PERSONAL COMPUTERS

Author: TAPIADOR MATEOS MARINO.
Year: 2005.
University: AUTÓNOMA DE MADRID [www.uam.es].
Place of defense: ESCUELA POLITÉCNICA SUPERIOR.
Place of preparation: ESCUELA POLITÉCNICA SUPERIOR.

Summary: This research has been carried out an analysis within the dynamic biometrics, in the area known as biometrics as of the writing. The approach is a holistic approach to the biometric capabilities of the human process of writing, covering both the phenomenon of writing as in handwritten form mecanográfica. Both processes are actually based on a single procedure generation, consisting of a cycle regulated where the generation of the idea, neuromuscular stimulation and monitoring of the movement, all of this on a continuous basis. The objective of this research was an analysis capabilities delas identifying the overall process of writing, addressing first the mechanical aspect of the process easier to deal with later on deeper analysis of the writing process manuscript, which enjoys greater complexity and information discriminatory. The investigation has been conducted primarily a study of the capacity of identification process click, click dynamics of the call, using keyboards writing mechanical or electronic standardized today in personal computers. It has generated as a result of the investigation a prototype system of this type for e-commerce type Business-to-Custonmer (B2C) called BIOWER, which implements a layered architecture client / server for the Internet. They are used open technologies and standardized, non-intrusive, for customers. The recognition module allows the use of a model based on grading patterns through fuzzy sets (fuzzy), either through a pure statistical model. Supports samples click using time between beats (interkey times), or using pulse times (holdkey times), or both simultaneously. Following this research led to the analysis of the capability of identifying handwritten script. For this, the research focuses on the most open of identification based on free and write off, iedigitalizaciones of original manuscripts written on paper and in terms of writing. Such identification is what the state of the prior art is known simply and generically as identifying writer (writer-ID). It generated as a result of a system of identification grafística ieidentificación individuals through writing handwritten environments or forensic police (individuals who do not cooperate). System called BIOGRAFO. This system allows recording of written characters samples obtained from digitized manuscripts, through which the system is able to model the alphabet staff of individuals, and make identifications with a single-mode module for data fusion at the level of decision, based on a combination of channels alógrafos. The proposed system is based on the use of vector gradient to infer the direction of the stroke at each point of the characters and make a response in terms of ranking candidates closer. These candidates are obtained against a person questioned, which is used as a reference to search the database of registered individuals who implements. This identification by ranking has been rigorously tested using standardized techniques in the field of biometrics for this purpose, namely, using the Rank Probability Mass. (RPM), the Cumulative Match Curve (CMC) and the average rank (E (r )).