Excavation equipments classification based on short-time frame acoustic energy ratio

Underground pipeline network suffered severe breakages caused by external road excavation equipments in the mainland of China in nowadays. The frequent underground pipeline network breakages caused serious economic loss to urban living and industry manufacturers. Therefore, designing a surveillance system which can automatically detect potential damages caused by excavation equipments is essential and urgent for underground pipeline network system protection. With this objective, we investigate two frequently used excavation equipments, namely, electric hammer and cutting machine, based on acoustic signal processing in this paper. The acoustic waves generated by these two devices are analyzed and a novel statistical feature named short-time frame acoustic energy ratio (SFER) is developed for signal characterization. The approximated distributions of the SFER for both electric hammer and cutting machine have been estimated with real recorded data. A likelihood ratio test based on Neyman-Pearson (NP) approach is then designed to perform signal classification. For performance verifications, real recorded acoustic signals for electric hammer and cutting machine on a metro construction site are conducted in experiments with our proposed signal classification approach. Experimental resutls on various set-ups have been provided to show the effectiveness of our proposed method.

[1]  Yu Hen Hu,et al.  Detection, classification, and tracking of targets , 2002, IEEE Signal Process. Mag..

[2]  Jagannath H. Nirmal,et al.  A unique approach in text independent speaker recognition using MFCC feature sets and probabilistic neural network , 2015, 2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR).

[3]  Li Liu.  Ground Vehicle Acoustic Signal Processing Based on Biological Hearing Models , 1999 .

[4]  Tuo Zhao,et al.  An enhance excavation equipments classification algorithm based on acoustic spectrum dynamic feature , 2017, Multidimens. Syst. Signal Process..

[5]  Sazali Yaacob,et al.  Malaysian English accents identification using LPC and formant analysis , 2011, 2011 IEEE International Conference on Control System, Computing and Engineering.

[6]  Mannur J. Sundaresan,et al.  Time-frequency characterization of lamb waves for material evaluation and damage inspection of plates , 2015 .

[7]  Jianzhong Wang,et al.  Acoustics recognition of construction equipments based on LPCC features and SVM , 2015, 2015 34th Chinese Control Conference (CCC).

[8]  Soodabeh Soleymani,et al.  Application of continuous wavelet transform for fault location in combined overhead line and cable distribution networks , 2011, 2011 7th International Conference on Electrical and Electronics Engineering (ELECO).

[9]  Jianzhong Wang,et al.  Excavation equipment classification based on improved MFCC features and ELM , 2017, Neurocomputing.

[10]  Li Jian.  Study on Feature Extraction Method for Typical Abnormal Events of Buried Pipelines , 2010 .

[11]  S. M. Ahadi,et al.  Environmental sound classification using spectral dynamic features , 2011, 2011 8th International Conference on Information, Communications & Signal Processing.