Threshold Based Image Compression Scheme for Improved Compressibility

In this paper, we introduce an improved performance of well-known lossy data compression techniques, namely Huffman and Arithmetic coding, using a customized scheme that involves threshold selection and conditional pixel probability mapping. The scheme shows efficiency in terms of compressibility with minimal information and similarity loss in the process. In addition to this, flexibility is exhibited by the proposed scheme to select required compression using a threshold. The outcomes of the research are evaluated and compared based on a variety of parameters such as compression ratio, MSE, SSIM, and PSNR. The proposed approach is capable of effectively handling information-varying images of different quality levels.