Pith. sign in

REVIEW 5 major objections 5 minor 71 references

Simultaneous Automatic Picking and Manual Picking Refinement for First-Break

T0 review · 5 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A latent-variable training loop learns first-break picking from noisy manual labels.

desk verdict A solid, honest application of latent-variable noisy-label learning to first-break picking, worth a serious referee but with a few reproducibility gaps. read the letter →

arxiv 2502.01474 v1 pith:R6SDE4ZW submitted 2025-02-03 cs.CV eess.IV

Haowen Bai , Zixiang Zhao , Jiangshe Zhang , Yukun Cui , Chunxia Zhang , Zhenbo Guo , Yongjun Wang This is my paper · ORCID
classification cs.CVeess.IV
keywords first-breakpickingmicroseismicnoisylabelslatentvariablelabelrefinementdeeplearningseismicdataUNet
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

First-break picking locates the arrival time of seismic energy, and deep-learning systems usually train on manual picks. This paper introduces SPR, which treats the true first-break as a latent variable that is updated alongside network weights, so the model learns a cleaner labeling rather than memorizing noisy manual picks. In experiments on two hard-rock seismic datasets, SPR improves exact-match hit rate and mean absolute error over five supervised baselines, and it outperforms them when labels are randomly shifted by Gaussian noise. The same latent variable can also refine manual picks at inference time. If the method holds up, noisy or partially mislabeled seismic labels no longer need a separate cleaning step before training.

What carries the argument

The central object is the latent first-break label $\tilde{y}$ combined with the alternating maximization of the log-likelihood in Eq. (5). That objective couples a Laplace labeling prior $P(y|\tilde{y})\propto\prod_k \exp(-\|t_k-s_k\|/\gamma)$ with the network's binary cross-entropy on $\tilde{y}$. Fixing $\tilde{y}$ makes the $W$-update ordinary supervised training; fixing $W$ makes the $\tilde{y}$-update pull latent picks toward a $\gamma$-weighted compromise between the manual pick and the network's current prediction. This single mechanism serves both automatic picking (Eq. (8)) and manual-pick refinement (Eq. (9)).

What would settle it

Create a synthetic or semi-synthetic seismic dataset with known true first-breaks, then inject label errors whose probability increases as trace signal-to-noise ratio decreases; if SPR's refined labels are noticeably less accurate on the low-SNR subset than on the clean subset, the independence assumption in Eq. (2) is violated and a conditional noise model would be needed.

Watch

Extended reading notes

Core claim

The paper's central claim is that maximizing the joint likelihood of manual labels and a latent first-break, rather than the likelihood of manual labels alone, produces a first-break picker that is more accurate and more robust to label noise. The model writes $P(y,\tilde{y}|x;W)=P(y|\tilde{y})P(\tilde{y}|x;W)$, with a Laplace prior on the gap between manual and latent picks and the network predicting the latent picks. Alternating updates—cross-entropy training on current latent picks for $W$, and a prior-plus-prediction compromise for $\tilde{y}$—are what allow the method to see through outliers and mislabels. On the Sudbury and Lalor datasets the trained model achieves higher hit rates and lower mean absolute error than five comparison networks, transfers better across sites, and, when trained on labels of which only 13.30% are correct, still recovers most true first-breaks.

Load-bearing premise

The method assumes that the chance a manual pick is wrong does not depend on the signal itself, only on the distance from the true first-break; if noisy traces are systematically mislabeled more often, the Laplace prior cannot represent that and refinement will inherit the network's own bias.

Editorial extensions

If this is right

  • Seismic datasets with outlier traces or partially wrong manual picks can be used for training without a prior cleaning stage, because the latent variable absorbs sparse errors.
  • Any segmentation-style network can be plugged into SPR, so the accuracy gain is available to existing picking architectures without redesigning them.
  • The trained model can double as a label-refinement tool, letting analysts correct or audit manual picks across large surveys.
  • Because the network learns a distribution over the true first-break rather than memorizing manual labels, cross-site generalization is improved, as the Sudbury-to-Lalor experiment shows.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If mislabeling is more common on low-signal-to-noise traces, the independence assumption $P(y|\tilde{y},x)=P(y|\tilde{y})$ will bias refined labels toward the network's own errors, so conditioning the noise prior on $x$ is a natural extension.
  • The alternating procedure is a form of generalized expectation-maximization, so convergence may depend on initializing $\tilde{y}=y$ and on the $\gamma$ schedule; annealing $\gamma$ during training is a testable way to reduce sensitivity to the fixed $\gamma=5$.
  • The same latent-label loop could be applied to other sparse, densely annotated geophysical labels, such as P- and S-wave arrival pairs or DAS event detections, wherever label noise is sparse rather than systematic.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The manuscript proposes SPR, a probabilistic latent-variable model for first-break picking. SPR treats the true first-break as a latent variable ey, models manual labels y as generated from ey via a Laplace prior, and trains a UNet to predict ey by alternating between cross-entropy updates of the network weights and an update of ey that balances the prior with the network output. At inference, the network output can be used directly (Eq. (8)), or combined with the prior to refine available manual labels (Eq. (9)). Experiments on the Sudbury and Lalor datasets compare SPR with five deep-learning baselines on picking accuracy, cross-site generalization, noisy signals, and noisy labels, and report that SPR generally achieves higher hit rates and lower MAE.

