A Clustering-Based Approach to Image Compression Using the K-Means Algorithm

Abstract

In the current Internet era, image information is extensively utilized across various sectors of society. Due to the substantial data volume associated with digital images, compression techniques are essential for efficient transmission and storage. The K-Means algorithm, one of the most widely adopted clustering methods in unsupervised learning, offers a viable approach to this challenge. This study applies the K-Means clustering algorithm to image compression, with implementation carried out via Python programming. Experimental results demonstrate that the K-Means clustering method effectively reduces image data volume while preserving essential visual information, thereby confirming its utility as a practical technique for image compression.