Fingerprints are a unique and long-lasting indicator of our identity. Partial fingermarks found at crime scenes have been used for over 100 years in criminal investigations for the purpose of personal identification. Despite the increasing automation of the forensic process, most of the work is still done manually by investigators. One of the major shortcomings in this process is the lack of dedicated software for guided fingermark analysis and enhancement, forcing investigators to use generic image processing software. In this thesis we will present a user interface for fingermark analysis and enhancement. The user interface allows the use of an open source framework for automated analysis, feature extraction and fingermark quality assessment. We will present its functionalities and demonstrate some of them with concrete examples. The tool will allow future forensic investigators to intuitively use more advanced computer vision methods for the purpose of processing fingerprint and fingermark images.
|
