Seismic Signal Adaptive Endpoint Detection Based on Human-Computer Interaction

Yue Wang, Xiaochun Lu, Juntao Fei, Seokwon Yeom, Yong-Hyun Woo, J.-t Zhao, C.-x Yu, S.-p Peng, G Beskardes, J Hole, K Wang, D Arosio, L Longoni, M Papini, Luz Garcia, Isaac Alvarez, Manuel Titos, Guillermo Cort S, M Benitez, Luz Garc A, J Rittgers, A Revil, T, Roee Diamant, Dror Kipnis, Michele Zorzi, Xuan Feng, Xuebing Zhang, Cai Liu, Elena Palahina, Mria Gamcov, Iveta Gladiov, Mengnan Wang, Wang Zhuang, Andreas Bulling, Kai Kunze, Anjana Ramkumar, Jan Pieter, Wiro Stappers, Niessen, Qingwei Han, Xi Chen, Kai Tang, David Rozado, Jason Niu, Martin Lochner, Leonardo Melo, Roza, Aline Oliveira, Cristina Antoneli De, Priscila Fagundes, Basto, Maria Roussou, Nikolaos Avouris, George Lepouras, Shaimaa Lazem, Susan Dray, H Sun, E Lennart, Nacke, Roman Hak, Tomas Zeman, Xiangyang Li, Zhili Zhang, Feng Liang, Carla Barreiros, Viktoria Pammer-Schindler, Eduardo Veas, Christian Reuter, Amanda Hughes, Marc-Andr Kaufhold, Xu Sun, Andrew May, Qingfeng Wang, Andreas Kolling, Phillip Walker, Nilanjan Chakraborty, Wei He, Yuhao Chen, Zhao Yin, Kamran Zargar-Shoshtari, Declan Murphy, Homayoun ZargarInternational Journal of Multimedia Computing 2022 · DOI

TL;DR

This paper introduces a high-precision, adaptive method for detecting the start and end points of seismic signals by leveraging "autocorrelation similarity distance." By integrating this algorithm into a human-computer interaction (HCI) framework, the researchers achieved a detection accuracy of 96.11%, significantly outperforming traditional manual inspection methods even in high-noise environments.

What Problem Does This Paper Address?

In seismology, "endpoint detection"—identifying exactly when a seismic wave (like a P-wave or S-wave) starts and stops—is the bread and butter of disaster response. Precise timing allows researchers to calculate the earthquake's source location and even distinguish between natural tectonic shifts and suspicious underground nuclear explosions.

Historically, this has been a manual, labor-intensive process. Experts look at time-domain waveforms and "eye-ball" the start and end points. However, seismic signals are "non-stationary" (their statistical properties change over time) and are often buried under significant environmental noise. Manual detection is prone to misjudgment and "leakage" (missing the signal entirely), especially when the signal-to-noise ratio (SNR) is low. The author seeks to automate this process using adaptive algorithms that can function within modern human-machine interfaces, such as those used by disaster-relief robots.

Methods

The core of the paper’s approach is the Autocorrelation Similarity Distance. Since seismic signals are non-stationary, the algorithm breaks the data into small, quasi-stationary "frames."

  1. Noise Modeling: The system first analyzes a segment of pure background noise to establish a baseline autocorrelation function.
  2. Similarity Distance ($\lambda$): As new signal frames come in, the algorithm calculates the "distance" between the current frame's autocorrelation and the noise model. If the distance is small, it’s just noise; if the distance jumps, a seismic event has likely begun.
  3. Adaptive Thresholding: Instead of a hard-coded limit, the system calculates a dynamic threshold ($E \pm 0.8\sigma$) based on the first 20 frames of data, allowing it to adapt to different environments.
  4. Phase Adaptive Correction: Different sensors (geophones) have different physical characteristics. The paper describes a three-parameter ($D_{01}, D_{02}, K$) iterative calculation to align the phases of different geophone records (e.g., correcting a 5Hz record to match a 40Hz standard), ensuring the data is consistent before detection begins.

Datasets

The study utilized two primary types of data for evaluation:

  • Simulated Seismic Signals: A synthetic dataset of 2,000 sampling points created by superimposing five frequency-modulated (FM) exponential functions. This allowed the researchers to test the algorithm against known "ground truth" endpoints at various noise levels (6dB, 0dB, and -3dB).
  • Real-World Seismic Records: Full-band vertical (BHZ) signals with a sampling frequency of 20 Hz, sourced from three-way seismic stations.

