Enhancements of Robust Transmission for Block Compressed Sensing with LASSO

Transmission of compressed multimedia forms an important part for practical applications. Block compressed sensing (BCS) is a newly developed research in multimedia compression studies. Due to the fact that compressed data are vulnerable to the effects caused by lossy channels, protection schemes like multiple description coding (MDC) may need to be applied in advance. To enhance the protected results, we apply the least absolute shrinkage and selection operator (LASSO) for obtaining better results for reconstruction of BCS data. Simulations have shown the potential of LASSO with other protection mechanisms for the transmission of block compressed sensing.

[1]  Michael T. Orchard,et al.  Multiple description coding using pairwise correlating transforms , 2001, IEEE Trans. Image Process..

[2]  P. P. Deepthi,et al.  Block compressive sensing: Individual and joint reconstruction of correlated images , 2017, J. Vis. Commun. Image Represent..

[3]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[4]  J. Romberg,et al.  Imaging via Compressive Sampling , 2008, IEEE Signal Processing Magazine.

[5]  Hsiang-Cheh Huang,et al.  Error Resilient Transmission for Block Compressed Sensing of Color Images , 2018, 2018 IEEE 7th Global Conference on Consumer Electronics (GCCE).

[6]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.