LR-Net for Weak Vibration Event Location and Recognition With DAS
Bibliographic record
Abstract
Distributed acoustic sensing (DAS) has been deployed across various large-scale infrastructures for safety monitoring and health maintenance operations. Among these applications, the localization and recognition of vibrations are common and critical post-processing tasks. Currently, multi-event vibration localization and recognition present a significant challenge for post-processing algorithms. Moreover, weak vibrations, which are marked by short durations and limited propagation ranges, further exacerbate the difficulty of the accurate location and recognition. Consequently, these challenges contribute to a high false alarm rate and missed detection rate across the DAS system. To address those challenges, this paper proposes a location and recognition convolutional neural network (LR-Net) that can achieves end-to-end and multi-event recognition and localization along the fiber within a single sample. In the model, we propose the location-attention-mechanism feature fusion and squeeze framework (LAMFS) and dynamic matching strategy (DMS) to enable the model to focus on weak vibration and enhance its fitting ability. In the field experiments conducted in three typical scenarios, LR-Net achieves a 99.1% mAP for six types of events with merely average location error of ±1.5m. Moreover, the Nuisance Alarm Rate (NAR) and the Missing Alarm Rate (MAR) were only 1.4% and 1.06% respectively. These results demonstrate superior performance compared with other deep learning models. Above all, the proposed algorithm possesses significant practical value and can be adapted to other scenarios such as pipeline leak detection, perimeter security, and protection of important facilities.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".