Infrared laser composite signal target recognition and positioning method based on YOLO positioning idea

Abstract

This invention discloses a method for target recognition and localization of infrared laser composite signals based on the YOLO localization concept, belonging to the field of signal recognition and localization technology. This invention innovatively proposes applying the localization concept of the YOLO series models to the field of one-dimensional data signals. Based on this concept, a network structure model is built and a sample dataset is constructed, enabling target recognition and localization of one-dimensional data signals. Experimental results show that this method not only improves the accuracy of composite signal recognition but also determines the position of the target signal within the data segment, achieving target localization.

Description

Technical Field

[0001] This invention relates to the field of signal recognition and positioning technology, specifically to a method for infrared laser composite signal target recognition and positioning based on the YOLO positioning concept. Background Technology

[0002] Convolutional neural networks (CNNs) are a typical end-to-end deep learning method that can directly learn target features from raw data and output prediction results without human intervention. In recent years, this method has also been gradually applied to the research of target recognition for terminally guided missile detection signals. Yan Guangli et al., targeting the characteristics of composite detection signals from terminally guided missiles, proposed three information fusion schemes and three different input methods for network structures, obtaining the optimal scheme through testing. Wu Jun'an et al. used a linear array lidar as a detector, building a convolutional neural network model to classify and identify targets in lidar range images. This method not only has high target recognition accuracy but also effectively identifies armored targets in complex backgrounds. Zhou Shenghui et al. proposed a composite detection signal recognition method based on a multi-channel convolutional neural network and an extreme gradient boosting decision tree. This method can reduce the false detection rate and false negative rate, achieving good recognition results.

[0003] While convolutional neural network models can be effectively used to identify composite detection signals from terminally guided munitions, a problem remains: for signals indicating a target, the location of the target cannot be determined, preventing the munition's control system from calculating the optimal detonation time. YOLO, a popular target detection algorithm, uses a single neural network model to simultaneously predict the location and category of multiple objects in an image, leveraging global information and a single scale to achieve efficient and accurate target detection. Although infrared or range images from terminally guided munitions can be used for target identification and localization via YOLO models, the large number of model parameters and computational demands prevent it from meeting the real-time target identification requirements of onboard computing systems. Summary of the Invention

[0004] The purpose of this invention is to provide a method for infrared laser composite signal target recognition and localization based on the YOLO localization concept. This method utilizes the YOLO model localization approach to achieve one-dimensional composite signal target recognition and localization. Specifically, this invention targets one-dimensional detection signals from lidar and infrared detectors. It constructs a YOLO-1D network model and sample set using the YOLO model for target localization to achieve target signal recognition and localization. The performance of this model is then experimentally tested.

[0005] The technical solution to achieve the purpose of this invention is: an infrared laser composite signal target recognition and localization method based on the YOLO positioning concept, comprising the following steps:

[0006] Step 1: Use an infrared laser composite detection system to detect the target and collect n sets of infrared laser data. Use a sliding window of length s to extract the infrared laser data and establish an infrared laser composite signal database. At the same time, construct labels (X1, X2, Class) with category information and location information based on the position of the infrared laser signal change when the target appears. Divide the infrared laser composite signal database into a training set and a test set in a 7:3 ratio.

[0007] Step 2: Construct the SE-YOLO network, which includes 5 CBR modules, an SE attention mechanism module, a feature parameter size integration module, and a feature selection module.

[0008] The CBR module includes Conv1d, BatchNorm1d, and ReLU activation functions; the SE attention mechanism module includes AdaptiveAvgPool1d, Linear, ReLU activation functions, and Sigmoid activation function; the feature parameter size integration module includes Conv1d convolutional layers; and the core of the feature selection module is the non-maximum suppression algorithm.

[0009] Step 3: Train the SE-YOLO network using the training set to obtain the trained SE-YOLO network model.