Significance. If the results hold, SPR would be a useful contribution to robust first-break picking, particularly for training with imperfect manual labels. The probabilistic framing is principled and the use of a latent variable to decouple manual labels from the learning target is a clean idea. The derivation of the objective is transparent, and the paper makes available all experimental details on public datasets. However, the validation has several gaps—an unspecified ey update, test-set hyperparameter selection, a single-direction generalization test, and a synthetic noise model that matches the model's assumptions—so the strength of the empirical claims is currently not commensurate with the conclusions.

major comments (5)
  1. [II-B, Eq. (7)] The paper does not specify how the update of ey is computed in Eq. (7). Since ey is a binary matrix with exactly one 1 per trace, the minimization involves a discrete search (or a closed-form per-trace selection), but Algorithm 1 simply states 'Update ey by Eq. (7)' without giving the procedure, any approximation, or initialization beyond ey=y. This is load-bearing because the alternating update is the core of SPR; without this detail the algorithm is not reproducible. Please provide the exact per-trace update rule (e.g., evaluating the objective for each candidate first-break position) and discuss any issues of local optima.
  2. [III-G, Table VI] The hyperparameter γ is selected by sweeping over γ on the Sudbury dataset, and the same dataset (or its test split) is then used for the reported results in Tables II, IV, and V. This is effectively test-set tuning: the best γ is chosen on the basis of the very metrics used to evaluate SPR in the main comparisons. The authors should instead use a separate validation split (or report results for all γ with a clear selection protocol) so that the reported numbers are not optimistically biased. This is important because Table VI shows that performance varies strongly with γ (e.g., HR0 from 55.11 to 74.20).
  3. [III-F and IV-A] The synthetic label-noise experiment uses Gaussian shifts with variance 3, i.e., input-independent, symmetric noise. This exactly instantiates the model's assumption in Eq. (2) that P(y|ey,x)=P(y|ey). The paper concedes in Section IV-A that real annotation errors may depend on x (low-SNR traces, different annotation standards, etc.). Without an experiment where label noise is correlated with trace characteristics, the claim that SPR 'refines misaligned manual annotations' is only validated under a favorable noise model. Adding an experiment with, e.g., larger label shifts on low-SNR traces would directly test the robustness of the refinement procedure.
  4. [Table III] The cross-site generalization experiment is one-directional (train on Sudbury, test on Lalor), and on the MAE metric SPR is not the best: ResUNet achieves 5.1749 while SPR achieves 6.2842. The paper's claim that SPR 'maintained a high degree of consistency with manual picking' and shows superior generalization is therefore only partially supported. Please report the reverse direction (Lalor→Sudbury) or otherwise justify why the one-directional test is sufficient, and address the HR/MAE trade-off in the discussion.
  5. [Tables II–V] All experimental results appear to be based on a single training run per configuration. Several improvements over the baselines are small (e.g., Lalor HR0: 92.60 vs 91.91 for MSNet), and without error bars or multiple seeds it is not possible to assess whether these differences are statistically meaningful. The authors should report mean and standard deviation over at least three runs (or provide an equivalent stability analysis) for the main comparisons.
minor comments (5)
  1. [III-C] In Section III-C, the text reads 'SPR exhibits a higher HR and a lower MSE', but the metric used throughout is MAE; please correct the terminology.
  2. [Table I] In Table I, the Lalor dataset is listed with 1001 sample points but an input shape of 192×1504; please reconcile this discrepancy (the text also states 1001 sampling points, while the input width 1504 suggests a different sample count).
  3. [Algorithm 1] The instruction 'Calculate the number of steps L based on the dataset size' is vague; please specify how L is determined (e.g., number of batches per epoch).
  4. [II-A, Eq. (5)] The phrase 'logging it' should be 'taking the logarithm' for clarity.
  5. [II-B, Eq. (7)] The two sums in Eq. (7) use different indices (k over traces, i,j over sample points); please make explicit that s_k is the index of the non-zero entry in ey_k, and that the Laplace term couples the two sums through this index.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: SPR's derivation is self-contained, and its benchmarks are external and not encoded in the model definition.

