Target position determination method, device and electronic equipment

Abstract

The present invention provides a target position determination method, device, and electronic device, relating to the field of image data processing technology. The method comprises: determining a prediction frame for each of a plurality of first missile launchers based on satellite image data; constructing at least one missile launcher formation based on the prediction frames of each of the plurality of first missile launchers, each missile launcher formation including the prediction frames of each of a plurality of second missile launchers, each of the plurality of second missile launchers belonging to the plurality of first missile launchers; determining, for each missile launcher formation, a formation configuration frame corresponding to the missile launcher formation based on the prediction frames of the plurality of second missile launchers; and determining the formation configuration frame as the target position if the formation configuration frame does not match a preset position library. This method enables target position determination using missile launchers, effectively improving the accuracy of detecting target positions without fortifications. Furthermore, the entire process is insensitive to background changes and has high universality.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a target position determination method, device and electronic equipment. Background Art

[0002] Due to the diverse equipment, complex shapes, and strong protective features of military facilities, detection and identification of these highly modular facilities is more challenging than for individual targets. Permanent positions typically have distinct features and are easy to locate. However, in wartime, missile launchers are mobile, and these permanent positions can serve as field or temporary positions, making their location difficult to determine.

[0003] At present, a missile position direct search method is used for permanent positions. First, a missile position detection algorithm model is used to detect large-area satellite image data. Then, a missile position segmentation model is used to extract the internal structure of the position to assist the missile position recognition model in identifying specific types of missile positions, thereby realizing the identification of permanent positions.

[0004] There is relatively little research on the discovery of field positions such as temporary launch sites during wartime. If the above-mentioned direct search method of missile positions for permanent positions is used to identify field positions, then due to the high mobility of field positions, random locations, small number of real samples, and the lack of fortifications found in permanent missile positions, the accuracy of the ultimately identified field positions will be low. Summary of the Invention

[0005] The present invention provides a target position determination method, device and electronic equipment to overcome the defects in the prior art of low accuracy in the final identification of field positions due to the high mobility of field positions, random locations, few real samples and the lack of fortifications in permanent missile positions. The method determines the target position by using a missile launch vehicle, which can effectively improve the accuracy of detecting target positions without fortifications. In addition, the entire process is insensitive to background changes and has high universality.

[0006] The present invention provides a target position determination method, comprising:

[0007] Determine prediction frames of respective first missile launch vehicles based on satellite image data;

[0008] constructing at least one missile launcher formation based on the respective prediction frames of the plurality of first missile launchers, each missile launcher formation including respective prediction frames of a plurality of second missile launchers, the plurality of second missile launchers belonging to the plurality of first missile launchers;

[0009] For each missile launcher formation, a formation configuration frame corresponding to the missile launcher formation is determined based on the respective prediction frames of the multiple second missile launchers; when the formation configuration frame does not match the preset position library, the formation configuration frame is determined as the target position.

[0010] According to a target position determination method provided by the present invention, the formation configuration frame corresponding to the missile launch vehicle formation is determined based on the prediction frames of each of the multiple second missile launch vehicles, including: determining the position area corresponding to the missile launch vehicle formation based on the prediction frames of each of the multiple second missile launch vehicles; obtaining a first area feature of the position area and a second area feature of the multiple missile launch vehicles; and determining the formation configuration frame based on the first area feature and the second area feature.

[0011] According to a target position determination method provided by the present invention, determining a formation configuration frame based on the first area features and the second area features includes: fusing the first area features and the second area features to obtain a fused feature; preprocessing the fused feature to obtain the formation configuration frame; wherein the preprocessing includes at least one of the following: normalization processing, convolution processing, and activation processing.

[0012] According to a target position determination method provided by the present invention, the first area feature and the second area feature are fused to obtain a fused feature; and the fused feature is pre-processed to obtain the formation configuration frame, comprising: determining the formation configuration frame according to a configuration formula; wherein the configuration formula is: F zd represents the formation configuration frame; X represents the second area feature; F pos (X) represents the first regional feature; Indicates that the first regional feature F pos (X) and the second regional feature X are fused; f(·) represents the preprocessing.

[0013] According to a target position determination method provided by the present invention, determining a formation configuration frame based on the first area characteristics and the second area characteristics includes: determining an initial formation configuration frame based on the first area characteristics and the second area characteristics; rotating and / or standardizing the initial formation configuration frame to obtain the formation configuration frame.

