Fast lossless images compression for synchrotron radiation facility using deep learning and hybrid architecture

Zhang Min-xing; Fu Shi-yuan; Gao Yu; Cheng Yao-dong; Abdulhafiz Ahmed Mustofa; Chen Gang

doi:10.1007/s41605-024-00490-9

Volume 8 Issue 4

Nov. 2024

Turn off MathJax

Article Contents

Abstract

References

Radiation Detection Technology and Methods > 2024 > 8(4): 1693-1703 > doi: 10.1007/s41605-024-00490-9 CSTR: 32043.14.s41605-024-00490-9

PDF (2078 KB)

Fast lossless images compression for synchrotron radiation facility using deep learning and hybrid architecture

1 Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049, China;
2 University of Chinese Academy of Sciences, Beijing 100049, China

More Information

Corresponding author:
Zhang Min-xing,E-mail:zhangmx@ihep.ac.cn
Received Date: March 24, 2024
Revised Date: July 17, 2024
Accepted Date: August 01, 2024
Published Date: August 22, 2024

Abstract

Abstract

Purpose The rapid growth in image data generated by high-energy photon sources poses significant challenges for storage and analysis, with conventional compression methods offering compression ratios often below 1.5.
Methods This study introduces a novel, fast lossless compression method that combines deep learning with a hybrid computing architecture to overcome existing compression limitations. By employing a spatiotemporal learning network for predictive pixel value estimation and a residual quantization algorithm for efficient encoding.
Results When benchmarked against the DeepZip algorithm, our approach demonstrates a 40% reduction in compression time while maintaining comparable compression ratios using identical computational resources. The implementation of a GPU + CPU + FPGA hybrid architecture further accelerates compression, reducing time by an additional 38%.
Conclusions This study presents an innovative solution for efficiently storing and managing large-scale image data from synchrotron radiation facilities, harnessing the power of deep learning and advanced computing architectures.
- Lossless compression,
- Deep learning,
- Heterogeneous architecture,
- Synchrotron radiation image

FullText(HTML)

References (20)

References

[1]	. Q. Fazhi, H. Qiulan, H. Hao et al., The design of science data platform for high energy photon source. Front. Data Comput. 2(2), 40–58 (2020)
[2]	. Libpng.org. PNG Specification: Filter Algorithms. 2021. Available online:http://www.libpng.org/pub/png/spec/1.2/PNG-Filters. html.
[3]	. M. Rabbani, Book review: JPEG2000: image compression fundamentals, standards and practice. J. Electron. Imaging 11(2), 286–287 (2002)
[4]	. M. Goyal, K. Tatwawadi, S. Chandak, et al. DeepZip: lossless data compression using recurrent neural networks, in Proceedings of the 2019 data compression conference (DCC). Piscataway: IEEE, pp. 575–575 (2019)
[5]	. H. Yu, H. Mei, X. Xu, H. Jia, Research review of image compression algorithms based on deep learning. Comput. Eng. Appl. 56(15), 15–23 (2021)
[6]	. F. Mentzer, E. Agustsson, M. Tschannen et al. Practical full resolution learned lossless image compression, in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. Pp. 10629–10638 (2019)
[7]	. F. Mentzer, L.V. Gool, M. Tschannen, Learning better lossless compression using lossy compression, in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 6638–6647 (2020)
[8]	. J. Ho, E. Lohn, P. Abbeel, Compression with flows via local bitsback coding. Adv. Neural Inform. Process. Syst., 32 (2019)
[9]	. S. Zhang, N. Kang, T. Ryder et al., iFlow: numerically invertible flows for efficient lossless compression via a uniform coder. Adv. Neural Inform. Process. Syst., 2021, 34 (2021)
[10]	. S. Reed, A. Oord, N. Kalchbrenner et al. Parallel multiscale autoregressive density estimation, in International conference on machine learning. PMLR, pp. 2912–2921 (2017)
[11]	. N. Khan, K. Iqbal, M.G. Martini, Time-aggregation-based lossless video encoding for neuromorphic vision sensor data. IEEE Internet Things J. 8(1), 596–609 (2020)
[12]	. NVIDIA Corporation. Tesla V100 Data Sheet. [EB/OL]. (2017) [2021–10–13]. https:// www. nvidia. cn/ conte nt/ dam/ en- zz/ Solut ions/Data-Center/tesla-product-literature/Tesla-V100-PCIe-Produ ct- Brief. pdf.
[13]	. Xilinx Corporation. (2020). A100 Accelerator Card Data Sheet. [EB/OL]. Retrieved October 14, 2021, from https:// docs. xilinx. com/r/ en- US/ ds962- u200- u250.
[14]	. Intel Corporation. (2017). Xeon Silver 4116 Processor Specifications. [EB/OL]. Retrieved October 14, 2021, from https:// www. intel.com/content/www/us/en/products/sku/120481/intel-xeon-silver- 4116- proce ssor- 16- 5m- cache-2- 10- ghz/ speci ficat ions. html? wapkw= 4116.
[15]	. F. Mentzer, E. Agustsson, M. Tschannen et al. Practical full resolution learned lossless image compression, in Proceedings of the 2019 IEEE/CVF conference on computer vision and pattern recognition. Piscataway: IEEE, pp. 10629–10638 (2019)
[16]	. J.H. Zar, Spearman rank correlation. Encyclopedia of biostatistics, 7 (2005)
[17]	. Z. Wang, A.C. Bovik, H.R. Sheikh et al., Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)
[18]	. V.V. Starovoytov, E.E. Eldarova, K.T. Iskakov, Comparative analysis of the SSIM index and the pearson coefficient as a criterion for image similarity. Eurasian J. Math. Comput. Appl. 8(1), 76–90 (2020)
[19]	. S. Bai, J.Z. Kolter, V. Koltun, An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv preprint arXiv: 1803. 01271, (2018)
[20]	. Y. Chen, H. Fan, B. Xu et al. Drop an octave: Reducing spatial redundancy in convolutional neural networks with octave convolution, in Proceedings of the IEEE/CVF international conference on computer vision. pp. 3435–3444 (2019)

Zhang Min-xing, Fu Shi-yuan, Gao Yu, et al. Fast lossless images compression for synchrotron radiation facility using deep learning and hybrid architecture[J]. Radiation Detection Technology and Methods, 2024, 8(4): 1693-1703. DOI: 10.1007/s41605-024-00490-9

Citation: