Automatic Face Recognition and Identification Tools in the Forensic Science Domain
| Autor: | Angelo Salici, Claudio Ciampini |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Computer science
business.industry Speech recognition 010401 analytical chemistry ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Proficiency test Certification Machine learning computer.software_genre 01 natural sciences Facial recognition system 0104 chemical sciences Domain (software engineering) Forensic science 03 medical and health sciences Identification (information) 0302 clinical medicine Experimental work 030216 legal & forensic medicine Artificial intelligence business computer |
| Zdroj: | Digital Communication. Towards a Smart and Secure Future Internet ISBN: 9783319676388 TIWDC |
| DOI: | 10.1007/978-3-319-67639-5_2 |
| Popis: | This paper describes an experimental work executed by Carabinieri Forensic Investigation Department (Italy) to explore the performance of the automatic face recognition systems in forensic domain. The main goal of the research is to survey the recognition ability and identification performance of these tools. The experiments are carried out using three commercial automatic facial recognition platforms. In our work we compare the difference between the forensic experts’ way of manual facial comparison with the machine outcome in two different scenarios; the first is a training and certification environment, a facial image comparisons proficiency test to verify the recognition capabilities. The second is a daily forensic caseworks scenario, formed by 130 real cases successfully investigated by forensic experts, to analyze the identification achievement. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|