full rationale

I walked the derivation from Eq. (2) through Eq. (9). The likelihood factorization P(y,ey|x;W)=P(y|ey)P(ey|x;W) is a stated modeling assumption (Section II-A), not a result derived from the target. The Laplace prior Eq. (3) uses manual picks as centers, but this is the intended noise model, not a hidden reuse of the evaluation labels: training-set manual picks are used for training; test-set manual picks are used only as evaluation reference (Section III-B). Eq. (6) and Eq. (7) are the standard alternating maximization of the log-likelihood Eq. (5); the network is trained on the latent ey, not directly on y. Inference Eq. (8) is the network's predictive posterior and Eq. (9) is a MAP refinement combining the prior and network; neither reduces to the manual labels by construction. The paper has no load-bearing self-citations: the only data source is the external Sudbury/Lalor benchmark [69], and the network architecture is the external UNet [68]. Section IV-A explicitly acknowledges the conditional-independence assumption P(y|ey,x)=P(y|ey) and lists x-dependent mislabeling as future work; that is a limitation with testable consequences, not a circular step. The gamma=5 choice is a hyperparameter investigated in Section III-G; even if it raises overfitting concerns on the Sudbury benchmark, it is not a fitted quantity renamed as a prediction. Hence no circularity by the standards of this review.

Assumptions & free parameters 1 free parameters · 5 assumptions · 1 invented entities

The central claim rests on a small set of modeling assumptions and one tuned hyperparameter. The latent variable ey is initialized from manual labels and regularized toward them, so the refinement result is partially anchored to the same labels used for evaluation.

free parameters (1)
  • gamma (Laplace scale) = 5
    Controls the strength of the label prior in Eq. (3). Chosen after a parameter sweep on Sudbury (Table VI) with no separate validation set; central to the method's behavior.
assumptions (5)
  • domain assumption Manual first-break labels are drawn from a Laplace distribution centered at the true first-break (Eq. 3).
    Underlies the prior P(y|ey); if real mislabeling is not Laplace, the refinement step is misspecified.
  • domain assumption Label noise is independent of the input signal x given the true first-break, P(y|ey,x;W)=P(y|ey).
    Simplifies Eq. (2); acknowledged in Sec. IV-A as potentially violated by expertise-related errors.
  • ad hoc to paper The alternating updates of ey and W converge to a local maximum of Eq. (5).
    No convergence proof is provided; Algorithm 1 simply alternates for a fixed number of epochs.
  • domain assumption A U-Net can adequately represent the posterior P(ey|x;W).
    The method is stated to work with any network, but experiments use U-Net only; representational capacity is assumed.
  • domain assumption Manual picks on test sets are treated as ground truth for evaluation.
    All reported metrics compare against manual picks, even though the method's premise is that manual picks can be noisy.
invented entities (1)
  • Latent true first-break ey
    purpose: Acts as a corrected version of the manual labels; updated during training to balance the manual pick and the network prediction.
    No falsifiable prediction outside the model; it is a trainable latent variable initialized from manual labels.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Simultaneous Automatic Picking and Manual Picking Refinement for First-Break." pith.science (2026). https://pith.science/paper/R6SDE4ZW

@misc{pith2026250201474,
  author       = {Pith},
  title        = {Pith review of: Simultaneous Automatic Picking and Manual Picking Refinement for First-Break},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R6SDE4ZW}},
  note         = {Machine review of arXiv:2502.01474}
}
read the original abstract

First-break picking is a pivotal procedure in processing microseismic data for geophysics and resource exploration. Recent advancements in deep learning have catalyzed the evolution of automated methods for identifying first-break. Nevertheless, the complexity of seismic data acquisition and the requirement for detailed, expert-driven labeling often result in outliers and potential mislabeling within manually labeled datasets. These issues can negatively affect the training of neural networks, necessitating algorithms that handle outliers or mislabeled data effectively. We introduce the Simultaneous Picking and Refinement (SPR) algorithm, designed to handle datasets plagued by outlier samples or even noisy labels. Unlike conventional approaches that regard manual picks as ground truth, our method treats the true first-break as a latent variable within a probabilistic model that includes a first-break labeling prior. SPR aims to uncover this variable, enabling dynamic adjustments and improved accuracy across the dataset. This strategy mitigates the impact of outliers or inaccuracies in manual labels. Intra-site picking experiments and cross-site generalization experiments on publicly available data confirm our method's performance in identifying first-break and its generalization across different sites. Additionally, our investigations into noisy signals and labels underscore SPR's resilience to both types of noise and its capability to refine misaligned manual annotations. Moreover, the flexibility of SPR, not being limited to any single network architecture, enhances its adaptability across various deep learning-based picking methods. Focusing on learning from data that may contain outliers or partial inaccuracies, SPR provides a robust solution to some of the principal obstacles in automatic first-break picking.

