A Fast and Efficient Near-Lossless Image Compression using Zipper Transformation

TL;DR

This paper introduces a novel near-lossless image compression method using Zipper Transformation (ZT), which leverages DFT properties to outperform traditional methods like DCT and FWHT in speed and compression quality.

Contribution

The paper proposes a new Zipper Transformation-based compression scheme with two configurations, demonstrating superior performance over existing transforms in efficiency and compression quality.

Findings

ZT-based method is near-lossless and achieves better compression.

ZT offers faster implementation compared to DCT and FWHT.

Interlacing and concatenating ZT yield similar results.

Abstract

Near-lossless image compression-decompression scheme is proposed in this paper using Zipper Transformation (ZT) and inverse zipper transformation (iZT). The proposed ZT exploits the conjugate symmetry property of Discrete Fourier Transformation (DFT). The proposed transformation is implemented using two different configurations: the interlacing and concatenating ZT. In order to quantify the efficacy of the proposed transformation, we benchmark with Discrete Cosine Transformation (DCT) and Fast Walsh Hadamard Transformation (FWHT) in terms of lossless compression capability and computational cost. Numerical simulations show that ZT-based compression algorithm is near-lossless, compresses better, and offers faster implementation than both DCT and FWHT. Also, interlacing and concatenating ZT are shown to yield similar results in most of the test cases considered.