Noise Diagnostics of Scooter Faults by Using MPEG-7 Audio Features and Intelligent Classification Techniques

ABSTRACT

A scooter fault diagnostic system that makes use of feature extraction and intelligent classification algorithms is presented in this paper. Sound features based on MPEG (Moving Picture Experts Group)-7 coding standard and several other features in the time and frequency domains are extracted from noise data and preprocessed prior to classification. Classification algorithms including the Nearest Neighbor Rule (NNR), the Artificial Neural Networks (ANN), the Fuzzy Neural Networks (FNN), and the Hidden Markov Models (HMM) are employed to identify and classify scooter noise. A training phase is required to establish a feature space template, followed by a test phase in which the audio features of the test data are calculated and matched to the feature space template. The proposed techniques were applied to classify noise data due to various kinds of scooter fault, such as belt damage, pulley damage, etc. The results reveal that the performance of methods is satisfactory, while varying slightly in performance with the algorithm and the type of noise used in the tests.