Weak magnetic positioning method, device, equipment and computer readable storage medium

Abstract

Embodiments of the present application provide a weak magnetic positioning method, device, equipment and computer-readable storage medium. The method includes obtaining geomagnetic sequence information; inputting the geomagnetic sequence information into a trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information; wherein the weak magnetic positioning model can be trained in the following manner: obtaining a reference map of the original geomagnetic; performing feature enhancement on the reference map to obtain a target reference map; processing the target reference map by a preset algorithm to obtain a geomagnetic sequence fingerprint database; training a weak magnetic positioning model based on the geomagnetic sequence fingerprint database. In this way, effective positioning results can be achieved in a wide range of outdoor weak magnetic areas, reducing a lot of time and labor costs in data acquisition, and being able to generate more realistic trajectories while overcoming the influence of noise.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of geomagnetic positioning technology, and in particular to weak magnetic positioning methods, devices, equipment, and computer-readable storage media. Background Art

[0002] With the continuous development of social science and technology, people's demand for positioning has increased. Various positioning technologies have been widely used. Indoor positioning technologies include technologies based on wireless signals such as WIFI, Bluetooth, UWB, RFID, etc., and outdoor positioning technologies include Beidou, GPS, inertial navigation and other technologies. However, the above methods have disadvantages such as high cost, susceptibility to interference of signals, and cumulative errors. Since geomagnetic positioning has the advantages of not being affected by environmental occlusion, stable geomagnetic signals, and no need to deploy additional infrastructure, it has broad application prospects in scenarios such as outdoor positioning, indoor positioning, and underground space positioning.

[0003] Traditional geomagnetic positioning methods mainly rely on manual feature extraction and statistical model-based methods, but these methods are often sensitive to environmental changes, and positioning accuracy and robustness are difficult to guarantee. In recent years, with the rapid development of deep learning technology, geomagnetic positioning methods based on deep learning have gradually become a research hotspot. Deep learning has powerful feature extraction and pattern recognition capabilities, and can effectively process complex nonlinear data relationships, making it show great potential in geomagnetic positioning.

[0004] At present, geomagnetic positioning technology based on deep learning has achieved certain success in laboratory environments, but the following problems still exist in practical applications:

[0005] 1) In some special geographical environments, the geomagnetic signal shows weak magnetic characteristics, that is, the geomagnetic changes are slow, and the geomagnetic characteristic information is not rich enough, resulting in a decrease in geomagnetic positioning accuracy;

[0006] 2) The noise existing during the geomagnetic signal collection will affect the accuracy of geomagnetic positioning;

[0007] 3) The training of deep learning models relies on a large amount of labeled data, and obtaining large-scale, high-quality geomagnetic field labeled data often requires huge time and manpower costs. Summary of the invention

[0008] According to the embodiments of the present application, a weak magnetic positioning solution is provided, which can achieve effective positioning results in a wide range of weak magnetic areas, while reducing a lot of time and labor costs in data acquisition, and the generated trajectory is more realistic and can overcome the influence of noise.

[0009] In a first aspect of the present application, a weak magnetic positioning method is provided. The method comprises:

[0010] Obtain geomagnetic sequence information;

[0011] Inputting the geomagnetic sequence information into a trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information;

[0012] The weak magnetic positioning model can be trained in the following way:

[0013] Obtain a reference map of the original geomagnetic field;

[0014] Performing feature enhancement on the reference image to obtain a target reference image;

[0015] Processing the target reference map by a preset algorithm to obtain a geomagnetic sequence fingerprint database;

[0016] Based on the geomagnetic sequence fingerprint database, a weak magnetic positioning model is trained.

[0017] Furthermore, the obtaining of the original geomagnetic reference map includes:

[0018] Based on the preset collection path, obtain the geomagnetic information of each collection point;

[0019] Based on the pre-divided grid size, the geomagnetic information is resampled to construct a reference map of the original geomagnetism.

[0020] Furthermore, the geomagnetic information includes a total geomagnetic field strength value, a three-dimensional geomagnetic vector and corresponding geographic location coordinates.

[0021] Further, the step of performing feature enhancement on the reference image to obtain a target reference image includes:

[0022] Based on the preset acquisition path and the spatial features of the reference image, the reference image is feature enhanced by a convolution enhancement operator to obtain a target reference image.

