A radar target detection method based on contrastive learning under multi-terrain conditions

Abstract

The present invention discloses a radar target detection method based on contrastive learning under multi-terrain conditions, comprising the following steps: constructing a data set after preprocessing and dividing a training set into a test set; constructing an enhanced sample set after data enhancement; and constructing positive and negative sample pairs for subsequent model training and radar target detection performance testing; network training; and obtaining a decision threshold based on the distribution of training clutter samples in a feature space according to the sorting of training scores; and making a final decision based on the relationship between the calculated score of the detection network using multi-terrain features and the decision threshold. By using clutter data under different terrains as different categories to train a deep network, the characteristic distribution of clutter under different terrains can be learned, the method can adapt to target detection under different terrains, and greatly improve the practical application performance of radar target detection under multi-terrain scenarios.

Description

Technical Field

[0001] The present invention relates to the technical field of radar target detection, and in particular to a radar target detection method based on contrast learning under multi-terrain conditions. Background Art

[0002] A high-resolution radar is a radar system capable of obtaining high-resolution information about the distribution of target scattering points along the range direction. The echo produced by a high-resolution radar is called a high-resolution range profile (HRRP). It is the vector sum of the complex sub-echoes of the target scattering points, acquired using a broadband radar signal, projected onto the radar ray. It provides detailed information about the distribution of target scattering points along the range direction and the differences in scattering characteristics between clutter and target. Furthermore, the target HRRP can reduce clutter energy per range unit, improving the target signal-to-clutter ratio. Therefore, the HRRP is often used for target detection.

[0003] In recent years, HRRP target detection technology based on machine learning has made great progress. Some manual feature extraction detection algorithms based on the idea of ​​anomaly detection have been used for target detection. This type of method first extracts manual features from clutter training data, then constructs a single classifier (such as convex hull learning, decision tree, support vector machine, etc.) to learn the characteristic distribution of clutter, establishes a decision hypersphere in the feature space to include the clutter features, and finally determines whether the sample to be detected is a target or clutter based on whether the features of the sample to be detected are within the hypersphere, thereby achieving target detection.

[0004] However, in multi-terrain scenarios, the clutter feature distributions under different terrains are significantly different, resulting in their distributions being far apart in the feature space. The decision hypersphere constructed by a single classifier containing all the clutter feature distributions will be very large. When the target sample's features are in the feature space of the non-clutter distribution area within it, they will be misclassified as clutter, resulting in a decrease in detection performance. Summary of the Invention

[0005] Due to the differences in clutter feature distributions across different terrains, existing algorithms based on manual feature extraction and a single classifier have poor applicability in multi-terrain scenarios and are difficult to form accurate decision boundaries, resulting in a significant reduction in the probability of radar target detection. To address the above drawbacks, the present invention provides a radar target detection method based on contrastive learning under multi-terrain conditions. By utilizing the concept of anomaly detection, clutter is considered a normal sample and targets are considered an abnormal sample. Clutter distributions under various terrains are clustered based on terrain type, forming multiple clutter feature distribution clusters in a feature space. Multiple decision thresholds are determined based on the size of the distribution clusters. Then, the relationship between the sample to be detected and each distribution cluster is determined based on the threshold, thereby determining whether the sample to be detected is a target or clutter, thereby achieving robust radar target detection under multi-terrain conditions. The method comprises the following steps:

[0006] S1. Acquire measured polarimetric high-resolution radar echo data of radar targets in multiple terrains. The acquired data includes clutter samples and samples to be tested. Complete the data set construction and perform preprocessing such as data slicing and amplitude normalization. After organizing and labeling the data, form a data set. Then, divide the data set into training and test sets to facilitate subsequent model training and radar target detection performance testing. Specifically:

[0007] First, a polarization high-resolution radar is used to collect measured data of radar targets under multiple terrains and perform data preprocessing.

[0008] Acquire radar echo data of measured targets and clutter under multiple terrains, and perform preprocessing operations such as segmentation and amplitude normalization on the radar echo data based on prior knowledge such as target position and physical size.

