Railway station area iron shoe anti-slip positioning monitoring method

Abstract

The invention relates to the technical field of target positioning, and discloses a railway station area iron shoe anti-slip positioning monitoring method comprising the following steps: arranging a plurality of cameras at different positions and an AOA positioning module; the original signal intensity and the original arrival angle of the target equipment are obtained, error correction is carried out on the original signal intensity, and the distance between the target equipment and the AOA positioning module is obtained; performing error correction on the original arrival angle to obtain a corrected arrival angle; obtaining a first position coordinate of the target device based on the distance between the target device and the AOA positioning module and the corrected arrival angle; collecting an image of target equipment, and obtaining a second position coordinate of the target equipment; and carrying out feature fusion on the first position coordinate and the second position coordinate, and taking a fused position feature vector as final positioning of the target equipment. According to the invention, through image identification and dual-mode data fusion of the angle of arrival, the positioning precision of the railway station area on the target is improved.

Description

Technical Field

[0001] The invention relates to the technical field of target positioning, and in particular to a method for positioning and monitoring an anti-slip iron shoe in a railway station area. Background Art

[0002] In order to better establish and improve the safety control of railway station areas and the dual prevention work mechanism of active defense and hidden danger investigation and treatment, supervise the implementation of the main responsibility for production safety, and effectively perform safety supervision duties, the on-site anti-slip iron shoe operation behavior is intelligently identified, real-time warning, and accurate portraits are carried out through artificial intelligence equipment control means to strengthen the control of operation links. Although video surveillance management systems now play an important role in the prevention and control of safety risks in railway station areas, with the advancement of technology, the shortcomings of existing video surveillance management systems are becoming increasingly obvious, such as low recognition and positioning accuracy. Summary of the invention

[0003] The purpose of the present invention is to improve the positioning accuracy of a target in a railway station area by fusing dual-modal data of image recognition and angle of arrival (AOA), and to provide a method for positioning and monitoring an iron shoe to prevent slipping in a railway station area.

[0004] In order to achieve the above-mentioned object of the invention, the embodiment of the present invention provides the following technical solutions:

[0005] The method for monitoring and positioning the anti-slipping iron shoes in the railway station area comprises the following steps:

[0006] Step 1, setting up multiple video surveillance systems at different locations in the railway station area, each of the video surveillance systems includes a camera and an AOA positioning module;

[0007] Step 2: Obtain the original signal strength and original arrival angle of the target device through the AOA positioning module, perform error correction on the original signal strength, and obtain the distance between the target device and the AOA positioning module based on the corrected error strength; perform error correction on the original arrival angle to obtain a corrected arrival angle; obtain the first position coordinates of the target device based on the distance between the target device and the AOA positioning module and the corrected arrival angle;

[0008] Step 3, collecting an image of the target device through a camera, and obtaining a second position coordinate of the target device through a three-dimensional reconstruction technology;

[0009] Step 4: Fuse the first position coordinates and the second position coordinates based on the attention mechanism, and use the fused position feature vector as the final positioning of the target device.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: the purpose of the present invention is to improve the positioning accuracy of anti-skid shoes in railway station areas through dual-modal data fusion of image recognition and angle of arrival (AOA). BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0013] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present invention.

[0014] It should be noted that similar numbers and letters represent similar items in the following figures, so once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance, or implying any such actual relationship or order between these entities or operations. In addition, the terms "connected", "connected", etc. can be directly connected between elements, or indirectly connected via other elements.

[0015] Embodiment 1:

[0016] The present invention is achieved through the following technical solutions: Figure 1 As shown, a method for monitoring and positioning anti-slipping shoes in a railway station area comprises the following steps:

[0017] Step 1: Set up multiple video surveillance systems at different locations in the railway station area, each of the video surveillance systems includes a camera and an AOA positioning module.

[0018] Cameras are set at different locations in the railway station area. Preferably, the images captured by each camera overlap with the images captured by at least another camera, so as to construct a complete image. These cameras have different heights from the standard horizontal plane and different shooting angles. It should be noted that in this embodiment, the heights and shooting angles of all cameras cannot be exactly the same, but the heights and shooting angles of some cameras can be the same. The standard horizontal plane refers to a manually defined horizontal plane, such as the ground, which is used to determine the height of the camera setting.