[0023] Furthermore, the target reference map is processed by a preset algorithm to obtain a geomagnetic sequence fingerprint database, which includes:

[0024] Based on the target reference map, construct an index of acquisition points;

[0025] Based on the index, randomly select multiple reference points as key trajectory points to generate a trajectory point set;

[0026] Based on the trajectory point set, a geomagnetic sequence fingerprint database is constructed.

[0027] Furthermore, constructing a geomagnetic sequence fingerprint database based on the trajectory point set includes:

[0028] Select two adjacent trajectory points from the trajectory point set, take a first trajectory point with a forward index among the two adjacent trajectory points as a starting point, and a second trajectory point with a backward index as an end point, and generate a first trajectory path between the first trajectory point and the second trajectory point;

[0029] Selecting a third trajectory point adjacent to the second trajectory point from the trajectory point set, and generating a second trajectory path between the second trajectory point and the third trajectory point;

[0030] If the angle between the first trajectory path and the second trajectory path is greater than or equal to a preset threshold, the second trajectory path is saved; if the angle is less than the threshold, a fourth trajectory point adjacent to the third trajectory point is selected, a third trajectory path between the second trajectory point and the fourth trajectory point is randomly generated, and the angle between the first trajectory path and the third trajectory path is compared with the threshold; if the angle is greater than or equal to the preset threshold, the third trajectory path is saved;

[0031] Repeat the above steps until all key trajectory points in the trajectory point set are traversed;

[0032] Based on the saved trajectory path, a geomagnetic sequence fingerprint database is constructed.

[0033] Furthermore, the weak magnetic positioning model is a multi-scale neural network positioning model.

[0034] In a second aspect of the present application, a weak magnetic positioning device is provided. The device comprises:

[0035] An acquisition module, used for acquiring geomagnetic sequence information;

[0036] A positioning module, used for inputting the geomagnetic sequence information into a trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information;

[0037] The weak magnetic positioning model can be trained in the following way:

[0038] Obtain a reference map of the original geomagnetic field;

[0039] Performing feature enhancement on the reference image to obtain a target reference image;

[0040] Processing the target reference map by a preset algorithm to obtain a geomagnetic sequence fingerprint database;

[0041] Based on the geomagnetic sequence fingerprint database, a weak magnetic positioning model is trained.

[0042] In a third aspect of the present application, an electronic device is provided, which includes a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the program, the method described above is implemented.

[0043] In a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method according to the first aspect of the present application is implemented.

[0044] The weak magnetic positioning method provided in the embodiment of the present application obtains geomagnetic sequence information; inputs the geomagnetic sequence information into a trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information; wherein the weak magnetic positioning model can be trained in the following manner: obtain a reference map of the original geomagnetic; perform feature enhancement on the reference map to obtain a target reference map; process the target reference map through a preset algorithm to obtain a geomagnetic sequence fingerprint database; and train a weak magnetic positioning model based on the geomagnetic sequence fingerprint database, which can achieve effective positioning results in a wide range of outdoor weak magnetic areas, greatly improving positioning accuracy while reducing time and labor costs.

[0045] It should be understood that the contents described in the Summary of the Invention are not intended to limit the key or important features of the embodiments of the present application, nor are they intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0047] Figure 1 is a flow chart of a weak magnetic positioning method according to an embodiment of the present application;

[0048] Figure 2 is a schematic diagram of a collection path according to an embodiment of the present application;

[0049] Figure 3 is a schematic diagram of a geomagnetic grid reference map according to an embodiment of the present application;

[0050] Figure 4 is a schematic diagram of row enhancement according to an embodiment of the present application;

[0051] Figure 5 is a schematic diagram of column enhancement according to an embodiment of the present application;

[0052] Figure 6 is a schematic diagram of regional enhancement according to an embodiment of the present application;

[0053] Figure 7 Schematic diagram of the overall flow of the random waypoint walking algorithm according to an embodiment of the present application;

[0054] Figure 8 A schematic diagram of a random walk algorithm flow in a random waypoint walk algorithm according to an embodiment of the present application;

[0055] Fig. 9 is a schematic diagram of the overall structure of a multi-scale neural network positioning model according to an embodiment of the present application;

[0056] Fig.10 Schematic diagram of three-level wavelet decomposition according to an embodiment of the present application;

[0057] Fig.11 Schematic diagram of the working principle of the channel attention mechanism according to an embodiment of the present application;

[0058] Fig.12 is a block diagram of a weak magnetic positioning device according to an embodiment of the present application;

[0059] Fig.13 It is a schematic diagram of the structure of a terminal device or a server suitable for implementing the embodiments of the present application. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0061] In addition, the term "and / or" in this article is only a description of the association relationship between the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0062] Figure 1 A flow chart of a weak magnetic positioning method according to an embodiment of the present disclosure is shown. The method comprises:

[0063] S110, obtaining geomagnetic sequence information.

