A semi-supervised DAS event detection method and system

Abstract

The present invention relates to the field of distributed optical fiber sensing technology, and discloses a semi-supervised DAS event detection method and system. The method comprises the following steps: selecting a data set collected through a DAS system and pre-processing it to obtain paired training data; building a semi-supervised DAS event detection model, including a core event detection network, a student network, and a teacher network; designing a loss function and training the semi-supervised DAS event detection network; inputting a new set of data to be detected into the trained event detection network to obtain event classification and positioning results, thereby realizing event detection. The semi-supervised architecture of the DAS event detection model designed by the present invention combines two semi-supervised technologies, pseudo-labeling and consistency regularization, to fully utilize a large amount of unlabeled DAS data and a small amount of labeled DAS data for joint training, thereby reducing dependence on labeled DAS data, significantly reducing labeling costs, and improving model performance.

Description

Technical Field

[0001] The present invention relates to the field of distributed optical fiber sensing technology, and in particular to a semi-supervised DAS event detection method and system. Background Art

[0002] Distributed fiber acoustic sensing (DAS) technology based on phase-sensitive optical time-domain reflectometry (OTDR) has the advantages of being able to provide long-distance, high-resolution acoustic signal monitoring, continuous large-scale detection, and positioning of acoustic signals acting on the sensing optical fiber, high sensitivity, and fast response. It has attracted close attention and has been widely used.

[0003] By analyzing changes in optical signals propagating through optical fibers, DAS systems can detect and locate events along the fiber path, such as intrusion, vibration, or temperature changes. Leveraging the extensive underground communications cables, various vibration sources can be sensed and located with high sensitivity and precision. Consequently, DAS systems have been applied in various underground and surface monitoring applications, such as natural disaster prediction for submarine seismic activity, volcanic events and earthquakes, energy exploration in the oil and gas industry, and civil infrastructure security monitoring for energy transmission pipelines, railways, perimeter security, and urban underground structures.

[0004] In recent years, with the continuous development of artificial intelligence (AI), deep learning-based security threat event detection technologies have emerged. Numerous studies have demonstrated that models built using deep learning methods can achieve good performance in security threat event detection. However, in practical applications, due to the wide range and high resolution of DAS monitoring, acquiring large amounts of labeled DAS data is extremely time-consuming. Acquiring sufficient labeled samples to train deep learning models is often impractical. Semi-supervised learning, a machine learning strategy that combines the advantages of supervised and unsupervised learning, can improve model performance by leveraging large amounts of unlabeled data. It reduces the cost of data annotation by leveraging a small amount of labeled data with a large amount of unlabeled data, while simultaneously improving the model's generalization and prediction accuracy. This approach is particularly suitable for scenarios where acquiring large amounts of labeled data is costly or time-consuming, such as medical diagnosis, natural language processing, and image recognition. The advantages of semi-supervised learning lie in its ability to more efficiently utilize data, reduce reliance on labeled data, improve data efficiency, enhance model robustness, and maintain scalability in the big data era. In this way, the model can better understand and predict the inherent structure and relationships of the data, thereby providing more accurate and reliable results in real-world applications. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a semi-supervised DAS event detection method and system.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A semi-supervised DAS event detection method is proposed. The DAS event detection model used includes a core event detection network, a student network and a teacher network formed by the core event detection network. The core event detection network includes a backbone network, a neck network and a detection head.

[0008] The student network and the teacher network form a semi-supervised architecture for training: the student network is trained using labeled DAS data, the learnable weights are updated through back-propagation, and the teacher network is gradually updated by applying exponential moving average; the teacher network predicts the weakly enhanced unlabeled DAS data, and the prediction results above the set event classification confidence threshold are added to the training as DAS event classification pseudo-labels, so as to train the DAS event classification branch on the unlabeled DAS data, and the weights of the student network are iteratively updated; the prediction results below the set interval regression variance threshold are used as DAS event pseudo-interval labels, so as to train the interval DAS event regression branch on the unlabeled DAS data, and the weights of the student network are iteratively updated; the student network is optimized by the supervision loss, the classification loss of the DAS event classification pseudo-labels, and the regression loss of the DAS event pseudo-interval labels, and the parameters of the teacher network are only updated by the student network through exponential moving average;

[0009] The DAS data is passed through the backbone network in sequence for feature extraction, and the obtained feature map is upsampled and fused through the neck network to obtain a preliminary prediction result. Finally, the final DAS event classification prediction result and DAS event positioning prediction result are obtained through the DAS event classification branch and DAS event regression branch of the detection head respectively.

[0010] Furthermore, by building a distributed optical fiber sensing data acquisition platform, monitoring signals of DAS events are acquired to obtain the DAS data.