[0009] S1-1. Construction of the data set and division of the training set and test set:

[0010] S1-1-1. Divide the radar echo data into a clutter sample set and a test sample set:

[0011] In the acquired radar echo data, several frames of clutter echo samples constitute the clutter data sample set X clutter ={x1, x2, ..., x M}, where X clutter The total number of frames in is recorded as M, X clutter The number of terrains contained in the data is recorded as T, X clutter Each frame signal x in m is a 4×N matrix, where m represents the frame number of the polarization high-resolution radar echo signal, m=1, 2, ..., M, N represents the number of range units, and the first dimension of the matrix represents the echo data of the four polarization channels HH, HV, VH, and VV. m (i)=[S HH , S HV , S VH , S VV ] T , i represents the distance unit number, i=1,2,...,N,x m (i) represents x m The element on the i-th distance unit, [·] T Indicates transposition, H indicates horizontal polarization, V indicates vertical polarization, S HH Indicates the echo data of horizontal transmission and horizontal reception, S HV Indicates the echo data transmitted horizontally and received vertically, S VH Indicates the echo data transmitted vertically and received horizontally, S VVRepresents the echo data of vertical transmission and vertical reception; for each clutter radar echo data sample x under the terrain m , each with a label y m ,y m ∈{1, 2, …, T} corresponds to the terrain where the corresponding echo is collected, and T is the total number of terrains where data is collected;

[0012] The target echo data collected under multiple terrains constitute the target data sample set X target ={x1, x2, ..., x O}, X target The total number of frames in is denoted as O, X target Each frame signal x n The matrix dimension size and the physical meaning of each dimension are respectively clutter The signal in each frame is the same.

[0013] S1-1-2. Divide the data set into training set and test set:

[0014] Since the target data sample set X target Do not participate in the training process, so the target data samples are all used as test samples, and the clutter data sample set X clutter The training set and test set are randomly divided in a 1:1 ratio to form the training clutter sample set X clutter_train And the test clutter sample set X clutter_test .

[0015] S1-2, data enhancement, construct an enhanced sample set, and construct positive and negative sample pairs.

[0016] Using contrastive learning method, the training clutter samples are enhanced by two data augmentation methods, cyclic shift and flipping, to form an enhanced sample set. The enhanced samples of the same training clutter sample are defined as positive sample pairs, and the enhanced samples of different training clutter samples are defined as negative sample pairs.

[0017] S1-2-1. Perform data enhancement on the training clutter samples to generate an enhanced sample set.

[0018] The training clutter sample set X clutter_train Divide into different batches, the batch size is recorded as P, consider a random batch For a sample x in this batch k , applying two independent data enhancement methods, we get two enhanced samples and Forming an augmented batch

[0019] S1-2-2. Construct positive and negative sample pairs to facilitate subsequent training.

[0020] For the same sample x k Enhanced sample and Enhanced sample labels and With the original label y k Stay consistent, and Constitutes a positive sample pair, enhanced sample or The enhanced samples obtained with other samples in the batch constitute negative sample pairs.

[0021] S2. Perform network training:

[0022] S2-1. Construct a multi-terrain feature extraction and detection network.

[0023] S2-1-1. Construction of encoder network:

[0024] The encoder network uses the classic ResNet50 network for feature extraction, and the network input sample is a clutter sample x of size 4×N m , after passing through the encoder, an N1×1 embedding representation is formed for feature extraction.

[0025] S2-1-2. Construction of projection head network:

[0026] The projection head network consists of a fully connected layer. After the N1×1 embedding representation is input into the projection head network, the network outputs an N2×1 feature vector, which is used to map the extracted features into the feature space.

[0027] The multi-terrain feature extraction and detection network is recorded as φ, then and The corresponding vector representation and

[0028] S2-2. Construct a loss function based on contrastive learning.

[0029] By introducing and adjusting the temperature hyperparameter, the loss function reduces the repulsive force of clutter samples under the same terrain in the feature space and increases the repulsive force of clutter samples under different terrains, thereby further improving the clustering effect of clutter sample features under the same terrain.

[0030] S2-2-1. Add the category label of the enhanced sample as a condition to the calculation of the training loss function, assign different temperature hyperparameters to samples of different categories, and for the sample feature z of the i-th sample and the j-th sample i and z j , the loss function l1(i, j) is as follows:

[0031]

[0032] where sim(z i , z j ) represents z i and z j The cosine similarity of is the indicator function, which takes 1 if the conditions in the brackets are met and 0 if they are not met. N is the total number of original samples. and To enhance the sample labels, τ is an adjustable temperature hyperparameter.

