3-dimensional image analysis of fatigue fracture surface.

Quantitative Analysis of Three-dimensional Fatigue Fracture Surface Reconstructed by Stereo Matching Method

“…Method In the template matching, the template size should be large enough for obtaining reasonable results of the matching 1,4,23) , although the amount of calculation increases with increasing the template (window) size or with increasing the searched area. Therefore, it is necessary to reduce time for template matching, especially on stereo pair with the image size of more than some hundred thousands pixels.…”

“…Two-dimensional analysis of actual fracture surface profiles can reveal microstructures such as debris, overhang and microcracks linked to the fracture surface, which cannot be detected by three-dimensional fracture surface analysis 10,11,14,15,27) . However, the geometrical information about a given fracture surface in a wide area can be obtained by the three-dimensional fracture surface analysis 1,2,[4][5][6][7][8] . Combination of three-dimensional and two-dimensional analyses may lead to the further understanding of the geometrical features of fracture surfaces and the fracture mechanism in materials.…”

“…1 It is necessary for reconstructing a three-dimensional image from a stereo pair to examine the pixel-to-pixel correspondence between the basic image and the tilted image. [1][2][3][4] The template matching was employed to detect the corresponding point (pixel) in these two images. The window of a given size (q pixelsϫr pixels) in the basic image was imposed on that of the same size in the searched area (k pixelsϫl pixels) of the tilted image in order to examine the similarity of these windows, as shown in Fig.…”

“…[1][2][3][4] Stereo matching method has also been applied to the three-dimensional image reconstruction of fracture surfaces in materials. 1,2,[4][5][6][7][8] Kobayashi and Shockey 5) and Stampfl et al 6) estimated the fracture toughness of steels from the reconstructed three-dimensional images. Kimura and Tanaka.…”

“…[1][2][3] The procedure of three-dimensional image reconstruction is principally composed of the detection of corresponding points in a stereo pair by the pattern recognition, and the calculation of height at all the corresponding points. [1][2][3][4] Stereo matching method has also been applied to the three-dimensional image reconstruction of fracture surfaces in materials. 1,2,[4][5][6][7][8] Kobayashi and Shockey 5) and Stampfl et al 6) estimated the fracture toughness of steels from the reconstructed three-dimensional images.…”