Efficient lossy compression for compressive sensing acquisition of images in compressive sensing imaging systems - PubMed Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Dec 5;14(12):23398-418.
doi: 10.3390/s141223398.

Efficient lossy compression for compressive sensing acquisition of images in compressive sensing imaging systems

Xiangwei Li  1 , Xuguang Lan  2 , Meng Yang  3 , Jianru Xue  4 , Nanning Zheng  5
Affiliations

Efficient lossy compression for compressive sensing acquisition of images in compressive sensing imaging systems

Xiangwei Li et al. Sensors (Basel). .

Abstract

Compressive Sensing Imaging (CSI) is a new framework for image acquisition, which enables the simultaneous acquisition and compression of a scene. Since the characteristics of Compressive Sensing (CS) acquisition are very different from traditional image acquisition, the general image compression solution may not work well. In this paper, we propose an efficient lossy compression solution for CS acquisition of images by considering the distinctive features of the CSI. First, we design an adaptive compressive sensing acquisition method for images according to the sampling rate, which could achieve better CS reconstruction quality for the acquired image. Second, we develop a universal quantization for the obtained CS measurements from CS acquisition without knowing any a priori information about the captured image. Finally, we apply these two methods in the CSI system for efficient lossy compression of CS acquisition. Simulation results demonstrate that the proposed solution improves the rate-distortion performance by 0.4~2 dB comparing with current state-of-the-art, while maintaining a low computational complexity.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
The framework of a typical CSI system. (a) CS acquisition process; (b) CS reconstruction process.
Figure 2.
Figure 2.
Comparison between l1 norm and l2 norm of DCT coefficients of 16 × 16 blocks in an image. (a) Lena (Figure 8a); (b) Cameraman (Figure 8b).
Figure 3.
Figure 3.
Reconstruction PSNR for different sampling rates on four test images. (a) SR = 1/4; (b) SR = 5/8.
Figure 4.
Figure 4.
Empirical truncation point model.
Figure 5.
Figure 5.
CS measurements histogram and the fitted curve. (a) Lena 256 × 256 (SR = 0.5); (b) Cameraman 256 × 256 (SR = 0.7).
Figure 6.
Figure 6.
Quantization mapping table for the given target bits R. (a) Quantization cells; (b) Quantization intervals.
Figure 7.
Figure 7.
The framework of the CSI system with the proposed methods. (a) CS acquisition process with the proposed methods; (b) CS reconstruction process.
Figure 8.
Figure 8.
Gray scale test images. (a) Lena; (b) Cameraman; (c) Boats. (d) Peppers; (e) Goldhill; (f) Bank; (g) House; (h) Baboon; (i) Fingerprint; (j) Jetplane; (k) Lake; (l) Pirate.
Figure 9.
Figure 9.
Subjective quality comparison on image Boats at SR = 1/4. (a) Proposed; (b) Method [19]; (c) Baseline.
Figure 10.
Figure 10.
Subjective quality comparison on image Cameraman at SR = 5/8. (a) Proposed; (b) Method [19]; (c) Baseline.
Figure 11.
Figure 11.
R-D performance for three methods after CS reconstruction. (SR = 0.7). (a) Lena; (b) Cameraman; (c) Boats; (d) Peppers.
Figure 12.
Figure 12.
Subjective comparison (PSNR) of portions of Bank (SR = 0.7, R = 5) after CS reconstruction. (a) no quantization (26.73 dB); (b) uniform quantization (23.34 dB); (c) PDF-based quantization (24.91 dB); (d) universal quantization (24.90 dB).
Figure 13.
Figure 13.
Subjective comparison (PSNR) of portions of Lena (SR = 0.7, R = 5) after CS reconstruction. (a) no quantization (27.92 dB); (b) uniform quantization (24.34 dB); (c) PDF-based quantization (26.32 dB); (d) universal quantization (26.13 dB).
Figure 14.
Figure 14.
R-D performance comparison. (a) Lena; (b) Cameraman; (c) Boats; (d) Peppers.
Figure 15.
Figure 15.
Reconstructed image with SR = 1/4 and R = 6. Lena: (a) Proposed CSI + DPCM (PSNR = 26.44 dB); (b) Baseline CSI + DPCM (PSNR = 23.78 dB); Boats: (c) Proposed CSI + DPCM (PSNR = 26.76 dB); (d) Baseline CSI + DPCM (PSNR = 23.92 dB).
Figure 16.
Figure 16.
Reconstructed image at SR = 1/4 and JPEG quality level = 1/10. Lena: (a) Proposed CSI + JPEG (PSNR = 23.99 dB); (b) Baseline CSI + JPEG (PSNR=23.05 dB); Boats: (c) Proposed CSI + JPEG (PSNR = 24.20 dB); (d) Baseline CSI + JPEG (PSNR = 23.32 dB).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Unser M. Sampling-50 years after Shannon. Proc. IEEE. 2000;88:569–587.
    1. Li C., Jiang H., Wilford P., Zhang Y., Scheutzow M. A new compressive video sensing framework for mobile broadcast. IEEE Trans. Broadcast. 2013;59:197–205.
    1. Akyildiz I.F., Melodia T., Chowdhury K.R. A survey on wireless multimedia sensor networks. Comput. Netw. 2007;51:921–960.
    1. Pudlewski S., Melodia T., Prasanna A. Compressed-sensing enabled video streaming for wireless multimedia sensor networks. IEEE Trans. Mob. Comput. 2012;11:1060–1072.
    1. Blanes I., Magli E., Serra-Sagrista J. A tutorial on image compression for optical space imaging systems. IEEE Trans. Geosci. Remote Sens. Mag. 2014;2:8–26.

LinkOut - more resources