[0033] In order to give different enhanced samples of the same category a larger temperature hyperparameter to reduce the repulsive force between them, the new loss function Defined as:

[0034]

[0035] S2-2-2. Adjust the temperature hyperparameter to improve clustering effect:

[0036] Although the repulsive force between similar samples is reduced, the above method is essentially similar to SimCLR and is still learning the unique characteristics of a single sample. After a long period of training, it is still difficult to form concentrated clusters. To further improve the clustering effect, the temperature hyperparameter between similar samples is adjusted to make different enhanced samples of the same sample attractive to each other.

[0037] For a sample x in a batch i An enhanced sample of The rest of the augmented samples from the same category in the same batch can be expressed as:

[0038]

[0039] The symbol \ represents the integer quotient operator. Based on the original loss function, this embodiment introduces an additional loss term:

[0040]

[0041] The l2 loss function makes the enhanced samples and Random positive sample augmentation batch Positive samples in Mutual attraction, Represents a uniform distribution.

[0042] S2-2-3. Integrate the loss function based on contrastive learning:

[0043] The overall network loss is as follows:

[0044]

[0045] Where λ represents the weight hyperparameter of l2 loss;

[0046] In order to ensure the effectiveness of l2 loss, α should satisfy the following formula:

[0047]

[0048] Where C represents the total number of training categories and N represents the training batch size.

[0049] S2-3. Input the enhanced sample set of training clutter samples into the multi-terrain feature extraction and detection network. The network is trained based on a contrastive loss function to construct clutter distribution clusters for different terrains. The enhanced samples are input into the network for feature extraction and feature mapping to obtain the final representation. The contrastive loss is then calculated based on the enhanced sample set. The network is trained based on the loss function, using a backpropagation algorithm to update weight parameters.

[0050] S3. Threshold determination and final judgment.

[0051] Based on the distribution of training clutter samples in the feature space, the decision threshold is obtained by taking the quantiles according to the sorting of training scores, and the final decision is made according to the relationship between the calculated scores of the sample to be tested through the encoder and projection head network and the decision threshold.

[0052] S3-1. Detection of threshold determination:

[0053] The present invention adopts the negative Mahalanobis distance -dM as the scoring function s, so for sample x, its final score s under the i-th type of terrain is i It can be expressed by the following formula:

[0054]

[0055] Where z is the feature vector obtained after sample x passes through the network, μ i and ∑ i The mean and covariance matrices calculated for all training samples of the i-th class, for i, i∈I={1,2,…,C main}.

[0056] S(x) contains the scores of each category of sample x. The present invention selects the category with the highest score as the final predicted label.

[0057] i*=argmax n∈I S n (x),

[0058]

[0059] In order to achieve target detection, the present invention needs to use Mahalanobis distance to obtain the detection threshold. The threshold of the i-th class containing samples is as follows:

[0060] First, all the augmented sample sets in class i are sorted in ascending order of their scores to form S i :S i =[S i1 , S i2 ,…,S iw-1 ]. Secondly, the value of the index n is set according to the adjustable parameter p and the number of feature samples w: n = round(w·(1-p)), where round(·) represents the rounding operator. Finally, by i Select the nth element in , and you can get δ i :δ i =S in In the above, p is the rejection rate of the training sample, and its value ranges from 0 to 1.

[0061] In order to test the false alarm probability of 10 -4 The detection probability under this condition is set to p=0.9999.

[0062] S3-2. Final judgment of the sample to be tested.

[0063] For the test samples, the following judgment rules are followed:

[0064]

[0065] Among them, S i* (x) represents the highest score of the test sample, δ i* is the threshold corresponding to the highest scoring class.

[0066] If the score of the test sample is greater than or equal to the threshold, it is classified as a clutter sample; conversely, if the score of the sample is less than the threshold, it is classified as a target sample.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] (1) This paper proposes a loss function based on contrastive learning. A temperature hyperparameter is introduced and adjusted in the loss function. The improved loss function takes the terrain of the clutter samples into account, transforming unsupervised learning into supervised learning. Different temperature hyperparameters are assigned to the cosine similarities between clutter-enhanced samples from the same and different terrains. By aggregating sample features based on the temperature hyperparameter and cosine similarity, compact clustering is achieved in the multi-terrain clutter feature space, constructing distribution clusters of clutter from different terrains.