[0010] The infrared laser composite signal from the training set is input into the SE-YOLO network. Features of the infrared laser composite signal are extracted layer by layer through three CBR modules to obtain the first feature vector P1. P1 is then input into the SE attention mechanism module to change the importance of each feature channel, causing the neural network to focus on channels with large weights, resulting in the second feature vector P2. The remaining two CBR modules are used to increase the depth of the SE-YOLO network model, further extracting features from P2 to obtain the third feature vector P3. Information fusion is performed on P3 through the feature parameter size integration module to obtain a fourth feature vector V with a size of 8×N. The feature filtering module filters V to obtain the predicted value (X'1, X'2, Clss') with the highest confidence. Error calculation is performed between the predicted value and its corresponding label to update the parameters of the SE-YOLO network, resulting in a trained SE-YOLO network model.

[0011] Step 4: Input the test set into the SE-YOLO network model to test the accuracy and localization precision of the SE-YOLO network model.

[0012] Compared with the prior art, the advantages of the present invention are as follows:

[0013] 1) This invention applies the positioning concept of the YOLO series models to one-dimensional data signals, which can not only realize the identification of infrared laser composite signals, but also realize the positioning of target signals.

[0014] 2) Introduce the SE attention mechanism module into the overall model, enabling the model to automatically learn which feature channels are more important, thereby enhancing the model's feature expression ability and contributing to improving the classification or recognition performance of the model. Brief Description of the Drawings

[0015] Figure 1 It is a flowchart of the infrared laser composite signal target recognition and positioning method based on the YOLO positioning idea of the present invention.

[0016] Figure 2 It is a framework diagram of the overall network model. Detailed Embodiment

[0017] To make the objectives, technical solutions and advantages of the present invention clearer, the following further describes the embodiments of the present invention in detail.

[0018] Combined with Figures 1-2 , an infrared laser composite signal target recognition and positioning method based on the YOLO positioning idea described in the present invention includes the following steps:

[0019] Step 1: Use an infrared laser composite detection system to detect the target, collect n groups of infrared laser data, where 1000 < n < 3000, intercept the infrared laser data using a sliding window with a length of s, s = 1024, establish an infrared laser composite signal database, and at the same time construct labels with category information and position information according to the position where the infrared laser signal changes when the target appears. The label format is (X1, X2, Class), where X1 is the position where the infrared laser composite signal target appears, X2 is the position where the infrared laser composite signal target disappears, and Class is the true category of the infrared laser composite signal. Then divide the infrared laser composite signal database into a training set and a test set according to a ratio of 7:3.

[0020] Step 2: Construct an SE-YOLO network, including 5 CBR modules, an SE attention mechanism module, a feature parameter size integration module, and a feature screening module, specifically as follows:

[0021] The CBR module includes Conv1d, BatchNorm1d, and a ReLU activation function, which are used to enhance the network's expression ability, improve the stability and efficiency of training at the same time. Stacking multiple CBR modules can increase the network depth, thereby extracting the features of the infrared laser composite signal layer by layer.

[0022] The SE attention mechanism module includes AdaptiveAvgPool1d, Linear, ReLU activation functions, and Sigmoid activation function. The SE attention mechanism module adds an attention mechanism in the channel dimension, which changes the importance of each feature channel, causing the neural network to focus on certain channels with large weight values.

[0023] The feature parameter size integration module includes a Conv1d convolutional layer. The feature parameter integration module integrates the features of the extracted infrared laser composite signal to extract higher-level features and integrates the higher-level feature size into 8×N, where N represents the number of columns of the output features.

[0024] The core of the feature selection module is the non-maximum suppression algorithm. The feature selection module is used to select features extracted by the SE-YOLO network model in the feature extraction stage in order to obtain the label prediction value with the highest confidence.

[0025] Step 3: Train the SE-YOLO network using the training set to obtain the trained SE-YOLO network model, as follows:

[0026] Step 3-1: Input the infrared laser composite signal from the training set into the SE-YOLO network, and extract the features of the infrared laser composite signal layer by layer through three CBR modules to obtain the first feature vector P1:

[0027] P1 = CBR 32 (CBR 16 (CBR 8 (D))) (1)

[0028] Where D is the input infrared laser composite signal, CBR t (·) is the feature extraction function with output channel size t, where t = 8, 16, 32.

