Acoustics recognition of construction equipments based on LPCC features and SVM

Abstract

To provide the reliability of power supply and ensure the personal security, cables in urban city are usually paved underground in nowadays. However, according to the statistical report from State Grid, around 53.4% of cable breakages are caused by external damages from 2009 to 2011. Among these external cable vandalisms, construction equipments are the main damage sources, including impact hammer, cutting machine, grab excavator, etc. Thus, designing a surveillance system which can automatically detect such potential destroy is highly desired. In this paper, we investigate the automatic recognition system based on classifying the acoustic waves generated by the three equipments. In the proposed recognition framework, the linear prediction cepstral coefficients (LPCC) for acoustic waves generated by the three equipments are extracted in small frames. These LPCC features are then fed to the support vector machine classifier for training and testing. To show the efficiency of the recognition system, real recorded acoustic waves are collected in varies construction using a cross microphone array for experiments. Experiments show that the recognition algorithm we developed is efficiency.

Automating the Estimation of Productivity Metrics for Construction Workers Using Deep Learning and Kinematics

“…Their algorithm was used to classify activities: nailing with nail-gun, hammering, table-saw cutting, and drilling. Others use only cepstral features (Yang et al, 2015) or only time features (Cheng et al, 2017). As the experiments differ in activities classified, experimental setup, and data quantities, it is impossible to compare these to each other, which is also why it is impossible to state which features are the most important when using audio-based methods for classifying activities.…”

Automatic Recognition of Construction Worker Activities Using Deep Learning Approaches and Wearable Inertial Sensors

“…Researchers [31][32][33][34][35] have also evaluated the benefit of kinematic-based approaches, including accelerometers and gyroscopes, since they may offer a more trustworthy data source for worker movement monitoring. Initial kinematic-based labor activity identification studies can be traced back to a survey [33] in which single-wired triaxial accelerometers were utilized to determine five operations of construction laborers: collecting bricks, twist placing, fetching mortar, spreading mortar, and trimming bricks.…”

“…Using accelerometers connected to the wristbands of construction laborers, [31] identified four of their operations: spreading mortar, hauling and placing bricks, correcting blocks, and discarding residual mortar. Using a smartphone accelerometer and auroscope sensors attached to the wrists and legs of rebar laborers, [35] developed a technique for recognizing eight operations, including standing, strolling, squatting, washing templates, setting rebar, lashing rebar, welding rebar, and cutting rebar. Generally, HAR is a time-series classification issue comprising sequential procedures such as sensor readings, data preparation, feature extraction, feature selection, feature annotation, supervised learning, and classifier model evaluation [20].…”

Correct Pronunciation Detection of the Arabic Alphabet Using Deep Learning

“…However, limited work has been done in Arabic ASR using pattern recognition and feature extraction techniques. Pattern recognition techniques used in ASR incorporate hidden Markov model (HMM) [4,17,18], Gaussian mixture model (GMM) [4,19], artificial neural network (ANN) [20], and multi-layer perceptron (MLP) [20] using different feature extraction techniques such as mel-frequency cepstral coefficient (MFCC), linear predictive cepstrum coefficients (LPCC) [21], and spectrogram.…”