Damage Detection of Metal Parts by Combining Information Entropy and Low-Rank Tensor Analysis

Peng Yang; Deer Liu; Ruixue Li; Jingyu Liu; Heyuan Zhang

doi:10.3788/LOP56.211006

Journals >Laser & Optoelectronics Progress >Volume 56 >Issue 21 >Page 211006 > Article

Laser & Optoelectronics Progress
Vol. 56, Issue 21, 211006 (2019)

Damage Detection of Metal Parts by Combining Information Entropy and Low-Rank Tensor Analysis

Peng Yang¹, Deer Liu^1、*, Ruixue Li¹, Jingyu Liu¹, and Heyuan Zhang²

Author Affiliations

¹School of Architectural and Surveying & Mapping Engineering, Jiangxi University of Science and Technology, Ganzhou, Jiangxi 341000, China;

²College of Chinese & Asean Arts, Chengdu University, Chengdu, Sichuan 610106, China;

show less

DOI: 10.3788/LOP56.211006 Cite this Article Set citation alerts

Peng Yang, Deer Liu, Ruixue Li, Jingyu Liu, Heyuan Zhang. Damage Detection of Metal Parts by Combining Information Entropy and Low-Rank Tensor Analysis[J]. Laser & Optoelectronics Progress, 2019, 56(21): 211006 Copy Citation Text

show less

Fig. 1. Data acquisition platform

Download full size

Fig. 2. Matrix amplification

Download full size

Fig. 3. Metal parts

Download full size

Fig. 4. Edge extraction. (a) Sobel algorithm; (b) Roberts algorithm; (c) Prewitt; (d) Log; (e) Canny; (f) G&C; (g) algorithm in Ref. [4]; (h) algorithm in Ref. [17]; (i) algorithm in Ref. [16]; (j) algorithm in Ref. [14]; (k) difference entropy matrix T_c of the proposed algorithm; (l) weighted difference entropy matrix T_c' of the proposed algorithm

Download full size

Fig. 5. Tensor analysis. (a) Sobel; (b) Roberts; (c) Prewitt; (d) Log; (e) Canny; (f) G&C; (g) algorithm in Ref. [4]; (h) algorithm in Ref. [17]; (i) algorithm in Ref. [16]; (j) algorithm in Ref. [14]; (k) T_c-tensor of the proposed algorithm; (l) T_c'-tensor of the proposed algorithm

Download full size

Method	R	R'	C_Y /%	C_Q
Sobel	1166	3711	31.4	0.116
Roberts	1540	5312	29.0	0.154
Prewitt	1187	3588	33.1	0.118
Log	1357	5305	25.6	0.135
Canny	1969	19742	10.0	0.196
G&C	1969	19742	10.0	0.196
D-S	3258	4048	80.5	0.325
D-QS	793	876	90.5	0.079
Ref. [4]	3449	4680	73.7	0.344
Ref. [17]	671	1962	34.2	0.067
Ref. [16]	593	1599	37.1	0.059
Ref. [14]	2918	6554	44.5	0.291

Table 1. Accuracy test

Noise	R	R'	C_Y /%	C_Q
0.022	3466	4431	78.2	0.346
0.032	3325	4430	75.1	0.332
0.039	3217	4370	73.6	0.321
0.045	3348	5615	59.6	0.334
0.050	2166	2861	75.7	0.216
0.055	3180	7662	41.5	0.317
0.059	2089	3479	60.0	0.208
0.063	2827	8271	34.2	0.282
0.067	2801	11996	23.3	0.280

Table 2. Robustness test

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information