[0029] Step 3-2: Input P1 into the SE attention mechanism module to change the importance of each feature channel, so that the neural network focuses on the channels with large weight values, and obtain the second feature vector P2:

[0030] P2=SE(P1) (2)

[0031] Step 3-3: Utilize the remaining two CBR modules to increase the depth of the SE-YOLO network, further extracting the third feature vector P3 from P2.

[0032] P3 = CBR 32 (CBR 32 (P2)) (3)

[0033] Steps 3-4: Information fusion is performed on P3 through the feature parameter size integration module to obtain a fourth feature vector V with a size of 8×N. ij The eigenvalues ​​contained in V:

[0034] V=C(P3) (4)

[0035] V = [v 11 , ..., v 1j , ..., v 1N ;

[0036] v 21 , ..., v 2j , ..., v 2N ; ...

[0038] v i1 , ..., v ij , ..., v iN ; ...

[0040] v 81 , ..., v 8j , ..., v 8N (5)

[0041] Where C(·) represents the Conv1d function, and V represents the fourth eigenvector, which contains 8 channels, each representing a different meaning: v 1j v represents the location where the first predicted bounding box target appears. 2j v represents the location where the first predicted bounding box disappears. 3j v represents the confidence level between the first predicted bounding box and the ground truth bounding box. 4j v represents the location where the second predicted bounding box target appears. 5j v represents the location where the second predicted bounding box target disappears. 6j v represents the confidence level between the second predicted bounding box and the ground truth bounding box. 7j v represents the probability of no target signal. 8j This represents the probability of having a target signal, with channel numbers i = 1, 2, 3... 8 and j = 1, 2,... N.

[0042] Steps 3-5: Use the feature filtering module to filter the fourth feature vector V to obtain the predicted value (X'1, X'2, Class') with the highest confidence.

[0043] Steps 3-6: Calculate the error between the predicted values ​​of the SE-YOLO network model and their corresponding labels to update the parameters of the SE-YOLO network, thus obtaining the trained SE-YOLO network model, as detailed below:

[0044] Step 3-6-1: Construct location loss function L1, category loss function L2, and confidence loss function L3 to approximate the predicted probability distribution and true probability distribution of the SE-YOLO network:

[0045]

[0046] L2 = L Cls (Class,Class') (7)

[0047] L3 = L Confi (IOU((X'1,X'2),(X1,X2)),1) (8)

[0048] Where, x m The location label representing the target's position in the infrared laser composite signal, x' m The value represents the predicted position of the target in the infrared laser composite signal, m represents the current prediction box number, and L represents the predicted position. Cls It is the cross-entropy loss function between the predicted class and the true class, L Confi It is the cross-entropy loss function of the IOU value between the predicted box and the ground truth box and 1; (X1, X2) represents the ground truth location interval, (X'1, X'2) represents the predicted location interval, Class represents the ground truth class of the infrared laser signal, Class' represents the predicted class of the infrared laser composite signal, and IOU represents the intersection calculation function.

[0049] Step 3-6-2: Construct the final loss function L:

[0050] L=λ1*L1+λ2*L Cls +λ3*L Confi (9)

[0051] Where λ1, λ2, and λ3 are the balancing parameters of the loss function.

[0052] Step 3-6-3: Obtain the trained SE-YOLO network model.

[0053] Step 4: Input the test set into the trained SE-YOLO network model to test the accuracy and localization precision of the trained SE-YOLO network model, as detailed below:

[0054] For the classification problem of infrared laser composite signals, the evaluation metrics are accuracy, precision, recall, and F1 score to measure the classification performance of the model. For the target localization problem of infrared laser composite signals, since localization is only meaningful on signals where a target is present, the mean of the predicted location interval (X'1, X'2) is calculated only on correctly classified signals containing a target. Whether it is within the range of the real position interval (X1, X2) is used as the criterion for successful positioning, and the evaluation index is calculated based on this. The specific principle is shown in the following formula:

[0055]

[0056] Among them, P loc represents the positioning accuracy rate, M represents the number of correctly predicted positive samples, and k represents the number of samples in the test set.

[0057] The meaning of the numerator is: if the predicted mean value of the target signal position is within the real value interval, it is recorded as 1, otherwise it is recorded as 0; the positioning accuracy rate is calculated by calculating the number of successful positionings.

