IEA/AIE 2007: New Trends in Applied Artificial Intelligence pp 195-206 | Cite as
Selecting an Appropriate Segmentation Method Automatically Using ANN Classifier
Abstract
In general, we can easily determine the manufacturing step that does not function properly by referring to the flaw type. However, a successful segmentation of flaws is the prerequisite for the success of the subsequent flaw classification. It is worth noticing that, different segmentation methods are needed for different types of images. In the study, a mechanism that is capable of choosing a proper segmentation method automatically has been proposed. The mechanism employed artificial neural networks to select a suitable segmentation method from three methods, i.e., Otsu, HV standard deviation, and Gradient Otsu. The selection is based on the four features extracted from an image including standard deviation of background image, variance coefficient, the ratio of the width to height of both foreground and background histograms. The results show the success of the proposed mechanism. The high segmentation rate reflects the fact that the four carefully selected features are adequate.
Keywords
Segmentation Feature Extraction Flaw Detection Flaw Classification BPN NetworkPreview
Unable to display preview. Download preview PDF.
References
- 1.Ng, H.F.: Automatic Thresholding for Defect Detection. In: Proc. 3rd Int. Conf. Image and Graphics, pp. 532–535 (2004)Google Scholar
- 2.Yeh, C., Perng, D.B.: A Reference Standard of Defect Compensation for Leather Transactions. The Int. J. of Advanced Manufacturing Technology 25(11-12), 1197–1204 (2005)CrossRefGoogle Scholar
- 3.Mery, D., Carrasco, M.: Automated Multiple View Inspection Based on Uncalibrated Image Sequences. In: Kalviainen, H., Parkkinen, J., Kaarna, A. (eds.) SCIA 2005. LNCS, vol. 3540, pp. 1238–1247. Springer, Heidelberg (2005)Google Scholar
- 4.Cheriet, M., Said, J.N., Suen, C.Y.: A Recursive Thresholding Technique for Image Segmentation. IEEE Trans. Image Processing 7(6), 918–921 (1998)CrossRefGoogle Scholar
- 5.Chan, F.H.Y., Lam, F.K., Hui, Z.: Adaptive Thresholding by Variational Method. IEEE Trans. Image Processing 7(3), 468–473 (1998)CrossRefGoogle Scholar
- 6.Mery, D.: Crossing Line Profile: A New Approach to Detecting Defects in Aluminum Die Castings. In: Bigun, J., Gustavsson, T. (eds.) SCIA 2003. LNCS, vol. 2749, pp. 725–732. Springer, Heidelberg (2003)Google Scholar
- 7.Gao, H., Siu, W.C., Hou, C.H.: Improved Techniques for Automatic Image Segmentation. IEEE Trans. Circuits Syst. Video Technol. 11(12), 1273–1280 (2001)CrossRefGoogle Scholar
- 8.Tancharoen, D., Jitapunkul, S., Chompun, S.: Spatial Segmentation based on Modified Morphological Tools. In: Proc. Int. Conf. Information Technology: Coding and Computing, pp. 478–482 (2001)Google Scholar
- 9.Bellon, O.R.P., Silva, L.: New Improvements to Range Image Segmentation by Edge Detection. IEEE Signal Processing Lett. 9(2), 43–45 (2002)CrossRefGoogle Scholar
- 10.Stojanovic, R., Mitropulos, P., Koulamas, C., Karayiannis, Y., Koubias, S., Papadopoulos, G.: Real-time Vision-based System for Textile Fabric Inspection. Real-Time Imaging. 7(6), 507–518 (2001)MATHCrossRefGoogle Scholar
- 11.Sauvola, J., Pietikainen, M.: Adaptive Document Image Binarization. Pattern Recognition 33, 225–236 (1999)CrossRefGoogle Scholar
- 12.Otsu, N.: Threshold Selection Method from Gray-Level Histograms. IEEE Trans. Syst. Man, Cybern. SMC-9(1), 62–66 (1979)Google Scholar
- 13.Freeman, J.A., Skapura, D.M.: Neural Networks Algorithms, Applications and Programming Techniques, 1st edn. pp. 89–128. Addison-Wesley, London (1991)MATHGoogle Scholar