Title :
Image Retrieval in Forensics: Tattoo Image Database Application
Author :
Lee, Jung-Eun ; Jin, Rong ; Jain, Anil K. ; Tong, Wei
Author_Institution :
Michigan State Univ., East Lansing, MI, USA
Abstract :
In this article, we took an unsupervised approach in designing appropriate similarity measures to explicitly address the challenge arising from low-quality tattoo image matching. In the future, we plan to improve the matching algorithm by exploring both super- vised and semisupervised learning algorithms. Besides tattoos, other types of soft forensic evidence can be collected and managed in the form of images, such as shoe prints and gang graffiti images. Although Tattoo-ID focuses on tattoo image matching and retrieval, the underlying techniques developed in the Tattoo-ID system can be adopted to other forensic image databases.15 Other types of soft forensic image evidence might include shoeprints and gang graffiti images. In the future, we plan to extend the Tattoo-ID system to different application domains.
Keywords :
forensic science; image matching; image retrieval; law administration; learning (artificial intelligence); forensics; gang graffiti images; image retrieval; semisupervised learning algorithm; shoe prints; similarity measures; soft forensic evidence; supervised learning algorithm; tattoo image database application; tattoo image matching; tattoo image retrieval; unsupervised approach; Content management; Digital forensics; Fingerprint recognition; Forensics; Image retrieval; Iris recognition; Tattoo-ID system; biometrics; forensic databases; multimedia; near-duplicate image retrieval; tattoo images;
Journal_Title :
MultiMedia, IEEE
DOI :
10.1109/MMUL.2011.59