Small sample underwater target identification method based on deep forest

Abstract

The invention provides a small sample underwater target identification method based on a deep forest. The method comprises the steps that the Mel-frequency cepstral coefficient MFCC features of an underwater target radiation noise signal is taken as the input data of a deep forest model; the deep forest model adopts a cascade structure of a forest to realize layer-by-layer processing of input data so as to perform characterization learning and perform prediction according to output of the last layer of the cascade forest, meanwhile, the number of layers of the cascade forest in the deep forest model is adaptively adjusted according to the data, the model complexity matches the data, and the method is suitable for small sample scenes. The method based on the deep forest provided by the invention can effectively improve the recognition accuracy of the underwater target under the small sample condition.

Description

technical field [0001] The invention relates to the field of information signal processing, in particular to theories such as underwater acoustic signal processing and non-neural network deep learning, and in particular to an underwater target recognition method. Background technique [0002] Resolutely safeguarding national maritime rights and interests and building a maritime power have been incorporated into the national strategic system as an important strategic goal. As one of the important research directions in the field of underwater acoustic signal processing, underwater target recognition is a key technology for developing modern underwater equipment and winning information-based local wars from the sea, which is of great significance to my country's marine national defense. [0003] At present, traditional machine learning methods are mostly used for underwater target recognition, such as support vector machines, random forests, and k-nearest neighbors. These tra...

AI Technical Summary

Problems solved by technology

Aiming at the problem that the recognition effect of the traditional machine learning method and the deep learning method based on the neural network are not ideal for underwater targets in the case of small samples, the present invention adopts the Mel frequency cepstral coefficient MFCC feature of the radiation noise signal of the underwater target as the deep forest model The input data of the deep forest model adopts the cascading structure of the forest to realize the layer-by-layer processing of the input data so as to perform representation learning and predict according to the output of the last layer of the cascading forest

Images