Wall detection method and apparatus therefor

Abstract

The application provides a wall detection method and device, and relates to the technical field of data processing.The method comprises the following steps: controlling an ultra-wideband radar detection device to move in a direction parallel to a wall to be detected to perform target detection, and obtaining a radar echo matrix; inputting the radar echo matrix into a trained target wall parameter identification model to perform identification, so as to obtain a wall thickness and a wall relative dielectric coefficient of the wall to be detected; and determining a detection distance between targets detected by the ultra-wideband radar detection device based on the wall thickness and the wall relative dielectric coefficient.The application can improve the estimation accuracy of wall parameters, and in turn, the accuracy and real-time performance of wall detection, avoid deviation in positioning of a detection target position and problems such as false targets, and can effectively estimate wall parameters, and improve the robustness and stability of wall detection.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a wall detection method and apparatus. Background Technology

[0002] In related technologies, when detecting life forms in complex environments, the weak vital signs, uneven walls, and variable environmental conditions necessitate methods and devices for estimating wall parameters (thickness, relative permittivity) and compensating for through-wall radar parameters to possess high stability and robustness. Furthermore, emergency rescue detection requires algorithms with strong real-time performance. The high accuracy requirements for wall parameters in the detection environment necessitate extensive algorithm iterations, limiting the real-time performance of wall detection. Therefore, improving the accuracy and flexibility of wall detection has become a crucial research direction. Summary of the Invention

[0003] This application aims to at least partially address one of the technical problems in the related art. Therefore, one objective of this application is to propose a wall detection method.

[0004] The second objective of this application is to propose a wall detection device.

[0005] The third objective of this application is to propose an electronic device.

[0006] The fourth objective of this application is to provide a non-transitory computer-readable storage medium.

[0007] The fifth objective of this application is to provide a computer program product.

[0008] To achieve the above objectives, a wall detection method is proposed in the first aspect of this application, comprising:

[0009] The ultra-wideband radar detection device is controlled to move in a direction parallel to the wall to be measured to detect the target and obtain the radar echo matrix.

[0010] The radar echo matrix is ​​input into the trained target wall parameter recognition model for identification in order to obtain the wall thickness and relative permittivity of the wall to be tested.

[0011] Based on the wall thickness and the relative permittivity of the wall, the detection distance between targets detected by the ultra-wideband radar detection device is determined.

[0012] In this embodiment, the radar echo matrix is ​​input into the trained target wall parameter recognition model for recognition, which can improve the estimation accuracy of wall parameters, thereby improving the accuracy and real-time performance of wall detection, avoiding problems such as deviation in target location and the appearance of false targets, effectively estimating wall parameters, and improving the robustness and stability of wall detection.

[0013] To achieve the above objectives, a second aspect of this application provides a wall detection device, comprising:

[0014] The first acquisition module is used to control the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured to detect the target and acquire the radar echo matrix.

[0015] The second acquisition module is used to input the radar echo matrix into the trained target wall parameter recognition model for recognition, so as to obtain the wall thickness and relative permittivity of the wall to be tested.

[0016] The determination module is used to determine the detection distance between targets detected by the ultra-wideband radar detection device based on the wall thickness and the wall's relative permittivity.

[0017] To achieve the above objectives, a third aspect of this application provides an electronic device comprising:

[0018] At least one processor; and

[0019] A memory that is communicatively connected to at least one processor; wherein,

[0020] The memory stores instructions that can be executed by at least one processor, which enables the at least one processor to perform the wall detection method provided in the first aspect embodiment of this application.

[0021] To achieve the above objectives, a fourth aspect of this application provides a computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are used to cause a computer to execute the wall detection method provided in the first aspect of this application.

[0022] To achieve the above objectives, a fifth aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the wall detection method provided in the first aspect of this application. Attached Figure Description

[0023] Figure 1 This is a flowchart of a wall detection method according to an embodiment of this application;

[0024] Figure 2 This is a flowchart of a wall detection method according to an embodiment of this application;

[0025] Figure 3 This is a schematic diagram of radar signal penetration through walls according to an embodiment of this application;

[0026] Figure 4 This is a schematic diagram of the signal wave intensity according to an embodiment of this application;

[0027] Figure 5 This is a schematic diagram of radar signal penetration through walls according to an embodiment of this application;

[0028] Figure 6 This is a flowchart of a wall detection method according to an embodiment of this application;

[0029] Figure 7 This is a flowchart of a wall detection method according to an embodiment of this application;