Figures

Figures reproduced from arXiv: 2502.01474 by the authors.

Figure 1
Figure 1. (a) Amplitude signals. Manual picking are marked with red circles. (b) [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The training process of SPR. The network parameters [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Comparison of picking results of different methods for trace 1005070-1005172 of dataset Sudbury. Manual picking and picking results of different [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Comparison of picking results of different methods for trace 2362428-2362573 of dataset Lalor. Manual picking and picking results of different [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Generalization experiments for first-break picking. The model is trained on data Sudbury and tested on data Lalor. The figure shows traces 298083-298257 [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Comparison of picking results of different methods for trace 1005070-1005172 of dataset Sudbury at noise level 0.05. Manual picking and picking [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Comparison of picking results of different methods for trace 1005070-1005172 of dataset Sudbury at noise level 0.1. Manual picking and picking [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Comparison of picking results of different methods for trace 1005070-1005172 of dataset Sudbury at noise level 0.2. Manual picking and picking [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: Comparison of noisy label refinement of different methods for trace 71347-71407 of dataset Sudbury. Manual picking and picking results of different [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: Comparison of noisy label picking of different methods for trace 1005070-1005172 of dataset Sudbury. Manual picking and picking results of different [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

71 extracted references · 68 canonical work pages

  1. [1]

    Linking microseismic event observations with geomechanical models to minimise the risks of storing co2 in geological formations,

    J. Verdon, J.-M. Kendall, D. White, and D. Angus, “Linking microseismic event observations with geomechanical models to minimise the risks of storing co2 in geological formations,” Earth and Planetary Science Letters, vol. 305, no. 1-2, pp. 143–152, 2011

    work page 2011

  2. [2]

    The microseismic response at the in salah carbon capture and storage (ccs) site,

    A. L. Stork, J. P. Verdon, and J.-M. Kendall, “The microseismic response at the in salah carbon capture and storage (ccs) site,” International Journal of Greenhouse Gas Control , vol. 32, pp. 159–171, 2015

    work page 2015

  3. [3]

    Technical potential of salt caverns for hydrogen storage in europe,

    D. G. Caglayan, N. Weber, H. U. Heinrichs, J. Linßen, M. Robinius, P. A. Kukla, and D. Stolten, “Technical potential of salt caverns for hydrogen storage in europe,” International Journal of Hydrogen Energy , vol. 45, no. 11, pp. 6793–6805, 2020

    work page 2020

  4. [4]

    Field trials of distributed acoustic sensing for geophysical monitoring,

    J. Mestayer, B. Cox, P. Wills, D. Kiyashchenko, J. Lopez, M. Costello, S. Bourne, G. Ugueto, R. Lupton, G. Solano et al. , “Field trials of distributed acoustic sensing for geophysical monitoring,” in Seg technical program expanded abstracts 2011. Society of Exploration Geophysicists, 2011, pp. 4253–4257

    work page 2011

  5. [5]

    Distributed sensing of microseisms and teleseisms with submarine dark fibers,

    E. F. Williams, M. R. Fern ´andez-Ruiz, R. Magalhaes, R. Vanthillo, Z. Zhan, M. Gonz ´alez-Herr´aez, and H. F. Martins, “Distributed sensing of microseisms and teleseisms with submarine dark fibers,” Nature communications, vol. 10, no. 1, p. 5778, 2019

    work page 2019

  6. [6]

    Microearthquakes at flathead lake, montana: A study using automatic earthquake processing,

    P. R. Stevenson, “Microearthquakes at flathead lake, montana: A study using automatic earthquake processing,” Bulletin of the Seismological Society of America , vol. 66, no. 1, pp. 61–80, 1976

    work page 1976

  7. [7]

    Automatic earthquake recognition and timing from single traces,

    R. V . Allen, “Automatic earthquake recognition and timing from single traces,” Bulletin of the seismological society of America , vol. 68, no. 5, pp. 1521–1532, 1978

    work page 1978

  8. [8]

    First-arrival automatic picking based on improved energy ratio method and outlier detection theory,

    Z. Qin, S. Pan, L. Hu, Q. Cui, and Q. Gou, “First-arrival automatic picking based on improved energy ratio method and outlier detection theory,” Acta Geophysica, vol. 69, no. 5, pp. 1667–1677, 2021

    work page 2021

Show all 71 references
  1. [9]

    Automatic phase pickers: Their present use and future prospects,

    R. Allen, “Automatic phase pickers: Their present use and future prospects,” Bulletin of the Seismological Society of America , vol. 72, no. 6B, pp. S225–S242, 1982

    1982

  2. [10]

    An automatic phase picker for local and teleseismic events,

    M. Baer and U. Kradolfer, “An automatic phase picker for local and teleseismic events,” Bulletin of the Seismological Society of America , vol. 77, no. 4, pp. 1437–1445, 1987

