SAPE: A program for semi-automatic parameter extraction for strain analysis

SAPE (Semi-Automatic Parameter Extraction) is a program for extracting information relevant to strain analysis from input digital images. Input images are manually produced by tracing the outlines of objects of interest. This is a less onerous, more objective and much faster task than manual measurement or digitisation of object data (i.e. aspect ratio Ri and orientation φi). SAPE rapidly extracts the required data by using a simple region-growing algorithm to identify regions of interest. Subsequently, the second moments are calculated for each region enabling the common strain analysis parameters to be readily computed. The performance of SAPE was tested on three samples (microphotographs of deformed sandstone and oolite as well as a meter scale photograph of a deformed conglomerate) using a non-overlap statistic. This statistic is a normalised measure of overlap between two regions and can be used to compare different fits applied to the same region. In each case SAPE out-performed the manual method. SAPE offers significant advantages in terms of speed, objectivity and robustness and should facilitate the collection of large datasets for strain analysis. © 2005 Elsevier Ltd. All rights reserved.

Duke Scholars

Author Kingshuk Roy Choudhury Biostatistics & Bioinformatics, Division of Biostatistics