SPIHT BASED IMAGE COMPRESSION USING QUADTREE DECOMPOSITION

Abstract

Contemporary lightweight devices such as mobile phones, cellular sensor systems, and high-consumption power devices, low-complexity picture compression techniques are critical. Low small bit rate and acceptable picture quality are vital criteria in such applications. This study offers low and average complexity methods for the picture mentioned above compression issue. The first will be level focused adaptive quantization coding, while the second will be a combination of discrete wavelet transforms and the depth established adaptive quantization coding method. Adaptive quantization coding provides a good peak signal to noise ratio (PSNR), but at the expense of a high little rate compared to other simple methods. The proposed methods generate a low small bit rate while maintaining an acceptable PSNR and picture quality. The acquired findings indicate a decent rate decrease with the same PSNR, or just a little less than any solo adaptive quantization coding algorithm's PSNR. Decomposing the type image using various wavelet filters and intensity focused adaptive quantization coding resulted in a bit of rate reduction. The suggested method includes several parameters that may be tweaked to alter the compressed image's performance. To improve picture compression, the Set Partitioning in Hierarchical Trees (SPIHT) method was recommended, based on the previous scanning of the coefficients around which there were more significant. Before being coded, the coefficients or items were ordered by the degree of significant coefficients. The more recent significant coefficients were preferable when the devices around us with significant coefficients were scanned. The findings of the experiments indicate that the approach can enhance PSNR and evident subjective experience compared to SPIHT.