.png){kind=icon}
.png){kind=icon}
Return
An efficient hybrid transform algorithm for image compression using a matrix rank-based optimization approach
.png){kind=icon}
PRE
AI
DOI：10.1080/20421338.2025.2601667.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The digital images play a major part in variety of fields such medical, satellite, remote sensing, etc. Advancements in information technology have resulted in a huge number of digital images to be stored and transmitted. Since the demand for storage capacity and communication bandwidth exceeds the available supply capacity, image compression techniques are carried out to overcome these limitations. Many compression algorithms have been proposed to meet these limitations; however, the computational complexity and faster compression rate issues are still challenging. In this research, a simple compression algorithm using Hybrid transformation such as fast wavelet transform (FWT) and discrete cosine transform (DCT) is used to decompose an image signal. Then a fast matrix rank algorithm (FMRA) using the magical graph method is used to compress and encode the image efficiently. The standard test images are used to evaluate the performance of the proposed algorithm in terms of CR, PSNR and MSE using MATLAB software. The proposed scheme works better in terms of its coding complexity and timing constraints compared with state of art algorithms. The findings imply that the proposed approach can be effectively applied in real-time systems such as medical imaging, satellite and wireless sensor networks where low memory use and high speed are essential, offering an important policy and application implication. The originality of this work lies in introducing a hybrid transform framework that integrates FWT, DCT and an FMRA to achieve high compression efficiency with reduced computational complexity.
Keywords:
fast wavelet transform
discrete cosine transform
fast matrix rank
magical graph
compression
Journal
A
IF:
1.2
Papers:
39
Citations:
0