[0019] Step 2: Obtain the original signal strength and original arrival angle of the target device through the AOA positioning module, perform error correction on the original signal strength, and obtain the distance between the target device and the AOA positioning module based on the corrected error strength; perform error correction on the original arrival angle to obtain a corrected arrival angle; and obtain the first position coordinates of the target device based on the distance between the target device and the AOA positioning module and the corrected arrival angle.

[0020] The step 2 specifically includes the following steps:

[0021] Step 2-1: Obtain the original signal strength and original arrival angle of the target device through the AOA positioning module.

[0022] The target device is a radio frequency tag equipped with AOA positioning technology, which is used to transmit radio frequency signals to the AOA positioning module. For example, the target that needs to be located in real time in the railway station area, such as anti-slip shoes for railways, staff, railway-related equipment, etc., wears the radio frequency tag.

[0023] AOA positioning module is M y ×M z The antenna array consists of antennas, and the distance between two adjacent antennas is D G , M y 、M z are the number of antennas in the y-axis direction and the z-axis direction of the antenna array respectively. The AOA positioning module can obtain the original signal strength and original arrival angle of the target device. Taking the positioning of any target device in the railway station area as an example, assuming that there are N cameras in total, an AOA positioning module is installed in each camera. The original signal strength of the target device obtained by the i-th AOA positioning module is RSS i , the original arrival angle is α i , i = 1, 2, ..., N; then the original signal strength of the target device obtained by all AOA positioning modules is RSS = [RSS 1 ,RSS 2 ,...,RSS N ], the original arrival angle is α=[α 1 ,α 2,...,α N ],have where θ i Represents the direction angle between the i-th AOA positioning module and the target device, Indicates the elevation angle between the i-th AOA positioning module and the target device.

[0024] Step 2-2: Perform error correction on the original signal strength, and obtain the distance between the target device and the AOA positioning module based on the corrected error strength.

[0025] The signal strength obtained by the i-th AOA positioning module is RSS i According to the Hata model, the relationship between the radio signal propagation distance and the received signal strength (RSS) is:

[0026]

[0027] Among them, d i Indicates the distance from the target device to the i-th AOA positioning module; RSS i (d i ) represents the signal strength received by the i-th AOA positioning module from the target device; d 0 Indicates unit distance, usually 1m; RSS i (d 0 ) represents the signal strength received by the i-th AOA positioning module at unit distance; n represents the path attenuation exponent; ε represents the environmental noise during the transmission process and the Gaussian white noise in the channel, that is, a random variable with a mean of 0 and a variance of σ.

[0028] The distance from the target device to the i-th AOA positioning module can be obtained by adding the arrival angle α i , the target device can be located. However, this solution takes into account that when the AOA positioning module receives signals, the scattering, reflection, and refraction of the signals by obstacles in the environment, as well as antenna array errors, will reduce the accuracy of the signal strength and arrival angle, ultimately resulting in poor positioning accuracy of the target device. Therefore, this solution needs to correct the errors of the original signal strength and the original arrival angle, thereby improving the positioning accuracy of the target device.

[0029] When performing error correction on the original signal strength, the power of the target device transmitting the signal and obstacles on the signal path are taken into account. The corrected signal strength is:

[0030]

[0031] in, Indicates the corrected signal strength; P irepresents the power of the signal transmitted by the target device to the i-th AOA positioning module; cosech(.) represents the hyperbolic cosecant function; Z i represents the number of obstacles on the path between the target device and the i-th AOA positioning module; γ 1 P i The weight of 2 is cosech(Z i )’s weight.

[0032] The ratio of the corrected signal strength to the original signal strength γ is obtained:

[0033]

[0034] Thus, the corrected distance from the target device to the i-th AOA positioning module is obtained:

[0035]

[0036] in, Indicates the corrected distance. In this way, the problem of reduced signal strength accuracy caused by scattering, reflection, and refraction of signals by obstacles in the environment can be eliminated.

[0037] Step 2-3: Perform error correction on the original arrival angle to obtain a corrected arrival angle.

[0038] The direction steering vector and pitch steering vector of the antenna array of the i-th AOA positioning module are:

[0039]

[0040] Among them, α i (θ i ), are the directional steering vector and the pitch steering vector respectively; j is the imaginary unit; λ is the wavelength; D G is the distance between two adjacent antennas; M y 、M z are the number of antennas in the y-axis direction and the z-axis direction of the antenna array respectively.