    1987

  3. [11]

    Characterization of global seismograms using an automatic-picking algorithm,

    P. S. Earle and P. M. Shearer, “Characterization of global seismograms using an automatic-picking algorithm,” Bulletin of the Seismological Society of America , vol. 84, no. 2, pp. 366–376, 1994

    1994

  4. [12]

    Identification and location method of microseismic event based on improved sta/lta algorithm and four-cell-square-array in plane algorithm,

    S.-c. Li, S. Cheng, L.-p. Li, S.-s. Shi, and M.-g. Zhang, “Identification and location method of microseismic event based on improved sta/lta algorithm and four-cell-square-array in plane algorithm,” International Journal of Geomechanics , vol. 19, no. 7, p. 04019067, 2019

    2019

  5. [13]

    Sta/lta fractal dimension algorithm of detecting the p-wave arrival,

    J. Zhang, Y . Tang, and H. Li, “Sta/lta fractal dimension algorithm of detecting the p-wave arrival,” Bulletin of the Seismological Society of America, vol. 108, no. 1, pp. 230–237, 2018

    2018

  6. [14]

    A multi-window algorithm for real-time automatic detection and picking of p-phases of microseismic events,

    Z. Chen and R. R. Stewart, “A multi-window algorithm for real-time automatic detection and picking of p-phases of microseismic events,” CREWES Res. Rep , vol. 18, pp. 1–9, 2006

    2006

  7. [15]

    Easy detection for the high-pass filter cut-off frequency of digital ground motion record based on sta/lta method: A case study in the 2008 wenchuan mainshock,

    X. Longjun and C. Yabin, “Easy detection for the high-pass filter cut-off frequency of digital ground motion record based on sta/lta method: A case study in the 2008 wenchuan mainshock,” Journal of Seismology , vol. 25, no. 5, pp. 1281–1300, 2021

    2008

  8. [16]

    Automated locating mining- induced microseismicity without arrival picking by weighted sta/lta traces stacking,

    Y . Jiang, P. Peng, L. Wang, and Z. He, “Automated locating mining- induced microseismicity without arrival picking by weighted sta/lta traces stacking,” Sustainability, vol. 12, no. 9, p. 3665, 2020

    2020

  9. [17]

    Improving the first-arrival picking in mine microseismic data using a stationary discrete wavelet denoising

    C. Mborah and M. Ge, “Improving the first-arrival picking in mine microseismic data using a stationary discrete wavelet denoising.”

  10. [18]

    Microseismic monitoring and hypocenter location,

    L. Han, “Microseismic monitoring and hypocenter location,” Ph.D. dissertation, University of Calgary, 2010

    2010

  11. [19]

    Improved modified energy ratio method using a multi-window approach for accurate arrival picking,

    M. Lee, J. Byun, D. Kim, J. Choi, and M. Kim, “Improved modified energy ratio method using a multi-window approach for accurate arrival picking,” Journal of Applied Geophysics , vol. 139, pp. 117–130, 2017

    2017

  12. [20]

    A multi-window algorithm for automatic picking of microseis- mic events on 3-c data,

    Z. Chen, “A multi-window algorithm for automatic picking of microseis- mic events on 3-c data,” in SEG International Exposition and Annual Meeting. SEG, 2005, pp. SEG–2005

    2005

  13. [21]

    First-break picking method based on the difference between multiwindow energy ratios,

    D. Kim, Y . Joo, and J. Byun, “First-break picking method based on the difference between multiwindow energy ratios,” IEEE Transactions on Geoscience and Remote Sensing , vol. 61, pp. 1–10, 2023

    2023

  14. [22]

    First-break timing; arrival onset times by direct correlation,

    J. B. Molyneux and D. R. Schmitt, “First-break timing; arrival onset times by direct correlation,” Geophysics, vol. 64, no. 5, pp. 1492–1501, 1999

    1999

  15. [23]

    Semiautomated relative picking of microseismic events,

    D. Raymer, J. Rutledge, and P. Jaques, “Semiautomated relative picking of microseismic events,” in SEG Technical Program Expanded Abstracts

  16. [24]

    A semi-automatic approach to identify first arrival time: the cross-correlation technique (cct),

    M. Senkaya and H. Karsli, “A semi-automatic approach to identify first arrival time: the cross-correlation technique (cct),” Earth Sciences Research Journal, vol. 18, no. 2, pp. 107–113, 2014

    2014

  17. [25]

    A stress test to evaluate the usefulness of akaike information criterion in short-term earthquake prediction,

    R. Tozzi, F. Masci, and M. Pezzopane, “A stress test to evaluate the usefulness of akaike information criterion in short-term earthquake prediction,” Scientific reports, vol. 10, no. 1, p. 21153, 2020

