Науковий вісник НГУ

Nonlinear dynamical analysis of abrupt welding texture change

• 
• 

User Rating:  / 0

PoorBest 

Details

Category: Information technologies, systems analysis and administration

Last Updated on 19 November 2016

Published on 17 November 2016

Hits: 4365

Authors:

Shuguang Wu, Department of Electronics and Information Technology, Jiangmen Polytechnic, Jiangmen, Guangdong, China

Yiqing Zhou, Department of Electronics and Information Technology, Jiangmen Polytechnic, Jiangmen, Guangdong, China

Abstract:

Purpose. Owing to its diversity, the texton of weld image is not very salient, and weld defects are difficult to detect automatically. The goal of this work is to identify the weld image texture change for such flaw detection and to determine the optimal number of elements, in particular, in a chaotic dynamic mode.

Methodology. The texture is characterized by the approximate entropy, which is calculated in phase space. The time series are reconstructed with the entropy of sub-image values for choosing the proper texture parameters. Applying the chaotic theory, we proposed the abrupt texture change area detection method.

Findings. We first get the approximate entropy in phase space, and then by using the abrupt texture change area detection method, we obtained the abrupt texture change area.

Originality. We pursued a study of the abrupt texture change area. We discussed a reconstruction of the principle of time series, approximate entropy mutation threshold determination. The research on this aspect has not been conducted before.

Practical value. In practice, it is essential to reconstruct the time series with the entropy of sub-image values in a first step, these results of approximate entropy in phase space are much more accurate within abrupt texture change areas.

References/Список літератури

1. Zhou, H., Zheng, J. and Wei, L., 2013. Texture aware image segmentation using graph cuts and active contours. Pattern Recognition, Vol. 46, No. 6, pp. 1719–1733.

2. Asha, V., Bhajantri, N. U. and Nagabhushan, P., 2012. Automatic Detection of Texture-defects using Texture-periodicity and Jensen-Shannon Divergence. Journal of Information Processing Systems, Vol. 8, No. 2, pp. 548–553.

3. Gernot Stübl, 2013. Periodicity Estimation of Nearly Regular Textures Based on Discrepancy Norm. Proc of SPIE, Vol. 8661, pp. 45–78.

4. Chen, Mi and Strobl, J., 2013. Multispectral textured image segmentation using a multi-resolution fuzzy Markov random field model on variable scales in the wavelet domain. International Journal of Remote Sensing, Vol. 34, No. 13, pp. 4550–4569.

5. Spampinato, C., Palazzo, S. and Kavasidis, I., 2013. A texton-based kernel density estimation approach for background modeling under extreme conditions. Computer Vision & Image Understanding, Vol. 122, No. 4, pp. 74–83.

6. Kennel, Pol, Fiorio, C. and F. Borne, 2015. Supervised image segmentation using Q-Shift Dual-Tree Complex Wavelet Transform coefficients with a texton approach. Formal Pattern Analysis & Applications, Vol. 18, No. 2, pp. 1–11.

7. Basha, S. R. and Reddy, P. K., 2015. Discrimination of Textures Using Texton Patterns. Global Journal of Computer Science and Technology, Vol. 15, No. 3, pp. 1–7.

05_2016_Shuguang
2016-11-15  789.01 KB  898

• < Prev
• Next >