[0011] Furthermore, the DAS data is sequentially passed through the backbone network for feature extraction, specifically including:

[0012] The DAS data undergoes downsampling operation N1 to output feature map group C1, which in turn undergoes multiple feature fusion downsampling operations N2 to output feature map groups C2, C3, and C4, and C4 undergoes convolutional attention fusion operation N3 to output feature map group C5.

[0013] Furthermore, the downsampling operation N1 consists of a one-dimensional convolutional layer Conv1d, batch normalization BatchNorm1d, and activation function SiLU; the feature fusion downsampling operation N2 consists of the downsampling operation N1 and the segmentation operation Split, the concatenation operation Concat, and the addition operation Add; the convolution attention fusion operation N3 consists of the downsampling operation N1, the maximum pooling layer Maxpooling, the attention module Attention, and the feedforward neural network FFN.

[0014] Furthermore, the obtained feature map is subjected to upsampling and feature fusion of the neck network to obtain a preliminary prediction result, specifically including:

[0015] After the upsampling operation N4, the feature map group C5 is merged with the feature map group C4 by channel, and the feature fusion N5 outputs the feature map group F4. After the upsampling operation N4, F4 is merged with the feature map group C3 by channel, and the preliminary prediction result P3 is output after N5. After the downsampling operation N1, P3 is merged with F4 by channel, and the preliminary prediction result P4 is output after N5. After the downsampling operation N1, P4 is merged with C5 by channel, and the preliminary prediction result P5 is output after the feature fusion N5.

[0016] Furthermore, the feature fusion operation N5 is composed of a one-dimensional convolution layer, batch normalization and SiLU activation function, as well as a segmentation operation, a splicing operation, and an addition operation.

[0017] Furthermore, the final event classification prediction result and event location prediction result obtained by the detection head specifically include:

[0018] The preliminary prediction results P3, P4, and P5 are respectively subjected to multi-dimensional convolution and compound loss optimization operation N6 to obtain the final event classification prediction results and event location prediction results; the multi-dimensional convolution and compound loss optimization operation N6 consists of downsampling operation N1, depth-wise separable convolution DWConv, and IOU loss, distribution focus loss, and binary cross entropy loss BCE.

[0019] Furthermore, the student network is optimized by the supervision loss, the classification loss of the DAS event classification pseudo-labels, and the regression loss of the DAS event pseudo-interval labels, and the parameters of the teacher network are only updated by the student network through the exponential moving average, specifically including:

[0020] The labeled DAS data is weakly enhanced and then input into the student network for supervised training, and the supervision loss L is calculated. s ;

[0021]

[0022] in, represents the θth labeled DAS data, is the classification loss of the DAS event classification label, is the regression loss of the DAS event interval label;

[0023] The unlabeled DAS data are processed with strong enhancement and weak enhancement respectively. The weakly enhanced DAS data are input into the teacher network for prediction. The prediction results higher than the set event classification confidence threshold are used as DAS event classification pseudo labels, and the classification loss of the unlabeled DAS data is calculated.

[0024]

[0025] in, The i-th unlabeled DAS data, express The corresponding DAS event pseudo labels generated by the teacher network for classification;

[0026] For the prediction results below the set interval regression variance threshold, they are used as pseudo interval labels for DAS events, and the regression loss of unlabeled DAS data is calculated.

[0027]

[0028] in express The corresponding DAS event pseudo interval labels for regression generated by the teacher network;

[0029] Finally, the total unsupervised loss is obtained

[0030]

[0031] In each iteration of the training phase, the student network passes through L s and L u The total loss L composed of weighted fusion is optimized:

[0032]

[0033] λ u Parameters contributing to controlling the unsupervised loss;

[0034] The parameters of the teacher network are updated by the student network through exponential moving average, with weights θ s Student Model M s and weight θ t Teacher Model M t ,θ s and θ tThe exponential moving average update process between is as follows:

[0035]

[0036] where the superscript i denotes the i-th iteration during training and α denotes the smoothing coefficient of the exponential moving average.

