Performance Evaluation of Image Compression Algorithms for Digital Communication Applications

Abstract

Digital communication systems increasingly transmit high-resolution visual information over channels with finite bandwidth, latency, storage, and energy constraints. This methodology paper develops a reproducible framework for evaluating image compression algorithms from both image-quality and communication-efficiency perspectives. The study focuses on the widely deployed discrete-cosine-transform JPEG codec and the wavelet-based JPEG 2000 codec, while drawing on earlier wavelet, SPIHT, EBCOT, and lossless coding research to establish the evaluation criteria. Five diverse 512 x 512 RGB test images are compressed at three operating profiles and assessed using compression ratio, peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), encoding time, decoding time, and estimated transmission time over a 1-Mbit/s link. A prototype benchmark shows that JPEG provides substantially lower computational latency and strong bandwidth reduction, whereas JPEG 2000 provides higher reconstruction quality at moderate and high-quality settings but requires greater encoding and decoding effort. The proposed methodology therefore evaluates compression as a multi-objective communication problem rather than selecting a codec solely from file size or PSNR. The framework is suitable for wireless imaging, telemedicine, surveillance, mobile multimedia, and sensor-network applications where quality, bandwidth, latency, and receiver complexity must be jointly balanced.