    2020

  18. [26]

    Multi-component autoregressive techniques for the analysis of seismograms,

    M. Leonard and B. Kennett, “Multi-component autoregressive techniques for the analysis of seismograms,” Physics of the Earth and Planetary Interiors, vol. 113, no. 1-4, pp. 247–263, 1999

    1999

  19. [27]

    A method for reading and checking phase times in autoprocessing system of seismic wave data,

    N. Maeda, “A method for reading and checking phase times in autoprocessing system of seismic wave data,” Zisin, vol. 38, pp. 365–379, 1985

    1985

  20. [28]

    Fast-aic method for automatic first arrivals picking of microseismic event with multitrace energy stacking envelope summation,

    Y . Long, J. Lin, B. Li, H. Wang, and Z. Chen, “Fast-aic method for automatic first arrivals picking of microseismic event with multitrace energy stacking envelope summation,” IEEE Geoscience and Remote Sensing Letters, vol. 17, no. 10, pp. 1832–1836, 2019

    2019

  21. [29]

    Comparison of manual and automatic onset time picking,

    M. Leonard, “Comparison of manual and automatic onset time picking,” Bulletin of the Seismological Society of America , vol. 90, no. 6, pp. 1384–1390, 2000

    2000

  22. [30]

    Enhancing manual p-phase arrival detection and automatic onset time picking in a noisy microseismic data in underground mines,

    C. Mborah and M. Ge, “Enhancing manual p-phase arrival detection and automatic onset time picking in a noisy microseismic data in underground mines,” International Journal of Mining Science and Technology , vol. 28, no. 4, pp. 691–699, 2018

    2018

  23. [31]

    Using a deep convolutional neural network to enhance the accuracy of first-break picking,

    Y . Hollander, A. Merouane, and O. Yilmaz, “Using a deep convolutional neural network to enhance the accuracy of first-break picking,” in SEG International Exposition and Annual Meeting . SEG, 2018, pp. SEG– 2018

    2018

  24. [32]

    Automatic waveform classification and arrival picking based on convolutional neural network,

    Y . Chen, G. Zhang, M. Bai, S. Zu, Z. Guan, and M. Zhang, “Automatic waveform classification and arrival picking based on convolutional neural network,” Earth and Space Science , vol. 6, no. 7, pp. 1244–1261, 2019

    2019

  25. [33]

    Convolutional neural networks for microseismic waveform classification and arrival picking,

    G. Zhang, C. Lin, and Y . Chen, “Convolutional neural networks for microseismic waveform classification and arrival picking,” Geophysics, vol. 85, no. 4, pp. W A227–W A240, 2020

    2020

  26. [34]

    Phasenet: A deep-neural-network-based seismic arrival-time picking method,

    W. Zhu and G. C. Beroza, “Phasenet: A deep-neural-network-based seismic arrival-time picking method,” Geophysical Journal International , vol. 216, no. 1, pp. 261–273, 2019

    2019

  27. [35]

    First arrival traveltime picking through 3-d u-net,

    S. Han, Y . Liu, Y . Li, and Y . Luo, “First arrival traveltime picking through 3-d u-net,” IEEE Geoscience and Remote Sensing Letters , vol. 19, pp. 1–5, 2021

    2021

  28. [36]

    First-break automatic picking with deep semisupervised learning neural network,

    K. C. Tsai, W. Hu, X. Wu, J. Chen, and Z. Han, “First-break automatic picking with deep semisupervised learning neural network,” in SEG Technical Program Expanded Abstracts 2018 . Society of Exploration Geophysicists, 2018, pp. 2181–2185

    2018

  29. [37]

    First-break refraction event picking and seismic data trace editing using neural networks,

    M. D. McCormack, D. E. Zaucha, and D. W. Dushek, “First-break refraction event picking and seismic data trace editing using neural networks,” Geophysics, vol. 58, no. 1, pp. 67–78, 1993

    1993

  30. [38]

    A deep learning- based automatic first-arrival picking method for ultrasound sound-speed tomography,

    X. Qu, G. Yan, D. Zheng, S. Fan, Q. Rao, and J. Jiang, “A deep learning- based automatic first-arrival picking method for ultrasound sound-speed tomography,” IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 68, no. 8, pp. 2675–2686, 2021

    2021

  31. [39]

    Aenet: Automatic picking of p-wave first arrivals using deep learning,

    C. Guo, T. Zhu, Y . Gao, S. Wu, and J. Sun, “Aenet: Automatic picking of p-wave first arrivals using deep learning,” IEEE Transactions on Geoscience and Remote Sensing , vol. 59, no. 6, pp. 5293–5303, 2020

    2020

  32. [40]

    First-arrival picking with a u-net convolutional network,

    L. Hu, X. Zheng, Y . Duan, X. Yan, Y . Hu, and X. Zhang, “First-arrival picking with a u-net convolutional network,” Geophysics, vol. 84, no. 6, pp. U45–U57, 2019

    2019

  33. [41]

    U-net convolutional networks for first arrival picking,

    L. Hu, X. Zheng, and Y . Duan, “U-net convolutional networks for first arrival picking,” in SEG 2018 Workshop: SEG Maximizing Asset Value Through Artificial Intelligence and Machine Learning, Beijing, China, 12 17-19 September 2018 . Society of Exploration Geophysicists and th...