[0064] In this embodiment, the execution entity (such as a server) of the weak magnetic positioning method can obtain the geomagnetic sequence information through a wired or wireless connection.

[0065] Furthermore, the above-mentioned execution subject may obtain geomagnetic sequence information sent by an electronic device (such as a mobile phone, PAD or other terminal device) that is in communication with it, or may obtain geomagnetic sequence information pre-stored locally.

[0066] S120, inputting the geomagnetic sequence information into a trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information.

[0067] The weak magnetic positioning model can be trained in the following way:

[0068] Obtain a reference map of the original geomagnetic field;

[0069] Performing feature enhancement on the reference image to obtain a target reference image;

[0070] Processing the target reference map by a preset algorithm to obtain a geomagnetic sequence fingerprint database;

[0071] Based on the geomagnetic sequence fingerprint database, a weak magnetic positioning model is trained.

[0072] Specifically, an original geomagnetic reference map of the target area is collected and constructed, and then the features of the original geomagnetic reference map are enhanced, and the enhanced features are constructed into a new geomagnetic reference map to obtain a target reference map.

[0073] like Figure 2 As shown in the figure, before starting the acquisition, design an acquisition path that covers the entire acquisition area (the preset acquisition path):

[0074] Continuous collection is performed based on the preset collection path, and the total geomagnetic field strength value F, three-dimensional geomagnetic vectors Mx, My, Mz and corresponding geographical location coordinates (X, Y) of each collection point are recorded.

[0075] like Figure 3 As shown, the acquisition area is divided into n grids, and the acquired data is resampled according to the size of the divided grids to form an original geomagnetic reference map.

[0076] Furthermore, the fourth-order difference convolution enhancement operator can be used to enhance the features of the original geomagnetic reference map:

[0077] The one-dimensional Laplace operator is defined as:

[0078] ;

[0079] in, is the second-order partial derivative of variable x;

[0080] δf is the magnetic field variable;

[0081] δx is a one-dimensional coordinate variable;

[0082] The original geomagnetic reference map is a discrete grid point. Based on this feature, the above formula is differentially transformed to obtain its discrete expression:

[0083] ;

[0084] Taking the second-order difference of the one-dimensional Lagrangian operator, we get:

[0085] ;

[0086] Assume that the geomagnetic field of the central grid point of the convolution processing area after convolution is g(n):

[0087] ;

[0088] in, is the geomagnetic value before convolution;

[0089] K is the enhancement coefficient, usually ranging from 1 to 3. The value of k represents the strength of the convolution difference, which can directly affect the enhancement effect of the convolution.

[0090] When k=1, the enhancement formula is as follows:

[0091] ;

[0092] It can be obtained that the one-dimensional fourth-order difference convolution enhancement operator is:

[0093] ;

[0094] Furthermore, according to the data collection route direction and the spatial characteristics of the geomagnetic reference map, the enhancement method can be divided into line enhancement reference Figure 4 , column enhancement reference Figure 5 And regional enhancement reference Figure 6 .

[0095] Among them, both row enhancement and column enhancement can use the above-mentioned one-dimensional fourth-order difference convolution enhancement operator for feature enhancement:

[0096] X direction:

[0097] ;

[0098] Y direction:

[0099] ;

[0100] In summary, we can get:

[0101] ;

[0102] Similarly, assuming that the geomagnetic field of the central grid point of the convolution processing area after convolution is g(x,y), then:

[0103] ;

[0104] When K=1, the enhancement formula is as follows:

[0105] ;

[0106] The two-dimensional fourth-order difference convolution enhancement operator is as follows:

[0107] ;

[0108] In some embodiments, in the target reference map, a preset random waypoint walk algorithm (self-developed algorithm) can be used to generate a large number of geomagnetic trajectory sequences and establish a geomagnetic sequence fingerprint database.