[0014] According to a target position determination method provided by the present invention, constructing at least one missile launcher formation based on the respective prediction frames of the multiple first missile launchers includes: determining the position information of each of the multiple first missile launchers, the number of the respective prediction frames of the multiple first missile launchers, the length and area of all the prediction frames, and the spacing between any two prediction frames; clustering the multiple first missile launchers according to all position information, all numbers, all lengths, all areas and all spacings to construct the at least one missile launcher formation.

[0015] According to a target position determination method provided by the present invention, the preset position library includes at least one preset position configuration frame, and when the formation configuration frame does not match the preset position library, the formation configuration frame is determined as the target position, including: determining the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame for each preset position configuration frame; when all intersection-and-union ratios are less than a preset threshold, determining that the formation configuration frame does not match the preset position library, and determining the formation configuration frame as the target position.

[0016] According to a target position determination method provided by the present invention, determining the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame includes: determining the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame according to an intersection-and-union ratio formula; wherein the intersection-and-union ratio formula is:

[0017] IOU=TruePositive / (TruePositive+FalsePositive+FalseNegative); IOU represents the intersection-over-union ratio; TruePositive represents the part of the formation configuration frame that matches the preset position configuration frame; FalsePositive represents the part of the formation configuration frame that should match the preset position configuration frame but does not match; FalseNegative represents the part of the preset position configuration frame that does not exist but the formation configuration frame exists.

[0018] The present invention also provides a target position determination device, comprising:

[0019] an acquisition module, configured to determine prediction frames of respective first missile launch vehicles based on satellite image data;

[0020] a launch vehicle formation determining module, configured to construct at least one missile launch vehicle formation based on the respective prediction frames of the plurality of first missile launch vehicles, each missile launch vehicle formation including respective prediction frames of a plurality of second missile launch vehicles, the plurality of second missile launch vehicles belonging to the plurality of first missile launch vehicles;

[0021] The target position determination module is used to determine the formation configuration frame corresponding to each missile launcher formation according to the respective prediction frames of the multiple second missile launchers; when the formation configuration frame does not match the preset position library, the formation configuration frame is determined as the target position.

[0022] 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 target position determination method described above is implemented.

[0023] 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 any of the target position determination methods described above.

[0024] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the target position determination methods described above.

[0025] The target position determination method, device, and electronic equipment provided by the present invention determine prediction frames for multiple first missile launchers based on satellite image data; construct at least one missile launcher formation based on the prediction frames of the multiple first missile launchers, each missile launcher formation including prediction frames for multiple second missile launchers, each of which belongs to the multiple first missile launchers; and determine, for each missile launcher formation, a formation configuration frame corresponding to the missile launcher formation based on the prediction frames of the multiple second missile launchers; and if the formation configuration frame does not match a preset position library, the formation configuration frame is determined as the target position. This method determines target positions using missile launchers, effectively improving the accuracy of detecting target positions without fortifications. Furthermore, the entire process is insensitive to background changes and has high universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to 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 any creative work.

[0027] Figure 1 It is a flow chart of the target position determination method provided by the present invention;

[0028] Figure 2 Schematic diagram of a scenario of the target position determination method provided by the present invention;

[0029] Figure 3 It is a structural schematic diagram of the target position determination device provided by the present invention;

[0030] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0031] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0032] It should be noted that the execution subject involved in the embodiment of the present invention can be a target position determination device or an electronic device. Optionally, the electronic device can include: a computer, a mobile terminal, a wearable device, etc.

[0033] The embodiment of the present invention will be further described below by taking an electronic device as an example.

[0034] like Figure 1 FIG. 1 is a flow chart of a target position determination method provided by the present invention, which may include:

[0035] 101. Determine prediction frames for each of the plurality of first missile launch vehicles based on satellite image data.

[0036] Among them, satellite image data refers to data with high-precision geographic coordinate information.

[0037] A missile launcher refers to a device that launches missiles.

[0038] It should be noted that there are multiple first missile launchers in the satellite image data, and the i-th missile launcher among these multiple first missile launchers can be used D i Indicates that the i-th missile launcher D i The prediction box can be used i express.

[0039] Optionally, the shape of the prediction box may include: rectangle, circle, diamond, etc. Generally, the electronic device will select a rectangular prediction box.

