Target detection method and device based on deep convolutional feature matrix manifold network

Abstract

The invention relates to a target detection method and device based on a deep convolutional feature matrix manifold network. The method comprises the following steps: using a simulation target signal driven by a Doppler steering vector in sea clutter data obtained in advance to obtain echo data of a simulation target containing different signal-to-noise ratios, and obtaining a time-frequency characteristic pattern of the echo data by using a time-frequency analysis mode, and extracting high-dimensional nonlinear features of the time-frequency feature map by adopting a pre-trained convolutional neural network to obtain a deep convolutional feature matrix, and training a pre-constructed deep manifold network by taking the deep convolutional feature matrix as a sample so as to enable the deep manifold network to detect a target in the echo data to be detected. The method can improve the radar target detection performance.

Description

Technical Field

[0001] The present application relates to the field of radar target detection technology, and in particular to a target detection method and device based on a deep convolutional feature matrix manifold network. Background Art

[0002] Radar systems are the core of modern surveillance and defense systems and are widely used in aviation, aerospace, and weather forecasting. The main challenge of radar target detection is to accurately detect and identify targets in complex environmental noise and interference. Clutter, which is the non-target reflected signal received by the radar system, is the main source of interference that affects radar detection performance. Different environments, such as sea, mountainous areas, and cities, will produce different types of clutter, which will affect radar target detection to varying degrees.

[0003] Traditional detection technology based on statistical characteristics has high requirements on the target signal-to-noise ratio, slow detection speed, and many false alarm and model mismatch problems. The detection performance cannot meet the needs of modern complex and diverse perception environments. Summary of the invention

[0004] Based on this, it is necessary to provide a target detection method and device based on a deep convolutional feature matrix manifold network to address the above technical problems.

[0005] A target detection method based on a deep convolutional feature matrix manifold network, the method comprising:

[0006] Using a simulated target signal driven by a Doppler steering vector in pre-acquired sea clutter data to acquire echo data of simulated targets with different signal-to-noise ratios; the Doppler steering vector is represented by a target signal amplitude parameter, the magnitude of which is related to an average magnitude of the sea clutter data, a signal-to-noise ratio, an initial phase, and a scattering coefficient;

[0007] A time-frequency feature map of the echo data is obtained by using a time-frequency analysis method, and a pre-trained convolutional neural network is used to extract high-dimensional nonlinear features of the time-frequency feature map to obtain a deep convolution feature matrix;

[0008] The deep convolution feature matrix is ​​used as a sample to train a pre-constructed deep manifold network, so that the deep manifold network can detect the target in the echo data to be detected.

[0009] In one embodiment, the method further includes: using Doppler-guided vector-driven simulated target signals in pre-acquired sea clutter data to be:

[0010]

[0011] α=β×mean(S)cosθ

[0012] Where S is the sea clutter data, s is the Doppler steering vector, α is the target signal amplitude parameter, θ represents the initial phase, β represents the scattering coefficient, N represents the number of target signal pulses, and f d represents the normalized Doppler frequency.

[0013] In one of the embodiments, it also includes: using STFT time-frequency analysis method to obtain the time-frequency characteristic diagram of the echo data.

[0014] In one of the embodiments, it also includes: inputting the time-frequency feature map into a pre-trained convolutional neural network, and using the convolution layer output of the convolutional neural network as a high-dimensional nonlinear convolution feature; symmetric positive definite matrices of all channels of the high-dimensional nonlinear convolution feature are averaged to obtain a deep convolution feature matrix.

[0015] In one embodiment, the method further includes: symmetric positive definite matrix of all channels of the high-dimensional nonlinear convolution feature is:

[0016]

[0017] Wherein, X0 represents the matrix of each channel in the high-dimensional nonlinear convolution feature, X2 represents the high-dimensional nonlinear convolution feature after positive definiteness, ε represents the correction threshold, I is the unit matrix, and U1 represents the orthogonal matrix;

[0018] The average value of the positively definite high-dimensional nonlinear convolution features is:

[0019]

[0020] Among them, S1(i,j,k) represents the high-dimensional nonlinear convolution feature with k channels after positive quantization, and S2(i,j) represents the deep convolution feature matrix.

[0021] In one of the embodiments, the number of convolutional layers of the convolutional neural network is 1 or 2.

[0022] In one embodiment, when the number of convolutional layers of the convolutional neural network is 1, the number of network layers of the deep manifold network when the signal-to-noise ratio is greater than the first signal-to-noise ratio is 4, the number of network layers when the signal-to-noise ratio is less than the first signal-to-noise ratio and greater than the second signal-to-noise ratio is 8, and the number of network layers when the signal-to-noise ratio is less than the second signal-to-noise ratio is 6; when the number of convolutional layers of the convolutional neural network is 2, the number of network layers of the deep manifold network is 8.

