Insulator Image Segmentation Based on Improved Unit-Linking Pulse-Coupled Neural Network

Xiaoyan Du; Jun Zhong

doi:10.3788/LOP56.151005

Journals >Laser & Optoelectronics Progress >Volume 56 >Issue 15 >Page 151005 > Article

Laser & Optoelectronics Progress
Vol. 56, Issue 15, 151005 (2019)

Insulator Image Segmentation Based on Improved Unit-Linking Pulse-Coupled Neural Network

Xiaoyan Du and Jun Zhong^*

Author Affiliations

School of Electrical Engineering, Sichuan University, Chengdu, Sichuan 610065, China

show less

DOI: 10.3788/LOP56.151005 Cite this Article Set citation alerts

Xiaoyan Du, Jun Zhong. Insulator Image Segmentation Based on Improved Unit-Linking Pulse-Coupled Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(15): 151005 Copy Citation Text

show less

Segmentation results of insulator images under different weather conditions. (a) Original images; (b) gray images; (c) segmentation results of MIEP; (d) segmentation results of algorithm in Ref. [11]; (e) segmentation results of algorithm in Ref. [4]; (f) segmentation results of proposed method

Fig. 1. Segmentation results of insulator images under different weather conditions. (a) Original images; (b) gray images; (c) segmentation results of MIEP; (d) segmentation results of algorithm in Ref. [11]; (e) segmentation results of algorithm in Ref. [4]; (f) segmentation results of proposed method

Download full size

Segmentation results of insulator images under different SNRs. (a) Gray images after adding noise; (b) segmentation results of MIEP; (c) segmentation results of algorithm in Ref. [11]; (d) segmentation results of algorithm in Ref. [4]; (e) segmentation results of proposed method

Fig. 2. Segmentation results of insulator images under different SNRs. (a) Gray images after adding noise; (b) segmentation results of MIEP; (c) segmentation results of algorithm in Ref. [11]; (d) segmentation results of algorithm in Ref. [4]; (e) segmentation results of proposed method

Download full size

Fig. 3. Comprehensive evaluation results under different SNRs

Download full size

Weather	Method	UM	SM	CM	Comprehensive evaluation	Runtime
Sunny	MIEP	0.9216	0.7021	0.6669	0.7943	1.882485
	Ref.[11]	0.9737	0.9118	0.9465	0.9440	1.667535
	Ref.[4]	0.6791	0.7286	0.7612	0.7230	1.810260
	Proposed	0.9789	0.9582	0.9943	0.9866	1.772112
Sunset glow	MIEP	0.6280	0.4901	0.4107	0.5096	1.735032
	Ref.[11]	0.9087	0.6185	0.7346	0.7539	1.431268
	Ref.[4]	0.9138	0.7028	0.7947	0.8038	1.887225
	Proposed	0.9791	0.9657	0.9641	0.9696	1.681153

Table 1. Comparison of segmentation performance of four methods under different weather conditions

Xiaoyan Du, Jun Zhong. Insulator Image Segmentation Based on Improved Unit-Linking Pulse-Coupled Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(15): 151005

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information