Published 12 issues per year
ISSN Print: 0040-2508
ISSN Online: 1943-6009
Indexed in
OPTIMAL OPERATION POINT IN 3D DCT- BASED LOSSY COMPRESSION OF COLOR AND MULTICHANNEL REMOTE SENSING IMAGES
ABSTRACT
A task of lossy compression of multichannel images corrupted by additive noise is considered. The main attention is paid to the three-dimensional (3D) compression based on discrete cosine transform (DCT). It is shown that optimal operation point (OOP) often exists where OOP relates to such compression parameters when a compressed image is maximally close to corresponding noise-free image according to a chosen quality metric. It is shown that OOP exists with higher probability if 3D compression is carried out compared to component-wise lossy compression. If noise variance is the same in all components, quantization step (QS) values that provide OOP for 3D and component-wise compression are practically the same. Meanwhile, 3D compression possesses obvious benefits in the sense of metric values and compression ratios (CRs) in OOP. Experiments are carried out for test color images and 3-component Landsat image. It is demonstrated that the situation becomes more complicated if noise variance is not equal in image components. Then, several approaches to compression are proposed and tested.
-
Kozhemiakin Ruslan, Lukin Vladimir, Vozel Benoit, Image quality prediction for DCT-based compression, 2017 14th International Conference The Experience of Designing and Application of CAD Systems in Microelectronics (CADSM), 2017. Crossref
-
Zemliachenko Alexander N., Kozhemiakin Ruslan A., Abramov Sergey K., Lukin Vladimir V., Vozel Benoit, Chehdi Kacem, Egiazarian Karen O., Prediction of Compression Ratio for DCT-Based Coders With Application to Remote Sensing Images, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11, 1, 2018. Crossref
-
Lukin Vladimir, Zemliachenko Alexander, Krivenko Sergey, Vozel Benoit, Chehdi Kacem, Lossy Compression of Remote Sensing Images with Controllable Distortions, in Satellite Information Classification and Interpretation, 2019. Crossref
-
Zemliachenko Alexander, Lukin Vladimir, Alhihi Mohammad, Analysis of approaches to hyperspectral data compression based on orthogonal transforms, 2018 14th International Conference on Advanced Trends in Radioelecrtronics, Telecommunications and Computer Engineering (TCSET), 2018. Crossref