[0023] A target detection device based on a deep convolutional feature matrix manifold network, the device comprising:

[0024] A data acquisition module, used for using a simulated target signal driven by a Doppler steering vector in pre-acquired sea clutter data to acquire echo data of a simulated target with different signal-to-noise ratios; the Doppler steering vector is represented by a target signal amplitude parameter, and the magnitude of the target signal amplitude parameter is related to the average magnitude of the sea clutter data, the signal-to-noise ratio, the initial phase and the scattering coefficient;

[0025] A deep convolution feature matrix construction module is used to obtain a time-frequency feature map of the echo data by using a time-frequency analysis method, and to extract high-dimensional nonlinear features of the time-frequency feature map by using a pre-trained convolutional neural network to obtain a deep convolution feature matrix;

[0026] The target detection module is used to train a pre-constructed deep manifold network using the deep convolution feature matrix as a sample, so that the deep manifold network can detect the target in the echo data to be detected.

[0027] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0028] Using a simulated target signal driven by a Doppler steering vector in pre-acquired sea clutter data to acquire echo data of simulated targets with different signal-to-noise ratios; the Doppler steering vector is represented by a target signal amplitude parameter, the magnitude of which is related to an average magnitude of the sea clutter data, a signal-to-noise ratio, an initial phase, and a scattering coefficient;

[0029] A time-frequency feature map of the echo data is obtained by using a time-frequency analysis method, and a pre-trained convolutional neural network is used to extract high-dimensional nonlinear features of the time-frequency feature map to obtain a deep convolution feature matrix;

[0030] The deep convolution feature matrix is ​​used as a sample to train a pre-constructed deep manifold network, so that the deep manifold network can detect the target in the echo data to be detected.

[0031] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0032] Using a simulated target signal driven by a Doppler steering vector in pre-acquired sea clutter data to acquire echo data of simulated targets with different signal-to-noise ratios; the Doppler steering vector is represented by a target signal amplitude parameter, the magnitude of which is related to an average magnitude of the sea clutter data, a signal-to-noise ratio, an initial phase, and a scattering coefficient;

[0033] A time-frequency feature map of the echo data is obtained by using a time-frequency analysis method, and a pre-trained convolutional neural network is used to extract high-dimensional nonlinear features of the time-frequency feature map to obtain a deep convolution feature matrix;

[0034] The deep convolution feature matrix is ​​used as a sample to train a pre-constructed deep manifold network, so that the deep manifold network can detect the target in the echo data to be detected.

[0035] The target detection method and device based on the deep convolution feature matrix manifold network use the Doppler steering vector-driven simulated target signal in the pre-acquired sea clutter data to obtain the echo data of the simulated target with different signal-to-noise ratios, obtain the time-frequency feature map of the echo data by time-frequency analysis, extract the high-dimensional nonlinear features of the time-frequency feature map by using a pre-trained convolutional neural network, obtain the deep convolution feature matrix, and train the pre-constructed deep manifold network with the deep convolution feature matrix as a sample, so that the deep manifold network can detect the target in the echo data to be detected. The present invention proposes to use a convolutional neural network to first extract the high-dimensional nonlinear features from the feature map, and then use a deep matrix manifold network to reduce the dimension and aggregate it to expand more features, thereby improving the radar target detection performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 1 is a flow chart of a target detection method based on a deep convolutional feature matrix manifold network in one embodiment;

[0037] Figure 2 1 is a diagram of sea clutter data in an embodiment; wherein (a) is a diagram of nearby sea clutter data, and (b) is a diagram of selected sea clutter data;

[0038] Figure 3 1 is a signal diagram after adding a simulation target in one embodiment; wherein (a) is a signal diagram after adding a 0dB simulation target, and (b) is a signal diagram after adding a -5dB simulation target;

[0039] Figure 4 A diagram of the SPDNet structure in one embodiment;

[0040] Figure 5 : is a STFT characteristic diagram of the signal-to-noise ratio in an embodiment; wherein (a) is no target, (b) is 0 dB, (c) is -5 dB, (d) is -10 dB, (e) is -15 dB, and (f) is -20 dB;

[0041] Figure 6 is a STFT characteristic matrix of the signal-to-noise ratio in one embodiment; wherein (a) is no target, (b) is 0 dB, (c) is -5 dB, (d) is -10 dB, (e) is -15 dB, and (f) is -20 dB;

[0042] Figure 7 A convolution feature reconstruction flow chart in one embodiment;

[0043] Figure 8 Schematic diagram of selecting the signal-to-noise ratio of a training fusion network in an embodiment; wherein (a) represents fusion network 1, and (b) represents fusion network 2;

[0044] Fig. 9 is a detection probability graph of different networks in one embodiment;

