Browsing by Author "Santos, Gil Melfe Mateus"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Biometric recognition in unconstrained environmentsPublication . Santos, Gil Melfe Mateus; Proença, Hugo Pedro Martins CarriçoEvery human being is entitled, by his very nature, to a set of physiological and behavioral features that characterize him. The study of such features led to the development of a considerable amount of systems and applications, referred as biometric systems. The use of biometric systems has been significantly growing over the last years, particularly in the field of security: authentication, access control, criminal identification, etc. Being a highly demanding sector, it is then natural that greater focus is placed on the biometric traits that are able to deliver high discrimination between subjects whilst being less prone to forgery. However, such constraints represent a significant impact on both system’s usability and flexibility, requiring from the user a significant amount of cooperation. In this context, the iris is a primordial trait. Existing biometric recognition systems based on the iris follow the pioneer approach proposed by John Daugman, that proved itself as an excellent option for cooperative scenarios where images are acquired in the near-infrared spectrum. However, not in every case user cooperation is expected and, when not, systems with such high acquisition constraints are of little or no use. Research is then focused on circumventing those issues, either by improving the existing methods or finding new and more fitting traits. On the later, the periocular region (i.e., the region surrounding the eye) is one of the most promising characteristics: it mimics a natural and spontaneous way of recognition employed by the human beings; has an advantageous localization in relation to the iris, making it easy to be simultaneously acquired; and has, as corroborated by the literature, a set of promising characteristics that can be used for recognition purposes. The main objective of this doctoral work is then to either adapt or develop a novel biometric recognition system, suited for in the wild environments. Such systems should preferably use the periocular region as biometric trait, due to its flexibility and ease of acquisition in adverse conditions, and keep the operation constraints as low as possible. Subjects can be imaged ata- distance, on-the-move, and under irregular lighting conditions, using cameras working in the visible wavelength. To accomplish such goal, a set of intermediate milestones was established. At first, the iris was studied as biometric trait, paying particular attention to the techniques for allowing its use on in the wild scenarios. The effects of the visible wavelength light on iris performance for biometric purposes should not be disregarded and, as so, this factor was also studied. After rolling out iris appropriateness as the main distinctive feature, different emerging traits were analyzed, with special attention being paid to the periocular region. The most relevant methods were implemented and tested against the same dataset. Ultimately, multiple contributions were proposed and accepted by the scientific community, with applicability on different in the wild environments, the last of which is the proposal of an actual biometric system, working in real challenging conditions.
- Non-cooperative iris recognitionPublication . Santos, Gil Melfe Mateus; Proença, Hugo Pedro Martins CarriçoThe dramatic growth in practical applications for iris biometrics has been accompanied by relevant developments in the underlying algorithms and techniques. Along with the research focused on near-infrared images captured with subject cooperation, e orts are being made to minimize the trade-o between the quality of the captured data and the recognition accuracy on less constrained environments, where images are obtained at the visible wavelength, at increased distances, over simpli ed acquisition protocols and adverse lightning conditions. At a rst stage, interpolation e ects on normalization process are addressed, pointing the outcomes in the overall recognition error rates. Secondly, a couple of post-processing steps to the Daugman's approach are performed, attempting to increase its performance in the particular unconstrained environments this thesis assumes. Analysis on both frequency and spatial domains and nally pattern recognition methods are applied in such e orts. This thesis embodies the study on how subject recognition can be achieved, without his cooperation, making use of iris data captured at-a-distance, on-the-move and at visible wavelength conditions. Widely used methods designed for constrained scenarios are analyzed.