The impact of fingermark analysis Guidelines and fingerprint quality assessment software on the evaluation of fingermark value within the Indonesian National Police
Year: 2026
Type: article
Abstract: Previous studies have identified variability in the assessment of mark suitability, particularly for borderline-quality marks. Similar variability has been observed within the Indonesian National Police. This study examined the impact of adhering to established guidelines upon the consensus reached by Indonesian fingerprint examiners on the sufficiency of borderline quality fingermarks. An online survey was conducted with 146 examiners within the Indonesian National Police. Participants in the Guidelines Group followed simplified guidelines and assessed fingermark properties before determining their value. Guidelines+ Group adhered to the same protocol, supplemented by the inclusion of a clarity map and the DFIQI score. No Guidelines Group assessed fingermark images without guidelines or software assistance. The agreement among examiners was higher in the Guidelines (κ = 0.29) and Guidelines+ Groups (κ = 0.21) than in the No Guidelines Group (κ = 0.03), suggesting there is an advantage of utilising a simplified guideline. Furthermore, among the five quality parameters employed to evaluate fingermarks, minutiae count and image contrast emerged as the most critical factors, whereas the proportion of clear ridge area proved to be insignificant. Evidence indicates that differing examiner interpretations of quality parameter definitions contributed to data variability. Additionally, the clarity map and DFIQI Value Quality Score did not significantly affect the quantity or accuracy of annotated minutiae by the examiners. These findings indicate the value of incorporating a guideline in the assessment of fingermarks and the potential to develop the existing training program to promote consistency in the assessment of fingermark value within the Indonesian National Police.
Source: Science & Justice
Institution: University College London
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,720
OpenAlex ID: W7138277894