Image Segmentation for rock fractures based on ARMA model

Rock fracture mapping is very important in many applications related to rock mechanics. The toughest task is the extraction of the fractures from the images of the rocks. Time series model has been used in this paper for segmentation of fractures from the rock images. The model is compared with orthodox edge detection algorithms. A first order autoregressive image model has been implemented. The model has been applied for both rough as well as smooth fractures. The model was observed to perform well specifically for

Keywords:

Image segmentation, ARMA, rock fracture,

PDF

___

[1]. W. Wang, “An edge based segmentation algorithm for rock fracture tracing,” Proceedings of the Computer Graphics, Imaging and vision: New trends, 2005. [2]. F.Zhao and W.Wang, “Rock fracture aperture characterization based on corner segment algorithm in image analysis,” IEEE, 2006, pp.1004-1007. [3]. A.Miyamoto,M.A.Konno,E.Bruhwiler, “Automatic crack recognition system from concrete structures using image processing approach,” Asian journal of InformationTechnology,Vol.6,no.5, 2007,pp. 553-561. [4]. W.Wang, H.Liao and Y.Huang, “Rock fracture tracing based on image processing and SVM,” Third International conference on Natural Computation, 2007. [5]. P.G.P.Ho and C.H.Chen, “Time Series Model Based Region Growing Method for Image Segmentation in Remote Sensing Images,” IEEE, 2004.