[0069] (2) The present invention proposes a threshold construction method based on the Mahalanobis distance. A score is generated based on the Mahalanobis distance between the clutter training sample features and the distribution cluster center of the corresponding terrain clutter. The decision threshold for the corresponding terrain is calculated based on the score distribution and the set false alarm probability. The relationship between the score of the test sample features relative to the center of each clutter distribution cluster and each decision threshold is used to determine whether the test sample is a target or clutter.

[0070] (3) By using clutter data under different terrains as different categories to train the deep network, the present invention can learn the characteristic distribution of clutter under different terrains, thereby adapting to target detection under different terrains and greatly improving the actual application performance of radar target detection in multi-terrain scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 Flowchart of radar target detection based on contrastive learning under multi-terrain conditions

[0072] Figure 2 Block diagram of radar target detection model based on contrastive learning under multi-terrain conditions;

[0073] Figure 3 : is the target detection performance curve of the vehicle 1 in the Gobi terrain in the embodiment.

[0074] Figure 4 : is the target detection performance curve of the vehicle 2 in the Gobi terrain in the embodiment.

[0075] Figure 5 : is the target detection performance curve of the vehicle 1 on the highway terrain in the embodiment.

[0076] Figure 6 : is the target detection performance curve of the vehicle 2 on the highway terrain in the embodiment. DETAILED DESCRIPTION

[0077] To facilitate understanding of the present invention, the apparatus of the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate embodiments of the apparatus. However, the apparatus can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0078] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "installed," "connected," and "disposed" should be understood in a broad sense. For example, they may refer to fixed connection or disposition, detachable connection or disposition, or integral connection or disposition. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0079] Example

[0080] like Figure 1-2 As shown, this embodiment provides a radar target detection method based on contrastive learning under multi-terrain conditions, including the following steps:

[0081] S1. Acquire measured polarimetric high-resolution radar echo data of radar targets in multiple terrains. The acquired data includes clutter samples and samples to be tested. Complete the data set construction and perform preprocessing such as data slicing and amplitude normalization. After organizing and labeling the data, form a data set. Then, divide the data set into training and test sets to facilitate subsequent model training and radar target detection performance testing. Specifically:

[0082] First, a polarization high-resolution radar is used to collect measured data of radar targets under multiple terrains and perform data preprocessing.

[0083] Acquire radar echo data of measured targets and clutter under multiple terrains, and perform preprocessing operations such as segmentation and amplitude normalization on the radar echo data based on prior knowledge such as target position and physical size.

[0084] S1-1. Construction of the data set and division of the training set and test set:

[0085] S1-1-1. Divide the radar echo data into a clutter sample set and a test sample set:

[0086] In the acquired radar echo data, several frames of clutter echo samples constitute the clutter data sample set X clutter ={x1, x2, ..., x M}, where X clutter The total number of frames in is recorded as M, X clutter The number of terrains contained in the data is recorded as T, X clutter Each frame signal x in mis a 4×N matrix, where m represents the frame number of the polarization high-resolution radar echo signal, m=1, 2, ..., M, N represents the number of range units, and the first dimension of the matrix represents the echo data of the four polarization channels HH, HV, VH, and VV. m (i)=[S HH , S HV , S VH , S VV ] T , i represents the distance unit number, i = 1, 2, ..., N, x m (i) represents x m The element on the i-th distance unit, [·] T Indicates transposition, H indicates horizontal polarization, V indicates vertical polarization, S HH Indicates the echo data of horizontal transmission and horizontal reception, S HV Indicates the echo data transmitted horizontally and received vertically, S VH Indicates the echo data transmitted vertically and received horizontally, S VV Represents the echo data of vertical transmission and vertical reception; for each clutter radar echo data sample x under the terrain m , each with a label y m ,y m ∈{1, 2, …, T} corresponds to the terrain where the corresponding echo is collected, and T is the total number of terrains where data is collected;

[0087] The target echo data collected under multiple terrains constitute the target data sample set X target ={x1, x2, ..., x O}, X target The total number of frames in is denoted as O, X target Each frame signal x n The matrix dimension size and the physical meaning of each dimension are respectively clutter The signal in each frame is the same.

[0088] S1-1-2. Divide the data set into training set and test set:

[0089] Since the target data sample set X target It does not participate in the training process, so the target samples to be tested are all used as test samples, and the clutter data sample set X clutter The training set and test set are randomly divided in a 1:1 ratio to form the training clutter sample set X clutter_train And the test clutter sample set X clutter_test .

