps S2021-S2023:
[0053] S2021. Keep the second spatial information and the third spatial information unchanged, change the first spatial information corresponding to the basic template image according to a first preset step size, and obtain a plurality of first intermediate template images corresponding to the basic template image.
[0054] S2022: Keep the first spatial information and the third spatial information unchanged, change the second spatial information corresponding to each first intermediate template image according to a second preset step length, and obtain a plurality of second intermediate template images corresponding to each first intermediate template image.
[0055] S2023 , keeping the first spatial information and the second spatial information unchanged, changing the third spatial information corresponding to each second intermediate template image according to a third preset step size, and obtaining a plurality of third intermediate template images corresponding to each second intermediate template image.
[0056] The third intermediate template image constitutes the reference template image; the first spatial information, the second spatial information and the third spatial information correspond one-to-one to the preset recognition direction, the preset recognition distance and the preset recognition height.
[0057] Specifically, the first spatial information corresponds to a preset identification direction, the second spatial information corresponds to a preset identification distance, and the third spatial information corresponds to a preset identification height; or, the first spatial information corresponds to a preset identification direction, the second spatial information corresponds to a preset identification height, and the third spatial information corresponds to a preset identification distance; or, the first spatial information corresponds to a preset identification distance, the second spatial information corresponds to a preset identification direction, and the third spatial information corresponds to a preset identification height; or, the first spatial information corresponds to a preset identification distance, the second spatial information corresponds to a preset identification height, and the third spatial information corresponds to a preset identification direction; or, the first spatial information corresponds to a preset identification height, the second spatial information corresponds to a preset identification distance, and the third spatial information corresponds to a preset identification direction; or, the first spatial information corresponds to a preset identification height, the second spatial information corresponds to a preset identification distance, and the third spatial information corresponds to a preset identification direction; or, the first spatial information corresponds to a preset identification height, the second spatial information corresponds to a preset identification direction, and the third spatial information corresponds to a preset identification distance.
[0058] Taking the case where the first spatial information corresponds to a preset recognition direction, the second spatial information corresponds to a preset recognition distance, and the third spatial information corresponds to a preset recognition height as an example, the preset recognition distance and the preset recognition height are kept unchanged, and the preset recognition direction corresponding to the basic template image is changed according to the first preset step length, and a plurality of first intermediate template images corresponding to the basic template image are obtained. The recognition direction and the preset recognition height are kept unchanged, and the preset recognition distance corresponding to each first intermediate template image is changed according to the second preset step length, and a plurality of second intermediate template images corresponding to each first intermediate template image are obtained. The recognition direction and the recognition distance are kept unchanged, and the preset recognition height corresponding to each second intermediate template image is changed according to the third preset step length, and a plurality of third intermediate template images corresponding to each second intermediate template image are obtained.
[0059] For example, according to task requirements, a preset identification target A to be identified is selected on an existing target area three-dimensional scene model, and its geographic location information is recorded.
[0060] With the selected preset recognition target A as the viewing angle center, on the existing target area three-dimensional scene model, a two-dimensional slice is obtained according to the preset recognition direction (for example: north), preset recognition distance (for example: 3km) and expected recognition height (for example: 800m) and saved as the basic template image A0.
[0061] For the basic template image A0 obtained above, keep its preset recognition distance and preset recognition height from the preset recognition target A unchanged, and perform angle rotation transformation in the recognition direction in the 360° direction with a step length of 1° to obtain the first intermediate template image after 360 angle transformations from 0° to 359°, and record A0 to A according to the angle value number. 359 ;
[0062] For the 360 first intermediate template images A0-A 359 Keeping the original recognition direction and preset recognition height unchanged, the preset recognition distance is pushed forward 500m to the range of 500m, with a step length of 100m, and the distance translation transformation is performed to obtain a second intermediate template image after 11×360 angle and distance transformations from -500m to 500m, and record A according to the angle + distance value number. 0-500 ~A 359+500 ;
[0063] For the 11×360 second intermediate template images A obtained above 0-500 ~A 359+500 , keep its original recognition direction and recognition distance unchanged, push up 300m from the preset recognition height to within the range of 300m, with a step length of 100m, perform height translation transformation, and obtain a third intermediate template image after 7×11×360 directions, distances and height transformations from -300m to 300m, and record A according to the direction+distance+height value number. 0-500-300 ~A 359+500+300 , forming the final reference template image.