    2018

  34. [42]

    Arrival-time picking of microseismic events based on msnet,

    G. Sheng, S. Yang, X.-G. Tang, and X. Guo, “Arrival-time picking of microseismic events based on msnet,” Geophysics, vol. 87, no. 2, pp. KS57–KS71, 2022

    2022

  35. [43]

    A robust first-arrival picking workflow using convolutional and recurrent neural networks,

    P. Yuan, S. Wang, W. Hu, X. Wu, J. Chen, and H. Van Nguyen, “A robust first-arrival picking workflow using convolutional and recurrent neural networks,” Geophysics, vol. 85, no. 5, pp. U109–U119, 2020

    2020

  36. [44]

    Seismic first break picking through swin transformer feature extraction,

    P. Jiang, F. Deng, X. Wang, P. Shuai, W. Luo, and Y . Tang, “Seismic first break picking through swin transformer feature extraction,” IEEE Geoscience and Remote Sensing Letters , vol. 20, pp. 1–5, 2023

    2023

  37. [45]

    Microseismic first-arrival picking using fine-tuning feature pyramid networks,

    N. Liu, J. Chen, H. Wu, F. Li, and J. Gao, “Microseismic first-arrival picking using fine-tuning feature pyramid networks,” IEEE Geosci. Remote. Sens. Lett. , vol. 19, pp. 1–5, 2022

    2022

  38. [46]

    Segnet-based first-break picking via seismic waveform classification directly from shot gathers with sparsely distributed traces,

    S.-Y . Yuan, Y . Zhao, T. Xie, J. Qi, and S.-X. Wang, “Segnet-based first-break picking via seismic waveform classification directly from shot gathers with sparsely distributed traces,” Petroleum Science, vol. 19, no. 1, pp. 162–179, 2022

    2022

  39. [47]

    First break picking with deep learning–evaluation of network architectures,

    P. Zwartjes and J. Yoo, “First break picking with deep learning–evaluation of network architectures,” Geophysical Prospecting, vol. 70, no. 2, pp. 318–342, 2022

    2022

  40. [48]

    First-break automatic picking with fully convolutional networks and transfer learning,

    T. Xie, Y . Zhao, X. Jiao, W. Sang, and S. Yuan, “First-break automatic picking with fully convolutional networks and transfer learning,” in SEG International Exposition and Annual Meeting . SEG, 2019, p. D043S152R001

    2019

  41. [49]

    First arrival picking of microseismic signals based on nested u-net and wasserstein generative adversarial network,

    J. Zhang and G. Sheng, “First arrival picking of microseismic signals based on nested u-net and wasserstein generative adversarial network,” Journal of Petroleum Science and Engineering , vol. 195, p. 107527, 2020

    2020

  42. [50]

    Seismic waveform classification and first-break picking using convolution neural networks,

    S. Yuan, J. Liu, S. Wang, T. Wang, and P. Shi, “Seismic waveform classification and first-break picking using convolution neural networks,” IEEE Geoscience and Remote Sensing Letters, vol. 15, no. 2, pp. 272–276, 2018

    2018

  43. [51]

    Intelligent location of microseismic events based on a fully convolutional neural network (fcnn),

    K. Ma, X. Sun, Z. Zhang, J. Hu, and Z. Wang, “Intelligent location of microseismic events based on a fully convolutional neural network (fcnn),” Rock Mechanics and Rock Engineering , vol. 55, no. 8, pp. 4801–4817, 2022

    2022

  44. [52]

    A new technique for first-arrival picking of refracted seismic data based on digital image segmentation,

    W. A. Mousa, A. A. Al-Shuhail, and A. Al-Lehyani, “A new technique for first-arrival picking of refracted seismic data based on digital image segmentation,” Geophysics, vol. 76, no. 5, pp. V79–V89, 2011

    2011

  45. [53]

    Method of automatically detecting the abnormal first arrivals using delay time (december 2020),

    Z. Qin, S. Pan, J. Chen, Q. Cui, and J. He, “Method of automatically detecting the abnormal first arrivals using delay time (december 2020),” IEEE Transactions on Geoscience and Remote Sensing , vol. 60, pp. 1–8, 2021

    2020

  46. [54]