[0109] Specifically, Figure 7 As shown, all grid points of the target reference map are traversed according to the index; N reference points are randomly selected as key trajectory points according to the index, and these key trajectory points are arranged in the order of selection and recorded as [P1, P2, P3, ..., PN];

[0110] like Figure 8 As shown, set P1 as the current point, compare the coordinates of P1 and P2, and set the coordinates of P1 to , the coordinates of P2 are , the coordinates of PN are ;

[0111] like ,but And update the current point coordinates;

[0112] like ,but And update the current point coordinates;

[0113] like , then compare the Y coordinates, similarly:

[0114] like ,but And update the current point coordinates;

[0115] like ,but And update the current point coordinates;

[0116] Until the current point reaches P2, that is, the coordinates of the current point are equal to the coordinates of P2;

[0117] Assume that the previous point of P2 is P2'. When moving from P2 to P3, first determine the angle formed by the line segment P2'P2 and the line segment P2P3.

[0118] The angle calculation formula is as follows:

[0119] Let the coordinates of P2' be ;

[0120] First calculate the vector ;

[0121] vector ;

[0122] Compute the length (magnitude) of two vectors:

[0123] ;

[0124] ;

[0125] Compute the dot product of two vectors:

[0126] ;

[0127] Based on the above, the radian value of the angle can be calculated:

[0128] ;

[0129] Convert radians to degrees:

[0130] ;

[0131] If the angle is greater than or equal to 90°, then a random walk is performed between P2 and P3;

[0132] If the angle is less than 90°, discard P3 and jump directly to the next key trajectory point P4 and compare the angle again;

[0133] Repeat the above steps until the current point coordinates reach the last key trajectory point PN.

[0134] It should be noted that each time a trajectory is generated, the key trajectory points are reselected to ensure that the trajectory generated each time is different.

[0135] The parameters that can be pre-set when generating trajectories may include the number of key trajectory points P for each trajectory. number , the number of trajectories T number And the moving step length Step; save each trajectory in an EXCEL file in a different Sheet.

[0136] In summary, the constructed fingerprint database is:

[0137] ;

[0138] in, Indicates the geomagnetic information contained in each trajectory point; including the total geomagnetic field strength value F, three-dimensional geomagnetic vectors Mx, My, Mz;

[0139] ;

[0140] Among them, X i and Y i Represents the position label corresponding to the i-th trajectory.

[0141] In some embodiments, based on the geomagnetic sequence fingerprint database, a weak magnetic positioning model is trained; the weak magnetic positioning model is a multi-scale neural network positioning model, and its structure is as follows: Fig. 9 shown.

[0142] Specifically, Fig.10 As shown in the figure, the input geomagnetic sequence is processed by multi-level wavelet decomposition technology. Each level of decomposition will produce approximate (low-frequency) coefficients and detail (high-frequency) coefficients. The formula is as follows:

[0143] ;

[0144] ;

[0145] in, is the approximate coefficient of the jth level;

[0146] is the detail coefficient of the jth level;

[0147] is the input geomagnetic sequence signal;

[0148] g[n] and h[n] are low-pass filter coefficient and high-pass filter coefficient respectively;

[0149] 2k-n is the downsampling operation.

[0150] Among them, the approximate coefficient contains the effective components of the signal, while the detail coefficient mainly contains useless components such as noise. Therefore, in the present disclosure, the approximate coefficient is retained and the detail coefficient is discarded.

[0151] The approximation coefficients of different levels are input into the local convolution module to learn and extract features of different scales;

[0152] Preferably, the local convolution module may include two convolution layers, a normalization layer and a maximum pooling layer;

[0153] The working process of the convolutional layer is as follows:

[0154] ;

[0155] Where y[i] is the value of the output feature vector at position i;

[0156] X is the input geomagnetic sequence;

[0157] W is the weight;

[0158] K is the convolution kernel size.

[0159] The maximum pooling layer is used to reduce the length of the feature vector. The formula is as follows:

[0160] ;

[0161] Among them, y[i] is the output of the maximum pooling layer;

[0162] x is the input of the maximum pooling layer;

[0163] s is the step size of the pooling window;

[0164] The activation function used in the network is the ReLU activation function, which is used to introduce nonlinearity, as shown below:

[0165] ;

[0166] The output feature vectors of each local convolution module are concatenated according to the channel dimension.

[0167] The introduction of the channel attention mechanism allows the model to learn the importance of each channel and assign weights according to the importance, so that the model focuses on feature channels with larger weight values, enhances the channels of feature vectors that are useful for the current task, and suppresses feature channels that are less useful for the current task.

[0168] like Fig.11 As shown in Figure 2, the channel attention mechanism works as follows:

[0169] The features of each channel are compressed into a real scalar through global average pooling;

[0170] ;

[0171] Where X is the input feature vector with the size of B×L×C;

[0172] is the global average pooling result of channel c;

[0173] B is the batch size;

[0174] L is the vector length;

[0175] C is the number of channels.