[0040] Optionally, the prediction box sizes of different first missile launchers may be the same or different, which is not specifically limited here.

[0041] Satellites can be equipped with various sensors to obtain initial data that can comprehensively, truly and objectively reflect the surface characteristics of the target area; then, the satellite uses remote sensing technology to process the initial data to obtain satellite image data, and then sends the satellite image data to electronic equipment associated with the satellite.

[0042] After receiving the satellite image data sent by the satellite, the electronic equipment can use the missile launcher detection algorithm model to extract the target from the satellite image data, obtain multiple first missile launchers and their respective prediction frames, and improve data support for the subsequent construction of the missile launcher formation.

[0043] Optionally, the missile launcher detection algorithm model may include: Faster Region-based Convolutional Neural Network (Faster RCNN), Single Shot MultiBox Detector (SDD), and Detection Transformer (DETR), etc.

[0044] Optionally, for the i-th missile launcher D i The prediction box can be used i ={x i ,y i , w i , h i} indicates that the predicted box Q i The corresponding rectangle can be used i ={x i1 ,y i1 , x i2 ,y i2 , x i3 ,y i3 , x i4 ,y i4} represents. Among them, {x i ,y i} represents the predicted box Q i The center point of {w i} represents the predicted box Q i The width of {h i} represents the predicted box Q i The height of , at this time, the prediction box Q i The calculation process of the corresponding rectangle is as follows: The coordinates of the first corner (such as the upper left corner): x i1 =(x i -y i ) / 2,y i1 =(y i -hi ) / 2; coordinates of the second corner (such as the upper right corner): x i2 =(x i +w i ) / 2,y i2 =(y i -h i ) / 2; coordinates of the third corner (such as the lower left corner): x i3 =(x i +w i ) / 2,y i3 =(y i +h i ) / 2; coordinates of the fourth corner (such as the lower right corner): x i3 =(x i -y i ) / 2,y i3 =(y i +h i ) / 2.

[0045] Optionally, before step 101, the method may further include but is not limited to one of the following implementations:

[0046] Implementation method 1: The electronic device obtains a plurality of first satellite image data; the electronic device determines the satellite image data based on the plurality of first satellite image data.

[0047] After receiving multiple first satellite image data sent by the satellite, the electronic device does not need to extract the target for each first satellite image data in order to improve data processing efficiency. At this time, one satellite image data can be first determined from the multiple first satellite image data, and then the prediction frames of the multiple first missile launchers can be determined.

[0048] Optionally, the electronic device determines the satellite image data based on multiple first satellite image data, which may include: the electronic device determines any first satellite image data among the multiple first satellite image data as satellite image data; or, the electronic device obtains the clarity of each of the multiple first satellite image data, and determines the first satellite image data corresponding to the highest clarity as the satellite image data.

[0049] When determining a single satellite image from a plurality of first satellite image data, if the clarity of each first satellite image is not considered, the electronic device may randomly select a single first satellite image from the plurality of first satellite image data and determine the randomly selected first satellite image as the satellite image. If the clarity of each first satellite image is considered, the electronic device may first obtain the clarity of all first satellite image data, then perform pairwise comparisons of all the clarity values and determine the highest clarity, and then determine the first satellite image corresponding to the highest clarity as the satellite image. This effectively improves the efficiency of subsequent processing of the satellite image data.

[0050] Implementation method 2: The electronic device obtains a second satellite image data, where the number of the second satellite image data is one; the electronic device performs image processing on the second satellite image data to obtain satellite image data.

[0051] The image processing may include but is not limited to: erosion operation, dilation operation, opening operation and closing operation, etc. The opening operation refers to performing an erosion operation on the second satellite image data first and then a dilation operation, and the closing operation refers to performing a dilation operation on the second satellite image data first and then an erosion operation.

[0052] After receiving a second satellite image data sent by a satellite, the electronic device may perform image processing on the second satellite image data to improve the clarity of the second satellite image data, that is, obtain satellite image data with better image quality, and thus obtain a prediction frame with higher accuracy, because the second satellite image data may have a problem of low clarity, that is, the image quality of the second satellite image data is poor, which is not conducive to the subsequent determination of the prediction frame of the first missile launcher.

[0053] 102. Construct at least one missile launcher formation based on the respective prediction frames of the plurality of first missile launchers.

[0054] Among them, each missile launcher formation may include respective prediction frames of multiple second missile launchers, and the multiple second missile launchers belong to the multiple first missile launchers.