    Learning from noisy labels with deep neural networks: A survey,

    H. Song, M. Kim, D. Park, and J. Lee, “Learning from noisy labels with deep neural networks: A survey,” CoRR, vol. abs/2007.08199, 2020

    2007 arXiv

  47. [55]

    Learning from massive noisy labeled data for image classification,

    T. Xiao, T. Xia, Y . Yang, C. Huang, and X. Wang, “Learning from massive noisy labeled data for image classification,” in CVPR. IEEE Computer Society, 2015, pp. 2691–2699

    2015

  48. [56]

    Making deep neural networks robust to label noise: A loss correction approach,

    G. Patrini, A. Rozza, A. K. Menon, R. Nock, and L. Qu, “Making deep neural networks robust to label noise: A loss correction approach,” in CVPR. IEEE Computer Society, 2017, pp. 2233–2241

    2017

  49. [57]

    Toward robustness against label noise in training deep discriminative neural networks,

    A. Vahdat, “Toward robustness against label noise in training deep discriminative neural networks,” in NIPS, 2017, pp. 5596–5605

    2017

  50. [58]

    Learning from noisy labels with distillation,

    Y . Li, J. Yang, Y . Song, L. Cao, J. Luo, and L. Li, “Learning from noisy labels with distillation,” in ICCV. IEEE Computer Society, 2017, pp. 1928–1936

    2017

  51. [59]

    Learning from noisy large-scale datasets with minimal supervision,

    A. Veit, N. Alldrin, G. Chechik, I. Krasin, A. Gupta, and S. J. Belongie, “Learning from noisy large-scale datasets with minimal supervision,” in CVPR. IEEE Computer Society, 2017, pp. 6575–6583

    2017

  52. [60]

    Iterative learning with open-set noisy labels,

    Y . Wang, W. Liu, X. Ma, J. Bailey, H. Zha, L. Song, and S. Xia, “Iterative learning with open-set noisy labels,” in CVPR. Computer Vision Foundation / IEEE Computer Society, 2018, pp. 8688–8696

    2018

  53. [61]

    Simultaneous edge alignment and learning,

    Z. Yu, W. Liu, Y . Zou, C. Feng, S. Ramalingam, B. V . K. V . Kumar, and J. Kautz, “Simultaneous edge alignment and learning,” in ECCV (3), ser. Lecture Notes in Computer Science, vol. 11207. Springer, 2018, pp. 400–417

    2018

  54. [62]

    Meta label correction for noisy label learning,

    G. Zheng, A. H. Awadallah, and S. T. Dumais, “Meta label correction for noisy label learning,” in AAAI. AAAI Press, 2021, pp. 11 053–11 061

    2021

  55. [63]

    Total variation regularization strategies in full-waveform inversion,

    E. Esser, L. Guasch, T. van Leeuwen, A. Y . Aravkin, and F. J. Herrmann, “Total variation regularization strategies in full-waveform inversion,”SIAM Journal on Imaging Sciences , vol. 11, no. 1, pp. 376–406, 2018

    2018

  56. [64]

    H. S. Aghamiry, A. Gholami, and S. Operto, “Implementing bound constraints and total-variation regularization in extended full-waveform inversion with the alternating direction method of multiplier: application to large contrast media,” Geophysical Journal International , vol....

    2019

  57. [65]

    Compound regularization of full-waveform inversion for imaging piecewise media,

    ——, “Compound regularization of full-waveform inversion for imaging piecewise media,” IEEE Transactions on Geoscience and Remote Sensing, vol. 58, no. 2, pp. 1192–1204, 2020

    2020

  58. [66]

    A sequential iterative deep learning seismic blind high-resolution inversion,

    H. Chen, J. Gao, Z. Gao, D. Chen, and T. Yang, “A sequential iterative deep learning seismic blind high-resolution inversion,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , vol. 14, pp. 7817–7829, 2021

    2021

  59. [67]

    Siahkoohi, G

    A. Siahkoohi, G. Rizzuti, and F. J. Herrmann, Weak deep priors for seismic imaging, 2020, pp. 2998–3002

    2020

  60. [68]

    U-net: Convolutional networks for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III

    2015

  61. [69]

    A multi-survey dataset and benchmark for first break picking in hard rock seismic exploration,

    S. Pierre-Luc, R. Bruno, G. Joumana, N. Jean-Philippe, B. Gilles, and S. Ernst, “A multi-survey dataset and benchmark for first break picking in hard rock seismic exploration,” in Proc. Neurips 2021 Workshop on Machine Learning for the Physical Sciences (ML4PS) , 2021

    2021

  62. [70]

    Springer, 2015, pp. 234–241

    2015

  63. [2008]

    1411–1414

    Society of Exploration Geophysicists, 2008, pp. 1411–1414

    2008

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.