Image Edge Information Aided Compressive Sampling Strategy

Jun Yang; Bo Pan; Li Chen; Yongan Zhu; Tao Jiang; Chen Cui

doi:10.3788/LOP57.081018

Journals >Laser & Optoelectronics Progress >Volume 57 >Issue 8 >Page 081018 > Article

Laser & Optoelectronics Progress
Vol. 57, Issue 8, 081018 (2020)

Image Edge Information Aided Compressive Sampling Strategy

Jun Yang^1、*, Bo Pan², Li Chen¹, Yongan Zhu¹, Tao Jiang¹, and Chen Cui³

Author Affiliations

¹College of Mathematics Physics and Information Engineering, Jiaxing University, Jiaxing, Zhejiang 314001, China

²Jiaxing Guodiantong New Energy Technology Co. LTD., Jiaxing, Zhejiang 314001, China

³School of Data Science and Technology, Heilongjiang University, Harbin, Heilongjiang 150080, China

show less

DOI: 10.3788/LOP57.081018 Cite this Article Set citation alerts

Jun Yang, Bo Pan, Li Chen, Yongan Zhu, Tao Jiang, Chen Cui. Image Edge Information Aided Compressive Sampling Strategy[J]. Laser & Optoelectronics Progress, 2020, 57(8): 081018 Copy Citation Text

show less

Fig. 1. Schematic for the ASR framework

Download full size

Fig. 2. Some test images. (a) Lena; (b) Fruits; (c) Boat; (d) House; (e) Cameraman; (f) Peppers

Download full size

Fig. 3. Comparison of effects of adaptive random sampling and completely random sampling

Download full size

Fig. 4. Recovery effect of completely random sampling and different adaptive random sampling when sampling rate is 35%. (a) Original image; (b) Φ_r: 27.2 dB; (c) Φ_m+M_c: 31.3 dB; (d) Φ_m+

\begin{matrix} M_{c}^{p} \end{matrix}

Download full size

Fig. 5. PSNR results of ASR and other CS recovery methods

Download full size

Fig. 6. Visual results of different algorithms with sampling ratio is 40%. (a)-(d) ASR; (e)-(h) BCS; (i)-(l) MARX; (m)-(p) TVAL3; (q)-(t) SRM

Download full size

Fig. 7. Optimal value estimation of adaptive sampling rate Φ_m+

\begin{matrix} M_{d, c}^{p} \end{matrix}

Download full size

Image	MARX	ASR	TVAL3	SRM	BCS
Lena	5820.0	8.7	1.7	0.5	15.6
Boat	4620.5	6.8	1.8	0.6	14.3
House	5220.1	7.0	1.5	0.5	16.4

Table 1. Running time of different algorithms for 256×256 image with sampling ratio is 35%s

Image	PSNR /dB
Image	ASR	BCS	MARX	TVAL3	SRM
Lena	32.06	31.09	36.08	29.55	29.33
House	37.19	34.67	38.46	28.01	33.26
Boat	29.04	26.67	29.87	27.53	25.52
Baboon	21.41	22.03	22.05	19.80	20.59

Table 2. PSNR values of different algorithms with sampling ratio is 40%

Jun Yang, Bo Pan, Li Chen, Yongan Zhu, Tao Jiang, Chen Cui. Image Edge Information Aided Compressive Sampling Strategy[J]. Laser & Optoelectronics Progress, 2020, 57(8): 081018

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information