[0055] For any missile launcher formation, D={D1,…,D j ,…,D N} means that N≥2.

[0056] Optionally, the number of prediction boxes included in any missile launcher formation may be the same or different, and is not specifically limited here.

[0057] In some embodiments, the electronic device constructs at least one missile launcher formation based on the respective prediction frames of multiple first missile launchers, which may include: the electronic device determines the respective position information of the multiple first missile launchers, the number of the respective prediction frames of the multiple first missile launchers, the length and area of all the prediction frames, and the spacing between any two prediction frames; the electronic device clusters the multiple first missile launchers based on all the position information, all the numbers, all the lengths, all the areas and all the spacings to construct at least one missile launcher formation.

[0058] Since satellite image data is data with high-precision geographic coordinate information, the electronic device can determine the position information of multiple first missile launchers based on the satellite image data; then, the electronic device determines the number of all prediction frames, the length and area of all prediction frames, and the distance between any two prediction frames based on the prediction frames of the multiple first missile launchers; then, the electronic device uses a clustering method to merge the first missile launchers that may belong to the same position into a formation based on all the position information, quantity, lengths, areas and distances, combined with expert knowledge, to obtain a corresponding missile launcher formation.

[0059] Among them, expert knowledge can include the neighborhood distance threshold ε and the minimum number of neighboring points MinPts between two prediction boxes, etc., to provide assistance for the merging formation of multiple first missile launchers.

[0060] In the auxiliary process, if N ε (D β ) is greater than or equal to the minimum number of neighboring points MinPts, then D β Join the missile launcher formation D = {D1, ..., D j ,…,D N}, where N ε (D β ) indicates that D β The number of first missile launchers within the range of radius centered on ε and with the neighborhood distance threshold ε as the radius.

[0061] It should be noted that the electronic device obtains the position information of each of the multiple first missile launchers, the number of prediction frames of each of the multiple first missile launchers, the length and area of all the prediction frames, and the spacing between any two prediction frames at any time.

[0062] 103. For each missile launcher formation, determine a formation configuration frame corresponding to the missile launcher formation based on the respective prediction frames of the plurality of second missile launchers; if the formation configuration frame does not match the preset position library, determine the formation configuration frame as the target position.

[0063] The formation configuration frame may include the orientation angle and rectangular frame boundary information corresponding to the missile launcher formation.

[0064] The preset position library refers to the configuration template library of permanent positions, which can include a large number of preset position configuration frames for permanent battle positions.

[0065] The target position is a field position, which refers to a defensive position mainly composed of field fortifications.

[0066] Since each missile launcher formation corresponds to a formation configuration frame, the electronic device can first determine the corresponding formation configuration frame for each missile launcher formation based on the predicted frames of the multiple second missile launchers in the formation. Since the preset position library includes a large number of preset position configuration frames for permanent battle sites, to reduce data interference from permanent battle sites, the electronic device can match the formation configuration frame with the preset position library. If the formation configuration frame does not match the preset position library, it is determined that the preset position configuration frame is not a permanent battle site and the formation configuration frame can be identified as the target position. This process enables the search and discovery of unfortified field positions, improving the accuracy of detecting unfortified target positions.

[0067] In some embodiments, the electronic device determines the formation configuration frame corresponding to the missile launcher vehicle formation based on the respective prediction frames of multiple second missile launchers, which may include: the electronic device determines the position area corresponding to the missile launcher vehicle formation based on the respective prediction frames of multiple second missile launchers; the electronic device obtains the first area feature of the position area and the second area feature of the multiple missile launchers; the electronic device determines the formation configuration frame based on the first area feature and the second area feature.

[0068] The position area refers to the position prediction outline determined by the respective prediction frames of multiple second missile launch vehicles.

[0069] Optionally, regional features may include: regional size, regional texture, and regional topography.

[0070] The electronic device can first determine the position area corresponding to the missile launch vehicle formation based on the respective prediction frames of multiple second missile launch vehicles. The shape of the position area is not limited and is generally a matrix; then, the electronic device obtains the first area feature of the position area and obtains the second area feature of the area where the multiple missile launch vehicles are located; then, the electronic device combines the first area feature and the second area feature to determine the formation configuration frame corresponding to the missile launch vehicle formation.

[0071] It should be noted that the timing of the electronic device acquiring the first area feature and the electronic device acquiring the second area feature is not limited.