[0058] Example 1

[0059] An infrared laser composite signal target recognition and positioning method based on the YOLO positioning idea described in the present invention includes the following steps:

[0060] Step 1: Use an infrared laser composite detection system to detect the target, collect n groups of infrared laser data, 1000 < n < 3000, intercept the infrared laser data using a sliding window with a length of s, s = 1024, establish an infrared laser composite signal database, the database is divided into 2 categories, there are target signal samples and non-target signal samples, a total of 3500 groups. At the same time, construct labels with category information and position information according to the position where the infrared laser signal changes when the target appears. The label format is (X1, X2, Class), X1 is the position where the infrared laser composite signal target appears, X2 is the position where the infrared laser composite signal target disappears, Class is the real category of the infrared laser composite signal, and divide the infrared laser composite signal database into a training set and a test set according to a ratio of 7:3. There are 2450 groups of infrared laser composite signals in the training set and 1050 groups of infrared laser composite signals in the test set.

[0061] Step 2: Construct a SE-YOLO network, including 5 CBR modules, a SE attention mechanism module, a feature parameter size integration module, and a feature screening module, specifically as follows:

[0062] The CBR module includes Conv1d, BatchNorm1d, and ReLU activation functions, which are used to enhance the expression ability of the network, improve the stability and efficiency of training at the same time. Stacking multiple CBR modules can increase the depth of the network, so as to extract the features of the infrared laser composite signal layer by layer.

[0063] The SE attention mechanism module includes AdaptiveAvgPool1d, Linear, ReLU activation functions, and Sigmoid activation function. The SE attention mechanism module adds an attention mechanism in the channel dimension, which changes the importance of each feature channel, causing the neural network to focus on certain channels with large weight values.

[0064] The feature parameter size integration module includes a Conv1d convolutional layer. The feature parameter integration module integrates the features of the extracted infrared laser composite signal to extract higher-level features and integrates the higher-level feature size into 8×N, where N represents the number of columns of the output features.

[0065] The core of the feature selection module is the non-maximum suppression algorithm. The feature selection module is used to select features extracted by the SE-YOLO network model in the feature extraction stage in order to obtain the label prediction value with the highest confidence.

[0066] Step 3: Train the SE-YOLO network using the training set to obtain the trained SE-YOLO network model, as follows:

[0067] Step 3-1: Input the infrared laser composite signal from the training set into the SE-YOLO network, and extract the features of the infrared laser composite signal layer by layer through 3 CBR modules to obtain the first feature vector P1.

[0068] P1 = CBR 32 (CBR 16 (CBR 8 (D))) (1)

[0069] Where D is the input infrared laser composite signal, CBR t (·) is the feature extraction function with output channel size t, where t = 8, 16, 32.

[0070] Step 3-2: Input P1 into the SE attention mechanism module to change the importance of each feature channel, so that the neural network focuses on the channel with the larger weight value, and obtain the second feature vector P2.

[0071] P2=SE(P1) (2)

[0072] Step 3-3: Use the remaining two CBR modules to increase the depth of the SE-YOLO network and further extract the third feature vector P3 from P2.

[0073] P3 = CBR 32 (CBR 32 (P2)) (3)

[0074] Steps 3-4: Information fusion is performed on P3 through the feature parameter size integration module to obtain a fourth feature vector V with a size of 8×N. ij Let V be the eigenvalues ​​contained in V.

[0075] V=C(P3) (4)

[0076] V = [v 11 , ..., v 1j , ..., v 1N ;

[0077] v 21 , ..., v 2j , ..., v 2N ; ...

[0079] v i1 , ..., v ij , ..., v iN ; ...

[0081] v 81 , ..., v 8j , ..., v 8N (5)

[0082] Where C(·) represents the Conv1d function, and V represents the fourth eigenvector, which contains 8 channels, each representing a different meaning: v 1j v represents the location where the first predicted bounding box target appears. 2j v represents the location where the first predicted bounding box disappears. 3j v represents the confidence level between the first predicted bounding box and the ground truth bounding box. 4j v represents the location where the second predicted bounding box target appears. 5j v represents the location where the second predicted bounding box target disappears. 6j v represents the confidence level between the second predicted bounding box and the ground truth bounding box. 7j v represents the probability of no target signal. 8j This represents the probability of having a target signal, with channel numbers i = 1, 2, 3... 8 and j = 1, 2,... N.