[0090] S1-2, data enhancement, construct an enhanced sample set, and construct positive and negative sample pairs.

[0091] like Figure 2 As shown in FIG, the contrastive learning method is used to enhance the training clutter samples through two data enhancement methods, cyclic shift and flipping, to form an enhanced sample set; the enhanced samples of the same training clutter sample are defined as positive sample pairs, and the enhanced samples of different training clutter samples are defined as negative sample pairs.

[0092] S1-2-1. Perform data enhancement on the training clutter samples to generate an enhanced sample set.

[0093] The training clutter sample set X clutter_train Divide into different batches, the batch size is recorded as P, consider a random batch For a sample x in this batch k , applying two independent data enhancement methods, we get two enhanced samples and Forming an augmented batch

[0094] S1-2-2. Construct positive and negative sample pairs to facilitate subsequent training.

[0095] For the same sample x k Enhanced sample and Enhanced sample labels and With the original label y k Stay consistent, and Constitutes a positive sample pair, enhanced sample or The enhanced samples obtained with other samples in the batch constitute negative sample pairs.

[0096] S2. Perform network training:

[0097] S2-1. Construct a multi-terrain feature extraction and detection network.

[0098] S2-1-1. Construction of encoder network:

[0099] The encoder network uses the classic ResNet50 network for feature extraction, and the network input sample is a clutter sample x of size 4×N m , after passing through the encoder, an N1×1 embedding representation is formed for feature extraction.

[0100] S2-1-2. Construction of projection head network:

[0101] The projection head network consists of a fully connected layer. After the N1×1 embedding representation is input into the projection head network, the network outputs an N2×1 feature vector, which is used to map the extracted features into the feature space.

[0102] The multi-terrain feature extraction and detection network is recorded as φ, then and The corresponding vector representation and

[0103] S2-2. Construct a loss function based on contrastive learning.

[0104] By introducing and adjusting the temperature hyperparameter, the loss function reduces the repulsive force of clutter samples under the same terrain in the feature space and increases the repulsive force of clutter samples under different terrains, thereby further improving the clustering effect of clutter sample features under the same terrain.

[0105] S2-2-1. Add the category label of the enhanced sample as a condition to the calculation of the training loss function, assign different temperature hyperparameters to samples of different categories, and for the sample feature z of the i-th sample and the j-th sample i and z j , the loss function l1(ij) is as follows:

[0106]

[0107] where sim(z i , z j ) represents z i and z j The cosine similarity of is the indicator function, which takes 1 if the conditions in the brackets are met and 0 if they are not met. N is the total number of original samples. and To enhance the sample labels, τ is an adjustable temperature hyperparameter.

[0108] In order to give different enhanced samples of the same category a larger temperature hyperparameter to reduce the repulsive force between them, the new loss function Defined as:

[0109]

[0110] Among them, n is the total number of samples of the original clutter training samples.

[0111] S2-2-2. Adjust the temperature hyperparameter to improve clustering effect:

[0112] Although the repulsive force between similar samples is reduced, the above method is essentially similar to SimCLR and is still learning the unique characteristics of a single sample. After a long period of training, it is still difficult to form concentrated clusters. To further improve the clustering effect, the temperature hyperparameter between similar samples is adjusted to make different enhanced samples of the same sample attractive to each other.

[0113] For a sample x in a batch i An enhanced sample of The rest of the augmented samples from the same category in the same batch can be expressed as:

[0114]

[0115] The symbol \ represents the integer quotient operator. Based on the original loss function, this embodiment introduces an additional loss term:

[0116]

[0117] The l2 loss function makes the enhanced samples and Random positive sample augmentation batch Positive samples in Mutual attraction, Represents a uniform distribution.

[0118] S2-2-3. Integrate the loss function based on contrastive learning:

[0119] The overall network loss is as follows:

[0120]

[0121] Where λ represents the weight hyperparameter of l2 loss;

[0122] In order to ensure the effectiveness of l2 loss, α should satisfy the following formula:

[0123]

[0124] Where C represents the total number of training categories and N represents the training batch size.