[0072] In some embodiments, the electronic device determines the formation configuration frame based on the first area feature and the second area feature, which may include but is not limited to at least one of the following implementations:

[0073] Implementation method 1: The electronic device fuses the first area feature and the second area feature to obtain a fused feature; the electronic device pre-processes the fused feature to obtain a formation configuration frame.

[0074] The preprocessing includes at least one of the following: normalization processing, convolution processing and activation processing.

[0075] The electronic device can use an attention configuration recognition network to fuse the first area features and the second area features to obtain a fused feature, and then pre-process the fused feature to obtain a formation configuration frame with higher accuracy.

[0076] Among them, the attention configuration recognition network refers to a configuration recognition network that adopts the attention mechanism. Based on the first region features and the second region features of the input, it can accurately identify the fusion features corresponding to the missile launcher formation.

[0077] In some embodiments, the electronic device fuses the first area feature and the second area feature to obtain a fused feature; preprocesses the fused feature to obtain a formation configuration frame, which may include: the electronic device determines the formation configuration frame according to a configuration formula.

[0078] Among them, the configuration formula is:

[0079] F zd Indicates the formation configuration frame; X indicates the second area feature; F pos (X) indicates the first region feature; Indicates that the first regional feature F pos (X) and the second region feature X are fused; f(·) represents preprocessing.

[0080] Based on the above configuration formula, the electronic equipment can accurately determine the formation configuration frame corresponding to any missile launcher formation.

[0081] Implementation method 2: The electronic device determines an initial formation configuration frame according to the first area feature and the second area feature; the electronic device performs rotation processing and / or normalization processing on the initial formation configuration frame to obtain a formation configuration frame.

[0082] After the electronic device determines the initial formation configuration frame based on the first area characteristics and the second area characteristics, since the initial formation configuration frame does not match the preset configuration frame and is inconvenient for subsequent data processing, the electronic device may rotate and / or standardize the initial formation configuration frame to obtain a formation configuration frame, which matches the preset configuration frame.

[0083] The preset configuration frame can constrain the direction and size of the initial formation configuration frame.

[0084] Optionally, during the process of rotating the initial formation configuration frame, the electronic device may mark the vertex coordinates of the initial formation configuration frame with a rotating rectangular frame in a clockwise or counterclockwise direction to obtain a plurality of marking information; then, the electronic device processes the plurality of marking information to obtain a mask image of the initial formation configuration frame; then, the electronic device marks each pixel in the mask image, and G={d t , d r , d b , d l , θ} indicates that the rectangular box rotation angle information and the rectangular box boundary information are obtained, and then the formation configuration frame is obtained, where θ∈[0,π).

[0085] Among them, d t , d r , d b , d l They represent the distances from the target pixel to the four sides of the rotated rectangular frame, and θ represents the rotation angle information of the rectangular frame, that is, the target orientation, which can be defined as the angle between the center line of the second missile launcher and the horizontal direction of the image.

[0086] Optionally, during the process of standardizing the initial formation configuration frame, the electronic device determines the size of the initial formation configuration frame and the size of the preset configuration frame. If they are different, it is necessary to standardize the size of the initial formation configuration frame to obtain a size that is the same as the size of the preset configuration frame, which facilitates subsequent processing of the formation configuration frame.

[0087] In some embodiments, the preset position library may include at least one preset position configuration frame. When the formation configuration frame does not match the preset position library, the electronic device determines the formation configuration frame as the target position. This may include: the electronic device determines the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame for each preset position configuration frame; when all intersection-and-union ratios are less than a preset threshold, the electronic device determines that the formation configuration frame does not match the preset position library, and determines the formation configuration frame as the target position.

[0088] The preset threshold value may be set before the electronic device leaves the factory or may be user-defined, and is not specifically limited here.

[0089] For each preset position configuration frame, the electronic device can first determine the similarity between the formation configuration frame and the preset position configuration frame, that is, determine the intersection-and-union ratio between the configuration frames. In this way, the electronic device will obtain as many intersection-and-union ratios as there are preset position configuration frames; then, the electronic device will compare the multiple intersection-and-union ratios obtained with the preset threshold one by one, and when all the intersection-and-union ratios are less than the preset threshold, it is determined that the formation configuration frame does not match the preset position library, that is, the formation configuration frame is not a permanent position. At this time, the electronic device can determine the formation configuration frame as the target position, that is, as a field position. Since the data interference of the permanent position is eliminated, the accuracy of the target position determined by the electronic device is higher.

