Classification And Recognition Of The Ground Disaster Infrasound Signal

Geological disasters are sudden,once they occur,there will be major casualties and property losses,and successful prediction of geological disasters has great practical significance.Studies have shown that by monitoring the infrasound signal generated by geological disasters,it is possible to achieve early warning of the disasters.The purpose of this paper is to find an effective classification and identification method for the infrasound signal of the disasters and to improve the accuracy of early-warning disasters,we must strive for time for related departments to ensure the safety of people's lives and property.By consulting a large number of documents and understanding the research on the classification of infrasound signal at home and abroad,the advantages and disadvantages of various methods are summarized and compared.On this basis,the main process of pattern recognition is referred to the design of a model which can improve the accuracy rate of the classification and recognition of infrasonic signal.The independent component analysis is used to extract the independent component of the ground disaster.The independent component is extracted and the interference of Gauss noise is removed.Then feature extraction is used to extract the eigenvector of the independent component.Finally,the extracted feature vector is used as the input signal of the classifier,and the effective classification method is selected according to the classification results.The infrasound data used in this paper come from the Beijing National Data Center of the Comprehensive Nuclear Test Ban Treaty.There are three infrasonic events of earthquakes,tsunamis and volcanoes.Four independent component analysis algorithms and three kinds of feature extraction and two classification algorithms will be used to classify and identify three infrasound events.According to the classification results of the test,the advantages and disadvantages of each algorithm are analyzed and the application situation is analyzed.It is found that the classification results of the infrasound signal processed by the Independent Component Analysis and the Fast Fourier Transform are best,the accuracy of the classification is about 97%,but the time of the signal processing is too long,and it is not suitable for the real-time monitoring of the infrasound signal.If the infrasound signal is processed by fast independent component analysis and Fast Fourier Transform,the classification result is reduced but the processing speed is raised hundreds times,and it can be applied to the real-time monitoring of the infrasound classification and recognition.This paper finds an ideal method for classification and recognition of infrasound signal,which provides a new theoretical basis for the research in this field.