ia-PNCC: Noise Processing Method For Underwater Target Recognition Convolutional Neural Network

Underwater targetrecognition is a key technology for underwater acousticcountermeasure. And how to classify and recognize underwater targetsaccording to the noise information of underwater targetshas been a hot topic in the field of underwater acousticsignals. In this paper, the deep learning model is applied in the underwater targetrecognition and the improved anti- noise Power-Normalized Cepstral Coefficients (ia-PNCC) is proposed based on PNCC oriented to underwater targetnoise features. In this coefficient, multitaperand normalized Gammatone filtergroup are used to improve the anti-noise capacity of PNCC in underwater targetrecognition, and it is combined with the convolutional neural networkto recognize the underwater target. The experiment results show that the acoustic feature presented by ia-PNCC is of higher anti-noise capacity and more adaptive to the underwater targetrecognition model of convolutional neural network. Compared with the combination of convolutional neural networkwith single acoustic feature such as MFCC ( Mel-scalefrequency cepstral coefficients) or LPCC ( Linear PredictionPLP) and so on, the combination of ia-PNCC with convolutional neural networkimproves the underwater targetrecognition accuracy greatly.