Music Classification Model based on Multi-Feature Image Fusion and Machine Learning Algorithms

How to effectively classify massive music data has become a key research topic in the field of music information retrieval. This study proposes a novel music classification model that combines multi-feature image fusion (MFIF) and machine learning (ML) algorithms to improve the accuracy and efficiency of music classification. The specific method is to first convert music signals into various visual representations such as spectrograms, Mel spectrograms, and chromaticity maps, and then combine these different visual features into a comprehensive feature map through image fusion technology. In addition, the study also introduced a deep learning framework, using Convolutional Neural Network (CNN) to learn these fused image features. In the absence of noise, the model exhibits high performance with an accuracy of 0.87 and an F1 score of 0.855, indicating that the model can perform music classification tasks well under ideal input conditions. This article provides technical support for related music recommendation systems and music copyright management.