[0176] Through two fully connected layers and nonlinear activation functions, the correlation between channels is constructed to generate attention weights:

[0177] ;

[0178] in, and is the weight matrix of the two fully connected layers;

[0179] is the ReLU activation function;

[0180] is the Sigmoid activation function;

[0181] is the attention weight vector.

[0182] The weights are combined with the original features to form weighted fusion features:

[0183] ;

[0184] in, is the attention weight of channel c;

[0185] is the weighted fusion feature vector.

[0186] Furthermore, the fused features are input into the global CNN module to obtain the global features after learning fusion.

[0187] Among them, the global CNN module contains two convolutional layers and one maximum pooling layer.

[0188] Furthermore, the global features are input into a positioning module to output geographic location coordinates; the positioning module may be composed of a plurality of fully connected layers.

[0189] It should be noted that the loss function used in the network is the mean square error (MSE) function:

[0190] ;

[0191] in, is the true value;

[0192] is the predicted value;

[0193] N is the sample size.

[0194] Furthermore, the positioning accuracy of the model can be measured by calculating the Euclidean distance between two points:

[0195] ;

[0196] Among them, (X,Y) is the label coordinate;

[0197] ( , ) are the predicted coordinates.

[0198] In some embodiments, the geomagnetic sequence obtained in step S110 is input into the weak magnetic positioning model trained in the above steps to obtain position information corresponding to the geomagnetic sequence information.

[0199] According to the embodiments of the present disclosure, the following technical effects are achieved:

[0200] Effective positioning results can be achieved in a large range of outdoor weak magnetic areas, which reduces a lot of time and labor costs in data acquisition and can generate more realistic trajectories while overcoming the influence of noise.

[0201] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0202] The above is an introduction to the method embodiment. The following is a further explanation of the scheme described in this application through an apparatus embodiment.

[0203] Fig.12 FIG. 4 shows a block diagram of a weak magnetic positioning device according to an embodiment of the present application, as shown in FIG. Fig.12 Shown include:

[0204] An acquisition module 1210 is used to acquire geomagnetic sequence information;

[0205] The positioning module 1220 is used to input the geomagnetic sequence information into the trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information;

[0206] The weak magnetic positioning model can be trained in the following way:

[0207] Obtain a reference map of the original geomagnetic field;

[0208] Performing feature enhancement on the reference image to obtain a target reference image;

[0209] Processing the target reference map by a preset algorithm to obtain a geomagnetic sequence fingerprint database;

[0210] Based on the geomagnetic sequence fingerprint database, a weak magnetic positioning model is trained.

[0211] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0212] Fig.13 A schematic diagram of the structure of a terminal device or server suitable for implementing an embodiment of the present application is shown.

[0213] like Fig.13 As shown, the terminal device or server includes a central processing unit (CPU) 1301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1302 or the program loaded from the storage part 1308 to the random access memory (RAM) 1303. In the RAM 1303, various programs and data required for the operation of the terminal device or server are also stored. CPU1301, ROM1302 and RAM1303 are connected to each other through a bus 1304. An input / output (I / O) interface 1305 is also connected to the bus 1304.

[0214] The following components are connected to the I / O interface 1305: an input section 1306 including a keyboard, a mouse, etc.; an output section 1307 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1308 including a hard disk, etc.; and a communication section 1309 including a network interface card such as a LAN card, a modem, etc. The communication section 1309 performs communication processing via a network such as the Internet. A drive 1310 is also connected to the I / O interface 1305 as needed. A removable medium 1311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1310 as needed, so that a computer program read therefrom is installed into the storage section 1308 as needed.

[0215] In particular, according to an embodiment of the present application, the above method flow steps can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a machine-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1309, and / or installed from the removable medium 1311. When the computer program is executed by the central processing unit (CPU) 1301, the above-mentioned functions defined in the system of the present application are executed.

[0216] It should be noted that the computer-readable medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0217] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of the code, and the aforementioned module, program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0218] The units or modules involved in the embodiments described in the present application may be implemented by software or hardware. The units or modules described may also be arranged in a processor. The names of these units or modules do not, in some cases, constitute limitations on the units or modules themselves.

[0219] As another aspect, the present application further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable storage medium stores one or more programs, and when the above programs are used by one or more processors to execute the method described in the present application.