[0037] A semi-supervised DAS event detection system adopts a DAS event detection model including a core event detection network, a student network and a teacher network formed by the core event detection network; the core event detection network includes a backbone network, a neck network and a detection head; the semi-supervised DAS event detection system specifically includes:

[0038] In the training module, the student network and the teacher network form a semi-supervised architecture for training: the student network is trained using labeled DAS data, the learnable weights are updated through back-propagation, and the teacher network is gradually updated by applying exponential moving average; the teacher network predicts the weakly enhanced unlabeled DAS data, and the results of the prediction above the set event classification confidence threshold are added to the training as DAS event classification pseudo-labels, so as to train the DAS event classification branch on the unlabeled DAS data, and the weights of the student network are iteratively updated; the results of the prediction below the set interval regression variance threshold are used as DAS event pseudo-interval labels, so as to train the interval DAS event regression branch on the unlabeled DAS data, and the weights of the student network are iteratively updated; the student network is optimized through the supervision loss, the classification loss of the DAS event classification pseudo-labels, and the regression loss of the DAS event pseudo-interval labels, and the parameters of the teacher network are only updated by the student network through exponential moving average;

[0039] The prediction module extracts features from the DAS data through the backbone network in sequence, and obtains preliminary prediction results through upsampling and feature fusion of the feature maps through the neck network. Finally, the final DAS event classification prediction results and DAS event positioning prediction results are obtained through the DAS event classification branch and DAS event regression branch of the detection head respectively.

[0040] The system of the present invention corresponds to the method, and the specific technical solution of the method is also applicable to the system.

[0041] Compared with the prior art, the beneficial technical effects of the present invention are:

[0042] The semi-supervised architecture of the DAS event detection model designed by the present invention combines two semi-supervised technologies, pseudo-labeling and consistency regularization, to make full use of a large amount of unlabeled DAS data and a small amount of labeled DAS data for joint training, reduce dependence on labeled DAS data, significantly reduce labeling costs, and improve model performance at the same time. In the pseudo-label part, the student network is used to predict the unlabeled data, generate pseudo-labels, and only the pseudo-labels with high prediction confidence are input into the teacher network through screening. Through the combination of strong enhancement and weak enhancement, the model's prediction consistency for the same data sample is maintained, so that the model can maximize the utilization of the potential of unlabeled data, further improving the classification accuracy and detection accuracy of the model. In the consistency regularization part, unlabeled data below the confidence threshold is input into the teacher network, forcing its prediction results to be close to the prediction of the teacher network, reducing the risk of overfitting, and improving the robustness, stability and generalization ability of the model. By using the average prediction results of the teacher network, the prediction oscillation of the model on unlabeled samples can be reduced, thereby stabilizing the training process. The detection accuracy is improved by the weighted averaging method, and the detection performance of the model is significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is the overall network framework diagram of the present invention.

[0044] Figure 2 Schematic diagram of the DAS event detection model in the present invention.

[0045] Figure 3 It is a schematic diagram of the semi-supervised architecture (student network and teacher network) in the present invention.

[0046] Figure 4 It is a flow chart of the downsampling operation N1, downsampling operation N2, and convolutional attention fusion operation N3 in the present invention; wherein, (a) is the downsampling operation N1, (b) is the downsampling operation N2, and (c) is the convolutional attention fusion operation N3.

[0047] Figure 5 It is a flow chart of the feature fusion operation N5, multi-dimensional convolution and compound loss optimization operation N6 in the present invention; wherein, (a) is the feature fusion operation N5, and (b) is the multi-dimensional convolution and compound loss optimization operation N6.

[0048] Figure 6 Schematic diagram of DAS data of five event types in an embodiment of the present invention; (a) is DAS data of a manual hoeing event type, (b) is DAS data of a manual drilling event type, (c) is DAS data of an excavator working event type, (d) is DAS data of a road breaker working event type, and (e) is DAS data of a vehicle driving event type.

[0049] Figure 7 It is the confusion matrix corresponding to each model in the data set of Table 2; Among them, (a) is the confusion matrix corresponding to the FPN-BiLSTM model, (b) is the confusion matrix corresponding to the CNN-BiGRU model, (c) is the confusion matrix corresponding to the STF-XGB model, (d) is the confusion matrix corresponding to the DSCDN model, (e) is the confusion matrix corresponding to the MTL model, and (f) is the confusion matrix corresponding to the DAS event detection model of the present invention.

[0050] Figure 8 This is a performance comparison diagram of different semi-supervised methods at 1%, 2%, 5%, and 10% label amounts.

[0051] Figure 9 2 is a schematic diagram of the detection results of DAS data of different event types according to the present invention; wherein (a) is the detection result of DAS data of manual hoeing event type, (b) is the detection result of DAS data of manual drilling event type, (c) is the detection result of DAS data of excavator working event type, (d) is the detection result of DAS data of road breaker working event type, and (e) is the detection result of DAS data of vehicle driving event type.

[0052] Figure 10 It is a schematic diagram of the effects of using different hyperparameters in the present invention. DETAILED DESCRIPTION

[0053] A preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0054] Example 1

[0055] The overall process of the present invention is as follows Figure 1 This embodiment discloses a semi-supervised DAS event detection method, including steps S1 to S4, as follows:

[0056] S1. Collect DAS data and preprocess it to obtain a training data set.