[0090] In some embodiments, the electronic device determines the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame, which may include: the electronic device determines the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame according to the intersection-and-union ratio formula.

[0091] The intersection-over-union ratio formula is:

[0092] IOU=TruePositive / (TruePositive+FalsePositive+FalseNegative);

[0093] IOU stands for intersection-over-union; TruePositive represents the part where the formation configuration frame matches the preset position configuration frame; FalsePositive represents the part where the formation configuration frame should match the preset position configuration frame but does not match; FalseNegative represents the part where the preset position configuration frame does not exist but the formation configuration frame exists.

[0094] Based on the above-mentioned intersection-and-combination ratio formula, the electronic equipment can accurately determine the intersection-and-combination ratio between the formation configuration frame and any preset position configuration frame in the preset position library.

[0095] In summary, for example, Figure 2 The figure is a scene diagram of the target position determination method provided by the present invention. Figure 2As can be seen in the figure, the electronic device first acquires satellite image data. Then, based on this satellite image data, it determines the prediction boxes for each of the four first missile launchers. These four first missile launchers are: first missile launcher D1, first missile launcher D2, first missile launcher D3, and first missile launcher D4. The prediction box for first missile launcher D1 is Q1, the prediction box for first missile launcher D2 is Q2, the prediction box for first missile launcher D3 is Q3, and the prediction box for first missile launcher D4 is Q4.

[0096] Then, the electronic device constructs a missile launch vehicle formation and the position area corresponding to the missile launch vehicle formation based on the four prediction frames, and then determines the initial formation configuration frame.

[0097] Next, the electronic device performs rotation processing and / or standardization processing on the initial formation configuration frame to obtain a formation configuration frame.

[0098] Finally, the electronic device matches the formation configuration frame with a preset configuration frame, and in the event of mismatch, determines the formation configuration frame as a target position.

[0099] In addition, based on Figure 2 ,The electronic device extracts the contour of the target position based on the contour,and compares it with the previously marked reference contour, and the,accuracy of the extraction of the contour of the target position reaches 95.1%.

[0100] It should be noted that, since the missile launcher is moving in real time, the entire process is based on the movement of the missile launcher, and the target position can also be determined in real time, and the accuracy of the target position is also relatively high.

[0101] In this embodiment of the present invention, prediction frames are determined for each of multiple first missile launchers based on satellite imagery data. Based on the prediction frames of each of the multiple first missile launchers, at least one missile launcher formation is constructed. For each missile launcher formation, a corresponding formation configuration frame is determined based on the prediction frames of each of the multiple second missile launchers. If the formation configuration frame does not match a preset position library, the formation configuration frame is determined as the target position. This method, which uses missile launchers to determine target positions, can effectively improve the accuracy of detecting unfortified target positions. Furthermore, the entire process is insensitive to background changes and has high applicability.

[0102] The target position determination device provided by the present invention is described below. The target position determination device described below and the target position determination method described above can be referenced to each other.

[0103] like Figure 3 FIG. 1 is a schematic diagram of the structure of the target position determination device provided by the present invention, which may include:

[0104] An acquisition module 301 is configured to determine prediction frames of respective first missile launch vehicles based on satellite image data;

[0105] A launch vehicle formation determining module 302 is configured to construct at least one missile launch vehicle formation based on the respective prediction frames of the plurality of first missile launch vehicles, wherein each missile launch vehicle formation includes respective prediction frames of a plurality of second missile launch vehicles, the plurality of second missile launch vehicles belonging to the plurality of first missile launch vehicles;

[0106] The target position determination module 303 is used to determine the formation configuration frame corresponding to each missile launcher formation based on the respective prediction frames of the multiple second missile launchers; if the formation configuration frame does not match the preset position library, the formation configuration frame is determined as the target position.

[0107] Optionally, the target position determination module 303 is specifically used to determine the position area corresponding to the missile launch vehicle formation based on the respective prediction frames of the multiple second missile launch vehicles; obtain the first area characteristics of the position area and the second area characteristics of the multiple missile launch vehicles; and determine the formation configuration frame based on the first area characteristics and the second area characteristics.

