Academia.eduAcademia.edu

Figure 9 – uploaded by ApaartiSebuahnama Saya

See full PDFdownloadDownload figure
Fig.8 Examples of templates for characteristic welding defects like a crack (upper right) and incomplete penetration of various sizes The SSDA method requires a set of templates representing characteristic welding defects. Although the defect size is important, we realized that a small number of templates is sufficient for obtaining good results. Examples of characteristic templates representing defects of incomplete penetration and a crack are given in Figure 8.

Figure 8 Examples of templates for characteristic welding defects like a crack (upper right) and incomplete penetration of various sizes The SSDA method requires a set of templates representing characteristic welding defects. Although the defect size is important, we realized that a small number of templates is sufficient for obtaining good results. Examples of characteristic templates representing defects of incomplete penetration and a crack are given in Figure 8.

Related Figures (10)

Fig.1 The total unsharpness in the object space, U/9, as a function of the magnification, M The behaviour of the total unsharpness U? as a function of the geometric magnification M, for RT and film radiography with a microfocus and fine focus X-ray system, is shown in Figure 1 (20 wm and 0.5 mm focal spots, respectively). For RT imaging with a fine focus X-ray system, the optimal magnification is found to be greater than 1. In fact, for a focal spot size of a = 0.5 mm, the optimal magnification is M,,, = 2. These results are
Fig.1 The total unsharpness in the object space, U/9, as a function of the magnification, M The behaviour of the total unsharpness U? as a function of the geometric magnification M, for RT and film radiography with a microfocus and fine focus X-ray system, is shown in Figure 1 (20 wm and 0.5 mm focal spots, respectively). For RT imaging with a fine focus X-ray system, the optimal magnification is found to be greater than 1. In fact, for a focal spot size of a = 0.5 mm, the optimal magnification is M,,, = 2. These results are
Fig.2 Areal-time X-ray image of an aluminium weld, with a defect of incomplete penetration. The image was taken with a geometrical magnification of 10
Fig.2 Areal-time X-ray image of an aluminium weld, with a defect of incomplete penetration. The image was taken with a geometrical magnification of 10
Fig. 3 Results for the fast-search algorithms applied to the image in Figure 2, for sub-images of 32 x 32 pixels. (a) Results from the first method — high frequencies contribution; (b) results from the second method — derivative function
Fig. 3 Results for the fast-search algorithms applied to the image in Figure 2, for sub-images of 32 x 32 pixels. (a) Results from the first method — high frequencies contribution; (b) results from the second method — derivative function
Fig. 4 The error vector Ep(/) as a function of / for: (a) good similarity; (b) — poor similarity. A few examples of applying the SSDA method to real-time X-ray images of Al welds are given in Figures 5—7. The welds in Figures 5 and 6 include defects of incomplete
Fig. 4 The error vector Ep(/) as a function of / for: (a) good similarity; (b) — poor similarity. A few examples of applying the SSDA method to real-time X-ray images of Al welds are given in Figures 5—7. The welds in Figures 5 and 6 include defects of incomplete
Fig. 6 An RTR image of a weld of 3 mm Al plates, including incomplete penetration defects of various sizes. Black dots represent points of identification by the SSDA method Fig. 5 An RTR image of a weld of 3 mm Al plates, including incomplete penetration defects of various sizes. Black dots represent points of identification by the SSDA method
Fig. 6 An RTR image of a weld of 3 mm Al plates, including incomplete penetration defects of various sizes. Black dots represent points of identification by the SSDA method Fig. 5 An RTR image of a weld of 3 mm Al plates, including incomplete penetration defects of various sizes. Black dots represent points of identification by the SSDA method
Fig. 7 An RTR image of a weld of 3 mm Al plates, including a crack of 0.1 mm width. Black dots represent points of identification by the SSDA method
Fig. 7 An RTR image of a weld of 3 mm Al plates, including a crack of 0.1 mm width. Black dots represent points of identification by the SSDA method
Fig. 9 Grey level profiles perpendicular to the welding seam including a crack: (a) raw image; (b) background image; (c) subtracted image
Fig. 9 Grey level profiles perpendicular to the welding seam including a crack: (a) raw image; (b) background image; (c) subtracted image
Fig. 10 An RTR image ( x 10 magnification) of a weld of 3 mm Al plates, including incomplete penetration defects: (a) raw image; (b) background image; (c) subtracted image; (d) thresholded image
Fig. 10 An RTR image ( x 10 magnification) of a weld of 3 mm Al plates, including incomplete penetration defects: (a) raw image; (b) background image; (c) subtracted image; (d) thresholded image
Fig. 11 An RTR image ( x 10 magnification) of a weld of 3 mm Al plates, including a crack: (a) raw image; (b) background image: (c) subtracted image; (d) thresholded image
Fig. 11 An RTR image ( x 10 magnification) of a weld of 3 mm Al plates, including a crack: (a) raw image; (b) background image: (c) subtracted image; (d) thresholded image
Related topics:
Engineering
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved