Answers are grounded in this paper and cite their sources.
A total of 1028 samples of architectural paint were collected from store patrons who purchased paint at a medium-sized building supply store in Edmonton, Alberta, Canada. Information about what surfaces the paint would be applied to was collected from each customer at the time of purchase and the data summarized. The samples were characterized by colour, Fourier transform infrared spectroscopy (FTIR), pyrolysis gas chromatograph-mass spectrometry (py-GC-MS), and scanning electron microscopy-energy dispersive spectrometry (SEM-EDS). After visual colour comparison and FTIR analysis alone, 700 samples could be uniquely identified while the remaining 328 samples were divided into 98 groups, representing 956 indistinguishable pairs; the results after application of these two techniques gave a calculated discrimination power of 99.82%. The application of py-GC-MS and SEM-EDS to the remaining indistinguishable pairs gave only a very modest improvement in further distinguishing the samples. Information about current colour trends and which objects are most likely to be painted, or re-painted, by the general public is also provided.
No takes yet. Share an insight, caveat, or question.
Sandercock et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: