RKRuslan A. Kozhemiakin

Ruslan A. Kozhemiakin

*This is an algorithmically generated author record based on publicly available information.

Ruslan A. Kozhemiakin has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Despeckling of Multitemporal Sentinel SAR Images and Its Impact on Agricultural Area Classification

    by Vladimir Lukin, Oleksii Rubel, Ruslan Kozhemiakin, Sergey Abramov, Andrii Shelestov, Mykola Lavreniuk, Mykola Meretsky, Benoit Vozel, Kacem Chehdi

    This chapter addresses an important practical task of classification of multichannel remote sensing data with application to multitemporal dual-polarization Sentinel radar images acquired for agricultural regions in Ukraine. We first consider characteristics of dual-polarization Sentinel radar images and discuss what kind of filters can be applied to such data. Several examples of denoising are presented with analysis of what properties of filters are desired and what can be provided in practice. It is also demonstrated that the use of preliminary denoising produces improvement of classification accuracy where despeckling that is more efficient in terms of standard filtering criteria results in better classification.


  • Prediction of Compression Ratio for DCT-Based Coders With Application to Remote Sensing Images

    by Alexander N. Zemliachenko, Ruslan A. Kozhemiakin, Sergey K. Abramov, Vladimir V. Lukin, Benoit Vozel, Kacem Chehdi, Karen O. Egiazarian

    A problem of predicting compression ratio (CR) for lossy image compression methods based on discrete cosine transform (DCT) is considered for remote sensing imaging as a main target application. We demonstrate that the noise presence in images leads to specific requirements to lossy image compression as well as to criteria used, and ways to meet these requirements. In particular, it is often desired to compress images in the neighborhood of optimal operation point (OOP) where a compressed image might be closer to the corresponding noise-free one compared to noisy (original, uncompressed) counterpart. In this paper, first, we consider a problem of predicting CR for compressing noisy images in OOP neighborhood. Several statistical parameters calculated in DCT domain in 8 × 8 pixel blocks are used for this. The factors that influence a prediction accuracy and the ways to improve this accuracy are discussed. Next, we show that there is a statistical parameter that does not require any a priori information on noise properties, and that can be used to predict CR with a high accuracy without any necessity to apply multiple image compression/decompression procedures. The proposed methods are thoroughly tested for two DCT-based image coders applied to component-wise lossy compression of hyperspectral data. Finally, a modification of the prediction approach for 3-D compression of multichannel images is proposed.