Fuzzy Clustering Remote Sensing Image Water Segmentation Algorithm Combined with Gravity Model

Qi Zhang; Guiqin Yang; Xiaopeng Wang

doi:10.3788/LOP202158.1410016

Journals >Laser & Optoelectronics Progress >Volume 58 >Issue 14 >Page 1410016 > Article

Laser & Optoelectronics Progress
Vol. 58, Issue 14, 1410016 (2021)

Fuzzy Clustering Remote Sensing Image Water Segmentation Algorithm Combined with Gravity Model

Qi Zhang, Guiqin Yang^*, and Xiaopeng Wang

Author Affiliations

School of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou, Gansu 730070, China

show less

DOI: 10.3788/LOP202158.1410016 Cite this Article Set citation alerts

Qi Zhang, Guiqin Yang, Xiaopeng Wang. Fuzzy Clustering Remote Sensing Image Water Segmentation Algorithm Combined with Gravity Model[J]. Laser & Optoelectronics Progress, 2021, 58(14): 1410016 Copy Citation Text

show less

Fig. 1. Algorithm flow chart

Download full size

Segmentation results of remote sensing image 1 by different algorithms. (a) Original image; (b) FCM segmentation result; (c) FLICM segmentation result; (d) KNLASC-FCM[9] segmentation result; (e) IIFCM[17] segmentation result; (f) NDFCM[7] segmentation result; (g) SNLS_IFCM[22] segmentation result; (h) segmentation result of our algorithm

Fig. 2. Segmentation results of remote sensing image 1 by different algorithms. (a) Original image; (b) FCM segmentation result; (c) FLICM segmentation result; (d) KNLASC-FCM^[9] segmentation result; (e) IIFCM^[17] segmentation result; (f) NDFCM^[7] segmentation result; (g) SNLS_IFCM^[22] segmentation result; (h) segmentation result of our algorithm

Download full size

Fig. 3. Segmentation results of remote sensing image 2 by different algorithms. (a) Original image; (b) FCM segmentation result; (c) FLICM segmentation result; (d) KNLASC-FCM^[9] segmentation result; (e) IIFCM^[17] segmentation result; (f) NDFCM^[7] segmentation result; (g) SNLS_IFCM^[22] segmentation result; (h) segmentation result of our algorithm

Download full size

Fig. 4. Segmentation results of remote sensing image 3 by different algorithms. (a) Original image; (b) FCM segmentation result; (c) FLICM segmentation result; (d) KNLASC-FCM^[9] segmentation result; (e) IIFCM^[17] segmentation result; (f) NDFCM^[7] segmentation result; (g) SNLS_IFCM^[22] segmentation result; (h) segmentation result of our algorithm

Download full size

Fig. 5. Segmentation results of remote sensing image 4 by different algorithms. (A) Original image; (b) FCM segmentation result; (c) FLICM segmentation result; (d) KNLASC-FCM^[9] segmentation result; (e) IIFCM^[17] segmentation result; (f) NDFCM^[7] segmentation result; (g) SNLS_IFCM^[22] segmentation result; (h) segmentation result of our algorithm

Download full size

Fig. 6. Segmentation results of remote sensing image 5 by different algorithms. (a) Original image; (b) FCM segmentation result; (c) FLICM segmentation result; (d) KNLASC-FCM^[9] segmentation result; (e) IIFCM^[17] segmentation result; (f) NDFCM^[7] segmentation result; (g) SNLS_IFCM^[22] segmentation result; (h) segmentation result of our algorithm

Download full size

Fig. 7. Segmentation results of remote sensing image 6 by different algorithms. (a) Original image; (b) FCM segmentation result; (c) FLICM segmentation result; (d) KNLASC-FCM^[9] segmentation result; (e) IIFCM^[17] segmentation result; (f) NDFCM^[7] segmentation result; (g) SNLS_IFCM^[22] segmentation result; (h) segmentation result of our algorithm

Download full size

Fig. 8. Reference images of water segmentation. (a) reference segmentation image of image 1; (b) reference segmentation image of image 2; (c) reference segmentation image of image 3; (d) reference segmentation image of image 4; (e) reference segmentation image of image 5; (f) reference segmentation image of image 6

Download full size

Method	Image 1	Image 2	Image 3	Image 4	Image 5	Image 6
FCM	87.3	79.7	75.4	86.4	82.7	84.3
FLICM	91.2	86.2	80.3	90.6	87.4	93.3
KNLASC-FCM^[9]	90.7	83.5	81.7	89.6	86.6	84.9
IIFCM^[17]	94.8	91.3	88.6	88.7	88.5	96.8
NDFCM^[7]	91.7	83.9	72.8	90.3	88.1	91.2
SNLS_IFCM^[22]	95.6	84.8	76.3	91.5	89.3	91.6
Proposed	96.3	94.8	89.2	93.4	90.8	97.1

Table 1. Accuracy of segmentation results unit: %

Method	Image 1	Image 2	Image 3	Image 4	Image 5	Image 6
FCM	27.3	37.9	35.7	26.8	24.3	18.2
FLICM	19.7	28.3	33.6	19.7	15.2	10.1
KNLASC-FCM^[9]	25.4	34.2	31.4	14.9	14.6	19.3
IIFCM^[17]	18.6	14.7	17.6	15.6	9.4	4.3
NDFCM^[7]	6.4	33.9	39.1	18.5	10.9	9.8
SNLS_IFCM^[22]	7.3	32.6	34.2	11.7	8.2	8.6
Proposed	5.2	7.9	13.8	9.3	7.7	3.7

Table 2. False alarm rate of segmentation results unit: %

Method	Image 1	Image 2	Image 3	Image 4	Image 5	Image 6
FCM	0.81	0.74	0.67	0.71	0.73	0.81
FLICM	0.87	0.82	0.73	0.84	0.81	0.88
KNLASC-FCM^[9]	0.83	0.76	0.71	0.79	0.77	0.82
IIFCM^[17]	0.91	0.88	0.82	0.77	0.84	0.93
NDFCM^[7]	0.89	0.79	0.64	0.82	0.82	0.89
SNLS_IFCM^[22]	0.91	0.81	0.72	0.83	0.85	0.91
Proposed	0.94	0.92	0.86	0.87	0.89	0.94

Table 3. Mean intersection of union of segmentation results

Method	Number of computational steps	Time complexity
FCM	N×c×T	O(n³)
FLICM	N×c×S×T	O(n⁴)
KNLASC-FCM^[9]	N×c×S×T	O(n⁴)
IIFCM^[17]	N+N×c×S×T	O(n⁴)
NDFCM^[7]	N×c×S×s×T	O(n⁵)
SNLS_IFCM^[22]	N×S+N×c×S×s×T	O(n⁵)
Proposed	N×S+N×c×S×T	O(n⁴)

Table 4. Time complexity of different methods

Qi Zhang, Guiqin Yang, Xiaopeng Wang. Fuzzy Clustering Remote Sensing Image Water Segmentation Algorithm Combined with Gravity Model[J]. Laser & Optoelectronics Progress, 2021, 58(14): 1410016

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information