Statistical estimation of blast fragmentation by applying stereophotogrammetry to block piles

• Blast fragmentation was analyzed with stereophotogrammetry and statistical methods.
• Stereophotogrammetry provided more accurate results than 2D image processing.
• Flat shape of blocks introduced an error factor in 2D image processing analysis.
• Performance of the statistical analysis was improved by input parameter improvement.
• The amount of error introduced by hidden parts in muckpile was assessed.

Stereophotogrammetry is the technique to extract the spatial information of an object by constructing a stereo-image from two or more photos. Additional information about the geometrical features of an object can be obtained with stereophotogrammetry than with 2D image processing techniques. In this study, stereophotogrammetry based on 3D modeling was used for the analysis of the surface blocks, and the information on the surface blocks in muckpile was used as the input parameter for the statistical estimation of blasted blocks. Monte Carlo simulation and Latin hypercube sampling were used as the statistical estimation methods for the blasted blocks. In the laboratory experiments, results with stereophotogrammetry and 2D image processing technique were compared with the physical measurements using a water tank. Finally, the applicability of stereophotogrammetry and the statistical estimation methods to the analysis of blast fragmentation was estimated through the field experiments.