[0108] Optionally, the target position determination module 303 is specifically used to fuse the first area feature and the second area feature to obtain a fused feature; and preprocess the fused feature to obtain the formation configuration frame; wherein the preprocessing includes at least one of the following: normalization processing, convolution processing, and activation processing.

[0109] Optionally, the target position determination module 303 is specifically configured to determine the formation configuration frame according to a configuration formula; wherein the configuration formula is: F zd represents the formation configuration frame; X represents the second area feature; F pos (X) represents the first region feature; Indicates that the first region feature F pos (X) and the second regional feature X are fused; f(·) represents the preprocessing.

[0110] Optionally, the target position determination module 303 is specifically configured to determine an initial formation configuration frame based on the first area feature and the second area feature; and perform rotation processing and / or standardization processing on the initial formation configuration frame to obtain the formation configuration frame.

[0111] Optionally, the launch vehicle formation determination module 302 is specifically used to determine the position information of each of the multiple first missile launch vehicles, the number of prediction boxes of each of the multiple first missile launch vehicles, the length and area of all prediction boxes, and the distance between any two prediction boxes; based on all the position information, all the numbers, all the lengths, all the areas and all the distances, the multiple first missile launch vehicles are clustered to construct at least one missile launch vehicle formation.

[0112] Optionally, the preset position library includes at least one preset position configuration frame, and the target position determination module 303 is specifically used to determine the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame for each preset position configuration frame; when all intersection-and-union ratios are less than a preset threshold, it is determined that the formation configuration frame does not match the preset position library, and the formation configuration frame is determined to be the target position.

[0113] Optionally, the target position determination module 303 is specifically used to determine the intersection-over-union ratio between the formation configuration frame and the preset position configuration frame according to the intersection-over-union ratio formula; wherein the intersection-over-union ratio formula is: IOU=TruePositive / (TruePositive+FalsePositive+FalseNegative); IOU represents the intersection-over-union ratio; TruePositive represents the part of the formation configuration frame that matches the preset position configuration frame; FalsePositive represents the part of the formation configuration frame that should match the preset position configuration frame but does not match; FalseNegative represents the part of the preset position configuration frame that does not exist but the formation configuration frame exists.

[0114] like Figure 4FIG. 4 is a schematic diagram of the structure of an electronic device provided by the present invention. The electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440. The processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 may call logic instructions in the memory 430 to execute a target position determination method, which includes: determining prediction frames of multiple first missile launchers based on satellite image data; constructing at least one missile launcher formation based on the prediction frames of the multiple first missile launchers, each missile launcher formation including prediction frames of multiple second missile launchers, the multiple second missile launchers belonging to the multiple first missile launchers; for each missile launcher formation, determining a formation configuration frame corresponding to the missile launcher formation based on the prediction frames of the multiple second missile launchers; and determining the formation configuration frame as the target position if the formation configuration frame does not match a preset position library.

[0115] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, 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. The computer software product is stored in a storage medium and includes several instructions for enabling 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: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0116] 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 target position determination method provided by the above-mentioned methods, the method including: determining the prediction frames of multiple first missile launchers based on satellite image data; constructing at least one missile launcher formation based on the prediction frames of the multiple first missile launchers, each missile launcher formation including the prediction frames of multiple second missile launchers, and the multiple second missile launchers belong to the multiple first missile launchers; for each missile launcher formation, determining the formation configuration frame corresponding to the missile launcher formation based on the prediction frames of the multiple second missile launchers; when the formation configuration frame does not match the preset position library, determining the formation configuration frame as the target position.

[0117] 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 target position determination method provided by the above-mentioned methods, the method comprising: determining the prediction frames of multiple first missile launchers based on satellite image data; constructing at least one missile launcher formation based on the prediction frames of the multiple first missile launchers, each missile launcher formation including the prediction frames of multiple second missile launchers, the multiple second missile launchers belonging to the multiple first missile launchers; for each missile launcher formation, determining the formation configuration frame corresponding to the missile launcher formation based on the prediction frames of the multiple second missile launchers; and determining the formation configuration frame as the target position when the formation configuration frame does not match the preset position library.

[0118] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0119] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling 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 certain parts of the embodiments.

[0120] 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 various embodiments of the present invention.

