A new algorithm used for rock pore segmentation is presented in this paper. Using the morphological erosion operator and hit-or-miss transform from mathematical morphology, the dividing lines’ possible positions are derived. The dividing lines’ direction can be further confirmed by dilation. The authenticity of the dividing lines can be verified using conditional dilation. Consequently, the accurate pore segmentation can be achieved. It has been proven by various practical applications that this algorithm is relatively accurate, rapidly executed, and insensitive to noise. In effect, the algorithm has excellent antinoise ability.
CITATION STYLE
Jiqun, Z., Chungjin, H., Xin, L., Dongmei, H., & Hua, L. (2015). An algorithm for rock pore image segmentation. Lecture Notes in Electrical Engineering, 339, 243–250. https://doi.org/10.1007/978-3-662-46578-3_28
Mendeley helps you to discover research relevant for your work.