[0064] S203: Extract image features from each reference template image to obtain a reference template feature library.
[0065] Each pixel in the reference template image is characterized by a first pixel.
[0066] For each first pixel in each reference template image, perform the following steps:
[0067] In a window of preset size around each first pixel of the current reference template image, the grayscale value of each first pixel in the window is compared with the grayscale value of the first center pixel of the window, and a result matrix for each window is obtained according to the comparison result; wherein, if the grayscale value of the first pixel is greater than the grayscale value of the first center pixel, it is marked as the first value, and if the grayscale value of the first pixel is not greater than the grayscale value of the first center pixel, it is marked as the second value; the overlapping pixels in adjacent windows are ORed to obtain a reference template feature map.
[0068] After obtaining the reference template feature map corresponding to each reference template image, the reference template feature maps corresponding to all reference template images form a reference template feature map library.
[0069] Continuing with the above example, for the 7×11×360 reference templates A obtained above 0-500-300 ~A 359+500+300 Perform image feature extraction to obtain the reference template feature library B 0-500-300 ~B 359+500+300 .
[0070] Taking the preset window size of 3×3, the first value of 1, and the second value of 0 as an example, the image feature extraction processing algorithm is as follows:
[0071] The following processing is performed on each reference template image, that is, starting from the second row and second column of the reference template image to the second to last row and second column of the reference template image, in a 3×3 window around each first pixel, the central pixel of the window is used as the threshold, and the grayscale values of its adjacent 8 pixels are compared with the grayscale value of the central pixel. If it is greater than the grayscale value of the central pixel, it is marked as 1, and if it is less than or equal to it, it is marked as 0, such as Figure 4 As shown in the figure, the grayscale value of the center pixel of the window is 5, so take 5 as the threshold and compare the grayscale values of the 8 adjacent first pixels with 5 to get Figure 5 The 3×3 result matrix shown in the figure can generate a Figure 5 The 3×3 result matrix shown; the overlapping parts of the 3×3 result matrices generated by adjacent first pixels are OR-operated.
[0072] In the present disclosure, in order to ensure the reliability and accuracy of automatic target recognition, when selecting a target reference template feature map from a reference template feature map library based on target recognition information, multiple target reference template feature maps that meet preset requirements with the target recognition information can be obtained from the reference template feature map library based on the target recognition information.
[0073] N images of the direction of the aircraft relative to the target to be identified, the distance between the aircraft and the target to be identified, and the flight altitude of the aircraft may be selected.
[0074] Continuing with the above example: Taking N as 3, the distance between the aircraft and the target to be identified is 1000 meters, then the target reference template feature maps corresponding to the identification distances of 900 meters, 1000 meters and 1100 meters can be selected. In this way, 27 target reference feature maps are selected for each target to be identified.
[0075] From the target reference template feature library B 0-500-300 ~B 359+500+300 27 target reference template feature maps are selected from the image, which are denoted as L1~L2, and the direction of the aircraft relative to the target to be identified, the distance between the aircraft and the target to be identified, and the flight altitude of the aircraft. 27 .
[0076] It is worth noting that in the present disclosure, the reference template feature library stores the preset recognition direction, preset recognition distance and preset recognition altitude. When in use, that is, when matching from the reference template feature library based on the real-time image acquired by the aircraft, what is matched is the direction of the aircraft relative to the target to be recognized, the distance between the aircraft and the target to be recognized, and the flight altitude of the aircraft. At this time, the direction of the aircraft relative to the target to be recognized matches the preset recognition direction in the reference template feature library, the distance between the aircraft and the target to be recognized matches the preset recognition distance in the reference template feature library, and the flight altitude of the aircraft matches the preset recognition altitude in the reference template feature library.
[0077] In step S104, image features are extracted from the real-time image to obtain a real-time feature map of the target area.
[0078] Each pixel in the real-time image is characterized by a second pixel.
[0079] Specifically, in a window of preset size around each second pixel of the real-time image, the grayscale value of each second pixel in the window is compared with the grayscale value of the second center pixel of the window, and a result matrix of each window is obtained according to the comparison result; wherein, if the grayscale value of the second pixel is greater than the grayscale value of the second center pixel, it is marked as the first value, and if the grayscale value of the second pixel is not greater than the grayscale value of the second center pixel, it is marked as the second value; the overlapping pixels in adjacent windows are ORed to obtain a real-time feature map of the target area.