[0045] Fig.10 1 is a detection probability diagram of a fusion network under different learning rates in one embodiment; wherein (a) represents fusion network 1, and (b) represents fusion network 2;

[0046] Fig.11 Detection probability diagram of the fusion network under different weight attenuation in one embodiment; (a) represents fusion network 1, and (b) represents fusion network 2;

[0047] Fig.12 Detection probability diagrams of fused networks with different numbers of network layers in one embodiment; (a) represents fused network 1, and (b) represents fused network 2;

[0048] Fig.13 is a structural block diagram of a target detection device based on a deep convolutional feature matrix manifold network in one embodiment;

[0049] Fig.14 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0051] In one embodiment, Figure 1 As shown, a target detection method based on a deep convolutional feature matrix manifold network is provided, comprising the following steps:

[0052] Step 102 : using a Doppler-steering vector-driven simulated target signal in pre-acquired sea clutter data to acquire echo data of a simulated target having different signal-to-noise ratios.

[0053] The Doppler steering vector is represented by the target signal amplitude parameter, and the size of the target signal amplitude parameter is related to the average size of the sea clutter data, the signal-to-noise ratio, the initial phase and the scattering coefficient.

[0054] Specifically, the radar of the present invention is an X-band solid-state power amplifier surveillance / navigation radar with a transmission time of 40ns-100us and a 360° all-round scanning in the horizontal plane. The radar technical parameters are shown in Table 1.

[0055] Table 1 Radar parameters

[0056]

[0057]

[0058] The present invention uses the "20191012112446_01_staring.mat" in the public data to construct a network data set, which contains 10,000 pulse trains and a total of 5,250 distance unit sea clutter data. The radar echo data are all pulse compressed data. In the present invention, the 186th to 190th distance units are selected as experimental clutter sample data, such as Figure 2 (a) The distance unit near the data selected by the present invention is intercepted, wherein the data in the red frame is the experimental data selected by the present invention. Figure 2 (b) shows the sea clutter data after selecting the range unit.

[0059] In this step, the signal-to-noise ratio is a variable in the range of [0dB, -20dB], the initial phase θ of the target is 0, and the scattering coefficient β is 0.8. The N of the simulation data set is 10000, and f d Set to 0.15.

[0060] Figure 3 (a) and (b) show the signal graphs after adding 0dB and -5dB simulated targets. It can be clearly seen that the data graph changes more after adding a large signal-to-noise ratio target, but the data graph hardly changes after adding a signal with a relatively small signal-to-noise ratio. The data graphs containing clutter and target signals are almost indistinguishable from the data graphs containing target signals under naked eye observation. Since the added steering vector is a phase factor, the specific position of the signal cannot be directly observed on the signal amplitude graph.

[0061] Step 104, using a time-frequency analysis method to obtain a time-frequency feature map of the echo data, using a pre-trained convolutional neural network to extract high-dimensional nonlinear features of the time-frequency feature map, and obtaining a deep convolution feature matrix.

[0062] In this step, for the matrix A n×n Diagonalization and eigenvalue decomposition can be performed:

[0063] U -1 AU=Σ

[0064] A=U -1 ΣU

[0065] The eigenvalues ​​and eigenvectors of a square matrix are the essential characteristics of a matrix. m×n It is impossible to perform eigenvalue decomposition, so the singular value decomposition method is used to decompose the non-square matrix B m×n Decompose. Since the singular value decomposition originates from the nature of the matrix, it is applicable to all real matrices:

[0066] BV=UΣ

[0067] B=UΣV -1 =UΣV T

[0068] Among them, U is an m×m orthogonal matrix, Σ is an m×n orthogonal matrix, and V is an n×n orthogonal matrix. The symmetric positive definite method of any matrix X0 is to transpose the matrix X0 and its conjugate When multiplied, the generated matrix X1 must have conjugate symmetry, and then by correcting the negative eigenvalues, the time-frequency characteristic matrix is ​​obtained.

[0069] Step 106 , using the deep convolution feature matrix as a sample, training a pre-constructed deep manifold network, so that the deep manifold network can detect the target in the echo data to be detected.

[0070] In the above-mentioned target detection method based on deep convolution feature matrix manifold network, a simulated target signal driven by a Doppler steering vector is used in the pre-acquired sea clutter data to obtain echo data of simulated targets with different signal-to-noise ratios, and a time-frequency feature map of the echo data is obtained by time-frequency analysis. A pre-trained convolutional neural network is used to extract high-dimensional nonlinear features of the time-frequency feature map to obtain a deep convolution feature matrix. The deep convolution feature matrix is ​​used as a sample to train a pre-constructed deep manifold network so that the deep manifold network detects targets in the echo data to be detected. The present invention proposes to use a convolutional neural network to first extract high-dimensional nonlinear features from the feature map, and then use a deep matrix manifold network to reduce dimensionality and aggregate, so as to expand more features, thereby improving radar target detection performance.