[0057] Step S1 specifically includes the following steps:

[0058] S11. Build a distributed fiber optic sensing data acquisition platform, including acquiring monitoring signals from mechanical excavation, manual hoeing, vehicle driving, and other events, and recording raw DAS data;

[0059] S12. Perform preprocessing such as denoising, labeling, weak enhancement, and strong enhancement on the collected original DAS data to obtain a training dataset and a test dataset.

[0060] S2. Build a semi-supervised DAS event detection model, including a core event detection network, a student network, and a teacher network.

[0061] Step S2 specifically includes the following steps:

[0062] S21. Design the core event detection network (Event Classification and Localization Networks, ECLNet):

[0063] like Figure 2 As shown in the figure, for a given input DAS data X, the feature map is obtained by extracting features through the backbone network Backbone, and then the preliminary prediction results are obtained through upsampling and feature fusion of the neck network Neck. Finally, the final classification prediction results and positioning prediction results are obtained through the detection head Head.

[0064] The backbone network is as follows: the original DAS data X is downsampled by N1 to output the feature map group C1, C1 is downsampled by multiple N2 in turn to output the feature map groups C2, C3, and C4, and C4 is downsampled by N3 to output the feature map group C5.

[0065] N1 consists of a one-dimensional convolutional layer (Conv1d), batch normalization (BatchNorm1d), and an activation function (SiLU). The one-dimensional convolution kernel size is 5×1, the stride is 2, and the padding is automatically calculated. N2 consists of a downsampling operation (N1), a split operation (Split), a concatenation operation (Concat), and an addition operation (Add). N3 consists of a downsampling operation (N1), a maximum pooling layer (Maxpooling), an attention module (Attention), and a feedforward neural network (FFN).

[0066] The specific structures of N1, N2 and N3 are shown in the table below. Figure 4 (a), (b), and (c) in the figure.

[0067] The overall Neck network is as follows: C5 is upsampled by N4 and merged with C4 by channel, and then N5 outputs the feature map group F4. F4 is upsampled by N4 and merged with C3 by channel, and then N5 outputs the preliminary prediction result P3. P3 is downsampled by N1 and merged with F4 by channel, and then N5 outputs the preliminary prediction result P4. P4 is downsampled by N1 and merged with C5 by channel, and then N5 outputs the preliminary prediction result P5.

[0068] N4 is the upsampling module. Figure 5 As shown in (a), N5 consists of a one-dimensional convolutional layer, batch normalization and SiLU activation function, as well as operations such as Split, Concat, and Add.

[0069] The structure of the detection head is as follows:

[0070] The preliminary prediction results P3, P4, and P5 are processed through N6 to obtain the final classification and localization results. The convolutions for classification and localization are decoupled. One convolution branch is the localization regression branch. The feature map for the localization interval has two channels, representing one event start position and one event end position. The classification channel count is the same as the number of categories. Depthwise separable convolution is used instead of conventional convolution to reduce computational complexity.

[0071] The preliminary prediction results P3, P4, and P5 are also feature maps, but they are more abstract after more upsampling and downsampling operations.

[0072] like Figure 5 As shown in (b), N6 consists of the downsampling operation N1, depth-wise separable convolution DWConv, as well as IOU loss, distribution focus loss, and binary cross entropy loss BCE.

[0073] Where IOU is:

[0074]

[0075] Among them, I pre is the prediction interval, I gt is the true interval.

[0076]

[0077] Among them, y i is the true label, p i is the predicted probability. DFL is obtained by adjusting p i The weight of makes the model pay more attention to samples that are difficult to classify.

[0078]

[0079] Among them, y i is the true label of sample i (0 or 1), p i is the probability that sample i is predicted to be positive, and N is the number of samples.

[0080] S22. Define the semi-supervised architecture (student network and teacher network):

[0081] First, the input DAS data is preprocessed to obtain labeled and unlabeled datasets. Basic weak data augmentation is performed on each DAS data point in the labeled dataset to increase the robustness of the model. Two different data augmentation strengths are used for each DAS data point in the unlabeled dataset: weak augmentation and strong augmentation. The teacher and student networks share the same main structure, both consisting of a core event detection network. The trained student network model classifies and predicts the location of each DAS data point in the unlabeled dataset, and calculates its classification probability. If the maximum probability value of the predicted classification result for an unlabeled DAS data point exceeds a preset threshold, the predicted result is added to the training dataset as a pseudo-label. Specifically, the pseudo-label generation rule is as follows: a minimum probability threshold τ is set, and only when the student network's predicted classification probability exceeds this threshold is the predicted result used as a valid pseudo-label. Subsequently, the DAS data with true labels and pseudo-labels are mixed, and the teacher network's weights are iteratively updated to improve the model's data classification accuracy. This part of the loss calculates the error between the DAS data in the unlabeled dataset and the generated pseudo-labels, using the same cross-entropy loss function as the classification loss for labeled data.

