Synthetic Aperture Radar (SAR) systems generate large data volumes that complicate onboard processing and transmission. This paper investigates the impact of several lossy SAR data compression techniques on radar image synthesis quality, including block adaptive quantization in quadrature and polar formats, vector quantization, and Daubechies D4 wavelet-based compression. The quantization signal-to-noise ratio (SQNR) is used as the main evaluation criterion. Simulation models were developed using ERS SAR data with 8-bit in-phase and quadrature components. Results show that for compression ratios of four or higher, the considered algorithms provide comparable performance and preserve the visual quality of synthesized images, while block adaptive quantization and vector quantization achieve higher SQNR at lower compression ratios. These findings confirm that efficient lossy compression can significantly reduce SAR data volume while maintaining acceptable image quality for practical applications.

In this work, we investigated the impact of data compression in space-based SAR on an image synthesis application. In the work defines a criterion for the quality of data compression algorithms, which makes it possible to evaluate the accuracy of data recovery after compression, relative to the original data: quantization signal-to-noise ratio (SQNR). The description of approaches to the implementation of models compression/decompression of SAR data based on block adaptive quantization (BAQ) algorithms in quadrature format, LHA in polar format, vector quantization (VC), Daubechies D4 wavelet transform (WP D4), which made it possible to obtain quantitative characteristics of the algorithms used for further analysis. In the article presents the characteristics of compression algorithms obtained by modeling, by which the possibility of using various approaches to data compression in the problem of synthesizing radar images is estimated, an example of synthesizing a radar image from data subjected to compression / decompression is shown. Simulation results using SAR ERS data show that all the considered algorithms, with compression ratios of 4 more times, have similar characteristics. In this case, noise is observed on the synthesized image, the brightest and most contrasting objects are clearly distinguishable. The image synthesized from the reconstructed data with compression less than 4 times visually practically does not differ from the image synthesized from the original data. At the same time, LHC algorithms in both formats and VK allow achieving quantization signal-to-noise ratios larger by 2-5 dB, compared with compression by the D4 IP method. LHC with quadrature representation of samples is the simplest of the considered algorithms in terms of practical implementation.

[1] J. C. Curlander and R. N. McDonough, Synthetic Aperture Radar: Systems and Signal Processing, New York, NY, USA: Wiley, 1991. [2] R. K. Raney, “Synthetic aperture imaging radar and moving targets,” IEEE Transactions on Aerospace and Electronic Systems, vol. AES-7, no. 3, pp. 499–505, May 1971. [3] C. Oliver and S. Quegan, Understanding Synthetic Aperture Radar Images, Raleigh, NC, USA: SciTech Publishing, 2004. [4] J. Max, “Quantizing for minimum distortion,” IRE Transactions on Information Theory, vol. 6, no. 1, pp. 7–12, Mar. 1960. [5] R. M. Gray and D. L. Neuhoff, “Quantization,” IEEE Transactions on Information Theory, vol. 44, no. 6, pp. 2325–2383, Oct. 1998. [6] A. Gersho and R. M. Gray, Vector Quantization and Signal Compression, Boston, MA, USA: Kluwer Academic Publishers, 1992. [7] D. Taubman and M. Marcellin, JPEG2000: Image Compression Fundamentals, Standards and Practice, Boston, MA, USA: Springer, 2002. [8] I. Daubechies, Ten Lectures on Wavelets, Philadelphia, PA, USA: SIAM, 1992. [9] I. G. Cumming and F. H. Wong, Digital Processing of Synthetic Aperture Radar Data, Norwood, MA, USA: Artech House, 2005. [10] G. Franceschetti and R. Lanari, Synthetic Aperture Radar Processing, Boca Raton, FL, USA: CRC Press, 1999.