[0080] The step of extracting image features from the real-time image here is similar to the step of extracting image features in the above S203, and will not be repeated here.
[0081] In step S105, a correlation matrix between the target area real-time feature map and the target reference template feature map is obtained.
[0082] When there is one target reference template feature map, obtain the correlation matrix between the target area real-time feature map and the target reference template feature map. If there are multiple target reference template feature maps, obtain the correlation matrix between the target area real-time feature map and each target reference template feature map.
[0083] Continuing with the above 27 target reference feature maps as an example, the selected 27 target reference feature maps are respectively subjected to traversal matching search operations with the target area real-time feature map to obtain 27 correlation arrays.
[0084] The 27 target reference template feature maps L1~L 27 Each of the images in the template is matched with the real-time feature map D of the target area to obtain 27 correlation arrays R1~R 27 , the values in each correlation matrix are floating point numbers between -1.0 and 1.0.
[0085] When obtaining the correlation matrix, you can first obtain the feature vector corresponding to the real-time feature map of the target area and the feature vector of the target reference template feature map, and then calculate the covariance matrix based on the feature vectors of the two images. After obtaining the covariance matrix, calculate the correlation coefficient based on the covariance matrix, and then get the correlation matrix based on the correlation coefficient.
[0086] In step S106, the position in the target area real-time feature map corresponding to the maximum value in each correlation matrix is determined as the position of the target to be identified.
[0087] Continue with the above example and add 27 related arrays R1 to R 27 Find the maximum value in the search and record its matrix position, which is the position of the attack target in the real-time image. The target recognition and positioning calculation is completed.
[0088] For the same ground fixed target, the target images obtained by the forward imaging of the aircraft from different entry directions, different distances and different flight altitudes are different, and there may be great differences, which brings great difficulties to automatic target recognition. In the present disclosure, based on the existing three-dimensional scene model of the target area, firstly, according to the selected preset recognition target and the preset recognition direction, preset recognition distance and preset recognition altitude, a reference template of the preset recognition target is intercepted and prepared from the existing three-dimensional scene model of the target area; then, the image feature analysis and extraction of the reference template prepared above is performed to obtain a reference template feature library; finally, the real-time image of the target area is obtained by using the imaging device mounted on the aircraft, and the same image feature analysis and extraction are performed to obtain a real-time feature map of the target area, and the obtained real-time feature map of the target area is matched with the reference template feature library, so that reliable and accurate recognition of the recognition target can be achieved.
[0089] Based on the same inventive concept, the embodiment of the present application also provides a ground fixed target identification device for implementing the ground fixed target identification method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more ground fixed target identification devices provided below can refer to the limitations of the ground fixed target identification method above, and will not be repeated here.
[0090] In an exemplary embodiment, Figure 6 As shown, a device for identifying a ground fixed target is provided, and the device for identifying a ground fixed target comprises:
[0091] The first acquisition module 11 is used to acquire the position information of the aircraft at the time of identification and the real-time image of the target area where the target to be identified is located;
[0092] A calculation module 12 is used to calculate target recognition information of the aircraft relative to the target to be recognized based on the position information at the recognition time and the position information of the target to be recognized in the target area, wherein the target recognition information includes: the direction of the aircraft relative to the target to be recognized, the distance between the aircraft and the target to be recognized, and the flight altitude of the aircraft;
[0093] The second acquisition module 13 is used to acquire a matching target reference template feature map from a reference template feature map library according to the target recognition information; the reference template feature library stores the corresponding relationship between different recognition information of each target to be recognized in the target area and the reference template feature map;
[0094] An extraction module 14 is used to extract image features from the real-time image to obtain a real-time feature map of the target area;
[0095] The third acquisition module 15 is used to acquire the correlation matrix between the target area real-time feature map and the target reference template feature map;
[0096] The determination module 16 is used to determine that the position in the target area real-time feature map corresponding to the maximum value in each of the correlation arrays is the position of the target to be identified.