[0071] It should be introduced that the deep manifold network is a deep learning algorithm that processes matrix manifolds. It uses matrix data as input to the neural network to perform nonlinear training on the matrix manifold. Among these networks, the deep manifold network (SPDNet) is a commonly used structure.

[0072] Figure 4It is the structural diagram of SPDNet. The generated SPD data is sent to the input layer, and then after being processed by multiple bilinear mapping layers and feature correction layers, the dimension of the SPD matrix gradually decreases and the features gradually aggregate. Finally, these matrices are mapped to Euclidean space through logarithmic Euclidean measurement and classified in space. The number of dimensionality reduction and aggregation iteration layers of the original network in the present invention is 6, namely three bilinear mapping layers and three feature correction layers.

[0073] Similar to the classic convolutional neural network, SPDNet contains two key functional layers: Bilinear Map Layer (BiMap Layer) and Eigenvalue Rectification Layer (ReEig Layer). After processing by these layers, the original SPD matrix will be converted into a new matrix with smaller dimensions and more significant features. This conversion enhances the distinction between targets and clutter on the matrix manifold.

[0074] The purpose of the bilinear mapping layer is to make the input SPD matrix smaller and more distinguishable. Through the bilinear mapping layer, the input on the original SPD manifold is converted into a new input to form a new SPD manifold. In other words, the data space on each bilinear mapping layer corresponds to an SPD manifold. The calculation formula of the bilinear mapping layer is defined as follows:

[0075]

[0076] Among them, X k-1 is the input matrix of the k-1th layer, X k is the output matrix of the kth layer, W k is a weight matrix with full row rank.

[0077] The purpose of the feature correction layer is to correct the input SPD matrix, correcting the negative eigenvalue to a positive number, ensuring the symmetric positive definiteness of the matrix, and thus retaining more significant features. The calculation formula of the eigenvalue correction layer is defined as follows:

[0078]

[0079] Among them, U k-1 and Σ k-1 By eigenvalue decomposition Obtained, ε is the correction threshold, and I is the identity matrix.

[0080] After passing through multiple bilinear mapping layers and eigenvalue correction layers, the SPD matrix enters the Log-Euclidean calculation layer (Eigenvalue Logarithm, LogEig Layer). The logarithmic Euclidean metric gives the SPD matrix manifold a Lie group structure that follows the characteristics of the Riemann manifold, and expands the SPD matrix manifold into a flat Euclidean space to form a vector

[48] . The calculation formula of the Log-Euclidean calculation layer is as follows:

[0081]

[0082] Among them, log(Σ k-1 ) is a diagonal matrix of logarithmic eigenvalues. After logarithmic Euclidean measurement processing, the input SPD sample matrix is ​​classified using a linear classification method.

[0083] In classification problems, the final output layer is generally a Softmax layer. The vectors that pass through the LogEig Layer are converted into a probability distribution in the range of [0,1] and a sum of 1 by the Softmax function. The calculation formula of Softmax is as follows:

[0084]

[0085] Among them, x i is the output of the ith node, C is the number of output nodes, that is, the number of classifications, where is the probability of having a target or not having a target, and the final detection probability is obtained from the classification result.

[0086] In addition to the forward propagation process, the back propagation process of SPDNet uses the Stochastic Gradient Descent (SGD) algorithm, which is a simple but very effective method that can be used for both classification and regression calculations. SGD randomly extracts a group from the samples, updates them once according to the gradient after training, and then extracts another group and updates them again. The randomness here means that during each iteration, the samples are randomly shuffled to effectively reduce the parameter update offset problem caused by the samples. The calculation formula is as follows:

[0087]

[0088] Among them, α is the network learning rate, θ j is the jth coefficient of the prediction function, x(i) is the input data, y(i) is the actual value of each sample in the data set, and h θ is the prediction function, and the predicted y(i) value is calculated according to θ for each input x(i).

[0089] In one embodiment, the simulated target signal using Doppler steering vector drive in pre-acquired sea clutter data is:

[0090]

[0091] α=β×mean(S)cosθ

[0092] Where S is the sea clutter data, s is the Doppler steering vector, α is the target signal amplitude parameter, θ represents the initial phase, β represents the scattering coefficient, N represents the number of target signal pulses, and f d represents the normalized Doppler frequency.

[0093] In another embodiment, the STFT time-frequency analysis method is used to obtain the time-frequency characteristic diagram of the echo data.

[0094] STFT time-frequency analysis method:

[0095] Specifically, after superimposing the simulated targets with different signal-to-noise ratios on the radar echo data of each range unit, a Hamming window with a length of 25, an overlap length of 24, and a sampling point number of 100 is used to perform short-time Fourier transform, and the size of the generated time-frequency graph is 100×76. Randomly taking the 77th time-frequency graph of the 187th range unit as an example, the following is obtained: Figure 5 Feature map of the signal-to-noise ratio of the parts without and with targets.

