Morphological operations such as erosion and opening are applied to both direct and spatially filtered images of test fabrics to identify defects. Detecting defects morpholog ically on spatially filtered images of fabrics produces better results, particularly when the fabric is fine and contains defects of small size. The diffraction pattern of the test fabric is obtained optically by illuminating it with a collimated laser beam. A spatial filter is placed at the Fourier plane to remove the periodic grating structure of the fabric from the image. Morphological operations with a critically selected structuring element are then applied to the image after suitable pre-processing.
Get full access to this article
View all access options for this article.
References
1.
Dougherty, E. R., Ed., "Mathematical Morphology in Image Processing," Marcel-Dekker, NY, 1992.
2.
Giardina, C.R., and Dougherty, E.R., Morphological Methods in Image and Signal Processing, Prentice-Hall, Englewood-Cliffs, NJ, 1988.
3.
Jasper, W.J. , and Palli, H., ImageAnalysis of Mispicks in Woven Fabric, Textile Res. J.65 (11), 683-692 (1995).
4.
Mallik (Goswami), B., and Datta, A.K., Optical Imaging Technique for Defect Detection in Fabric, Ind. J. Fibre Textile Res. 23 (12), 277-280 (1998).
5.
Mallik (Goswami), B., and Datta, A.K., Fast Fourier Transform Technique for Identification of the Structural Properties and Irregularities in Fabric , J. Inst. Eng. India, 79 (10), 1-4 (1998).
6.
Ribolzi, S., Mercele, J., and Gresser, J., Real-Time Fault Detection on Textiles Using Opto-electronic Processing , Textile Res. J.63 (2), 61-71 (1993).
