Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array Recordings
Slope instabilities caused by heavy rainfall, man-made activity or earthquakes can be characterised by seismic events. To minimise mortality and infrastructure damage, a good understanding of seismic signal properties characterising slope failures is therefore crucial to classify seismic events reco...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/1/243 |
| _version_ | 1827617225440231424 |
|---|---|
| author | Jiangfeng Li Lina Stankovic Vladimir Stankovic Stella Pytharouli Cheng Yang Qingjiang Shi |
| author_facet | Jiangfeng Li Lina Stankovic Vladimir Stankovic Stella Pytharouli Cheng Yang Qingjiang Shi |
| author_sort | Jiangfeng Li |
| collection | DOAJ |
| description | Slope instabilities caused by heavy rainfall, man-made activity or earthquakes can be characterised by seismic events. To minimise mortality and infrastructure damage, a good understanding of seismic signal properties characterising slope failures is therefore crucial to classify seismic events recorded from continuous recordings effectively. However, there are limited contributions towards understanding the importance of feature selection for the classification of seismic signals from continuous noisy recordings from multiple channels/sensors. This paper first proposes a novel multi-channel event-detection scheme based on Neyman–Pearson lemma and Multi-channel Coherency Migration (MCM) on the stacked signal across multi-channels. Furthermore, this paper adapts graph-based feature weight optimisation as feature selection, exploiting the signal’s physical characteristics, to improve signal classification. Specifically, we alternatively optimise the feature weight and classification label with graph smoothness and semidefinite programming (SDP). Experimental results show that with expert interpretation, compared with the conventional short-time average/long-time average (STA/LTA) detection approach, our detection method identified 614 more seismic events in five days. Furthermore, feature selection, especially via graph-based feature weight optimisation, provides more focused feature sets with less than half of the original number of features, at the same time enhancing the classification performance; for example, with feature selection, the Graph Laplacian Regularisation classifier (GLR) raised the rockfall and slide quake sensitivities to 92% and 88% from 89% and 85%, respectively. |
| first_indexed | 2024-03-09T09:40:55Z |
| format | Article |
| id | doaj.art-9a75b3b7fd364c80946ba33a70a450a7 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T09:40:55Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-9a75b3b7fd364c80946ba33a70a450a72023-12-02T00:54:38ZengMDPI AGSensors1424-82202022-12-0123124310.3390/s23010243Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array RecordingsJiangfeng Li0Lina Stankovic1Vladimir Stankovic2Stella Pytharouli3Cheng Yang4Qingjiang Shi5Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1XW, UKDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1XW, UKDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1XW, UKDepartment of Civil and Environmental Engineering, University of Strathclyde, Glasgow G1 1XJ, UKCollege of Electronics and Information Engineering, Shanghai University of Electric Power, Shanghai 200090, ChinaSchool of Software Engineering, Tongji University, Shanghai 201804, ChinaSlope instabilities caused by heavy rainfall, man-made activity or earthquakes can be characterised by seismic events. To minimise mortality and infrastructure damage, a good understanding of seismic signal properties characterising slope failures is therefore crucial to classify seismic events recorded from continuous recordings effectively. However, there are limited contributions towards understanding the importance of feature selection for the classification of seismic signals from continuous noisy recordings from multiple channels/sensors. This paper first proposes a novel multi-channel event-detection scheme based on Neyman–Pearson lemma and Multi-channel Coherency Migration (MCM) on the stacked signal across multi-channels. Furthermore, this paper adapts graph-based feature weight optimisation as feature selection, exploiting the signal’s physical characteristics, to improve signal classification. Specifically, we alternatively optimise the feature weight and classification label with graph smoothness and semidefinite programming (SDP). Experimental results show that with expert interpretation, compared with the conventional short-time average/long-time average (STA/LTA) detection approach, our detection method identified 614 more seismic events in five days. Furthermore, feature selection, especially via graph-based feature weight optimisation, provides more focused feature sets with less than half of the original number of features, at the same time enhancing the classification performance; for example, with feature selection, the Graph Laplacian Regularisation classifier (GLR) raised the rockfall and slide quake sensitivities to 92% and 88% from 89% and 85%, respectively.https://www.mdpi.com/1424-8220/23/1/243feature engineeringmulti-channel seismic events detectiongraph feature weight optimisation and classification |
| spellingShingle | Jiangfeng Li Lina Stankovic Vladimir Stankovic Stella Pytharouli Cheng Yang Qingjiang Shi Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array Recordings Sensors feature engineering multi-channel seismic events detection graph feature weight optimisation and classification |
| title | Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array Recordings |
| title_full | Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array Recordings |
| title_fullStr | Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array Recordings |
| title_full_unstemmed | Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array Recordings |
| title_short | Graph-Based Feature Weight Optimisation and Classification of Continuous Seismic Sensor Array Recordings |
| title_sort | graph based feature weight optimisation and classification of continuous seismic sensor array recordings |
| topic | feature engineering multi-channel seismic events detection graph feature weight optimisation and classification |
| url | https://www.mdpi.com/1424-8220/23/1/243 |
| work_keys_str_mv | AT jiangfengli graphbasedfeatureweightoptimisationandclassificationofcontinuousseismicsensorarrayrecordings AT linastankovic graphbasedfeatureweightoptimisationandclassificationofcontinuousseismicsensorarrayrecordings AT vladimirstankovic graphbasedfeatureweightoptimisationandclassificationofcontinuousseismicsensorarrayrecordings AT stellapytharouli graphbasedfeatureweightoptimisationandclassificationofcontinuousseismicsensorarrayrecordings AT chengyang graphbasedfeatureweightoptimisationandclassificationofcontinuousseismicsensorarrayrecordings AT qingjiangshi graphbasedfeatureweightoptimisationandclassificationofcontinuousseismicsensorarrayrecordings |