Compressive Sensing for DCT Image

The Shannon/Nyquist sampling theorem claim that when capturing a signal, one must sample at least two times faster than the signal bandwidth in order to avoid losing information. Nowadays, compressive sensing, as a big idea in signal processing, is a new method to capture and represent compressible signals at a rate significantly below the Nyquist rate. In this paper, compressive sensing is applied in DCT image. 1-D and 2-D DCT are adopted respectively and the corresponding schemes are designed to match the transform. Experimental results shows that for 2-D images, compressive sensing with 2-D DCT can achieve better performance than 1-D DCT whether in PSNR values or visual quality.