[0220] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of application involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned application concept. For example, the above features are replaced with (but not limited to) technical features with similar functions applied in the present application.

Claims

1. A weak magnetic positioning method, characterized in that: include: Obtain geomagnetic sequence information; Inputting the geomagnetic sequence information into a trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information; The weak magnetic positioning model can be trained in the following way: Obtain a reference map of the original geomagnetic field; Performing feature enhancement on the reference image to obtain a target reference image; The target reference map is processed by a preset algorithm to obtain a geomagnetic sequence fingerprint database, including: Based on the target reference map, construct an index of acquisition points; Based on the index, randomly select multiple reference points as key trajectory points to generate a trajectory point set; Based on the trajectory point set, a geomagnetic sequence fingerprint database is constructed, including: Select two adjacent trajectory points from the trajectory point set, take a first trajectory point with a forward index among the two adjacent trajectory points as a starting point, and a second trajectory point with a backward index as an end point, and generate a first trajectory path between the first trajectory point and the second trajectory point; Selecting a third trajectory point adjacent to the second trajectory point from the trajectory point set, and generating a second trajectory path between the second trajectory point and the third trajectory point; If the angle between the first trajectory path and the second trajectory path is greater than or equal to a preset threshold, the second trajectory path is saved; if the angle is less than the threshold, a fourth trajectory point adjacent to the third trajectory point is selected, a third trajectory path between the second trajectory point and the fourth trajectory point is randomly generated, and the angle between the first trajectory path and the third trajectory path is compared with the threshold. If the angle is greater than or equal to the preset threshold, the third trajectory path is saved; Repeat the above steps until all the key trajectory points in the trajectory point set are traversed; Based on the saved trajectory path, a geomagnetic sequence fingerprint database is constructed; based on the geomagnetic sequence fingerprint database, a weak magnetic positioning model is trained.

2. The method according to claim 1, characterized in that The reference map for obtaining the original geomagnetism comprises: Based on the preset collection path, obtain the geomagnetic information of each collection point; Based on the pre-divided grid size, the geomagnetic information is resampled to construct a reference map of the original geomagnetism.

3. The method according to claim 2, characterized in that The geomagnetic information includes the total geomagnetic field strength value, the three-dimensional geomagnetic vector and the corresponding geographical location coordinates.

4. The method according to claim 3, characterized in that The performing feature enhancement on the reference image to obtain a target reference image comprises: Based on the preset acquisition path and the spatial features of the reference image, the reference image is feature enhanced by a convolution enhancement operator to obtain a target reference image.

5. The method according to claim 4, characterized in that The weak magnetic positioning model is a multi-scale neural network positioning model.

6. A weak magnetic positioning device, characterized in that: include: An acquisition module, used for acquiring geomagnetic sequence information; A positioning module, used for inputting the geomagnetic sequence information into a trained weak magnetic positioning model to obtain position information corresponding to the geomagnetic sequence information; The weak magnetic positioning model can be trained in the following way: Obtain a reference map of the original geomagnetic field; Performing feature enhancement on the reference image to obtain a target reference image; The target reference map is processed by a preset algorithm to obtain a geomagnetic sequence fingerprint database, including: Based on the target reference map, construct an index of acquisition points; Based on the index, randomly select multiple reference points as key trajectory points to generate a trajectory point set; Based on the trajectory point set, a geomagnetic sequence fingerprint database is constructed, including: Select two adjacent trajectory points from the trajectory point set, take a first trajectory point with a forward index among the two adjacent trajectory points as a starting point, and a second trajectory point with a backward index as an end point, and generate a first trajectory path between the first trajectory point and the second trajectory point; Selecting a third trajectory point adjacent to the second trajectory point from the trajectory point set, and generating a second trajectory path between the second trajectory point and the third trajectory point; If the angle between the first trajectory path and the second trajectory path is greater than or equal to a preset threshold, the second trajectory path is saved; if the angle is less than the threshold, a fourth trajectory point adjacent to the third trajectory point is selected, a third trajectory path between the second trajectory point and the fourth trajectory point is randomly generated, and the angle between the first trajectory path and the third trajectory path is compared with the threshold. If the angle is greater than or equal to the preset threshold, the third trajectory path is saved; Repeat the above steps until all the key trajectory points in the trajectory point set are traversed; Based on the saved trajectory path, a geomagnetic sequence fingerprint database is constructed; Based on the geomagnetic sequence fingerprint database, a weak magnetic positioning model is trained.

7. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images