[0082] For DAS data with a prediction probability lower than the threshold, according to the consistency regularization principle, the model is required to generate consistent prediction results for the same unlabeled DAS data on different transformed versions after weak enhancement and strong enhancement. That is, for each unlabeled DAS data, the model's prediction should be consistent with its weak enhancement version and strong enhancement version. If the prediction results are inconsistent, a penalty term is added to encourage the model to be more robust to the transformation of the input data. In this process, the outputs of the teacher network and the student network are compared using consistency loss to ensure that the student network can obtain accurate guidance information from the teacher network during the learning process. Specifically, by calculating the difference between the outputs of the teacher and student networks and minimizing it, the student network can gradually approach the output of the teacher network.

[0083] S3. Define the loss function and perform network training.

[0084] Step S3 specifically includes the following steps:

[0085] S31. Define supervision loss L lab :

[0086] Supervision loss L labThe computation consists of two branches: classification and regression. The classification branch uses BCE Loss. The regression branch uses Distribution Focal Loss (DFL) and DIoU loss. The supervised loss is obtained by weighting these three losses with a certain ratio.

[0087] S32. Define training strategy:

[0088] In the training phase, the labeled DAS data is weakly enhanced and then input into the student network for supervised training, and the supervision loss L is calculated. s ;

[0089]

[0090] in, represents the i-th labeled DAS data, is the classification loss of the DAS event classification label, is the regression loss of the DAS event interval label;

[0091] The unlabeled DAS data are processed with strong enhancement and weak enhancement respectively. The weakly enhanced DAS data are input into the teacher network for prediction. The prediction results higher than the set event classification confidence threshold are used as DAS event classification pseudo labels, and the classification loss of the unlabeled DAS data is calculated.

[0092]

[0093] in, The i-th unlabeled DAS data, express The corresponding DAS event pseudo labels generated by the teacher network for classification;

[0094] For the prediction results below the set interval regression variance threshold, they are used as pseudo interval labels for DAS events, and the regression loss of unlabeled DAS data is calculated.

[0095]

[0096] in express The corresponding DAS event pseudo interval labels for regression generated by the teacher network;

[0097] Finally, the total unsupervised loss is obtained

[0098]

[0099] In each iteration of the training phase, the student network passes through L s and L u The total loss L composed of weighted fusion is optimized:

[0100]

[0101] λ u Parameters contributing to controlling the unsupervised loss;

[0102] The parameters of the teacher network are updated by the student network through the exponential moving average (EMA). The present invention has a student model Ms with a weight of θs and a teacher model Mt with a weight of θt. The EMA update process between θs and θt is as follows:

[0103]

[0104] where the superscript i represents the i-th iteration during training and α represents the smoothing coefficient of EMA.

[0105] S4. Model application:

[0106] After the training phase is completed, the DAS data in the test dataset is directly processed by the core event detection network, producing the final event classification and event localization results without the need for additional processing steps.

[0107] Example 2

[0108] The present invention also discloses a semi-supervised DAS event detection system. The DAS event detection model used includes a core event detection network, and a student network and a teacher network formed by the core event detection network. The core event detection network includes a backbone network, a neck network, and a detection head. The semi-supervised DAS event detection system specifically includes:

[0109] The training module uses a semi-supervised architecture composed of a student network and a teacher network for training. The student network predicts unlabeled DAS data. Event classification prediction results with prediction probabilities higher than a set value are added to the training as pseudo-labels, and the weights of the teacher network are iteratively updated. For DAS data with prediction probabilities lower than a set value, consistency regularization is used, requiring the student network and teacher network to have consistent prediction results for weakly augmented and strongly augmented versions of unlabeled DAS data, respectively. During the training phase, the student network and teacher network are optimized using supervision loss, pseudo-label cross-entropy loss, and consistency loss.

[0110] The prediction module extracts features from the DAS data through the backbone network in sequence, and obtains preliminary prediction results through upsampling and feature fusion of the feature maps through the neck network. Finally, the final event classification prediction results and event location prediction results are obtained through the detection head.

[0111] The system of the present invention corresponds to the method, and the specific technical solution of the method is also applicable to the system.

[0112] Experimental results:

[0113] The event signal samples collected on site are as follows: Figure 6 As shown, Figure 6 (a), (b), (c), (d), and (e) are five events: manual hoeing, manual drilling, excavator operation, road breaker operation, and vehicle driving. The database used for the experiment is constructed as shown in Table 1.

