Variability and Subjectivity in the Grading Process for Evaluating the Performance of Latent Fingermark Detection Techniques

Abstract

When assessing latent fingermark development methods, forensic researchers commonly evaluate treated samples using a grading scale. However, the subjective nature of these evaluation methods leaves the results of such investigations open to criticism for potential grader bias. Assessment of fingermark development quality is ultimately dependent on an individual's background and experience. A pilot study was conducted as a preliminary stage of a large-scale international collaboration. A set of 80 fingermark samples was developed with 1,2-indanedione-zinc chloride. Grades for photographic images of the developed fingermarks were assigned independently by 11 fingermark researchers. Sixty-seven percent of the scores given to each individual sample were the same as the median grade, and 99% of the scores were within 1 grade. The researchers were also assessed on their consistency by including 20 duplicate images to be graded. Seventy-eight percent of the grades given were identical to their original scores. These results indicate that a small group of independent fingermark graders is sufficient to produce reliable and consistent data in projects requiring the assessment of fingermark quality.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.