[0030] Figure 8 This is a schematic diagram of a candidate radar echo matrix according to an embodiment of this application;

[0031] Figure 9 This is a schematic diagram of a candidate radar echo matrix according to an embodiment of this application;

[0032] Figure 10 This is a schematic diagram of a candidate radar echo matrix according to an embodiment of this application;

[0033] Figure 11 This is a schematic diagram of a candidate radar echo matrix according to an embodiment of this application;

[0034] Figure 12 This is a schematic diagram of a wall detection method according to an embodiment of this application;

[0035] Figure 13 This is a schematic diagram illustrating the model training effect of one embodiment of this application;

[0036] Figure 14 This is a schematic diagram of the wall thickness prediction result according to one embodiment of this application;

[0037] Figure 15 This is a schematic diagram of the prediction result of the relative permittivity of the wall according to an embodiment of this application;

[0038] Figure 16 This is a schematic diagram of a wall detection method according to an embodiment of this application;

[0039] Figure 17 This is a structural block diagram of a wall detection device according to an embodiment of this application;

[0040] Figure 18 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0041] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0042] The wall detection method and apparatus of this application are described below with reference to the accompanying drawings.

[0043] Figure 1 This is a flowchart of a wall detection method according to an embodiment of this application, as follows: Figure 1 As shown, the method includes:

[0044] S101 controls the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured to detect the target and obtain the radar echo matrix.

[0045] In this embodiment, the ultra-wideband radar detection device can be a bio-radar (IR-UWB radar). Bio-radar is a technology that extracts signals related to the target from radar echoes to achieve non-contact, long-range, and penetrating target detection and localization. Its principle is that the radar emits radar signals towards the target; these signals are modulated by the target and reflected back to the radar receiving antenna; the radar receives the radar echo signals and obtains the radar echo matrix. Optionally, the target can be a human body.

[0046] Alternatively, the ultra-wideband radar detection device can also be a radar life detector, a signal-based life detector, a pulse radar life detector, etc.

[0047] In some implementations, the ultra-wideband radar detection device moves in a direction parallel to the wall to be measured to detect the target. The original radar echo matrix obtained not only contains the wall and the target, but also a large number of other interference clutter, such as static background clutter, additive white noise, unstable fast-time DC components, linear trends generated on the slow time axis due to amplitude instability during radar system acquisition, harmonics and signal distortion caused by dynamic targets, etc. Optionally, some clutter and noise interference can be removed by preprocessing methods to obtain a denoised radar echo matrix.

[0048] Optionally, the signal-to-noise ratio of the detected target can be improved by preprocessing methods such as channel signal subtraction and time averaging subtraction.

[0049] S102, input the radar echo matrix into the trained target wall parameter recognition model for recognition, in order to obtain the wall thickness and relative permittivity of the wall to be tested.

[0050] Optionally, in the embodiments of this application, the target wall parameter identification model can be a neural network model, such as a long short-term memory neural network model.

[0051] In this embodiment of the application, a wall parameter recognition model based on a long short-term memory neural network containing 5 layers is used as the target wall parameter recognition model for description. The target wall parameter recognition model contains 3 long short-term memory (LSTM) layers (i.e., the first LSTM layer is LSTM1, the second LSTM layer is LSTM2, and the third LSTM layer is LSTM3), one fully connected layer (Dense), and one output layer (Output).

[0052] The radar echo matrix is ​​input into the wall parameter identification model based on long short-term memory neural network to obtain the wall parameter prediction results, and the wall thickness and relative permittivity of the wall to be tested are obtained.

[0053] S103, based on the wall thickness and the relative permittivity of the wall, determine the detection distance between targets detected by the ultra-wideband radar detection device.

[0054] In this embodiment, the amplitude attenuation coefficient of the radar signal of the ultra-wideband radar detection device when it propagates in the wall under test can be obtained according to the wall thickness and the relative permittivity of the wall. Then, the propagation time of the radar signal in the wall under test can be obtained according to the amplitude attenuation coefficient. Finally, the detection distance between the targets detected by the ultra-wideband radar detection device can be determined based on the propagation time.

[0055] In this embodiment, the radar echo matrix is ​​input into the trained target wall parameter recognition model for recognition, which can improve the estimation accuracy of wall parameters, thereby improving the accuracy and real-time performance of wall detection, avoiding problems such as deviation in target location and the appearance of false targets, effectively estimating wall parameters, and improving the robustness and stability of wall detection.