[0096] Observe the time-frequency characteristic diagram with and without targets. From the perspective of amplitude, the clutter is concentrated in the low-frequency part, and the clutter spectrum line distribution amplitude is large. The target is distributed at the frequency f d = 0.15, the amplitude decreases as the signal-to-noise ratio decreases. From the graphical characteristics, the clutter is irregularly distributed, with all frequencies distributed, while the time-frequency distribution containing the target is a straight line. In the STFT experimental results, the target and clutter frequency distributions show a corresponding relationship.

[0097] For the feature graphs with different signal-to-noise ratios, as the signal-to-noise ratio decreases, the target's time-frequency feature amplitude decreases significantly. When the signal-to-noise ratio drops to -10dB, only a small part of the signal can be seen, and if the signal-to-noise ratio continues to decrease, it is no longer possible to distinguish the target from the clutter in the image with the naked eye.

[0098] The obtained time-frequency graph is symmetrically positively defined to convert it into a time-frequency feature matrix with a size of 100×100, such as Figure 6 shown.

[0099] It can be seen from the STFT feature matrix that on the matrix diagonal, the center point is low-frequency clutter, and the upper left point is the target. Corresponding to the feature map, when the signal-to-noise ratio is reduced to -10dB, the target can still be distinguished from the image color depth by the naked eye, but when the signal-to-noise ratio is further reduced, it is difficult to distinguish the target from the clutter from the feature matrix.

[0100] In one of the embodiments, the time-frequency feature map is input into a pre-trained convolutional neural network, and the convolution layer output of the convolutional neural network is used as a high-dimensional nonlinear convolution feature; the matrices of all channels of the high-dimensional nonlinear convolution feature are symmetrically positive-defined and then averaged to obtain a deep convolution feature matrix.

[0101] Specifically, the convolutional pooling layer in the convolutional neural network is used to extract high-dimensional nonlinear features of the target, while similar structures in the deep matrix manifold network: bilinear mapping layer and feature correction layer, only reduce the dimension of the matrix to remove redundant information. However, for weak targets, blindly reducing the dimension and aggregating cannot achieve better detection performance. Therefore, it is proposed to use a convolutional neural network to first extract high-dimensional nonlinear features from the feature map, and then use a deep matrix manifold network to reduce the dimension and aggregate to expand more features.

[0102] The convolution module of any convolutional neural network can extract high-dimensional nonlinear features. In order to carry out the experiment of the proposed method, GoogleNet with fewer network parameters is taken as an example. GoogleNet has two convolutional pooling modules before entering the Inception layer. The corresponding input and output sizes are shown in Table 2. To facilitate the research, the trained GoogleNet is used for preliminary feature extraction of the generated STFT time-frequency graph through transfer learning, and the output of these two convolutional pooling layers is used as the extracted high-dimensional nonlinear convolutional features. The process is as follows: Figure 7 As shown in Figure 3, the convolutional feature extraction module is also applicable to other convolutional neural networks.

[0103] Table 2 GoogleNet convolutional layer input and output size

[0104]

[0105] Fusion network 1 refers to the process of directly putting features into SPDNet for processing after extracting features through the Pool_1 layer, while fusion network 2 refers to the process of extracting features through the Pool_1 layer and the Pool_2 layer and then putting them into SPDNet for processing. The difference between the two is a convolutional pooling module.

[0106] The sample data selected in the experiment are all the extracted STFT time-frequency feature maps as the target of convolution feature extraction. After the convolution operation of fusion network 1 and fusion network 2, all the time-frequency feature maps are extracted as convolution feature matrices.

[0107] For fusion network 1, after training on multiple signal-to-noise ratio data, the fusion network 1 trained with the target and clutter data set with a signal-to-noise ratio of -7dB has the best detection performance, such as Figure 8 As shown in (a). For fusion network 2, the detection performance of the target and clutter data set with a signal-to-noise ratio of -8dB is the best, as shown in Figure 8 (b) As the signal-to-noise ratio continues to increase, the network still fails to converge and stops training. This experimental phenomenon proves that the proposed method can break through the training lower limit of -5dB signal-to-noise ratio samples, thereby improving the detection lower limit of the network under weak targets.

[0108] The convolution feature matrix corresponding to the optimal network signal-to-noise ratio is classified into 40% training samples and 20% validation samples. The remaining 40% test samples of all signal-to-noise ratio data are used to test the detection probability distribution of fusion network 1 and fusion network 2.