[0083] Steps 3-5: Use the feature filtering module to filter the fourth feature vector V to obtain the predicted value (X'1, X'2, Class') with the highest confidence.

[0084] Steps 3-6: Calculate the error between the predicted values ​​of the SE-YOLO network model and their corresponding labels to update the parameters of the SE-YOLO network, thus obtaining the trained SE-YOLO network model, as detailed below:

[0085] Step 3-6-1: Construct location loss function L1, category loss function L2, and confidence loss function L3 to approximate the predicted probability distribution and true probability distribution of the SE-YOLO network:

[0086]

[0087] L2 = L Cls (Class,Class') (7)

[0088] L3 = L Confi (IOU((X'1,X'2),(X1,X2)),1) (8)

[0089] Where, x m The location label representing the target's position in the infrared laser composite signal, x' m The value represents the predicted position of the target in the infrared laser composite signal, m represents the current prediction box number, and L represents the predicted position. CLs It is the cross-entropy loss function between the predicted class and the true class, L Confi It is the cross-entropy loss function of the IOU value between the predicted box and the ground truth box and 1; (X1, X2) represents the ground truth location interval, (X'1, X'2) represents the predicted location interval, Class represents the ground truth class of the infrared laser signal, Class' represents the predicted class of the infrared laser composite signal, and IOU represents the intersection calculation function.

[0090] Step 3-6-2: Construct the final loss function L:

[0091] L=λ1*L1+λ2*L Cls +λ3*L Confi (9)

[0092] Where λ1, λ2, and λ3 are the balancing parameters of the loss function.

[0093] Step 3-6-3: Obtain the trained SE-YOLO network model.

[0094] Step 4: Input the test set into the trained SE-YOLO network model to test the accuracy and localization precision of the trained SE-YOLO network model, as detailed below:

[0095] For the classification problem of infrared laser composite signals, the evaluation metrics are accuracy, precision, recall, and F1 score to measure the classification performance of the model. For the target localization problem of infrared laser composite signals, since localization is only meaningful on signals where a target is present, the mean of the predicted location interval (X'1, X'2) is calculated only on correctly classified signals containing a target. Whether the location falls within the actual location interval (X1, X2) is used as the criterion for successful positioning, and the evaluation index is calculated based on this. The specific principle is shown in the following formula:

[0096]

[0097] Among them, P loc The value represents the localization accuracy, M represents the number of correctly predicted positive samples, and k represents the number of samples in the test set.

[0098] The meaning of the numerator is: if the predicted mean of the target signal position is within the range of the true value, it is recorded as 1; otherwise, it is recorded as 0. The positioning accuracy is calculated by calculating the number of successful positioning.

[0099] The method proposed in this invention was tested on an Nvidia 1050Ti GPU host using Python and PyTorch programming languages ​​to build a network framework. The learning rate of the training network was set to 1e. -4 To accelerate network training, the Momentum optimization strategy was chosen, and the Adam optimization algorithm was used. The batch size was set to 16. In the experiment, to prevent network overfitting, the Dropout value was set to 0.5, the λ1 value was set to 1, the λ2 value was set to 0.5, and the λ3 value was set to 0.5.

[0100] Table 1 Comparative Experiment

[0101]

[0102] To test the real-world performance of infrared laser composite signals as much as possible, the experiments in this invention did not process the original data signals in any way. To demonstrate the superior performance of the algorithm of this invention, a comparative experiment was also conducted between the classic algorithm and the algorithm of this invention. The results of the comparative experiment are shown in Table 1. The experimental results show that the algorithm of this invention has a significant improvement in recognition rate and can achieve target localization with a localization accuracy of 85.03%.

