V.V.Krizhanovskii, V.V.Krizhanovskii (jr.) An adaptive detection and visualization of local changes of acoustical characteristics of the objects |
Acoustic bulletin, Vol. 12 ¹ 1, (2009) |
| The problem of synthesizing the adaptive algorithm for detecting and visualizing the changes of the objects' informative acoustic characteristics has been considered within the framework of the statistical approach. The estimations of power spectral density of the acoustic signals recorded on the object's surface are chosen as determining informative characteristics. Training data samples are used for estimating the signal's and interference's characteristics. The structure of the synthesized algorithm has been analyzed and variants of its modification have been offered. It is shown that the important component of the algorithm is the front-end signal processing, that provides the preliminary adaptive noise filtration. The results of experimental check of efficiency of the offered statistics based on the methodology of numerical and imitation modeling are presented. The developed sufficient statistics are shown to provide forming of the acoustic portraits of the signals registered on the surfaces of investigated objects. Changes in the objects' informative acoustic characteristics may be detected and regions of these artifacts localization may be determined on the basis of the objective analysis of these statistics. |
|
KEY WORDS: acoustic artifacts, detection, visualization, an adaptive algorithm, sufficient statistics |
| TEXT LANGUAGE: Russian |