[0114] Table 1. Database constructed for five types of target event recognition:

[0115] Label Event Type training set Validation set Test set Total dataset 1 Manual hoeing 7221 1444 1444 10109 2 Manual drilling 7193 1439 1439 10070 3 Excavator work 7441 1488 1488 10417 4 Road breaking machine working 7389 1478 1478 10345 5 Vehicle driving 7346 1469 1469 10284

[0116] The method of the present invention is compared with various methods to analyze the contribution of the present invention. It is worth mentioning that, in theory, the semi-supervised framework of the present invention is applicable to tasks such as event classification and detection. However, current research on DAS data is limited to event classification, and few methods for implementing localization tasks have been proposed. This work can be said to be the first study to truly use DAS data for event detection. Therefore, in the performance comparison of supervised learning, the present invention only compares the classification performance of the model's core test event detection network ECLGNet.

[0117] In Table 2, the performance of the proposed algorithm in terms of accuracy, precision, recall and F1 value is illustrated.

[0118]

[0119]

[0120]

[0121] Where FP, TP, FN and TN represent the number of false positive samples, true positive samples, false negative samples and true negative samples respectively.

[0122] Results show that the network introduced in this paper demonstrates excellent performance across various evaluation metrics in the field of multi-label event classification. Notably, the proposed method outperforms other classification networks in accuracy, precision, recall, and F1 score, achieving 96.31%, 96.33%, 96.32%, and 96.33%, respectively. These results underscore the effectiveness of the proposed method specifically for multi-label event classification.

[0123] Table 2, performance comparison with other classification models:

[0124]

[0125]

[0126] To illustrate the performance of the model classification in more detail, the confusion matrix is ​​plotted on the constructed dataset, as shown in Figure 7 As shown. The method of the present invention is evaluated by comparing it with the above-mentioned state-of-the-art models. It is obvious that the method proposed in the present invention outperforms other state-of-the-art multi-label classification networks. Regarding the average accuracy of each event category, this paper achieves the best performance in the five event categories of manual hoeing, manual drilling, excavator work, road breaker work and vehicle driving, with average accuracy scores of 97.78%, 97.50%, 95.16%, 95.33% and 95.85% respectively. Among them, Figure 7 (a), (b), (d), (d), and (e) are the confusion matrices corresponding to the FPN-BiLSTM model, CNN-BiGRU model, STF-XGB model, DSCDN model, MTL model, and the DAS event detection model of the present invention, respectively.

[0127] Next, the proposed method is compared with several current advanced semi-supervised methods on datasets with 1%, 2%, 5%, and 10% annotation labels, and the performance is compared using mAP as the evaluation indicator.

[0128]

[0129]

[0130] Where P is the precision, R is the recall, and N is the number of categories.

[0131] The results are shown in Table 3, which reveals the effectiveness of the semi-supervised method in utilizing unlabeled data. The results are visualized as follows Figure 8As shown in Figure 2, the performance of the method of the present invention is better than all the comparison methods in all label quantities. It is worth mentioning that in order to make a fair comparison with other semi-supervised methods, the ECLNet designed by the present invention is uniformly used as the default detector framework to illustrate the method of the present invention. In addition, a method similar to the object detection task is used to visualize the results of the present invention, as shown in Figure 2. Figure 9 As shown, in order to more intuitively feel and compare the performance of the method of the present invention. Figure 9 (a), (b), (c), (d), and (e) represent manual hoeing, manual drilling, excavator operation, road breaker operation, and vehicle driving, respectively.

[0132] Table 3, performance comparison of 1%, 2%, 5%, and 10% label amounts with other semi-supervised models:

[0133]

[0134]

[0135] In order to analyze the effectiveness of the key components in the method designed by the present invention, the present invention conducted an ablation study to verify the key design of the present invention. All ablation experiments were performed on the DAS dataset with 10% labels provided by the present invention.

[0136] 1) The impact of each component: The present invention removes some components of the semi-supervised method and observes the changes in model performance. The present invention eliminates the influence of exponential moving average, classification pseudo-labels and pseudo-interval labels. The results are shown in Table 4. By applying the classification pseudo-label strategy, the method of the present invention achieves 65.28mAP. Then, by using the exponential moving average strategy of the teacher-student model to update the teacher model with the student model, the method of the present invention further improves by 14.85 points. Further application of pseudo-interval labels achieves 85.44mAP, which is 5.31 points better than the classification pseudo-label plus exponential moving average.