Claims

1. A method for infrared laser composite signal target recognition and localization based on the YOLO positioning concept, characterized in that, It includes the following steps: Step 1: Detect the target using an infrared laser composite detection system, collect n groups of infrared laser data, intercept the infrared laser data through a sliding window with a length of s, establish an infrared laser composite signal database, and at the same time construct labels (X1, X2, Class) with existence category information and position information according to the position where the infrared laser signal changes when the target appears, and divide the infrared laser composite signal database into a training set and a test set according to a ratio of 7:3; Step 2: Construct an SE-YOLO network, which includes 5 CBR modules, an SE attention mechanism module, a feature parameter size integration module, and a feature screening module; Among them, the CBR module includes Conv1d, BatchNorm1d, and a ReLU activation function; the SE attention mechanism module includes AdaptiveAvgPool1d, Linear, a ReLU activation function, and a Sigmoid activation function; the feature parameter size integration module includes a Conv1d convolutional layer; the core of the feature screening module is the non-maximum suppression algorithm; Step 3: Train the SE-YOLO network using the training set to obtain a trained SE-YOLO network model; Input the infrared laser composite signal in the training set into the SE-YOLO network, and layer by layer extract the features of the infrared laser composite signal through 3 CBR modules to obtain the first feature vector P1; input P1 into the SE attention mechanism module to change the importance of each feature channel, so that the neural network focuses on the channels with large weight values to obtain the second feature vector P2; use the remaining 2 CBR modules to increase the depth of the SE-YOLO network model, and further extract P2 to obtain the third feature vector P3; perform information fusion on P3 through the feature parameter size integration module to obtain a fourth feature vector V with a size of 8×N; use the feature screening module to screen V to obtain the predicted value (X’1, X’2, Class’) with the highest confidence; calculate the error between the predicted value and its corresponding label to update the parameters of the SE-YOLO network and obtain a trained SE-YOLO network model; Step 4: Input the test set into the SE-YOLO network model to test the accuracy and positioning accuracy of the SE-YOLO network model.

2. The infrared laser composite signal target recognition and localization method based on YOLO positioning concept according to claim 1, characterized in that, In Step 1, 1000 < n < 3000, s = 1024, the label format is (X1, X2, Class), X1 is the position where the infrared laser composite signal target appears, X2 is the position where the infrared laser composite signal target disappears, and Class is the true category of the infrared laser composite signal.

3. The infrared laser composite signal target recognition and localization method based on YOLO positioning concept according to claim 2, characterized in that, In Step 2, construct an SE-YOLO network, including 5 CBR modules, an SE attention mechanism module, a feature parameter size integration module, and a feature screening module, specifically as follows: The CBR module includes Conv1d, BatchNorm1d, and ReLU activation functions, which are used to enhance the network's expressive power and improve training stability and efficiency. Multiple CBR modules can be stacked to increase the network depth, thereby extracting features of infrared laser composite signals layer by layer. The SE attention mechanism module includes AdaptiveAvgPool1d, Linear, ReLU activation functions, and Sigmoid activation function. The SE attention mechanism module adds an attention mechanism in the channel dimension, which changes the importance of each feature channel, causing the neural network to focus on certain channels with large weight values. The feature parameter size integration module includes a Conv1d convolutional layer. The feature parameter integration module integrates the features of the extracted infrared laser composite signal to extract higher-level features and integrates the higher-level feature size into 8×N, where N represents the number of columns of the output features. The core of the feature selection module is the non-maximum suppression algorithm. The feature selection module is used to select features extracted by the SE-YOLO network model in the feature extraction stage in order to obtain the label prediction value with the highest confidence.

