KOMPRESI CITRA BERBASIS MULTI-LEVEL THRESHOLDING MENGGUNAKAN DIFFERENTIAL EVOLUTION

Abstract

Compression in image data needs to be done to manage storage space efficiently. In this study, image compression is performed using the Differential Evolution (DE) algorithm as an approach to achieve an optimal threshold for pixel clustering in images. The performance of this image compression method is evaluated based on the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) parameters. The PSNR of DE image compression ranges from 27 dB to 40 dB with threshold levels between 10 to 40. Meanwhile, the SSIM ranges from 0.83 to 0.97 at threshold levels of 10 to 40. The PSNR and SSIM of DE image compression are both at a good level, resulting in compressed image visuals that closely resemble the original image as the threshold levels increase.