[0137] Table 4, Ablation studies:

[0138]

[0139]

[0140] 2) Influence of hyperparameters: This section is mainly used to explore the influence of hyperparameters used in the method of the present invention. Figure 10 shown.

[0141] Figure 10(a) in the figure shows the impact of different event classification confidence thresholds. The best performance is achieved when the threshold is set to 0.9, and lower or higher thresholds will lead to a significant drop in performance. When the threshold varies from 0 to 0.9, as expected, the number of pseudo labels generated increases as the threshold decreases. Models using too high a threshold (such as 0.95) cannot obtain satisfactory results because the number of pseudo labels generated is very small. On the other hand, models using low thresholds (such as 0.7) also cannot obtain good results because the model generates too many pseudo labels, which are likely to be false positive instances. The present invention also observed that if the threshold is lower than 0.5, the model cannot even converge.

[0142] like Figure 10 As shown in (b), the present invention studies the influence of the interval regression variance threshold. The best performance is achieved when the threshold is set to 0.02.

[0143] The present invention also studies the influence of EMA rate α. The present invention uses various EMA rates from 0.5 to 0.9999 to evaluate the model and Figure 10 (c) in Figure 3 shows the mAP results of the teacher model. We observe that with a smaller EMA rate (e.g., α = 0.5), the model has a lower mAP because the student contributes more to the teacher model in each iteration. This means that the teacher model may be adversely affected by noisy pseudo-labels. As α increases, this unstable learning curve can be stabilized and improved. When α reaches 0.999, it exhibits the best mAP. However, if α continues to increase, the teacher model will grow too slowly because the teacher model mainly derives the next model weights from the previous teacher model weights.

[0144] To check the unsupervised loss weight λ u The present invention changes the unsupervised loss weight from 1.0 to 8.0 in the case of 10% labeled data. Figure 10 As shown in (d) in Figure 3, the model performs poorly at lower unsupervised loss weights. On the other hand, we observed that the model performs best when the unsupervised loss weight is 5.0. However, when the weight is increased to 8.0, the model training fails to converge.

[0145] Experimental results show that these components are crucial to improving model performance, and their fusion significantly improves the model's detection performance.

[0146] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. It is intended that all variations within the meaning and range of equivalents of the claims be embraced herein, and any reference signs in the claims should not be construed as limiting the claims to which they relate.

[0147] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A semi-supervised DAS event detection method, characterized in that: The DAS event detection model used includes a core event detection network, a student network and a teacher network composed of the core event detection network; the core event detection network includes a backbone network, a neck network and a detection head; The student network and teacher network form a semi-supervised architecture for training: the student network is trained using labeled DAS data, updating its learnable weights through backpropagation, and gradually updating the teacher network using exponential moving average. The teacher network makes predictions on weakly augmented unlabeled DAS data, and predictions above a set event classification confidence threshold are added to the training as pseudo-labels for DAS event classification. This allows the DAS event classification branch to be trained on the unlabeled DAS data, and the weights of the student network are iteratively updated. The prediction results below the set interval regression variance threshold are used as pseudo interval labels for DAS events to train the interval DAS event regression branch on unlabeled DAS data and iteratively update the weights of the student network. The student network is optimized using the supervision loss, the classification loss of the DAS event classification pseudo labels, and the regression loss of the DAS event pseudo interval labels. The parameters of the teacher network are only updated by the student network through exponential moving average. The DAS data is passed through the backbone network in sequence for feature extraction, and the obtained feature map is upsampled and fused through the neck network to obtain a preliminary prediction result. Finally, the final DAS event classification prediction result and DAS event positioning prediction result are obtained through the DAS event classification branch and DAS event regression branch of the detection head respectively.

2. The semi-supervised DAS event detection method according to claim 1, characterized in that By building a distributed optical fiber sensing data acquisition platform, monitoring signals of DAS events are acquired to obtain the DAS data.

3. The semi-supervised DAS event detection method according to claim 1, characterized in that The DAS data is sequentially passed through the backbone network for feature extraction, specifically including: The DAS data undergoes downsampling operation N1 to output feature map group C1, which in turn undergoes multiple feature fusion downsampling operations N2 to output feature map groups C2, C3, and C4, and C4 undergoes convolutional attention fusion operation N3 to output feature map group C5.

4. The semi-supervised DAS event detection method according to claim 3, characterized in that: The downsampling operation N1 consists of a one-dimensional convolutional layer Conv1d, batch normalization BatchNorm1d, and activation function SiLU; the feature fusion downsampling operation N2 consists of the downsampling operation N1 and the segmentation operation Split, the concatenation operation Concat, and the addition operation Add; the convolution attention fusion operation N3 consists of the downsampling operation N1, the maximum pooling layer Maxpooling, the attention module Attention, and the feedforward neural network FFN.

