Multithresholding-Guided Region-of-Interest SPIHT for Embedded Wavelet Image Compression
International Journal of Advanced Research in Science Communication and Technology · 09.07.2026 · 256
DOI: 10.48175/ijarsct-37629 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 09.07.2026. Verfasst von Prabhjot Kaur. Die Arbeit nennt 24 Literaturangaben.
Abstract (Originalsprache)
Embedded wavelet coders such as SPIHT transmit image content in decreasing order of energy, but they are agnostic to which spatial regions carry perceptual importance. This paper proposes a region-of-interest (ROI) extension of SPIHT in which the ROI is identified automatically by multilevel Otsu thresholding of a local-activity map, and the wavelet coefficients supporting the ROI are scaled by a power of two before encoding, granting them a fixed bit-plane head start in the embedded stream. The binary ROI mask is transmitted at reduced resolution, adding only 1024 bits of side information. The method preserves the fully embedded, progressive nature of SPIHT: a single encoding serves every bitrate, and any truncation now favours the perceptually important content. Experiments on 512 × 512 grayscale test images at 0.125–1.0 bpp show that the proposed coder improves ROI fidelity over standard SPIHT in all tested conditions, by up to 7.14 dB (moon, 0.5 bpp) and 6.02 dB (coffee, 1.0 bpp), and exceeds JPEG's ROI fidelity by 3.5–12.7 dB at identical total bitrate, while the whole-image cost remains a controlled 0.8–2.7 dB borne by the background. An ablation of the scaling exponent identifies s = 2 as a robust operating point balancing ROI gain against whole-image cost, and the study characterizes an inverse relationship between ROI size and achievable gain. The approach is training-free, adds negligible computational overhead, and is attractive for progressive transmission in bandwidth-constrained applications such as telemedicine and remote sensing
Zitieren
APA
Kaur, P. (2026). Multithresholding-Guided Region-of-Interest SPIHT for Embedded Wavelet Image Compression. International Journal of Advanced Research in Science Communication and Technology, 256. https://doi.org/10.48175/ijarsct-37629
Vancouver
Kaur P. Multithresholding-Guided Region-of-Interest SPIHT for Embedded Wavelet Image Compression. International Journal of Advanced Research in Science Communication and Technology. 2026:256. doi:10.48175/ijarsct-37629
BibTeX
@article{Kaur2026,
author = {Kaur, Prabhjot},
title = {Multithresholding-Guided Region-of-Interest SPIHT for Embedded Wavelet Image Compression},
journal = {International Journal of Advanced Research in Science Communication and Technology},
year = {2026},
pages = {256},
doi = {10.48175/ijarsct-37629}
}
EUVIMED