[0056] Figure 2 This is a flowchart of a wall detection method according to an embodiment of this application, as follows: Figure 2 As shown, the method includes:

[0057] S201 controls the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured to detect the target and obtain the radar echo matrix.

[0058] S202, input the radar echo matrix into the trained target wall parameter recognition model for identification, in order to obtain the wall thickness and relative permittivity of the wall to be tested.

[0059] For details regarding steps S201 to S202, please refer to the relevant descriptions in the above embodiments; they will not be repeated here.

[0060] S203, based on the wall thickness and the relative permittivity of the wall, obtain the propagation time of the radar signal of the ultra-wideband radar detection device within the wall to be tested.

[0061] Analysis of the electromagnetic wave propagation characteristics of ultra-wideband through-wall radar reveals that the propagation path and speed of radar signals change when a wall is present. This change leads to discrepancies between the actual location of the detected target and the location data obtained under the assumption that the radar wave propagates in free space, necessitating correction of the target's location information.

[0062] In some implementations, the amplitude attenuation coefficient of the radar signal from the ultra-wideband radar detection device propagating within the wall under test is obtained based on the wall thickness and the wall's relative permittivity. The propagation time of the radar signal within the wall under test is then determined based on the amplitude attenuation coefficient.

[0063] The process of obtaining the amplitude attenuation coefficient includes: determining the wall parameters of the detection wall based on the wall thickness and relative permittivity; obtaining the first medium parameters and incident wave intensity of the first medium in which the ultra-wideband radar detection device is located; obtaining the second medium parameters and transmitted wave intensity of the second medium in which the target is located; obtaining the first transmission coefficient and second transmission coefficient based on the first medium parameters and incident wave intensity, the second medium parameters and transmitted wave intensity, and the wall parameters; and determining the amplitude attenuation coefficient based on the first transmission coefficient and second transmission coefficient.

[0064] In practice, there is a nonlinear relationship between the relative permittivity of the wall, the wall thickness, and the transmitted signal. For example... Figure 3 As shown, the radar wave is emitted by the transmitting antenna in medium 1, passes through mediums 1, 2, and 3 in sequence, and is finally received by the receiving antenna in medium 3. (That is to say, the first medium in which the ultra-wideband radar detection device is located is medium 1, and the second medium in which the target is located is medium 3.) The total field quantity in mediums 1, 2, and 3 can be expressed as follows:

[0065] Total field quantity in medium 1

[0066]

[0067]

[0068] Total field quantity in medium 2

[0069]

[0070]

[0071] Total field quantity in medium 3

[0072]

[0073]

[0074] in:

[0075] E n H n (n = 1, 2, 3) represent the electric field strength and magnetic field strength in media 1, 2, and 3, respectively;

[0076] E ni H ni (n = 1, 2, 3) represent the incident wave field strength and the reflected wave field strength in media 1, 2, and 3, respectively;

[0077] k0 and η0 represent the wave number and wave impedance in medium 1 (air) and medium 3 (air), respectively. k and η represent the wave number and wave impedance in medium 2 (the wall).

[0078] ε、ε r ε represents the dielectric constant and relative dielectric constant of medium 2 (the wall), ε0 represents the dielectric constant of air, and ε = ε0ε r ω represents the angular frequency of the transmitted signal;

[0079] μ0 is the magnetic permeability in medium 1 (air) and medium 3 (air), and μ is the magnetic permeability in medium 2 (wall). This application does not consider the magnetic permeability of the wall, i.e., μ = μ0.

[0080] Based on the boundary conditions when z = 0 and z = d, we can obtain:

[0081] E1| z＝0 =E2| z＝0 H1| z＝0 =H2| z＝0

[0082] E2| z＝d =E3| z＝d H2| z＝d =H3| z＝d

[0083] When z = 0:

[0084] E 1i +E 1r =E 2i +E 2r

[0085]

[0086] When z = d:

[0087]

[0088]

[0089] The transmission coefficient can then be expressed as:

[0090]

[0091] Where: E 1i Let E be the incident wave field strength of medium 1. 3i Let be the transmitted wave field intensity of medium 3, and P and Q be the first and second transmission coefficients, respectively.

[0092]

[0093]

[0094] Therefore, the amplitude attenuation coefficient S can be calculated as follows:

[0095]

[0096] Among them, A t A represents the amplitude of the transmitted wave. i This represents the amplitude of the incident wave.