Claims

1. A method for determining a target position, characterized in that: include: Determine prediction frames of respective first missile launch vehicles based on satellite image data; constructing at least one missile launcher formation based on the respective prediction frames of the plurality of first missile launchers, each missile launcher formation including respective prediction frames of a plurality of second missile launchers, the plurality of second missile launchers belonging to the plurality of first missile launchers; For each missile launcher formation, a formation configuration frame corresponding to the missile launcher formation is determined based on the respective prediction frames of the multiple second missile launchers; when the formation configuration frame does not match the preset position library, the formation configuration frame is determined as the target position.

2. The method according to claim 1, characterized in that Determining a formation configuration frame corresponding to the missile launcher formation based on the respective prediction frames of the plurality of second missile launchers includes: determining a position area corresponding to the missile launch vehicle formation according to the respective prediction frames of the plurality of second missile launch vehicles; Acquire a first area feature of the position area and a second area feature of the plurality of missile launch vehicles; A formation configuration frame is determined according to the first area characteristics and the second area characteristics.

3. The method according to claim 2, characterized in that The determining of the formation configuration frame according to the first area feature and the second area feature includes: Fusing the first region feature and the second region feature to obtain a fused feature; Preprocessing the fusion features to obtain the formation configuration frame; The preprocessing includes at least one of the following: normalization processing, convolution processing and activation processing.

4. The method according to claim 3, characterized in that fusing the first region feature and the second region feature to obtain a fused feature; Preprocessing the fusion features to obtain the formation configuration frame includes: Determining the formation configuration frame according to the configuration formula; Among them, the configuration formula is: F zd =f(X⊕F pos (X)); F zd represents the formation configuration frame; X represents the second area feature; F pos (X) represents the first regional feature; X⊕F pos (X) represents the first regional feature F pos (X) and the second regional feature X are fused; f(·) represents the preprocessing.

5. The method according to any one of claims 2 to 4, characterized in that: The determining of the formation configuration frame according to the first area feature and the second area feature includes: determining an initial formation configuration frame according to the first area characteristics and the second area characteristics; The initial formation configuration frame is subjected to rotation processing and / or normalization processing to obtain the formation configuration frame.

6. The method according to any one of claims 1 to 4, characterized in that The step of constructing at least one missile launcher formation according to the respective prediction frames of the plurality of first missile launchers comprises: Determining the position information of each of the plurality of first missile launchers, the number of prediction frames of each of the plurality of first missile launchers, the length and area of each of the prediction frames, and the distance between any two prediction frames; The plurality of first missile launchers are clustered according to all position information, all quantities, all lengths, all areas and all spacings to construct the at least one missile launcher formation.

7. The method according to any one of claims 1 to 4, characterized in that The preset position library includes at least one preset position configuration frame, and when the formation configuration frame does not match the preset position library, determining the formation configuration frame as the target position includes: For each preset position configuration frame, determining an intersection-over-union ratio between the formation configuration frame and the preset position configuration frame; In the case that all intersection-over-union ratios are less than a preset threshold, it is determined that the formation configuration frame does not match the preset position library, and the formation configuration frame is determined as the target position.

8. The method according to claim 7, characterized in that Determining the intersection-over-union ratio between the formation configuration frame and the preset position configuration frame includes: Determining the intersection-and-union ratio between the formation configuration frame and the preset position configuration frame according to an intersection-and-union ratio formula; The intersection-over-union ratio formula is: IOU=TruePositive / (TruePositive+FalsePositive+FalseNegative); IOU represents the intersection-over-union ratio; TruePositive represents the part of the formation configuration frame that matches the preset position configuration frame; FalsePositive represents the part of the formation configuration frame that should match the preset position configuration frame but does not match; FalseNegative represents the part of the preset position configuration frame that does not exist but the formation configuration frame exists.

9. A target position determination device, characterized in that: include: an acquisition module, configured to determine prediction frames of respective first missile launch vehicles based on satellite image data; a launch vehicle formation determining module, configured to construct at least one missile launch vehicle formation based on the respective prediction frames of the plurality of first missile launch vehicles, each missile launch vehicle formation including respective prediction frames of a plurality of second missile launch vehicles, the plurality of second missile launch vehicles belonging to the plurality of first missile launch vehicles; The target position determination module is used to determine the formation configuration frame corresponding to each missile launcher formation according to the respective prediction frames of the multiple second missile launchers; when the formation configuration frame does not match the preset position library, the formation configuration frame is determined as the target position.

10. 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 target position determination method according to any one of claims 1 to 8 is implemented.

Images