Region Growing Based Optical Fiber Panel Shadow Detection

Abstract:

Automatic defect detection of light image is very important in optical fiber panel (OFP) research now. In order to achieve detection of shadow defect automatically, proposed a new region growing algorithm. Using gray characteristics and fuzzy connectedness of image, realized automatic seeds selection by first selecting seed window then selecting seed points. Realized region growing algorithm with adaptive threshold by using maximum between-class variance method (OTSU). Proposed one OFP shadow detection evaluation operator, evaluation results showed that algorithm proposed in this paper was more accurate positioning of the shadow, achieved significant reduction in redundant information, and improved segmentation quality of the image effectively.

• Periodical: Advanced Materials Research (Volumes 403-408)
• Pages: 1914-1917
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.403-408.1914
• Online since: November 2011
• Authors: Ming Ji Wang, Yun Wu, Dong Hua Fu
• Keywords: Evaluation Operator, Fuzzy Connectedness, Otsu

References

[1] WANG Ming-ji, WU Yun, ZHOU Xi-hong: Optical fiber shadow detection based on edge operator and mathematical morphology, Optical Instruments, vol. 30 (1), 2008, pp.24-28.

[2] WuYun, Cao Zhimin: Optical fiber panel shadow detection and optimization, ISTAI'2008, vol. 1, 2008, pp.199-202.

[3] WU Yun, WANG Ming-ji, CAO Zhi-min: Optical fiber panel shadow self-adaptive detection, Optical Instruments, vol. 32 (4), 2010, pp.9-13.

[4] Weihong Cui, Zequn Guan, Zhiyi Zhang: An Improved Region Growing Algorithm for Image Segmentation, CSSE 2008, vol. 891, pp.93-96.

[5] Wangkai Deng, Wei Xiao, He Deng, Jianguo Liu: MRI Brain Tumor Segmentation With Region Growing Method Based On The Gradients And Variances Along And Inside Of The Boundary Curve, BMEI IEEE, 2010, pp.393-396.

DOI: 10.1109/bmei.2010.5639536

[6] HE Xiang-qian, DENG Shi-xiong, ZHOU Li-hua: The Estimation of Approximate Area of Cell within Legal Medical Image Based on Region Growing Algorithm, Microcomputer Information, vol. 25 (3), 2009, pp.293-295.

[7] JOHN F. HADDON, JAMES F. BOYCE: Image Segmentation by Unifying Region and Boundary Information, IEEE Transactions on pattern analysis and machine intelligence, vol. 12(10), 1990, pp.929-948.

DOI: 10.1109/34.58867

[8] WANG Hai-yang, PAN De-lu, XIA De-Shen: A Fast Algorithm for Two-dimensional Otsu Adaptive Threshold Algorithm, Acta Automatica Sinica, vol. 9, 2007, pp.968-971.

[9] WANG Xiang-hui, Zeng Ming: A new metric for objectively assessing the quality of enhanced images based on human visual perception, Journal of Optoelectronics and Laser, vol. 19(2), 2008, pp.258-262.

[10] GB1990-03-01, Fiber Panel Test Method.