Underwater acoustic target recognition based on automatic feature and contrastive coding

Underwater acoustic target recognition (UATR) technology based on deep learning and automatic encoding has become an important research direction in the underwater field recent years. However, existing methods do not have favourable self-adaptability for different data because of complex changeable environment, which easily leads to unsatisfactory effect. The concept contrastive is introduced into UATR a model named Contrastive Coding (CCU) proposed. Based unsupervised framework, been modified field. Thus, CCU can generate adaptable features according data. experimental test shows that superior other models achieved excellent performance datasets.