eprintid: 8652 rev_number: 8 eprint_status: archive userid: 2 dir: disk0/00/00/86/52 datestamp: 2023-09-05 23:30:12 lastmod: 2023-09-05 23:30:13 status_changed: 2023-09-05 23:30:12 type: article metadata_visibility: show creators_name: Rustam, Furqan creators_name: Ishaq, Abid creators_name: Hashmi, Muhammad Shadab Alam creators_name: Siddiqui, Hafeez Ur Rehman creators_name: Dzul Lopez, Luis creators_name: Castanedo Galán, Juan creators_name: Ashraf, Imran creators_id: creators_id: creators_id: creators_id: creators_id: luis.dzul@unini.edu.mx creators_id: juan.castanedo@uneatlantico.es creators_id: title: Railway Track Fault Detection Using Selective MFCC Features from Acoustic Data ispublished: pub subjects: uneat_eng divisions: uneatlantico_produccion_cientifica divisions: unincol_produccion_cientifica divisions: uninimx_produccion_cientifica divisions: uninipr_produccion_cientifica divisions: unic_produccion_cientifica full_text_status: public keywords: vehicle automation; railway track fault detection; mel frequency cepstral coefficient; acoustic data; machine learning abstract: Railway track faults may lead to railway accidents and cause human and financial loss. Spatial, temporal, and weather elements, and wear and tear, lead to ballast, loose nuts, misalignment, and cracks leading to accidents. Manual inspection of such defects is time-consuming and prone to errors. Automatic inspection provides a fast, reliable, and unbiased solution. However, highly accurate fault detection is challenging due to the lack of public datasets, noisy data, inefficient models, etc. To obtain better performance, this study presents a novel approach that relies on mel frequency cepstral coefficient features from acoustic data. The primary objective of this study is to increase fault detection performance. As well as designing an ensemble model, we utilize selective features using chi-square(chi2) that have high importance with respect to the target class. Extensive experiments were carried out to analyze the efficiency of the proposed approach. The experimental results suggest that using 60 features, 40 original features, and 20 chi2 features produces optimal results both regarding accuracy and computational complexity. A mean accuracy score of 0.99 was obtained using the proposed approach with machine learning models using the collected data. Moreover, this performance was significantly better than that of existing approaches; however, the performance of models may vary in real-world settings. date: 2023-08 publication: Sensors volume: 23 number: 16 pagerange: 7018 id_number: doi:10.3390/s23167018 refereed: TRUE issn: 1424-8220 official_url: http://doi.org/10.3390/s23167018 access: open language: en citation: Artículo Materias > Ingeniería Universidad Europea del Atlántico > Investigación > Producción Científica Fundación Universitaria Internacional de Colombia > Investigación > Producción Científica Universidad Internacional Iberoamericana México > Investigación > Producción Científica Universidad Internacional Iberoamericana Puerto Rico > Investigación > Producción Científica Universidad Internacional do Cuanza > Investigación > Producción Científica Abierto Inglés Railway track faults may lead to railway accidents and cause human and financial loss. Spatial, temporal, and weather elements, and wear and tear, lead to ballast, loose nuts, misalignment, and cracks leading to accidents. Manual inspection of such defects is time-consuming and prone to errors. Automatic inspection provides a fast, reliable, and unbiased solution. However, highly accurate fault detection is challenging due to the lack of public datasets, noisy data, inefficient models, etc. To obtain better performance, this study presents a novel approach that relies on mel frequency cepstral coefficient features from acoustic data. The primary objective of this study is to increase fault detection performance. As well as designing an ensemble model, we utilize selective features using chi-square(chi2) that have high importance with respect to the target class. Extensive experiments were carried out to analyze the efficiency of the proposed approach. The experimental results suggest that using 60 features, 40 original features, and 20 chi2 features produces optimal results both regarding accuracy and computational complexity. A mean accuracy score of 0.99 was obtained using the proposed approach with machine learning models using the collected data. Moreover, this performance was significantly better than that of existing approaches; however, the performance of models may vary in real-world settings. metadata Rustam, Furqan; Ishaq, Abid; Hashmi, Muhammad Shadab Alam; Siddiqui, Hafeez Ur Rehman; Dzul Lopez, Luis; Castanedo Galán, Juan y Ashraf, Imran mail SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, luis.dzul@unini.edu.mx, juan.castanedo@uneatlantico.es, SIN ESPECIFICAR (2023) Railway Track Fault Detection Using Selective MFCC Features from Acoustic Data. Sensors, 23 (16). p. 7018. ISSN 1424-8220 document_url: http://repositorio.unic.co.ao/id/eprint/8652/1/sensors-23-07018.pdf