Automated tracing of rock mass discontinuities from digital images

• Published in: International journal of rock mechanics and mining sciences (Oxford, England : 1997) Vol. 34; no. 3-4; pp. 256.e1 - 256.e19
• Main Authors: Reid, T.R., Harrison, J.P.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.04.1997
• Subjects: automated; characterisation; Digital; discontinuity; geometry; photogrammetry; scanline; spacing; trace length; image; automated; spacing; characterisation; photogrammetry; scanline; Digital; geometry; trace length; discontinuity
• ISSN: 1365-1609; 1873-4545
• DOI: 10.1016/S1365-1609(97)00281-5

The successful design of an engineered structure, either on or within a discontinuous rock mass, is largely dependent on our ability to reliably characterise discontinuity and rock mass geometry. Customary sampling methods for the collection of discontinuity measurements, using in situ scanline and window techniques, are inadequate in providing reliable characterisation and slow. Digital photogrammetry provides the means for the rapid collection of large quantities of discontinuity data. This paper presents initial results from a novel algorithm, coded in C, to automatically detect discontinuities from digital images of rock exposures. The basis of the method is that a digital image of a rock exposure is considered as a grey scale topographic surface, on which dark pixels represent low elevation and light pixels high elevation. Generally, discontinuities which intersect a near-planar rock face form dark linear features, and these can be regarded in three dimensions as topographic ravines. The algorithm we have developed considers the curvature of this surface and using this detects pixels which coincide with discontinuities.