[0097] like Figure 4 As shown, in some implementations, the echo intensity of the ultra-wideband radar detection device is compensated based on the amplitude attenuation coefficient to obtain the maximum echo intensity. The first transmission time of maximum intensity in the transmitted wave and the second transmission time of maximum echo intensity are obtained. Based on the first and second transmission times, the transmission time is obtained. Figure 4 As shown in the embodiments of this application, t(ε) r d) is the first transmission time, S(ε) r The time at which d) occurs is the second transmission time.

[0098] S204, Based on propagation time, determine the detection distance between targets detected by the ultra-wideband radar detection device.

[0099] The propagation distance is obtained based on the propagation time and wave velocity. The propagation distance is then corrected based on the wall thickness and relative permittivity to arrive at the detection distance. For example... Figure 5 As shown, the target coordinates are (x1, y1), the radar antenna coordinates are (x2, y2), the wall thickness is d, and the relative permittivity of the wall is ε. rCompared to the time delay of radar wave signal propagation in free space, the presence of a wall has more factors affecting the signal propagation time delay. When the wall thickness d is very small compared to the distance L from the target to the wall, L... m It is approximately equal to d.

[0100] The detection distance L can be obtained as:

[0101]

[0102] Where, τ k，mn c is the propagation time, and c is the wave speed.

[0103] In this embodiment, the radar echo matrix is ​​input into the trained target wall parameter recognition model for recognition, which can improve the estimation accuracy of wall parameters, thereby improving the accuracy and real-time performance of wall detection, avoiding problems such as deviation in target location and the appearance of false targets, effectively estimating wall parameters, and improving the robustness and stability of wall detection.

[0104] Figure 6 This is a flowchart of a wall detection method according to an embodiment of this application, as follows: Figure 6 As shown, the method includes:

[0105] S601 controls the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured, and acquires the original radar echo matrix.

[0106] For details regarding step S601, please refer to the relevant descriptions in the above embodiments; they will not be repeated here.

[0107] S602 preprocesses the original radar echo matrix to remove interference information and obtain the radar echo matrix.

[0108] Static background clutter is removed from the original radar echo matrix to obtain the first radar echo matrix. The DC component is then removed from the first echo matrix to obtain the second radar echo matrix. Automatic gain control is applied to the second radar echo matrix based on the wall thickness to obtain the final radar echo matrix.

[0109] For example, in actual detection scenarios, the original radar echo matrix contains not only walls and the target being detected, but also a large amount of other interference clutter. Therefore, a model is created for the original radar echo matrix under real measurement conditions:

[0110] R[m,n]=r[m,n]+c[n]+w[m,n]+d[m,n]+l[m,n]+t[m,n]

[0111] Where R[m,n] is the original radar echo matrix obtained under real-world conditions; r[m,n] is the target signature signal; c[n] is static background clutter; w[m,n] is additive white noise; d[m,n] is the unstable fast-time DC component; l[m,n] is the linear trend on the slow-time axis caused by amplitude instability during radar system acquisition; and t[m,n] is the harmonics and signal distortion caused by dynamic targets. Preprocessing methods are used to remove some clutter and noise interference.

[0112] First, after removing the static background c[n] using the channel signal subtraction method, the first radar echo matrix obtained is as follows:

[0113] R′[m, 1] = R[m, 1]

[0114] R'[m, n]=R[m, n]-R[m, n-1], (m=1,...,M; n=2,...,N)

[0115] Where M and N are positive integers greater than 3, then, after filtering out the DC component d[m] using time-averaged subtraction, the second radar echo matrix is ​​obtained as follows:

[0116]

[0117] To improve the signal-to-noise ratio of detected target features, automatic gain control is used to amplify weak target feature signals in the fast time direction. The obtained radar echo matrix X(m,i) is as follows:

[0118]

[0119]

[0120] X(m, n) = g mask (m, n)R″[m, n]

[0121] After the above preprocessing steps, the noise-reduced radar echo matrix is ​​obtained.

[0122] S603 inputs the radar echo matrix into the trained target wall parameter recognition model for identification, in order to obtain the wall thickness and relative permittivity of the wall to be tested.

[0123] S604, based on the wall thickness and the relative permittivity of the wall, determines the detection distance between targets detected by the ultra-wideband radar detection device.

[0124] For details regarding steps S603 to S604, please refer to the relevant descriptions in the above embodiments; they will not be repeated here.

[0125] In this embodiment, the radar echo matrix is ​​input into the trained target wall parameter recognition model for recognition, which can improve the estimation accuracy of wall parameters, thereby improving the accuracy and real-time performance of wall detection, avoiding problems such as deviation in target location and the appearance of false targets, effectively estimating wall parameters, and improving the robustness and stability of wall detection.

