Urban sound event classification method based on dual-feature 2-DenseNet in parallel

Abstract

The invention provides an urban sound event classification method based on dual-feature 2-DenseNet in parallel. The method has the advantages of being more efficient in fusion capability for feature information, higher in classification accuracy and stronger in generalization capability. The method comprises the following steps: S1, acquiring and processing to-be-processed audio data and outputting an audio frame sequence; S2, performing time domain and frequency domain analysis on the audio frame sequence, and respectively outputting a Mayer frequency cepstrum coefficient feature vector sequence and a gammatone cepstrum coefficient feature vector sequence; S3, constructing a classification model, wherein the classification model comprises a network model constructed by combining a two-order Markov model on the basis of a DenseNet model; the classification model taking a two-order DenseNet model as a basis to construct a basis network, arranging the basic network to be two parallel paths, and training the classification model to obtain a well-trained classification model; and S4, after processing the feature vector sequences output in the step S2, dividing the feature vector sequences output in the step S2 into two paths in a dual-feature mode to input into the well-trained classification model for classification and recognition so as to obtain a classification result of soundevents.

Description

technical field [0001] The invention relates to the technical field of sound recognition, in particular to an urban sound event classification method based on dual-feature 2-DenseNet parallel connection. Background technique [0002] Building a smart city complex in modern society is an important trend in urban development. Using a huge sensor network to collect audio data such as traffic conditions and noise levels in the target city, and analyzing the data to guide urban design and technical decision-making, It is one of the current ideas for building smart cities. The classification of urban sound events is mainly used in noise monitoring, urban security, soundscape assessment, multimedia information retrieval, etc. In the prior art, network models such as SVM, VGG, and DCNN are used in urban sound event classification technology. In 2014, the Iustin Salamon team used the Mel cepstral coefficient feature and the support vector machine model to formulate a baseline, and i...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem of needing a sound classification method with higher accuracy in existing practical applications, the present invention provides an urban sound event classification method based on dual-feature 2-DenseNet parallel connection, which has a more efficient fusion ability for feature information and can obtain more High classification accuracy and stronger generalization ability

Images