[0097] As an optional implementation, the device further includes:
[0098] A fourth acquisition module is used to acquire, in the three-dimensional scene model, a two-dimensional slice corresponding to each preset recognition target in the target area according to preset recognition information as a basic template image; the preset recognition information includes: a preset recognition direction, a preset recognition distance and a preset recognition height;
[0099] A fifth acquisition module, configured to perform spatial position transformation on the basic template image to obtain reference template images corresponding to the basic template image at different spatial positions;
[0100] The sixth acquisition module is used to extract image features from each of the reference template images to obtain the reference template feature library.
[0101] As an optional implementation manner, the fifth acquisition module is specifically configured to:
[0102] Keeping the second spatial information and the third spatial information unchanged, changing the first spatial information corresponding to the basic template image according to a first preset step length, and acquiring a plurality of first intermediate template images corresponding to the basic template image;
[0103] Keeping the first spatial information and the third spatial information unchanged, changing the second spatial information corresponding to each of the first intermediate template images according to a second preset step length, and acquiring a plurality of second intermediate template images corresponding to each of the first intermediate template images;
[0104] Keeping the first spatial information and the second spatial information unchanged, changing the third spatial information corresponding to each of the second intermediate template images according to a third preset step length, and acquiring a plurality of third intermediate template images corresponding to each of the second intermediate template images;
[0105] The third intermediate template image constitutes the reference template image; the first spatial information, the second spatial information and the third spatial information correspond one-to-one to the preset recognition direction, the preset recognition distance and the preset recognition height.
[0106] As an optional implementation manner, the sixth acquisition module is specifically configured to:
[0107] For each first pixel in each of the reference template images, the following steps are performed:
[0108] In a window of a preset size around each of the first pixels of the current reference template image, the grayscale value of each of the first pixels in the window is compared with the grayscale value of the first central pixel of the window, and a result matrix of each window is obtained according to the comparison result; wherein, if the grayscale value of the first pixel is greater than the grayscale value of the first central pixel, it is marked as a first value, and if the grayscale value of the first pixel is not greater than the grayscale value of the first central pixel, it is marked as a second value; and overlapping pixels in adjacent windows are ORed to obtain the reference template feature map;
[0109] The reference template feature graphs corresponding to all the reference template images constitute the reference template feature graph library.
[0110] As an optional implementation manner, the third acquisition module is specifically used to:
[0111] According to the target recognition information, a plurality of target reference template feature maps that meet preset requirements with the target recognition information are obtained from the reference template feature map library.
[0112] As an optional implementation manner, the extraction module is specifically used to:
[0113] In a window of a preset size around each second pixel of the real-time image, the grayscale value of each second pixel in the window is compared with the grayscale value of the second center pixel of the window, and a result matrix of each window is obtained according to the comparison result; wherein, if the grayscale value of the second pixel is greater than the grayscale value of the second center pixel, it is marked as a first value, and if the grayscale value of the second pixel is not greater than the grayscale value of the second center pixel, it is marked as a second value;
[0114] The overlapping pixels in adjacent windows are ORed to obtain the real-time feature map of the target area.
[0115] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for identifying a ground fixed target is implemented.
[0116] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0117] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.
[0118] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0119] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0120] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0121] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0122] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.
[0123] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0124] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A method for identifying a ground fixed target, characterized in that: The method for identifying the ground fixed target comprises: Obtain the position information of the aircraft at the time of identification and the real-time image of the target area where the target to be identified is located; Calculating target recognition information of the aircraft relative to the target to be recognized based on the position information at the recognition time and the position information of the target to be recognized in the target area, wherein the target recognition information includes: the direction of the aircraft relative to the target to be recognized, the distance between the aircraft and the target to be recognized, and the flight altitude of the aircraft; Obtaining a matching target reference template feature map from a reference template feature map library according to the target recognition information; the reference template feature library stores the corresponding relationship between different recognition information of each target to be recognized in the target area and the reference template feature map; Performing image feature extraction on the real-time image to obtain a real-time feature map of the target area; Acquire a correlation matrix between the target area real-time feature map and the target reference template feature map; The position in the target area real-time feature map corresponding to the maximum value in each of the correlation arrays is determined as the position of the target to be identified.
2. The method for identifying a ground fixed target according to claim 1, characterized in that: The method for identifying the ground fixed target also includes: In the three-dimensional scene model, a two-dimensional slice corresponding to each preset recognition target in the target area is obtained as a basic template image according to preset recognition information; the preset recognition information includes: a preset recognition direction, a preset recognition distance and a preset recognition height; Transforming the basic template image in spatial position to obtain reference template images corresponding to the basic template image in different spatial positions; Image features are extracted for each of the reference template images to obtain the reference template feature library.