[0126] Figure 7 This is a flowchart of a wall detection method according to an embodiment of this application, as follows: Figure 7 As shown, the training process of the target wall parameter recognition model includes:

[0127] S701 controls the ultra-wideband radar detection device to move in a direction parallel to multiple sample walls in order to obtain multiple sets of candidate radar echo matrices.

[0128] To achieve refined perception of target characteristics, multiple candidate radar echo matrices were acquired using IR-UWB radar under various experimental environments (indoor and outdoor, with and without obstacles, static and dynamic targets, single and multiple targets, and various detection distances) under different wall thicknesses, relative permittivity, target number, and target location. Specifically, when detection is performed without penetrating the wall, the wall parameters are set to 0 (wall thickness d = 0 and relative permittivity ε = 0). Furthermore, during data acquisition, sample diversity was maximized to ensure significant variations in wall parameters. Based on these considerations, target characteristic information was obtained under various experimental environments, resulting in a u*v input matrix, where u is the matrix length and v is the matrix width, constituting the data sample. The acquired candidate radar echo matrices are shown below. Figure 8 , Figure 9 , Figure 10 , Figure 11 As shown, where, Figure 8 The candidate radar echo matrix is ​​given by d = 2 and ε = 1.8. Figure 9 The candidate radar echo matrix is ​​given by d = 3.5 and ε = 4.5. Figure 10 The candidate radar echo matrix is ​​given by d = 2.5 and ε = 3.2. Figure 11 The candidate radar echo matrix is ​​given when d = 0 and ε = 0.

[0129] S702, for each group of candidate radar echo matrices, based on the wall thickness and relative permittivity of the sample wall corresponding to the group of candidate radar echo matrices, the group of candidate radar echo matrices is labeled to obtain sample radar echo data.

[0130] Record the wall parameters (including wall thickness d and relative permittivity ε) for each set of candidate radar echo matrices.

[0131] S703, based on sample radar echo data, trains a wall parameter recognition model based on a long short-term memory neural network to obtain a target wall parameter recognition model.

[0132] like Figure 12 As shown in the embodiments of this application, a wall parameter recognition model based on a long short-term memory neural network containing 5 layers is used as the target wall parameter recognition model for explanation. The target wall parameter recognition model includes a hidden layer, namely 3 long short-term memory (LSTM) layers (i.e., the first LSTM layer is LSTM1, the second LSTM layer is LSTM2, and the third LSTM layer is LSTM3), a fully connected layer (Dense), and an output layer (Output). The key parameters of each layer of the wall parameter recognition model based on the long short-term memory neural network are shown in Table 1.

[0133] Table 1

[0134] Layer name Input Dimensions Output Dimension Hidden number of layers Activation function Dropout Masking [X,200] [10,200] / / / LSTM1 [10,200] [10,128] 128 tanh 0.2 LSTM2 [10,128] [10,256] 256 tanh 0.2 LSTM3 [10,256] [10,256] 256 tanh 0.2 Dense 256 256 / Relu 0.3 Output 256 2 / / /

[0135] In this embodiment, based on the detection mechanism and sampling frequency of the ultra-wideband radar detection device, the number of time series features is set to 200. The input data can be a variable-length sequence. All input data first undergoes a masking layer to process the variable-length input sequence data. After processing, the variable-length input sequence is integrated into the feature length of the LSTM1 input sequence. The number of hidden layers in the three LSTM layers are 128, 256, and 256, respectively. Random deactivation (Dropout) and L2 regularization mechanisms are set to prevent overfitting. The activation function of all LSTM layers is set to the tanh function. A fully connected layer is set after all LSTM layers, with 256 features and the ReLU function as the activation function. Finally, an output layer is set with two output values, corresponding to the two feature parameters of the wall (including the wall thickness d and the relative permittivity ε).

[0136] Based on the established long short-term memory neural network-based wall parameter recognition model, the algorithm is trained. All data samples are randomly mixed and normalized, then divided into several batches. One batch of data is used for model training at a time. For each input data x... i After model computation, its output is a K-dimensional vector containing the predicted value of each output feature, i.e.:

[0137]

[0138] In this embodiment, the predicted value includes two output features (wall thickness d and relative permittivity ε), i.e., K = 2. After each update iteration of the wall parameter recognition model based on the long short-term memory neural network, a loss function needs to be calculated. Based on the task characteristics of wall parameter prediction in this project, the loss function is defined as:

[0139]

[0140] In the loss function, the first term is the RMSE (mean squared error) of the prediction result, and the second term is the regularization penalty term. Here, N is the batch size, K is the number of output features, and y... 1,k Let r represent the predicted value of the k-th output for the i-th sample. i,k This represents the true value of the k-th output of the i-th sample. λ is the regularization parameter, a small constant. The weights represent the weight parameters of each layer in the model, where l represents the layer number and n represents the number of weights in that layer. The introduction of a regularization penalty term allows the model to continuously train and optimize while taking into account the universality of the weight parameters, thus reducing the risk of overfitting.