5. The semi-supervised DAS event detection method according to claim 3, characterized in that: The obtained feature map is subjected to upsampling and feature fusion of the neck network to obtain a preliminary prediction result, specifically including: After the upsampling operation N4, the feature map group C5 is merged with the feature map group C4 by channel, and the feature fusion N5 outputs the feature map group F4. After the upsampling operation N4, F4 is merged with the feature map group C3 by channel, and the preliminary prediction result P3 is output after N5. After the downsampling operation N1, P3 is merged with F4 by channel, and the preliminary prediction result P4 is output after N5. After the downsampling operation N1, P4 is merged with C5 by channel, and the preliminary prediction result P5 is output after the feature fusion N5.

6. The semi-supervised DAS event detection method according to claim 5, characterized in that: The feature fusion operation N5 is composed of a one-dimensional convolution layer, batch normalization and SiLU activation function, as well as a segmentation operation, a splicing operation, and an addition operation.

7. The semi-supervised DAS event detection method according to claim 5, characterized in that: The final event classification prediction results and event location prediction results obtained by the detection head specifically include: The preliminary prediction results P3, P4, and P5 are respectively subjected to multi-dimensional convolution and compound loss optimization operation N6 to obtain the final event classification prediction results and event location prediction results; the multi-dimensional convolution and compound loss optimization operation N6 consists of downsampling operation N1, depth-wise separable convolution DWConv, and IOU loss, distribution focus loss, and binary cross entropy loss BCE.

8. The semi-supervised DAS event detection method according to claim 1, characterized in that: The student network is optimized by the supervision loss, the classification loss of the DAS event classification pseudo-label, and the regression loss of the DAS event pseudo-interval label. The parameters of the teacher network are only updated by the student network through the exponential moving average, specifically including: The labeled DAS data is weakly enhanced and then input into the student network for supervised training, and the supervision loss L is calculated. s ; in, represents the i-th labeled DAS data, is the classification loss of the DAS event classification label, is the regression loss of the DAS event interval label; The unlabeled DAS data are processed with strong enhancement and weak enhancement respectively. The weakly enhanced DAS data are input into the teacher network for prediction. The prediction results higher than the set event classification confidence threshold are used as DAS event classification pseudo labels, and the classification loss of the unlabeled DAS data is calculated. in, The i-th unlabeled DAS data, express The corresponding DAS event pseudo labels generated by the teacher network for classification; For the prediction results below the set interval regression variance threshold, they are used as pseudo interval labels for DAS events, and the regression loss of unlabeled DAS data is calculated. in express The corresponding DAS event pseudo interval labels for regression generated by the teacher network; Finally, the total unsupervised loss is obtained In each iteration of the training phase, the student network passes through L s and L u The total loss L composed of weighted fusion is optimized: λ u Parameters contributing to controlling the unsupervised loss; The parameters of the teacher network are updated by the student network through exponential moving average, with weights θ s Student Model M s and weight θ t Teacher Model M t ,θ s and θ t The exponential moving average update process between is as follows: where the superscript i denotes the i-th iteration during training and α denotes the smoothing coefficient of the exponential moving average.

9. A semi-supervised DAS event detection system, characterized in that: The DAS event detection model used includes a core event detection network, as well as a student network and a teacher network composed of the core event detection network. The core event detection network includes a backbone network, a neck network, and a detection head. The semi-supervised DAS event detection system specifically includes: In the training module, the student network and the teacher network form a semi-supervised architecture for training: the student network is trained using labeled DAS data, the learnable weights are updated through back-propagation, and the teacher network is gradually updated by applying exponential moving average; the teacher network predicts the weakly enhanced unlabeled DAS data, and the results of the prediction above the set event classification confidence threshold are added to the training as DAS event classification pseudo-labels, so as to train the DAS event classification branch on the unlabeled DAS data, and the weights of the student network are iteratively updated; the results of the prediction below the set interval regression variance threshold are used as DAS event pseudo-interval labels, so as to train the interval DAS event regression branch on the unlabeled DAS data, and the weights of the student network are iteratively updated; the student network is optimized through the supervision loss, the classification loss of the DAS event classification pseudo-labels, and the regression loss of the DAS event pseudo-interval labels, and the parameters of the teacher network are only updated by the student network through exponential moving average; The prediction module extracts features from the DAS data through the backbone network in sequence, and obtains preliminary prediction results through upsampling and feature fusion of the feature maps through the neck network. Finally, the final DAS event classification prediction results and DAS event positioning prediction results are obtained through the DAS event classification branch and DAS event regression branch of the detection head respectively.

Images