Advancing Image Compression Through Clustering Techniques: A Comprehensive Analysis

Image compression is a critical area of research aimed at optimizing data storage and transmission while maintaining image quality. This paper explores the application of clustering techniques as a means to achieve efficient and high-quality image compression. We systematically analyze nine clustering methods: K-Means, BIRCH, Divisive Clustering, DBSCAN, OPTICS, Mean Shift, GMM, BGMM, and CLIQUE. Each technique is evaluated across a variety of parameters, including block size, number of clusters, and other method-specific attributes, to assess their impact on compression ratio and structural similarity index. The experimental results reveal significant differences in performance among the techniques. K-Means, Divisive Clustering, and CLIQUE emerge as reliable methods, balancing high compression ratios and excellent image quality. In contrast, techniques like Mean Shift, DBSCAN, and OPTICS demonstrate limitations, particularly in compression efficiency. Experimental validation using benchmark images from the CID22 dataset confirms the robustness and applicability of the proposed methods in diverse scenarios.