Underwater acoustic signal recognition based on combination of multi-scale convolutional neural network and constant-Q transform

This article proposes a multi-scale deep learning network to classify different underwater acoustic signal sources. The proposed network is cleverly designed with multiple branches, creating a multi-scale block which allows learning various spatial features of Constant-Q Transform spectrograms.
Nội dung trích xuất từ tài liệu:
Underwater acoustic signal recognition based on combination of multi-scale convolutional neural network and constant-Q transform