Digital image processing based approach for tunnel excavation faces

The New Austrian Tunneling Method (NATM) is the most popular construction method in rock tunnel projects. Geological logging and rock mass classification are two major tasks prior to NATM tunnel excavation. Excavation face conditions are presently determined and logged by manual visual investigation for most, if not all tunnel projects. 2D geotechnical maps may then be drawn to show the locations and directions of weak planes. It is a costly and time-consuming process. To overcome the limitations of conventional approaches, image processing and information management techniques were adopted for tunnel face image storage, management, processing, interpolation, reconstruction, and visualization. An information system was used to manage and manipulate tunnel face related information, including geological description, rock mass rating and excavation face images. Image processing techniques in both spatial and frequency domains were applied to analyze and identify significant geological features (such as faults, joints and shear zones) from excavation faces. Shape-based image interpolation was then adopted to interpolate inter-slices between two neighboring captured images. Finally, a 3D image reconstruction and visualization environment was established to effectively assist geologists and tunnel project engineers in analyzing and evaluating geological characteristics of tunnel excavation faces.