[0125] S2-3. Input the enhanced sample set of training clutter samples into the multi-terrain feature extraction and detection network. The network is trained based on a contrastive loss function to construct clutter distribution clusters for different terrains. The enhanced samples are input into the network for feature extraction and feature mapping to obtain the final representation. The contrastive loss is then calculated based on the enhanced sample set. The network is trained based on the loss function, using a backpropagation algorithm to update weight parameters.

[0126] S3. Threshold determination and final judgment.

[0127] Based on the distribution of training clutter samples in the feature space, the decision threshold is obtained by taking the quantiles according to the sorting of training scores, and the final decision is made according to the relationship between the calculated scores of the sample to be tested through the encoder and projection head network and the decision threshold.

[0128] S3-1. Detection of threshold determination:

[0129] The present invention adopts the negative Mahalanobis distance -dM as the scoring function s, so for sample x, its final score s under the i-th type of terrain is i It can be expressed by the following formula:

[0130]

[0131] Where z is the feature vector obtained after sample x passes through the network, μ i and ∑ i The mean and covariance matrices calculated for all training samples of the i-th class, for i, i∈I={1,2,…,C main}.

[0132] S(x) contains the scores of each category of sample x. The present invention selects the category with the highest score as the final predicted label.

[0133] i*=argmax n∈I S n (x),

[0134]

[0135] In order to achieve target detection, the present invention needs to use Mahalanobis distance to obtain the detection threshold. The threshold of the i-th class containing samples is as follows:

[0136] First, all the augmented sample sets in class i are sorted in ascending order of their scores to form S i :S i =[S i1 , S i2 ,…,S iw-1 ]. Secondly, according to the adjustable parameter and the number of feature samples w, the value of the index n is set: n = round (w (1-p)), where round (·) represents the rounding operator. Finally, by i Select the nth element in , and you can get δ i :δ i =S in In the above, p is the rejection rate of the training sample, and its value ranges from 0 to 1.

[0137] In order to test the false alarm probability of 10-4 The detection probability under this condition is set to p=0.9999.

[0138] S3-2. Final judgment of the sample to be tested.

[0139] For the test samples, the following judgment rules are followed:

[0140]

[0141] Among them, S i* (x) represents the highest score of the test sample, δ i* is the threshold corresponding to the highest scoring class.

[0142] If the score of the test sample is greater than or equal to the threshold, it is classified as a clutter sample; conversely, if the score of the sample is less than the threshold, it is classified as a target sample.

[0143] At the same time, this embodiment uses the following data for further explanation:

[0144] Clutter data from two different terrain types, Gobi and highway, was used as training data, and two vehicle targets were used as test data to test its detection performance under different terrains. This dataset was collected using a wideband radar operating in the W-band frequency range. Furthermore, high-resolution range profile (HRRP) data was collected for the two vehicle targets in eight different poses under two different terrain conditions. Each pose corresponds to a different angle of rotation of the vehicle around its own axis, and the data was collected at equal intervals from 0 to 360 degrees.

[0145] Network parameter settings:

[0146] ResNet50 was used as the backbone network. Cyclic shift and flipping were used to obtain enhanced samples for each sample. This example sets the batch size to 1024, α to 0.008, τ to 0.07, and λ to 1. Furthermore, this example uses the SGD optimizer, with a learning rate of 0.075, a decay of 1e-6, and a momentum of 0.9. A cosine annealing scheduler was used to adjust the learning rate.

[0147] Preprocessing of experimental data:

[0148] To better verify detection performance under varying signal-to-noise ratios (SNRs), this example preprocessed the collected vehicle target data. First, this example used target segmentation to obtain the target segment data required for the experiment. The peak power of the target HRRP segment was used as the signal power, and the average power of the clutter HRRP segment was used as the clutter signal power. The SNR can be defined as:

[0149]

[0150] Where N is the length of the range unit of the clutter segment, represents the target peak power, = represents the power of the clutter segment. Finally, this embodiment generates target data with a signal-to-noise ratio of -10-20dB to test the detection performance under different signal-to-noise ratio conditions. The specific experimental training and test data division is shown in the table.

[0151] Table 1 Experimental data description

[0152]

[0153] The final experiment obtained the target detection probability curves of vehicle 1 and vehicle 2 in the Gobi and highway terrains at -10-20dB. The results are as follows Figure 3-Figure 6 shown.

[0154] The comparison methods used in the experiment are the classic binary accumulation detection algorithm (M / N detection algorithm) and the generalized likelihood ratio detection algorithm (GLRT detection algorithm), etc. These detection algorithms have good effects in HRRP target detection.