[0109] Since the output of the convolution layer is a feature matrix with a certain channel depth, it is necessary to compress the three-dimensional feature depth into a two-dimensional feature map. After the input image enters the convolutional neural network, although the convolution kernel of each layer is the same, the number of convolution calculations is different, the feature maps generated are also different, and the final output is the superposition of the feature maps of all channels. The matrix of each channel output by the convolution layer is symmetric and positively defined, and then the matrices of each channel are summed and averaged, so that the three-dimensional feature matrix is ​​compressed from a size of 57×57×64 or 57×57×192 to a two-dimensional feature matrix of size 57×57.

[0110] In one embodiment, the matrix of all channels of the high-dimensional nonlinear convolution feature is symmetric and positively defined as follows:

[0111]

[0112] Among them, X0 represents the matrix of each channel in the high-dimensional nonlinear convolution feature, X2 represents the high-dimensional nonlinear convolution feature after positive definiteness, ε represents the correction threshold, I is the unit matrix, and U1 represents the orthogonal matrix; the average value of the high-dimensional nonlinear convolution feature after positive definiteness is:

[0113]

[0114] Among them, S1(i,j,k) represents the high-dimensional nonlinear convolution feature with k channels after positive quantization, and S2(i,j) represents the deep convolution feature matrix.

[0115] Among them, S1 is the three-dimensional feature matrix with channel depth, S2 is the processed two-dimensional feature matrix, and k is the channel depth.

[0116] From the feature matrix diagram of fusion network 1, we can see that the center point on the diagonal of the matrix is ​​low-frequency clutter, and the other points on the diagonal are targets. However, since the generated feature map is too abstract after convolution extraction, when the signal-to-noise ratio is less than -10dB, it is difficult to clearly distinguish clutter and target with the naked eye.

[0117] From the feature matrix diagram of the higher signal-to-noise ratio in fusion network 2, we can see that the clutter is in the center of the matrix, while the target features are mainly distributed at both ends of the central diagonal. The feature matrix of fusion network 2 after one more convolution layer has a greater average depth than fusion network 1, showing more abstract features, and it is impossible to distinguish between targets and clutter even at a higher signal-to-noise ratio.

[0118] In one embodiment, the number of convolutional layers of the convolutional neural network is 1 or 2.

[0119] In one embodiment, the number of network layers of the deep manifold network is 4 when the signal-to-noise ratio is greater than the first signal-to-noise ratio, the number of network layers is 8 when the signal-to-noise ratio is less than the first signal-to-noise ratio and greater than the second signal-to-noise ratio, and the number of network layers is 6 when the signal-to-noise ratio is less than the second signal-to-noise ratio; when the number of convolutional layers of the convolutional neural network is 2, the number of network layers of the deep manifold network is 8.

[0120] In summary, based on the above method, the results are analyzed:

[0121] The initial network training hyperparameters of fusion network 1 and fusion network 2 are shown in Table 3. The two fusion network trainings use a larger learning rate to ensure that the fusion network can converge better with the trained SPDNet when other parameters are the same.

[0122] Table 3. Fusion network model training hyper parameters

[0123]

[0124] from Fig. 9 It can be seen that the target detection performance of both fusion network 1 and fusion network 2 is significantly better than that of SPDNet. When the detection probability is 70%, the corresponding signal-to-noise ratio of fusion network 1 is -9.46dB, fusion network 2 is -9.3dB, and SPDNet is only -8.12dB. Therefore, under the same conditions, the detection performance of fusion network 2 is 1.34dB higher than that of SPDNet. For the two fusion networks, because of one more convolution pooling process, the time-frequency features are further extracted and aggregated, and the detection performance of fusion network 2 is 0.16dB higher than that of fusion network 1.

[0125] The experimental results show that both fusion network 1 and fusion network 2, combined with the ability to extract high-dimensional nonlinear features in convolutional neural networks, increase the feature information in the features, break through the lower limit of the signal-to-noise ratio of target detection only through dimensionality reduction aggregation, and prove that the detection performance of the proposed fusion network method is better than that of the deep manifold network SPDNet.

[0126] Learning Rate:

[0127] Under the condition of ensuring the convergence of the fusion network training loss function, the influence of the learning rate on the fusion network detection performance is studied. Fig.10 As shown in the figure, for fusion network 1, the detection performance is best under all signal-to-noise ratios when the learning rate is 0.01. If the learning rate is too large, the network will overfit. If the learning rate is too small, the network may underfit. Both will reduce the detection performance of fusion network 1. For fusion network 2, when the signal-to-noise ratio is greater than -9dB, the detection performance is better when the learning rate is 0.01. When the signal-to-noise ratio is less than -9dB, the detection performance is better when the learning rate is 0.005. When the learning rate is 0.02, there will be a certain overfitting problem, so the detection performance is not as good as a smaller learning rate.