4. The infrared laser composite signal target recognition and localization method based on YOLO positioning concept according to claim 3, characterized in that, In step 3, the SE-YOLO network is trained using the training set to obtain the trained SE-YOLO network model, as detailed below: Step 3-1: Input the infrared laser composite signal from the training set into the SE-YOLO network, and extract the features of the infrared laser composite signal layer by layer through 3 CBR modules to obtain the first feature vector P1; P1=CBR 32 (CBR 16 (CBR 8 (D))) (1) Where D is the input infrared laser composite signal, CBR t (·) is the feature extraction function with output channel size t, where t = 8, 16, 32; Step 3-2: Input P1 into the SE attention mechanism module to change the importance of each feature channel, so that the neural network focuses on the channel with the larger weight value, and obtain the second feature vector P2. P2=SE(P1) (2) Step 3-3: Use the remaining two CBR modules to increase the depth of the SE-YOLO network and further extract the third feature vector P3 from P2; P3=CBR 32 (CBR 32 (P2)) (3) Steps 3-4: Information fusion is performed on P3 through the feature parameter size integration module to obtain a fourth feature vector V with a size of 8×N. ij The eigenvalues ​​contained in V; V=C(P3) (4) V=[v 11 ,...,v 1j ,...,v 1N ; v 21 ,...,v 2j ,...,v 2N ; ... v i1 ,...,v ij ,...,v iN ; ... v 81 ,...,v 8j ,...,v 8N ] (5) Where C(·) represents the Conv1d function, and V represents the fourth eigenvector, which contains 8 channels, each representing a different meaning: v 1j v represents the location where the first predicted bounding box target appears. 2j v represents the location where the first predicted bounding box disappears. 3j v represents the confidence level between the first predicted bounding box and the ground truth bounding box. 4j v represents the location where the second predicted bounding box target appears. 5j v represents the location where the second predicted bounding box target disappears. 6j v represents the confidence level between the second predicted bounding box and the ground truth bounding box. 7j v represents the probability of no target signal. 8j This represents the probability of having a target signal; channel number i = 1, 2, 3... 8, j = 1, 2,... N; Steps 3-5: Use the feature filtering module to filter the fourth feature vector V to obtain the predicted value (X'1, X'2, Class') with the highest confidence. Steps 3-6: Calculate the error between the predicted values ​​of the SE-YOLO network model and their corresponding labels in order to update the parameters of the SE-YOLO network and obtain the trained SE-YOLO network model.

5. The infrared laser composite signal target recognition and localization method based on YOLO positioning concept according to claim 3, characterized in that, In steps 3-6, the error between the predicted values ​​of the SE-YOLO network model and their corresponding labels is calculated to update the parameters of the SE-YOLO network, thus obtaining the trained SE-YOLO network model, as detailed below: Step 3-6-1: Construct location loss function L1, category loss function L2, and confidence loss function L3 to approximate the predicted probability distribution and true probability distribution of the SE-YOLO network: L2=L Cls (Class,Class') (7) L3=L Confi (IOU((X'1,X'2),(X1,X2)),1) (8) Where, x m The location label representing the target's position in the infrared laser composite signal, x' m The value represents the predicted position of the target in the infrared laser composite signal, m represents the current prediction box number, and L represents the predicted position. Cls It is the cross-entropy loss function between the predicted class and the true class, L Confi It is the cross-entropy loss function of the IOU value between the predicted box and the ground truth box and 1; (X1, X2) represents the ground truth location interval, (X'1, X'2) represents the predicted location interval, Class represents the ground truth class of the infrared laser signal, Class' represents the predicted class of the infrared laser composite signal, and IOU represents the intersection calculation function. Step 3-6-2: Construct the final loss function L: L=λ1*L1+λ2*L Cls +λ3*L Confi (9) Where λ1, λ2, and λ3 are the balancing parameters of the loss function; Step 3-6-3: Obtain the trained SE-YOLO network model.

6. The infrared laser composite signal target recognition and localization method based on YOLO positioning concept according to claim 4, characterized in that, In step 4, the test set is input into the trained SE-YOLO network model to test the accuracy and localization precision of the trained SE-YOLO network model, as detailed below: For the classification problem of infrared laser composite signals, the evaluation metrics are accuracy, precision, recall, and F1 score to measure the classification performance of the model. For the target localization problem of infrared laser composite signals, since localization is only meaningful on signals where a target is present, the mean of the predicted location interval (X'1, X'2) is calculated only on correctly classified signals containing a target. Whether the location falls within the actual location interval (X1, X2) is used as the criterion for successful positioning, and the evaluation index is calculated based on this. The specific principle is shown in the following formula: Among them, P loc M represents the localization accuracy, where M represents the number of correctly predicted positive samples and k represents the number of samples in the test set. The meaning of the numerator is: if the predicted mean of the target signal position is within the range of the true value, it is recorded as 1; otherwise, it is recorded as 0. The positioning accuracy is calculated by calculating the number of successful positioning.

Images