We report on new image-analysis techniques that, for the first time, provide a practical solution to the problem of fully automated counting of fission tracks in natural minerals, a long-desired goal in fission-track dating. Specific challenges to be overcome have been the discrimination of fission tracks from non-track defects, polishing scratches, etc.; resolving multiple track overlaps; and reliable identification of small tracks amongst a similarly sized background of surface defects, fluid inclusions, etc. Most previous attempts at automated image analysis have failed in one or more of these tasks. The central component of our system is called ‘coincidence mapping’ and utilizes two images of the same tracks obtained in transmitted and reflected light. The complementary nature of the information in these two images allows a powerful discrimination of true fission tracks from most non-track features. The much smaller average track size in the reflected light image allows the resolution of most track overlaps apparent in transmitted light. The discrimination is achieved by segmenting the two images using a custom-developed thresholding routine and extracting the coincidence of features in the two binary images. The analysis is computationally efficient and takes only a few seconds to complete the processing of images that may contain up to many hundreds of tracks. Preliminary indications are that error rates are about the same as, or better than, those achieved by a human operator using normal counting conditions in transmitted light. The performance is even better at high track densities (>107cm−2) giving the potential for measuring track densities up to an order of magnitude greater than a human operator can count. Automated counting should significantly increase the speed and consistency of analysis and improve data quality in fission-track dating through better counting statistics, increased objectivity and measurement of additional track description parameters that are not currently determined.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.