International Research Journal of Education and Technology

An image-based music recommendation application that leverages the visual content of images to enhance the recommendation process. The application utilizes feature extraction techniques to analyze the visual elements of images, such as color, texture, shape, and composition, and convert them into numerical representations. These extracted features are then used to compare and match images with similar visual characteristics, allowing for the association of relevant music genres, artists, or songs. By incorporating image analysis and feature extraction, the application aims to provide more accurate and personalized music recommendations based on users' visual preferences. The system utilizes deep learning- based methods and traditional computer vision algorithms to extract meaningful features from images, enabling the identification of visually similar images and their corresponding music recommendations.

Key Words:  Information retrieval, Metadata, Controlled vocabulary, Search optimization, Taxonomy, Thesaurus, Classification.