Adaptive Compressive Sensing of Images Using Spatial Entropy

Table 2

SSIM comparison of various CS-based codec for test images at different measurement rates.

Test image	BCS		V-ABCS [6]		S-ABCS [7]		Proposed
	SSIM	ΔSSIM	SSIM	ΔSSIM	SSIM	ΔSSIM	SSIM
= 0.1
Lenna	0.5903	−0.2387	0.5447	−0.2843	0.5169	−0.3121	0.8290
Barbara	0.4823	−0.2299	0.5886	−0.1236	0.5618	−0.1504	0.7122
Peppers	0.5589	−0.2700	0.5521	−0.2768	0.5141	−0.3148	0.8289
Goldhill	0.5314	−0.2392	0.5218	−0.2488	0.5037	−0.2669	0.7706
Mandrill	0.3312	−0.259	0.3094	−0.2808	0.3100	−0.2802	0.5902
Avg.	0.4988	−0.2474	0.5033	−0.2429	0.4813	−0.2649	0.7462
= 0.3
Lena	0.8850	−0.0696	0.8556	−0.099	0.9082	−0.0464	0.9546
Barbara	0.8482	−0.0171	0.8292	−0.0361	0.8630	−0.0023	0.8653
Peppers	0.8826	−0.0607	0.8462	−0.0971	0.8939	−0.0494	0.9433
Goldhill	0.8207	−0.1055	0.7939	−0.1323	0.8334	−0.0928	0.9262
Mandrill	0.6334	−0.1995	0.6285	−0.2044	0.6312	−0.2017	0.8329
Avg.	0.8140	−0.0905	0.7907	−0.1138	0.8259	−0.0785	0.9045
= 0.5
Lenna	0.9515	−0.0285	0.9170	−0.063	0.9570	−0.023	0.9800
Barbara	0.9367	0.0044	0.9125	−0.0198	0.9433	0.011	0.9323
Peppers	0.9412	−0.0273	0.9029	−0.0656	0.9422	−0.0263	0.9685
Goldhill	0.9164	−0.0512	0.8722	−0.0954	0.9218	−0.0458	0.9676
Mandrill	0.7964	−0.1239	0.7843	−0.136	0.8066	−0.1137	0.9203
Avg.	0.9084	−0.0453	0.8778	−0.0760	0.9142	−0.0396	0.9537