[0128] Weight decay:

[0129] like Fig.11 As shown in the figure, for fusion network 1, the detection performance is the best under all signal-to-noise ratios when the weight decay is 0.0005. However, for fusion network 2, the network training does not produce overfitting, and the network training effect is better.

[0130] Network layers:

[0131] This chapter selects comparative experiments with 4, 6, 8 and 10 network layers to study the ideal number of network layers for fusion network 1 and fusion network 2, ensuring that the final dimension reduction aggregation matrix size of all network layers is the same. The dimension reduction matrix size corresponding to the network layer number is shown in Table 3.

[0132] Table 4.3 Fusion network layer number reduction matrix

[0133]

[0134] like Fig.12 As shown, for fusion network 1, the performance difference of different network layers is small, but when the signal-to-noise ratio is greater than -7.67dB, the detection performance of network layer 4 is the best, when the signal-to-noise ratio is less than -8dB and greater than -7.61dB, the detection performance of network layer 8 is the best, and when the signal-to-noise ratio is less than -8dB, the detection performance of network layer 6 is the best. For fusion network 2, the detection performance of network layer 8 is the best under all signal-to-noise ratios.

[0135] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in the present invention, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0136] In one embodiment, Fig.13 As shown, a target detection device based on a deep convolution feature matrix manifold network is provided, comprising: a data acquisition module 1302, a deep convolution feature matrix construction module 1304 and a target detection module 1306, wherein:

[0137] The data acquisition module 1302 is used to use the simulated target signal driven by the Doppler steering vector in the pre-acquired sea clutter data to acquire the echo data of the simulated target with different signal-to-noise ratios; the Doppler steering vector is represented by the target signal amplitude parameter, and the magnitude of the target signal amplitude parameter is related to the average magnitude of the sea clutter data, the signal-to-noise ratio, the initial phase and the scattering coefficient;

[0138] A deep convolution feature matrix construction module 1304 is used to obtain a time-frequency feature map of the echo data by using a time-frequency analysis method, and to extract high-dimensional nonlinear features of the time-frequency feature map by using a pre-trained convolutional neural network to obtain a deep convolution feature matrix;

[0139] The target detection module 1306 is used to train a pre-constructed deep manifold network using the deep convolution feature matrix as a sample, so that the deep manifold network can detect the target in the echo data to be detected.

[0140] In one embodiment, the data acquisition module 1302 is further configured to use the simulated target signal driven by the Doppler steering vector in the pre-acquired sea clutter data as:

[0141]

[0142] α=β×mean(S)cosθ

[0143] Where S is the sea clutter data, s is the Doppler steering vector, α is the target signal amplitude parameter, θ represents the initial phase, β represents the scattering coefficient, N represents the number of target signal pulses, and f d represents the normalized Doppler frequency.

[0144] In one of the embodiments, the deep convolution feature matrix construction module 1304 is also used to obtain the time-frequency feature map of the echo data by adopting the STFT time-frequency analysis method.

[0145] In one embodiment, the deep convolution feature matrix construction module 1304 is also used to input the time-frequency feature map into a pre-trained convolutional neural network, and use the convolution layer output of the convolutional neural network as a high-dimensional nonlinear convolution feature; the matrices of all channels of the high-dimensional nonlinear convolution feature are symmetrically positive-defined and then averaged to obtain a deep convolution feature matrix.

[0146] In one embodiment, the deep convolution feature matrix construction module 1304 is also used to symmetric positively define the matrices of all channels of the high-dimensional nonlinear convolution feature as follows:

[0147]

[0148] Among them, X0 represents the matrix of each channel in the high-dimensional nonlinear convolution feature, X2 represents the high-dimensional nonlinear convolution feature after positive definiteness, ε represents the correction threshold, I is the unit matrix, and U1 represents the orthogonal matrix; the average value of the high-dimensional nonlinear convolution feature after positive definiteness is:

[0149]

[0150] Among them, S1(i,j,k) represents the high-dimensional nonlinear convolution feature with k channels after positive quantization, and S2(i,j) represents the deep convolution feature matrix.

[0151] In one of the embodiments, the number of convolutional layers of the convolutional neural network is 1 or 2.

[0152] In one embodiment, when the number of convolutional layers of the convolutional neural network is 1, the number of network layers of the deep manifold network when the signal-to-noise ratio is greater than the first signal-to-noise ratio is 4, the number of network layers when the signal-to-noise ratio is less than the first signal-to-noise ratio and greater than the second signal-to-noise ratio is 8, and the number of network layers when the signal-to-noise ratio is less than the second signal-to-noise ratio is 6; when the number of convolutional layers of the convolutional neural network is 2, the number of network layers of the deep manifold network is 8.

