Analyzing bacteriological growth using wavelet transform

This paper addresses the problem of extracting the important information from a complex response from an electronic tongue sensor. A wavelet transform is used in this approach and the approximation coefficients are extracted as features and classified using a minimum distance classifier (MDC). Two experimental setups have been tested, water and milk, and the bacteriological growth are monitored. Using only the approximation coefficients, the amount of data to be analyzed can be significantly reduced without loss of important information.