Cartoonification of Real-World Images Using Image Processing and K- Means Clustering

AbstractCartoonification refers to the process of transforming real-world images into cartoon-likevisuals by simplifying textures, enhancing edges, and reducing color gradients. This studypresents a hybrid approach that integrates classical image processing techniques withunsupervised machine learning to achieve visually appealing cartoon effects. The methodinvolves converting the input image to grayscale, followed by median blurring and adaptivethresholding for effective edge detection. To preserve edge sharpness while smoothingtextures, bilateral filtering is applied. Subsequent morphological operations further refine theimage structure. K-means clustering is then employed to reduce the color palette and enhancethe cartoon-like appearance [2] [3]. The effectiveness of the proposed method isquantitatively evaluated using Mean Squared Error (MSE), Peak Signal-to-Noise Ratio(PSNR), and Structural Similarity Index Measure (SSIM). The results demonstrate thepotential of the combined approach in generating high-quality cartoonified images withimproved structural fidelity and visual appeal.