[0153] For the specific limitations of the target detection device based on the deep convolution feature matrix manifold network, please refer to the limitations of the target detection method based on the deep convolution feature matrix manifold network mentioned above, which will not be repeated here. Each module in the above-mentioned target detection device based on the deep convolution feature matrix manifold network can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0154] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Fig.14 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a target detection method based on a deep convolution feature matrix manifold network is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0155] Those skilled in the art will understand that Fig.14 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0156] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the above embodiment when executing the computer program.

[0157] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method in the above embodiment are implemented.

[0158] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0159] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0160] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A target detection method based on deep convolutional feature matrix manifold network, characterized in that: The method comprises: Using a simulated target signal driven by a Doppler steering vector in pre-acquired sea clutter data to acquire echo data of simulated targets with different signal-to-noise ratios; the Doppler steering vector is represented by a target signal amplitude parameter, the magnitude of which is related to an average magnitude of the sea clutter data, a signal-to-noise ratio, an initial phase, and a scattering coefficient; A time-frequency feature map of the echo data is obtained by using a time-frequency analysis method, and a pre-trained convolutional neural network is used to extract high-dimensional nonlinear features of the time-frequency feature map to obtain a deep convolution feature matrix; The deep convolution feature matrix is ​​used as a sample to train a pre-constructed deep manifold network, so that the deep manifold network can detect the target in the echo data to be detected.

2. The method according to claim 1, characterized in that Simulated target signals driven using Doppler-steering vectors in pre-acquired sea clutter data, including: The simulated target signal using Doppler-steering vector drive in pre-acquired sea clutter data is: α=β×mean(S)cosθ Where S is the sea clutter data, s is the Doppler steering vector, α is the target signal amplitude parameter, θ represents the initial phase, β represents the scattering coefficient, N represents the number of target signal pulses, and f d represents the normalized Doppler frequency.

3. The method according to claim 1, characterized in that The method of obtaining the time-frequency characteristic diagram of the echo data by using the time-frequency analysis method includes: The STFT time-frequency analysis method is used to obtain the time-frequency characteristic diagram of the echo data.

4. The method according to claim 1, characterized in that: A pre-trained convolutional neural network is used to extract the high-dimensional nonlinear features of the time-frequency feature map to obtain a deep convolution feature matrix, including: Inputting the time-frequency feature map into a pre-trained convolutional neural network, and using the convolutional layer output of the convolutional neural network as a high-dimensional nonlinear convolutional feature; The matrices of all channels of the high-dimensional nonlinear convolution feature are symmetrically positive-defined and then averaged to obtain a deep convolution feature matrix.

5. The method according to claim 4, characterized in that The matrices of all channels of the high-dimensional nonlinear convolution feature are symmetrically positive-defined and then averaged to obtain a deep convolution feature matrix, including: The matrix of all channels of the high-dimensional nonlinear convolution feature is symmetric and positively defined as follows: Wherein, X0 represents the matrix of each channel in the high-dimensional nonlinear convolution feature, X2 represents the high-dimensional nonlinear convolution feature after positive definiteness, ε represents the correction threshold, I is the unit matrix, and U1 represents the orthogonal matrix; The average value of the positively definite high-dimensional nonlinear convolution features is: Among them, S1(i,j,k) represents the high-dimensional nonlinear convolution feature with k channels after positive quantization, and S2(i,j) represents the deep convolution feature matrix.

6. The method according to claim 1, characterized in that The number of convolutional layers of the convolutional neural network is 1 or 2.

7. The method according to claim 6, characterized in that When the number of convolutional layers of the convolutional neural network is 1, the number of network layers of the deep manifold network when the signal-to-noise ratio is greater than the first signal-to-noise ratio is 4, the number of network layers when the signal-to-noise ratio is less than the first signal-to-noise ratio and greater than the second signal-to-noise ratio is 8, and the number of network layers when the signal-to-noise ratio is less than the second signal-to-noise ratio is 6; when the number of convolutional layers of the convolutional neural network is 2, the number of network layers of the deep manifold network is 8.

8. A target detection device based on a deep convolutional feature matrix manifold network, characterized in that: The device comprises: A data acquisition module, used for using a simulated target signal driven by a Doppler steering vector in pre-acquired sea clutter data to acquire echo data of a simulated target with different signal-to-noise ratios; the Doppler steering vector is represented by a target signal amplitude parameter, and the magnitude of the target signal amplitude parameter is related to the average magnitude of the sea clutter data, the signal-to-noise ratio, the initial phase and the scattering coefficient; A deep convolution feature matrix construction module is used to obtain a time-frequency feature map of the echo data by using a time-frequency analysis method, and to extract high-dimensional nonlinear features of the time-frequency feature map by using a pre-trained convolutional neural network to obtain a deep convolution feature matrix; The target detection module is used to train a pre-constructed deep manifold network using the deep convolution feature matrix as a sample, so that the deep manifold network can detect the target in the echo data to be detected.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images