[0041] This solution is to correct the arrival angle and construct the first guidance vector

[0042]

[0043] Constructing the second steering vector

[0044]

[0045] in, Indicates wave control direction; Indicates the antenna error correction distance, which is:

[0046]

[0047] Among them, c y 、c z are the position errors measured along the y-axis and z-axis of the antenna array, respectively. It should be noted that the y-axis of the antenna array is along the line axis, and the z-axis is perpendicular to the line axis.

[0048] Based on the first steering vector and the second steering vector, the corrected arrival angle is obtained:

[0049]

[0050] in, represents the corrected arrival angle; This eliminates the problem of reduced arrival angle accuracy caused by antenna array errors.

[0051] Step 2-4: Based on the distance between the target device and the AOA positioning module and the corrected arrival angle, obtain the first position coordinates of the target device.

[0052] Based on the corrected distance and the corrected angle of arrival The first position coordinate u of the target device obtained by the i-th AOA positioning module can be obtained ti =(x ti ,y ti ,z ti ). Similarly, the first position coordinate of the target device obtained by N AOA positioning modules is u t =[u t1 ,u t2 ,...,u tN ].

[0053] Step 3: Capture an image of the target device through a camera, and obtain the second position coordinates of the target device through three-dimensional reconstruction technology.

[0054] The picture taken by the i-th camera is img i Since the images captured by each camera overlap with the images captured by at least another camera, it is possible to obtain feature points of fixed objects (such as the ground, stairs, buildings, fixed equipment, etc.) to match the feature points of the two overlapping images, and convert the two-dimensional coordinates of the feature points into three-dimensional coordinates in the world coordinate system, so as to reconstruct the railway station area in three dimensions and obtain a three-dimensional map 3DM=[(dp 1 ,img 1 ),(dp2 ,img 2 ),...,(dp N ,img N )], where dp i Represents image img i The second position coordinate of the target device in the three-dimensional map is u 3DM =(x 3DM ,y 3DM ,z 3DM ).

[0055] Step 4: Fuse the first position coordinates and the second position coordinates based on the attention mechanism, and use the fused position feature vector as the final positioning of the target device.

[0056] Convert the first position coordinates to the first position coordinate vector u t =[u t1 ,u t2 ,...,u tN ], convert the second position coordinates into the second position coordinate vector u 3DM =(x 3DM ,y 3DM ,z 3DM ), based on the attention mechanism, the first position coordinate vector and the second position coordinate vector are feature fused:

[0057] U=f 1 ×u 3DM +[f 2 ×u t1 +f 3 ×u t2 +...+f N+1 ×u tN ];

[0058] Among them, f 1 、f 2 , ..., f N+1 Represent weights respectively, and f 1 +f 2 +...+f N+1 =1; U represents the fused position feature vector, which is constrained by the objective function:

[0059]

[0060] Where J represents the objective function; Represents the real position coordinate vector of the target device; ∑ U Represents the covariance matrix, which is:

[0061]

[0062] in, The fused position feature vector U is used as the final positioning of the target device, which has a significantly improved accuracy compared to the positioning of the target device obtained by using the traditional AOA positioning technology alone or the video surveillance technology alone.

[0063] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for monitoring and positioning the anti-slipping iron shoes in railway station areas, characterized in that: The following steps are involved: Step 1, setting up multiple video surveillance systems at different locations in the railway station area, each of the video surveillance systems includes a camera and an AOA positioning module; Step 2: Obtain the original signal strength and original arrival angle of the target device through the AOA positioning module, perform error correction on the original signal strength, and obtain the distance between the target device and the AOA positioning module based on the corrected error strength; Perform error correction on the original arrival angle to obtain a corrected arrival angle; obtain the first position coordinates of the target device based on the distance between the target device and the AOA positioning module and the corrected arrival angle; Step 3, collecting an image of the target device through a camera, and obtaining a second position coordinate of the target device through a three-dimensional reconstruction technology; Step 4: Fuse the first position coordinates and the second position coordinates based on the attention mechanism, and use the fused position feature vector as the final positioning of the target device.

2. The method for monitoring and controlling the anti-slipping of iron shoes in railway station areas according to claim 1, characterized in that: In step 2, the step of obtaining the original signal strength and original arrival angle of the target device through the AOA positioning module includes: Assume that there are N cameras in total, and an AOA positioning module is installed in each camera. The original signal strength of the target device obtained by the i-th AOA positioning module is RSS i , the original arrival angle is α i , i=1,2,...,N, there are where θ i Represents the direction angle between the i-th AOA positioning module and the target device, Indicates the elevation angle between the i-th AOA positioning module and the target device.