[0155] The detection results show that the proposed method outperforms the comparison method in both Gobi and highway terrains for both vehicle 1 and vehicle 2, demonstrating the effectiveness of the proposed algorithm. Furthermore, compared to the GLRT detection algorithm, the proposed method requires a 1-4 dB reduction in signal-to-clutter ratio (SCR) to achieve a 90% detection probability.

[0156] It should be noted that the structure described in the present invention can be implemented in a variety of different forms and is not limited to the described embodiments. Any equivalent transformations made by ordinary technicians in this field using the contents of the present invention description and drawings, or directly or indirectly applied to other related technical fields, such as the loading and unloading of other items, are included in the scope of protection of the present invention.

Claims

1. A radar target detection method based on contrastive learning under multi-terrain conditions, characterized by: The following steps are involved: S1. Obtain measured polarimetric high-resolution radar clutter samples and test samples for radar targets in multiple terrains. After preprocessing, construct a dataset and divide it into training and test sets. After data enhancement, construct an enhanced sample set and construct positive and negative sample pairs for subsequent model training and radar target detection performance testing. S2. Network training: S2-1. Complete the construction of the multi-terrain feature extraction and detection network by establishing an encoder network and a projection head network, which is used to map the extracted features into the feature space; S2-2. The category labels of the clutter samples in the training set are added as conditions to the calculation of the training loss function. Different temperature hyperparameters are assigned to samples of different categories. The temperature hyperparameters are then adjusted to improve the clustering effect. The loss function based on contrastive learning is obtained by integration. S2-3, inputting the enhanced sample set of training clutter samples into the network, training the network based on the contrast loss function, and constructing clutter distribution clusters for different terrains; S3. Based on the distribution of the training clutter samples in the feature space, the Mahalanobis distance is used to obtain the detection threshold, and the final decision is made based on the relationship between the calculated score of the sample to be tested through the encoder and projection head network and the decision threshold.

2. The radar target detection method based on contrastive learning under multi-terrain conditions according to claim 1, characterized in that: The preprocessing in step S1 includes segmentation and normalization operations.

3. The radar target detection method based on contrastive learning under multi-terrain conditions according to claim 1, characterized in that: In step S1, the test samples obtained from multiple frames under multiple terrains constitute a target data sample set to be tested, and the clutter echo samples from multiple frames constitute a clutter data sample set. Since the target data sample set to be tested does not participate in the training process, the target data samples are all used as test samples. The clutter data sample set is randomly divided into a training set and a test set in a 1:1 ratio to form a training clutter sample set and a test clutter sample set.

4. The radar target detection method based on contrastive learning under multi-terrain conditions according to claim 1, characterized in that: In step S1, the contrastive learning method is used to enhance the training clutter samples through two data enhancement methods, cyclic shift and flipping, to form an enhanced sample set; the enhanced samples of the same clutter sample are defined as positive sample pairs, and the enhanced samples of different clutter samples are defined as negative sample pairs.

5. The radar target detection method based on contrastive learning under multi-terrain conditions according to claim 1, characterized in that: In step S2-1, the encoder network uses the classic ResNet50 network for feature extraction, and the projection head network is composed of a fully connected layer for mapping the extracted features into the feature space; In step S2-2, the loss function introduces and adjusts the temperature hyperparameter so that the repulsive force of clutter samples under the same terrain in the feature space is reduced, and the repulsive force of clutter samples under different terrains is increased, so as to improve the clustering effect of clutter sample features under the same terrain; In step S2-3, the enhanced samples are input to the network for feature extraction and feature mapping to obtain the final representation, and the contrast loss is calculated based on the enhanced sample set. At the same time, the network is trained based on the loss function, and the network training adopts the back propagation algorithm to update the weight parameters.

6. The radar target detection method based on contrastive learning under multi-terrain conditions according to claim 1, characterized in that: The step S3 includes using the Mahalanobis distance to obtain a detection threshold, and defining the final decision of the sample to be tested according to the following decision rule: ； in, represents the highest score of the test sample, is the threshold corresponding to the highest scoring category; If the score of the sample to be tested is greater than or equal to the threshold, it is classified as a clutter sample; on the contrary, if the score of the sample is less than the threshold, it is classified as a target sample.

Images