Image Segmentation Method Based on Dual Feature Markov Random Field

Mingyi Duan; Yinju Lu; Yu Su

doi:10.3788/LOP57.221014

Journals >Laser & Optoelectronics Progress >Volume 57 >Issue 22 >Page 221014 > Article

Laser & Optoelectronics Progress
Vol. 57, Issue 22, 221014 (2020)

Image Segmentation Method Based on Dual Feature Markov Random Field

Mingyi Duan¹, Yinju Lu^1、2、*, and Yu Su¹

Author Affiliations

¹College of Information Engineering, Zhengzhou University of Technology, Zhengzhou, Henan 450044, China

²School of Optical Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

show less

DOI: 10.3788/LOP57.221014 Cite this Article Set citation alerts

Mingyi Duan, Yinju Lu, Yu Su. Image Segmentation Method Based on Dual Feature Markov Random Field[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221014 Copy Citation Text

show less

$G-L fractional differential operator template$

Fig. 1. G-L fractional differential operator template

Download full size

Fig. 2. Original images. (a) Cameraman; (b) synthetic image

Download full size

Fig. 3. Segmentation results under different windows. (a) 3×3; (b) 5×5; (c) 7×7; (d) 9×9

Download full size

Fig. 4. Image segmentation results under different temperature coefficients. (a) β=1; (b) β=2; (c) β=4; (d) β=8

Download full size

Fig. 5. Variation curve of image sequence energy function U(x)

Download full size

Fig. 6. Segmentation effect of different algorithms. (a1)(b1) MRF; (a2)(b2) FCN; (a3)(b3) GMM; (a4)(b4) D_MRF

Download full size

Statistics	Symbol	Calculation formula
Entropy	t₁	E_EN=- $\overset{L}{\sum_{i = 1}} \overset{L}{\sum_{j = 1}}$ P(i,j)ln $[P (i, j)]$
Energy	t₂	E= $\overset{L}{\sum_{i = 1}} \overset{L}{\sum_{j = 1}} {[P (i, j)]}^{2}$
Correlation	t₃	C_Cov= $\frac{\overset{L}{\underset{i = 1}{Σ}} \overset{L}{\underset{i = 1}{Σ}} ijP (i, j) - μ_{i} μ_{j}}{σ_{i} σ_{j}}$
Contrast	t₄	C= $\overset{L}{\sum_{i, j}}$ (i-j)²E
IDM	t₅	E_EE=- $\overset{L}{\sum_{i = 1}} \overset{L}{\sum_{j = 1}} \frac{P (i, j)}{1 + {(i, j)}^{2}}$

Table 1. Commonly used statistical characteristics in co-occurrence matrix

Sample	Segmentation algorithm	R_J			R_CCR
Sample	Segmentation algorithm	WM	GM	CSF	R_CCR
Cameraman	MRF	0.8436	0.8271	0.8341	0.8231
	FCN	0.9063	0.8972	0.8845	0.8601
	GMM	0.8272	0.8193	0.8062	0.8023
	D_MRF	0.9373	0.9285	0.9263	0.9389
Synthetic image	MRF	0.8386	0.8476	0.8462	0.8406
	FCN	0.8872	0.9165	0.8964	0.8903
	GMM	0.8116	0.8273	0.8012	0.8208
	D_MRF	0.9466	0.9373	0.9342	0.9391

Table 2. Comparison of results obtained by different algorithms

Mingyi Duan, Yinju Lu, Yu Su. Image Segmentation Method Based on Dual Feature Markov Random Field[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221014

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information