Key Insights and Ideas

  • The Power of Autocorrelation: While raw waveforms are messy, the autocorrelation function is remarkably resilient to Gaussian white noise. By focusing on the similarity of the signal's structure rather than just its amplitude, the algorithm can "see" a seismic wave even when it's visually submerged in noise.
  • HCI as a Design Philosophy: The paper frames this not just as a standalone math problem, but as a component of Human-Computer Interaction. The goal is to reduce the "cognitive load" on human operators (like robot pilots in disaster zones) by providing an active, automated service rather than a passive tool.
  • Hardware Agnosticism: The inclusion of an adaptive phase correction method acknowledges a messy reality: not all sensors are created equal. By mathematically "calibrating" different geophones on the fly, the system becomes much more flexible for field use.

Key Results

The proposed algorithm demonstrated a clear advantage over manual inspection:

  • Accuracy: The algorithm reached a peak accuracy of 96.11%.
  • Robustness to Noise: In simulation tests with a -3dB signal-to-noise ratio (where the noise is louder than the signal), the "Article Method" identified the starting point at 796 (vs. the true 750), while manual inspection lagged significantly at 821.
  • Comparison: In a head-to-head with two other standard algorithms, the proposed method provided the highest accuracy, though it required a slightly longer training time (3.29 seconds compared to Algorithm 1’s 0.031 seconds).

Limitations and Future Work

The author notes that the model training time is currently "unsatisfactory" compared to simpler, less accurate algorithms. While 3.29 seconds is negligible for a single analysis, it could become a bottleneck in massive, real-time sensor networks.

Future work is hinted at in the conclusion, suggesting that while the current "seismic phase energy ratio" criterion is effective, further optimization of the training speed could make the system even more desirable for high-stakes, real-time earthquake monitoring and disaster-relief robotics.

Summary updated

Abstract

With the development of artificial intelligence technology, robot technology has become more mature, and human-computer interaction technology has become more widely used. China is a country with frequent earthquakes. The application of robot technology to seismic signal detection is of great significance to solve various problems in earthquake disasters. The research purpose of this paper is to study adaptive endpoint detection of seismic signals based on human-computer interaction technology. Based on the analysis of adaptive noise reduction of signal autocorrelation function, this paper references the concept of autocorrelation similarity distance, and proposes human-machine interaction technology Seismic signal adaptive endpoint detection, discusses the description of noise and noisy signals based on autocorrelation similarity distance, and summarizes the adaptive endpoint detection method of seismic signals based on human-computer interaction technology, and gives the specific implementation of the algorithm Compared with the experimental detection results obtained by this method and the time-domain waveform envelope and the method of manually detecting the time-table seismic signal endpoint detection, the research results in this paper show that the accuracy of this algorithm is as high as 96.11%, Under the condition of noise ratio, the end position of the seismic signal can still be detected more accurately by using this algorithm.
BibTeX
@article{wang2022seismic,
  author = {Yue Wang and Xiaochun Lu and Juntao Fei and Seokwon Yeom and Yong-Hyun Woo and J.-T Zhao and C.-X Yu and S.-P Peng and G Beskardes and J Hole and K Wang and D Arosio and L Longoni and M Papini and Luz Garcia and Isaac Alvarez and Manuel Titos and Guillermo Cort S and M Benitez and Luz Garc A and J Rittgers and A Revil and T and Roee Diamant and Dror Kipnis and Michele Zorzi and Xuan Feng and Xuebing Zhang and Cai Liu and Elena Palahina and Mria Gamcov and Iveta Gladiov and Mengnan Wang and Wang Zhuang and Andreas Bulling and Kai Kunze and Anjana Ramkumar and Jan Pieter and Wiro Stappers and Niessen and Qingwei Han and Xi Chen and Kai Tang and David Rozado and Jason Niu and Martin Lochner and Leonardo Melo and Roza and Aline Oliveira and Cristina Antoneli De and Priscila Fagundes and Basto and Maria Roussou and Nikolaos Avouris and George Lepouras and Shaimaa Lazem and Susan Dray and X Lu and X Chen and H Sun and E Lennart and Nacke and Roman Hak and Tomas Zeman and Xiangyang Li and Zhili Zhang and Feng Liang and Carla Barreiros and Viktoria Pammer-Schindler and Eduardo Veas and Christian Reuter and Amanda Hughes and Marc-Andr Kaufhold and Xu Sun and Andrew May and Qingfeng Wang and Andreas Kolling and Phillip Walker and Nilanjan Chakraborty and Wei He and Yuhao Chen and Zhao Yin and Kamran Zargar-Shoshtari and Declan Murphy and Homayoun Zargar},
  title = {{Seismic Signal Adaptive Endpoint Detection Based on Human-Computer Interaction}},
  journal = {International Journal of Multimedia Computing},
  year = {2022},
  doi = {10.38007/ijmc.2022.030405},
}
Figure