FilterNet: A CNN-RNN based filter model used for raw tunnel lining GPR data
Bibliographic record
Abstract
Ground-Penetrating Radar (GPR) technology, with its characteristics of being fast, non-destructive, and high-resolution, has become an important tool for detecting underground structures. However, GPR data inevitably suffer from environmental noise and electromagnetic interference during the acquisition process, leading to decreased data quality and increased complexity in data processing. Traditional filtering algorithms have limitations such as low discrimination between noise and signal, poor adaptability, and inability to process data in real time. This paper proposes a filtering model based on deep neural networks, called FilterNet. FilterNet combines Convolution Neural Networks (CNN) and recurrent neural networks (RNN) for processing multi-channel data. It can perform end-to-end filtering directly on the raw tunnel lining GPR data, achieving functions such as removing air reflection waves, denoising, and automatic gain. Using PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity Index) as statistical indicators, it is shown that the FilterNet model improves filtering precision. The SSIM of all three models is 0.997, and the PSNR of FilterNet1D and FilterNet are 19.06 and 19.41, respectively. Furthermore, tests on the model's processing efficiency indicate that FilterNet requires less memory and is more efficient than the UNet model. FilterNet's parameters are only 48 % of those of UNet. Its GFLOPS (Giga Floating Point Operations Per Second) is only one-third of UNet's, and it can process data in real time. Additionally, FilterNet performs exceptionally well in suppressing random noise.
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 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".