3. The method for monitoring and controlling the anti-slipping of iron shoes in railway station areas according to claim 2 is characterized in that: In step 2, the step of performing error correction on the original signal strength and obtaining the distance between the target device and the AOA positioning module based on the corrected error strength includes: The signal strength obtained by the i-th AOA positioning module is RSS i , the relationship between the radio signal propagation distance and the received signal strength is: Among them, d i Indicates the distance from the target device to the i-th AOA positioning module; RSS i (d i ) represents the signal strength received by the i-th AOA positioning module from the target device; d0 represents the unit distance, usually 1m; RSS i (d0) represents the signal strength received by the i-th AOA positioning module at a unit distance; n represents the path attenuation index; ε represents the environmental noise during the transmission process and the Gaussian white noise in the channel; The corrected signal strength is: in, Indicates the corrected signal strength; P i represents the power of the signal transmitted by the target device to the i-th AOA positioning module; cosech(.) represents the hyperbolic cosecant function; Z i represents the number of obstacles on the path between the target device and the i-th AOA positioning module; γ1 is P i The weight of γ2 is cosech(Z i )’s weight; The ratio of the corrected signal strength to the original signal strength γ is obtained: Thus, the corrected distance from the target device to the i-th AOA positioning module is obtained: in, Indicates the corrected distance.

4. The method for monitoring and controlling the anti-slipping positioning of iron shoes in railway station areas according to claim 3 is characterized in that: In step 2, the step of performing error correction on the original arrival angle to obtain a corrected arrival angle includes: The direction steering vector and pitch steering vector of the antenna array of the i-th AOA positioning module are: Among them, α i (θ i ), are the directional steering vector and the pitch steering vector respectively; j is the imaginary unit; λ is the wavelength; D G is the distance between two adjacent antennas; M y 、M z are the number of antennas in the y-axis direction and the z-axis direction of the antenna array respectively; Constructing the first steering vector Constructing the second steering vector in, Indicates wave control direction; Indicates the antenna error correction distance, which is: Among them, c y 、c z are the position errors of the antenna array measured along the y-axis and z-axis directions respectively; Based on the first steering vector and the second steering vector, the corrected arrival angle is obtained: in, represents the corrected arrival angle; is the inner product symbol.

5. The method for monitoring and controlling the anti-slipping positioning of iron shoes in railway station areas according to claim 4 is characterized in that: In step 2, based on the distance between the target device and the AOA positioning module and the corrected angle of arrival Get the first position coordinate of the target device as u ti =(x ti ,y ti ,z ti ); The first position coordinate of the target device obtained by N AOA positioning modules is u t =[u t1 ,u t2 ,...,u tN ].

6. The method for monitoring and controlling the anti-slipping positioning of iron shoes in railway station areas according to claim 5, characterized in that: The step 3 specifically comprises the following steps: The image captured by each camera overlaps with the image captured by at least one other camera. The image captured by the i-th camera is img i , match the feature points of the two overlapping images, convert the two-dimensional coordinates of the feature points into three-dimensional coordinates in the world coordinate system, and then reconstruct the railway station area in three dimensions to obtain a three-dimensional map 3DM = [(dp1, img1), (dp2, img2), ..., (dp N ,img N )], where dp i Represents image img i Depth information; The second position coordinate of the target device in the three-dimensional map is u 3DM =(x 3DM ,y 3DM ,z 3DM ).

7. The method for monitoring and controlling the anti-slipping positioning of iron shoes in railway station areas according to claim 6, characterized in that: The step 4 specifically comprises the following steps: Convert the first position coordinates to the first position coordinate vector u t =[u t1 ,u t2 ,...,u tN ], convert the second position coordinates into the second position coordinate vector u 3DM =(x 3DM ,y 3DM ,z 3DM ), based on the attention mechanism, the first position coordinate vector and the second position coordinate vector are feature fused: U=f1×u 3DM +[f2×u t1 +f3×u t2 +...+f N+1 ×u tN ]; Among them, f1, f2, ..., f N+1 They represent weights respectively, f1+f2+...+f N+1 =1; U represents the fused position feature vector, and the fused position feature vector U is used as the final positioning of the target device.

Images