CISIM 2010


Piotr Porwik, Associate Professor

Biography:
Piotr Porwik
Piotr Porwik is a Professor of Computer Science in the Computer System Department, Computer Science and Materials Science Faculty, Katowice, University of Silesia, Poland. At present he works as a head of Computer Systems Department in University of Silesia.
Nowadays, his scientific researches focus on the computer networks, bometrics and biometric classifier systems, image processing, biomedical imaging and spectral methods of boolean function analysis. He received PhD and DSc (habilitation) degrees in Computer Science from AGH University of Science and Technology in Krakow in 1986 and 2006, respectively.
He has published over 100 scientific papers, which were published in Polish and foreign journals as well as numerous national and international conference presentations and talks. He has also published 2 books and edited 5 books.
He was also invited to many program committees of numerous national and international conferences. Professor Porwik has taken part in the capacity of special guest and was invited as a speaker in conferences and congresses in Poland, Japan and India.
Since 2007 he is the editor in chief of Journal of Medical Informatics & Technologies.
P. Porwik is a reviewer of numerouns international journals and fellowships including IEEE Int. J. Circuits, Devices & Systems, IEEE Transactions on Neural Networks, Transactions on Transport Systems Telematics, Int. Journal of Biometrics, American Mathematical Society.
Keynote Title:
The modern techniques of latent fingerprint imaging

Abstract:
Fingerprint matching is one of the most popular and reliable biometric techniques which are used in personal identification. This technique is based on the human print feature extraction which is nowadays one of the most important tasks in the Automatic Fingerprint Identification Systems (AFIS). With the advent of electronic banking and security of information, automatic personal identification is a very crucial topic. Credible, automatic personal identification is performed in police and civilian areas including special control protocols, passports control, credit cards, special identifiers, etc.
Many problems of the fingerprint analysis have been successfully solved. On the other hand modern dactyloscopy methods are still developed. For example registration and acquisition of visible impressions are well-known, but acquisition of latent fingerprints is more complex even by using high-tech devices and techniques.
Many biometric technologies are used in specialized police departments only. Several of them are used in the various applications. Automatically working recognition systems indicate some number of errors. We can observe a lot of scientific and industry works, where number of faults in biometric systems is constantly decreased. Unfortunately, faultless systems do not exist. In the paper different techniques (also tested in the Fingerprint Department of the Central Forensic Laboratory of the Polish Police) will be presented. These techniques concern both fingerprint disclosure and imprints enhancement.