A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification model
With the increased frequency and intensity of landslides in recent years, there is growing research on timely detection of the underlying subsurface processes that contribute to these hazards. Recent advances in machine learning have introduced algorithms for classifying seismic events associated wi...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-06-01
|
| Series: | Science of Remote Sensing |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666017224000737 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841545948363227136 |
|---|---|
| author | Jiaxin Jiang David Murray Vladimir Stankovic Lina Stankovic Clement Hibert Stella Pytharouli Jean-Philippe Malet |
| author_facet | Jiaxin Jiang David Murray Vladimir Stankovic Lina Stankovic Clement Hibert Stella Pytharouli Jean-Philippe Malet |
| author_sort | Jiaxin Jiang |
| collection | DOAJ |
| description | With the increased frequency and intensity of landslides in recent years, there is growing research on timely detection of the underlying subsurface processes that contribute to these hazards. Recent advances in machine learning have introduced algorithms for classifying seismic events associated with landslides, such as earthquakes, rockfalls, and smaller quakes. However, the opaque, “black box” nature of deep learning algorithms has raised concerns of reliability and interpretability by Earth scientists and end-users, hesitant to adopt these models. Leveraging on recent recommendations on embedding humans in the Artificial Intelligence (AI) decision making process, particularly training and validation, we propose a methodology that incorporates data labelling, verification, and re-labelling through a multi-class convolutional neural network (CNN) supported by Explainable Artificial Intelligence (XAI) tools, specifically, Layer-wise Relevance Propagation (LRP). To ensure reproducibility, a catalogue of training events is provided as supplementary material. Evaluation from the French Seismologic and Geodetic Network (Résif) dataset, gathered in the Alps in France, demonstrate the effectiveness of the proposed methodology, achieving a recall/sensitivity of 97.3% for rockfalls and 68.4% for quakes. |
| format | Article |
| id | doaj-art-f4306b14203d4f49ba68fe0fda6111fa |
| institution | Kabale University |
| issn | 2666-0172 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Science of Remote Sensing |
| spelling | doaj-art-f4306b14203d4f49ba68fe0fda6111fa2025-01-11T06:41:58ZengElsevierScience of Remote Sensing2666-01722025-06-0111100189A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification modelJiaxin Jiang0David Murray1Vladimir Stankovic2Lina Stankovic3Clement Hibert4Stella Pytharouli5Jean-Philippe Malet6Department Electronic and Electrical Engineering, University of Strathclyde, Glasgow, UK; Corresponding author.Department Electronic and Electrical Engineering, University of Strathclyde, Glasgow, UKDepartment Electronic and Electrical Engineering, University of Strathclyde, Glasgow, UKDepartment Electronic and Electrical Engineering, University of Strathclyde, Glasgow, UKInstitut Terre & Environnement de Strasbourg, University of Strasbourg, Strasbourg, FranceDepartment Civil and Environmental Engineering, University of Strathclyde, Glasgow, UKInstitut Terre & Environnement de Strasbourg, University of Strasbourg, Strasbourg, FranceWith the increased frequency and intensity of landslides in recent years, there is growing research on timely detection of the underlying subsurface processes that contribute to these hazards. Recent advances in machine learning have introduced algorithms for classifying seismic events associated with landslides, such as earthquakes, rockfalls, and smaller quakes. However, the opaque, “black box” nature of deep learning algorithms has raised concerns of reliability and interpretability by Earth scientists and end-users, hesitant to adopt these models. Leveraging on recent recommendations on embedding humans in the Artificial Intelligence (AI) decision making process, particularly training and validation, we propose a methodology that incorporates data labelling, verification, and re-labelling through a multi-class convolutional neural network (CNN) supported by Explainable Artificial Intelligence (XAI) tools, specifically, Layer-wise Relevance Propagation (LRP). To ensure reproducibility, a catalogue of training events is provided as supplementary material. Evaluation from the French Seismologic and Geodetic Network (Résif) dataset, gathered in the Alps in France, demonstrate the effectiveness of the proposed methodology, achieving a recall/sensitivity of 97.3% for rockfalls and 68.4% for quakes.http://www.sciencedirect.com/science/article/pii/S2666017224000737Seismic signal analysisMicroseismic signal classificationDeep learningExplainable artificial intelligenceData annotationModel training |
| spellingShingle | Jiaxin Jiang David Murray Vladimir Stankovic Lina Stankovic Clement Hibert Stella Pytharouli Jean-Philippe Malet A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification model Science of Remote Sensing Seismic signal analysis Microseismic signal classification Deep learning Explainable artificial intelligence Data annotation Model training |
| title | A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification model |
| title_full | A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification model |
| title_fullStr | A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification model |
| title_full_unstemmed | A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification model |
| title_short | A human-on-the-loop approach for labelling seismic recordings from landslide site via a multi-class deep-learning based classification model |
| title_sort | human on the loop approach for labelling seismic recordings from landslide site via a multi class deep learning based classification model |
| topic | Seismic signal analysis Microseismic signal classification Deep learning Explainable artificial intelligence Data annotation Model training |
| url | http://www.sciencedirect.com/science/article/pii/S2666017224000737 |
| work_keys_str_mv | AT jiaxinjiang ahumanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT davidmurray ahumanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT vladimirstankovic ahumanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT linastankovic ahumanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT clementhibert ahumanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT stellapytharouli ahumanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT jeanphilippemalet ahumanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT jiaxinjiang humanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT davidmurray humanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT vladimirstankovic humanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT linastankovic humanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT clementhibert humanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT stellapytharouli humanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel AT jeanphilippemalet humanontheloopapproachforlabellingseismicrecordingsfromlandslidesiteviaamulticlassdeeplearningbasedclassificationmodel |