Wavelet Transform Domain for Deep Image Compression Using High Frequency Sub-Band Prediction

Main Article Content

Umang Garg

Abstract

This procedure demonstrates a Two Dimensional Discrete Cosine Transform (2D-DCT) system with quantization and zigzag layout. The JPEG picture compression approach uses this process as the foundation and main channel. Utilizing the 2D- DCT property, the calculation divides the overall process into compression and decompression. Even though the data in the images remained the same, the pixel was reduced predicated on the quantization and zigzag process. Division operation is used throughout the quantization process. As a result, this procedure minimizes data loss during decompression. The theoretical research offered in this study offers some fresh perspectives on how local variation behaves when JPEG compression is applied. Additionally, it might be utilised in a few aspects of image processing and analysis, including picture enhancement, image quality evaluation, and image filtering. MATLAB is used for this project, and improved vector quantization is used. Using the compression ratio, MSE, and PSNR, we can produce compressed pictures.

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

Article Details

How to Cite
Garg, U. . (2019). Wavelet Transform Domain for Deep Image Compression Using High Frequency Sub-Band Prediction. Turkish Journal of Computer and Mathematics Education (TURCOMAT), 10(1), 612–617. https://doi.org/10.17762/turcomat.v10i1.13556
Section
Research Articles