3. The method for identifying a ground fixed target according to claim 2, characterized in that: The spatial position transformation of the basic template image to obtain a reference template image corresponding to the basic template image at a different spatial position includes: Keeping the second spatial information and the third spatial information unchanged, changing the first spatial information corresponding to the basic template image according to a first preset step length, and acquiring a plurality of first intermediate template images corresponding to the basic template image; Keeping the first spatial information and the third spatial information unchanged, changing the second spatial information corresponding to each of the first intermediate template images according to a second preset step length, and acquiring a plurality of second intermediate template images corresponding to each of the first intermediate template images; Keeping the first spatial information and the second spatial information unchanged, changing the third spatial information corresponding to each of the second intermediate template images according to a third preset step length, and acquiring a plurality of third intermediate template images corresponding to each of the second intermediate template images; The third intermediate template image constitutes the reference template image; the first spatial information, the second spatial information and the third spatial information correspond one-to-one to the preset recognition direction, the preset recognition distance and the preset recognition height.
4. The method for identifying a ground fixed target according to claim 2, characterized in that: The step of extracting image features from each of the reference template images to obtain the reference template feature library comprises: For each first pixel in each of the reference template images, the following steps are performed: In a window of a preset size around each of the first pixels of the current reference template image, the grayscale value of each of the first pixels in the window is compared with the grayscale value of the first central pixel of the window, and a result matrix of each window is obtained according to the comparison result; wherein, if the grayscale value of the first pixel is greater than the grayscale value of the first central pixel, it is marked as a first value, and if the grayscale value of the first pixel is not greater than the grayscale value of the first central pixel, it is marked as a second value; and overlapping pixels in adjacent windows are ORed to obtain the reference template feature map; The reference template feature graphs corresponding to all the reference template images constitute the reference template feature graph library.
5. The method for identifying a ground fixed target according to claim 1, characterized in that: The step of acquiring a matching target reference template feature map from a reference template feature map library according to the target recognition information comprises: According to the target recognition information, a plurality of target reference template feature maps that meet preset requirements with the target recognition information are obtained from the reference template feature map library.
6. The method for identifying a ground fixed target according to claim 1, characterized in that: The step of extracting image features from the real-time image to obtain a real-time feature map of the target area includes: In a window of a preset size around each second pixel of the real-time image, the grayscale value of each second pixel in the window is compared with the grayscale value of the second center pixel of the window, and a result matrix of each window is obtained according to the comparison result; wherein, if the grayscale value of the second pixel is greater than the grayscale value of the second center pixel, it is marked as a first value, and if the grayscale value of the second pixel is not greater than the grayscale value of the second center pixel, it is marked as a second value; The overlapping pixels in adjacent windows are ORed to obtain the real-time feature map of the target area.
7. A ground fixed target identification device, characterized in that: The identification device of the ground fixed target comprises: The first acquisition module is used to acquire the position information of the aircraft at the time of identification and the real-time image of the target area where the target to be identified is located; a calculation module, configured to calculate target recognition information of the aircraft relative to the target to be recognized based on the position information at the recognition time and the position information of the target to be recognized in the target area, wherein the target recognition information includes: the direction of the aircraft relative to the target to be recognized, the distance between the aircraft and the target to be recognized, and the flight altitude of the aircraft; A second acquisition module is used to acquire a matching target reference template feature map from a reference template feature map library according to the target recognition information; the reference template feature library stores the corresponding relationship between different recognition information of each target to be recognized in the target area and the reference template feature map; An extraction module, used to extract image features from the real-time image to obtain a real-time feature map of the target area; A third acquisition module is used to acquire a correlation matrix between the target area real-time feature map and the target reference template feature map; The determination module is used to determine that the position in the target area real-time feature map corresponding to the maximum value in each of the correlation arrays is the position where the target to be identified is located.
8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for identifying a ground fixed target according to any one of claims 1 to 6.
9. A 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 method for identifying a ground fixed target described in any one of claims 1 to 6 are implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for identifying a ground fixed target described in any one of claims 1 to 6 are implemented.