[0141] As shown in Table 2, 80% of all training samples were used for model training and 20% for model testing, with a maximum of 80 epochs. The optimizer (Adam algorithm) was selected for optimization, with an initial learning rate of 0.003. After 10 weight parameter updates, the learning rate was decayed by a factor of 0.8 to slow down the gradient of weight parameter updates and allow for more refined model optimization. As the model continues to optimize, the loss values ​​and prediction accuracy of the training and test sets change after each epoch as shown below. Figure 13 As shown.

[0142] Table 2

[0143] parameter value Number of samples per batch 64 Total number of samples 8400 Training set ratio 80％ test set ratio 20％ Maximum number of iterations 80 Optimizer Adam Initial learning rate 0.003 Learning rate decay period 10 Learning rate decay factor 0.8 Iteration termination condition Loss (loss function) < 0.15

[0144] Following the steps outlined above, after training the wall parameter recognition model based on a long short-term memory neural network, a subset of test data is selected to test the model. The test samples are input into the model to obtain predicted wall parameters (including wall thickness d and relative permittivity ε). These predicted parameters are then compared with the actual values ​​from the test samples to verify the model's prediction effectiveness. The predicted wall thickness is shown below. Figure 14 As shown, the predicted results of the relative permittivity of the wall are as follows: Figure 15 As shown, Real represents the actual value, and Predicted represents the predicted value.

[0145] In this embodiment, a wall parameter identification model based on a long short-term memory neural network is trained based on sample radar echo data to obtain a target wall parameter identification model. This can improve the estimation accuracy of wall parameters, thereby improving the accuracy and real-time performance of wall detection, avoiding problems such as deviations in target location and the appearance of false targets. It can effectively estimate wall parameters and improve the robustness and stability of wall detection.

[0146] Figure 16 This is a schematic diagram of a wall detection method according to an embodiment of this application, as shown below. Figure 16 As shown in this embodiment, the ultra-wideband radar detection device is controlled to move in a direction parallel to the wall to be measured to obtain the original radar echo matrix. The original radar echo matrix is ​​preprocessed to remove interference information, resulting in the final radar echo matrix. This radar echo matrix is ​​then input into a trained target wall parameter recognition model for identification, yielding the wall parameter recognition results, namely the wall thickness and relative permittivity of the wall to be measured. Based on the wall thickness and relative permittivity, the amplitude attenuation coefficient of the radar signal from the ultra-wideband radar detection device is obtained as it propagates within the wall to be measured. Based on the amplitude attenuation coefficient, the propagation time of the radar signal within the wall to be measured is obtained. Based on the propagation time and in conjunction with the radar signal propagation characteristics, the detection distance between targets detected by the ultra-wideband radar detection device is determined.

[0147] Based on the same concept, this application also provides a wall detection device.

[0148] Figure 17 This is a structural block diagram of a wall detection device according to an embodiment of this application, as shown below. Figure 17 As shown, the wall detection device 1700 of this application embodiment includes:

[0149] The first acquisition module 1710 is used to control the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured to detect the target and acquire the radar echo matrix.

[0150] The second acquisition module 1720 is used to input the radar echo matrix into the trained target wall parameter recognition model for recognition, so as to obtain the wall thickness and relative permittivity of the wall to be tested.

[0151] The determination module 1730 is used to determine the detection distance between targets detected by the ultra-wideband radar detection device based on the wall thickness and the relative permittivity of the wall.

[0152] In some implementations, module 1730 is also used for:

[0153] Based on the wall thickness and the relative permittivity of the wall, the propagation time of the radar signal of the ultra-wideband radar detection device in the wall under test is obtained.

[0154] Based on propagation time, the detection distance between targets detected by the ultra-wideband radar detection device is determined.

[0155] In some implementations, module 1730 is also used for:

[0156] Based on the wall thickness and the relative permittivity of the wall, the amplitude attenuation coefficient of the radar signal of the ultra-wideband radar detection device when it propagates in the wall under test is obtained.

