Martin Drahanský

Influence of Skin Diseases on Fingerprint Quality and Recognition

Rozpoznavani osob na zakladě otisků prstů patři k nejrozsiřenějsim na celem světě. Někteři lide, jejichž prsty jsou zasaženy kožnim onemocněnim nemohou tuto technologii použivat a jsou tak diskriminovani, protože nemohou použivat otisky svých prstů pro autentizacni ucely. Tato kapitola popisuje vliv kožnich onemocněni na proces rozpoznavani osob podle otisků prstů a veskerou podpůrnou a souvisejici praci. Tento výzkum vyžaduje kooperaci specialistů z různých oborů: lekaři, daktyloskopicti experti, programatoři, atp. Po lehkem uvodu do rozpoznavani otisků prstů je představeno několik kožnich onemocněni, ktere mohou mit negativni vliv na proces rozpoznavani otisků prstů. Tato onemocněni byla vybrana z obsahle lekařske literatury za pomoci blizke spoluprace odborných lekařů. Jsou zde představena různa onemocněni, ktera mohou mit vliv na funkcionalitu systemů založených na rozpoznavani otisků prstů v podobě kompletniho souhrnu těchto onemocněni, ktera jsou pro tyto ucely rozdělena do tři kategorii. Tento přehled je nasledovan lehkým uvodem do problematiky daktyloskopických senzorů a popisem vývoje specialni stanice pro snimani otisků prstů zasažených kožnimi onemocněnimi. Posledni cast teto kapitoly se zabýva popisem algoritmu pro vylepseni obrazu otisku prstu a procesem odhadu kvality u nově vytvořene databaze otisků prstů poskozených kožnimi onemocněnimi.

Anatomy of biometric passports.

Travelling is becoming available for more and more people. Millions of people are on a way every day. That is why a better control over global human transfer and a more reliable identity check is desired. A recent trend in a field of personal identification documents is to use RFID (Radio Frequency Identification) technology and biometrics, especially (but not only) in passports. This paper provides an insight into the electronic passports (also called e-passport or ePassport) implementation chosen in the Czech Republic. Such a summary is needed for further studies of biometric passports implementation security and biometric passports analysis. A separate description of the Czech solution is a prerequisite for a planned analysis, because of the uniqueness of each implementation. (Each country can choose the implementation details within a range specified by the ICAO (International Civil Aviation Organisation); moreover, specific security mechanisms are optional and can be omitted).

Secure Collection Tree Protocol for Tamper-Resistant Wireless Sensors

We proposed modification of Collection Tree Protocol suitable for wireless sensors with tamper resistant module. This platform provides better security, however ordinary protocols cannot utilize its features. Our goal was to offer secure routing protocol with similar behavior and efficiency to the original protocol. Both protocols were simulated to prove that adding security to protocols does not necessarily lead to higher demands to data transfer and thus power consumption.

Simple EEG Driven Mouse Cursor Movement

This paper dwells on suggestion of direction control of computer mouse by electroencephalographic (EEG) device incorporated, which makes it possible for the user to affect the direction of the cursor’s movement on the screen by the frequency of brain’s oscillation. The motivation for solution of this problem is the effort to help to handicapped people to communicate with surrounding world. Described approach uses technique called operant conditioning [1] and in the simpliest version is based on comparison of signal magnitude in two narrow neighbouring bands. Computed difference influences direction and speed of cursor’s motion. Promising results of one axis control were achieved and described approach will be considered in more complex brain-computer interface.

Challenges for Fingerprint Recognition—Spoofing, Skin Diseases, and Environmental Effects

This chapter tries to find answers to the questions whether the fingerprint recognition is really so reliable and secure. The most biometric systems based on fingerprint recognition have very low error rates, but are these error rates really telling us everything about the quality of such a biometric system? What happens when we use spoofs to deceive the biometric system? What happens when the genuine user has any kind of skin disease on his fingertips? And could we acquire a fingerprint with acceptable quality if there are some distortions on a finger or there are some environmental effects influencing the scanning technology? Reading this chapter brings you an introduction of preparation of finger fakes (spoofs), spoof detection methods, summarization of skin diseases and their influence on papillary lines, and finally the environmental effects are discussed at the end.

3D face recognition based on the hierarchical score-level fusion classifiers

This paper describes the 3D face recognition algorithm that is based on the hierarchical score-level fusion clas-sifiers. In a simple (unimodal) biometric pipeline, the feature vector is extracted from the input data and subsequently compared with the template stored in the database. In our approachm, we utilize several feature extraction algorithms. We use 6 different image representations of the input 3D face data. Moreover, we are using Gabor and Gauss-Laguerre filter banks applied on the input image data that yield to 12 resulting feature vectors. Each representation is compared with corresponding counterpart from the biometric database. We also add the recognition based on the iso-geodesic curves. The final score-level fusion is performed on 13 comparison scores using the Support Vector Machine (SVM) classifier.

Einfluss von Hauterkrankungen auf den biometrischen Erkennungsprozess

ZusammenfassungFirmen und Institutionen wollen heutzutage vermehrt biometrische Systeme für Computersysteme und Datenschutz benutzen. Diese Systeme haben viele Vorteile, aber sie haben auch einige Nachteile. Eine gute biometrische Charakteristik soll (u.a.) folgende drei Eigenschaften erfüllen: Universalität, Dauerhaftigkeit und Erfassbarkeit. Mit anderen Worten sollte jede Person diese Charakteristik aufweisen, die Charakteristik soll sich nicht ändern und die Charakteristik soll einfach messbar sein. Viele Menschen leiden jedoch unter Fingerhauterkrankungen. Diese Krankheiten verändern meistens, oder zerstören sogar, ein Fingerabdruckmuster insofern, dass eine Person mit Fingerhauterkrankung einen veränderten Fingerabdruck aufweist, bzw. der Fingerabdruck gänzlich fehlt. Diese betroffenen Personen fühlen sich berechtigter Weise diskriminiert. Die Analyse und Vermeidung des Einflusses von Hauterkrankungen auf den biometrischen Erkennungsprozess ist jedoch eine große Aufgabe für Ingenieure, Entwickler und Wissenschaftler, die sich mit den Fingerabdrucksystemen beschäftigen.

Accelerometer Based Digital Video Stabilization for Security Surveillance Systems

This paper is devoted to accelerometer based image stabilization for video based security surveillance systems. At the beginning an introduction to the image stabilization is presented. Short description of the actual state of common algorithms for image stabilization follows, including our solution with some partial optimizations. At the end, we present a suitable hardware platform having a built-in accelerometer, which is responsible for advanced stabilization.

Gesichtsdetektion und -erkennung in Videos aus öffentlichen Kamerasystemen

ZusammenfassungKönnen wir das Gesicht des Täters in jeder Situation erkennen? Was benötigen wir dazu? Das Ziel dieses Artikels ist es, die Leser mit den Aufsichtskamerasystemen bekannt zu machen und deren jeweiligen Einsatz in der Praxis sowie die Anwendbarkeit für polizeiliche Zwecke (z. B. Ermittlungen bei Straftaten) zu diskutieren.

An algorithm for retina features extraction based on position of the blood vessel bifurcation

This paper deals with a proposal of the algorithm for features extraction from the blood vessels on the back of eyeball. Automated extraction is based on the bifurcations (points where a single vessel splits into two vessels) in retina. Each particular bifurcation is subsequently specified by its own position given by the coordinate system derived from the mutual position of the optic disc and fovea. The complete algorithm has been developed in C,, language and uses OpenCV library for image processing. This proposed method was evaluated on available images of the STARE and DRIVE databases.