Faculty Mentor

Subrata Acharya


Illicit drug evidence constitutes a vast majority of chemical evidence collected from crime scenes. However, determining which drug is seized is not a trivial task as most are white powders. Since their introduction, colorimetric chemical detection tests, also referred to as presumptive drug tests based on their tentative determination of unknown substances, aid in the on-scene differentiation of drug material with a rapid color change within 1-2 minutes. These colorimetric tests are an important tool used in crime scene investigation and for obtaining a search warrant to find illegal drug labs and drug distributors. However, both positive and negative color interpretation is often reported differently depending upon the user and two analysts may describe the same color differently, e.g., "brilliant greenish blue" vs. "strong greenish blue". Moreover, the high rate of false positives and working color memory limit the effectiveness of these manual tests. To this effect, this research reduces the subjective interpretation and reporting with regard to color in these tests by offering users with a new platform/technology in the form of a Raspberry Pi (standalone) application that "reads" the color of the presumptive drug test, searches and matches the color using a pre-built library database, and reports accuracy (%) matches for further laboratory evaluation.

Cover Page Note

This research has been supported by the School of Emerging Technologies, Towson University.