[0157] Based on the amplitude attenuation coefficient, the propagation time of the radar signal within the wall under test is obtained.

[0158] In some implementations, module 1730 is also used for:

[0159] The echo intensity of the ultra-wideband radar detection device is compensated based on the amplitude attenuation coefficient in order to obtain the maximum echo intensity.

[0160] Obtain the first transmission moment of maximum intensity in the transmitted wave and the second transmission moment of maximum echo intensity;

[0161] The transmission time is obtained based on the first transmission time and the second transmission time.

[0162] In some implementations, module 1730 is also used for:

[0163] The wall parameters of the detection wall are determined based on the wall thickness and the relative permittivity of the wall.

[0164] Obtain the first medium parameters and incident wave field strength of the first medium in which the ultra-wideband radar detection device is located;

[0165] Obtain the second medium parameters and transmitted wave field intensity of the second medium in which the target is located;

[0166] The first transmission coefficient and the second transmission coefficient are obtained based on the first medium parameter and the incident wave field strength, the second medium parameter and the transmitted wave field strength, and the wall parameters.

[0167] The amplitude attenuation coefficient is determined based on the first and second transmission coefficients.

[0168] In some implementations, module 1730 is also used for:

[0169] The propagation distance is obtained based on the propagation time and wave speed;

[0170] The propagation distance is corrected based on the wall thickness and the wall's relative permittivity to obtain the detection distance.

[0171] In some implementations, the first acquisition module 1710 is also used for:

[0172] Control the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured, and obtain the original radar echo matrix;

[0173] The original radar echo matrix is ​​preprocessed to remove interference information, thus obtaining the radar echo matrix.

[0174] In some implementations, the first acquisition module 1710 is also used for:

[0175] The first radar echo matrix is ​​obtained by removing static background clutter from the original radar echo matrix.

[0176] The second radar echo matrix is ​​obtained by removing the DC component from the first echo matrix;

[0177] Based on the wall thickness, automatic gain control is performed on the second radar echo matrix to obtain the radar echo matrix.

[0178] In some implementations, the wall detection device 1700 also includes a training module 1740 for:

[0179] The ultra-wideband radar detection device is controlled to move in a direction parallel to multiple sample walls in order to obtain multiple sets of candidate radar echo matrices.

[0180] For each set of candidate radar echo matrices, the wall thickness and relative permittivity of the sample wall corresponding to the set of candidate radar echo matrices are labeled to obtain sample radar echo data.

[0181] Based on sample radar echo data, a wall parameter recognition model based on a long short-term memory neural network is trained to obtain a target wall parameter recognition model.

[0182] In this embodiment, the radar echo matrix is ​​input into the trained target wall parameter recognition model for recognition, which can improve the estimation accuracy of wall parameters, thereby improving the accuracy and real-time performance of wall detection, avoiding problems such as deviation in target location and the appearance of false targets, effectively estimating wall parameters, and improving the robustness and stability of wall detection.

[0183] Based on the same concept, embodiments of this application also provide an electronic device.

[0184] Figure 18 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 18As shown, the electronic device 1800 includes a memory 1801, a processor 1802, and a computer program product stored in the memory 1801 and capable of running on the processor 1802. When the processor executes the computer program, it implements the aforementioned wall detection method.

[0185] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0186] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0187] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0188] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0189] Based on the same concept, embodiments of this application also provide a computer-readable storage medium storing computer instructions thereon, wherein the computer instructions are used to cause a computer to execute the wall detection method in the above embodiments.

[0190] Based on the same concept, this application also provides a computer program product, including a computer program that, when executed by a processor, describes the wall detection method in the above embodiments.

[0191] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. This application can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0192] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0193] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0194] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A wall detection method, characterized in that, include: The ultra-wideband radar detection device is controlled to move in a direction parallel to the wall to be measured to detect the target and obtain the radar echo matrix. The radar echo matrix is ​​input into the trained target wall parameter recognition model for recognition, so as to obtain the wall thickness and relative permittivity of the wall to be tested; Based on the wall thickness and the relative permittivity of the wall, the detection distance between the targets detected by the ultra-wideband radar detection device is determined; The step of determining the detection distance between targets detected by the ultra-wideband radar detection device based on the wall thickness and the relative permittivity of the wall includes: obtaining the propagation time of the radar signal of the ultra-wideband radar detection device within the wall to be tested according to the wall thickness and the relative permittivity of the wall; and determining the detection distance between targets detected by the ultra-wideband radar detection device based on the propagation time. The step of obtaining the propagation time of the radar signal of the ultra-wideband radar detection device within the wall under test based on the wall thickness and the relative permittivity of the wall includes: obtaining the amplitude attenuation coefficient of the radar signal of the ultra-wideband radar detection device propagating within the wall under test based on the wall thickness and the relative permittivity of the wall; and obtaining the propagation time of the radar signal within the wall under test based on the amplitude attenuation coefficient. The step of obtaining the propagation time of the radar signal within the wall under test based on the amplitude attenuation coefficient includes: compensating the echo intensity of the ultra-wideband radar detection device based on the amplitude attenuation coefficient to obtain the maximum echo intensity; obtaining a first transmission time of the maximum intensity in the transmitted wave and a second transmission time of the maximum echo intensity; and obtaining the propagation time based on the first transmission time and the second transmission time.

