Surface granularity as a discriminating feature of illicit tablets.

In this paper we propose an innovative methodology for automated profiling of illicit tablets by their surface granularity; a feature previously unexamined for this purpose. We make use of the tiny inconsistencies at the tablet surface, referred to as speckles, to generate a quantitative granularity profile of tablets. Euclidian distance is used as a measurement of (dis)similarity between granularity profiles. The frequency of observed distances is then modelled by kernel density estimation in order to generalize the observations and to calculate likelihood ratios (LRs). The resulting LRs are used to evaluate the potential of granularity profiles to differentiate between same-batch and different-batches tablets. Furthermore, we use the LRs as a similarity metric to refine database queries. We are able to derive reliable LRs within a scope that represent the true evidential value of the granularity feature. These metrics are used to refine candidate hit-lists form a database containing physical features of illicit tablets. We observe improved or identical ranking of candidate tablets in 87.5% of cases when granularity is considered.

[1]  Indra Gupta,et al.  Comparison of Pnn-Pca with Svm-Bdt and Moment Based Techniques for Leaf Shape Recognition and Classification , 2010, IPCV.

[2]  Zeno Geradts,et al.  Content based information retrieval in forensic image databases. , 2002, Journal of forensic sciences.

[3]  Cedric Neumann,et al.  New perspectives in the use of ink evidence in forensic science Part II. Development and testing of mathematical algorithms for the automatic comparison of ink samples analysed by HPTLC. , 2009, Forensic science international.

[4]  Jitendra Malik,et al.  Shape matching and object recognition using shape contexts , 2010, 2010 3rd International Conference on Computer Science and Information Technology.

[5]  Linda G. Shapiro,et al.  Computer and Robot Vision , 1991 .

[6]  Laura Aalberg,et al.  Drug intelligence based on MDMA tablets data I. Organic impurities profiling. , 2008, Forensic science international.

[7]  Michael D Cole,et al.  Development of a harmonised method for the profiling of amphetamines: IV. Optimisation of sample preparation. , 2007, Forensic science international.

[8]  Laura Aalberg,et al.  Drug intelligence based on MDMA tablets data: 2. Physical characteristics profiling. , 2008, Forensic science international.

[9]  Heung-Kyu Lee,et al.  A Ranking Algorithm Using Dynamic Clustering for Content-Based Image Retrieval , 2002, CIVR.

[10]  Franco Taroni,et al.  Statistics and the Evaluation of Evidence for Forensic Scientists , 2004 .

[11]  Odd Hordvin,et al.  The Drug Situation in Norway 2011 : Annual report to the European Monitoring Centre for Drugs and Drug Addiction - EMCDDA , 2009 .

[12]  Pierre Esseiva,et al.  Different likelihood ratio approaches to evaluate the strength of evidence of MDMA tablet comparisons. , 2009, Forensic science international.

[13]  C. Zingg,et al.  The analysis of ecstasy tablets in a forensic drug intelligence perspective , 2005 .

[14]  Anil K. Jain,et al.  Pill-ID: Matching and retrieval of drug pill images , 2012, Pattern Recognit. Lett..

[15]  Pierre Esseiva,et al.  The profiling of MDMA tablets: a study of the combination of physical characteristics and organic impurities as sources of information. , 2009, Forensic science international.

[16]  Dirk P. Kroese,et al.  Kernel density estimation via diffusion , 2010, 1011.2602.