2. The method according to claim 1, characterized in that, The process of obtaining the amplitude attenuation coefficient includes: The wall parameters of the wall to be tested are determined based on the wall thickness and the relative permittivity of the wall. Obtain the first medium parameters and incident wave field strength of the first medium in which the ultra-wideband radar detection device is located; Obtain the second medium parameters and transmitted wave field strength of the second medium in which the target is located; The first transmission coefficient and the second transmission coefficient are obtained based on the first medium parameters and the incident wave field strength, the second medium parameters and the transmitted wave field strength, and the wall parameters. The amplitude attenuation coefficient is determined based on the first transmission coefficient and the second transmission coefficient.

3. The method according to claim 1, characterized in that, Determining the detection distance between targets detected by the ultra-wideband radar detection device based on the propagation time includes: The propagation distance is obtained based on the propagation time and wave speed. The propagation distance is corrected based on the wall thickness and the wall's relative permittivity to obtain the detection distance.

4. The method according to any one of claims 1-3, characterized in that, The step of controlling the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured and acquiring the radar echo matrix includes: The ultra-wideband radar detection device is controlled to move in a direction parallel to the wall to be measured, and the original radar echo matrix is ​​obtained. The original radar echo matrix is ​​preprocessed to remove interference information, thus obtaining the radar echo matrix.

5. The method according to claim 4, characterized in that, The preprocessing of the original radar echo matrix to remove interference information and obtain the radar echo matrix includes: The first radar echo matrix is ​​obtained by removing static background clutter from the original radar echo matrix. The second radar echo matrix is ​​obtained by removing the DC component from the first radar echo matrix. Based on the wall thickness, automatic gain control is performed on the second radar echo matrix to obtain the radar echo matrix.

6. The method according to any one of claims 1-3, characterized in that, The training process of the target wall parameter recognition model includes: The ultra-wideband radar detection device is controlled to move in a direction parallel to multiple sample walls in order to obtain multiple sets of candidate radar echo matrices. For each set of candidate radar echo matrices, based on the wall thickness and relative permittivity of the sample wall corresponding to the set of candidate radar echo matrices, the set of candidate radar echo matrices is labeled to obtain sample radar echo data; Based on the sample radar echo data, the wall parameter recognition model based on long short-term memory neural network is trained to obtain the target wall parameter recognition model.

7. A wall detection device, characterized in that, include: The first acquisition module is used to control the ultra-wideband radar detection device to move in a direction parallel to the wall to be measured to detect the target and acquire the radar echo matrix. The second acquisition module is used to input the radar echo matrix into the trained target wall parameter recognition model for recognition, so as to obtain the wall thickness and relative permittivity of the wall to be tested. The determination module is used to determine the detection distance between targets detected by the ultra-wideband radar detection device based on the wall thickness and the wall relative permittivity. The determining module is further configured to: obtain the propagation time of the radar signal of the ultra-wideband radar detection device within the wall to be tested based on the wall thickness and the relative permittivity of the wall; and determine the detection distance between the targets detected by the ultra-wideband radar detection device based on the propagation time. The determining module is further configured to: obtain the amplitude attenuation coefficient of the radar signal of the ultra-wideband radar detection device when it propagates in the wall under test according to the wall thickness and the relative permittivity of the wall; and obtain the propagation time of the radar signal in the wall under test based on the amplitude attenuation coefficient. The determining module is further configured to: compensate the echo intensity of the ultra-wideband radar detection device based on the amplitude attenuation coefficient to obtain the maximum echo intensity; obtain the first transmission time of the maximum intensity in the transmitted wave and the second transmission time of the maximum echo intensity; and obtain the propagation time based on the first transmission time and the second transmission time.

Images