Paola Giordano

Fantappiè's group as an extension of special relativity on ε(∞) Cantorian space–time

In this paper we will analyze the Fantappie group and its properties in connection with Cantorian space-time. Our attention will be focused on the possibility of extending special relativity. The cosmological consequences of such extension appear relevant, since thanks to the Fantappie group, the model of the Big Bang and that of stationary state become compatible. In particular, if we abandon the idea of the existence of only one time gauge, since we do not see the whole Universe but only a projection, the two models become compatible. In the end we will see the effects of the projective fractal geometry also on the galactic and extra-galactic dynamics.

Image watermarking via wavelet approach and face biometrics

In this paper, we consider a new highly-performing system to establish the intellectual property of a file by using advanced Information Fusion (IF) techniques based on Face Biometrics (FB) and Wavelet Multiresolution Analysis (WMA). IF is used in different depending on the fact that it can be a textual file, an image or multimedial; indeed, the further step consists in creating blue biometric codes with different watermarking mechanisms based on: i) face features, ii) wavelets and wavelet packets respectively. It seems clear that such a system can be useful for all that concerns files/contents distributed online for what regards both the intellectual property and the access/modification/use of information. At the end we also show a new method, based on information entropy, to track the access and use of a specific file by different users. Indeed, thanks to our approach we can also know the sequence of the users.

An Optimization technique for image compression

In this work we consider an optimization method to image file compression based on fuzzy logic. This optimization is realized thanks to wavelet transform in a pre-compression phase. Thanks to this approach in the pre-processing phase the wavelet transform is used to fix some relevant parameters like as the best compression box and other tolerance parameter (locally defined). With respect to the traditional approach by using wavelet this method is better from time consuming point of view. While with respect to a standard fuzzy logic approach the presented method has the advantage to perform compression with different ratio on different box optimized with respect to each subimage.

Biometric Watermarking via Wavelet Approach, Reverse Engineering in Info-Security and Dematerialization of Public Administration

Abstract This research wants to present a highly-performing system to establish the intellectual property of a file by using advanced Information Fusion (IF) techniques. IF is different depending on the fact that it can be a textual file, an image or multimedial; indeed, the further step consists in creating blue biometric codes with different watermarking mechanisms based on wavelet and wavelet packet respectively. Thanks to reverse engineering algorithms which this work have generated, such a system seems particularly appropriate to contexts like Public Administration or production companies which present complex industrial/documental processes where it is relevant to trace the access to information, and the sequence of accesses made by operators/users. Hence it seems also clear that such a system can be useful for all that concerns files/contents distributed online for what regards both the intellectual property and the access/modification/use of information.

Image mining and massive analysis for large vision machines in astronomy

In this paper we present the architecture, the algorithm and the implementation of an advanced imaging system used in the context of very large vision machines. The challenge was to build a fully automated and real time system to perform image acquisition, processing, and analysis on 4 ÷ 256 Mpixels images coming from astronomical telescopes. The system has two levels of trigger to select automatically relevant events and to detect, and to recognize specially events into a set of images.

Information fusion based on Fuzzy Fourier Transform of fingerprints for image watermarking

This paper presents an efficient system which, by means of Information Fusion based on Wavelet techniques and Fuzzy Fourier Transform Fingercoding, is able to establish the intellectual property of a file. Due to the proliferation of online tools for sharing documents and multimedia data, additional security mechanisms are needed for ensuring data consistency, integrity and intellectual property. We propose to create a fingercode, obtained through Fuzzy Fourier Transform of fingerprint data; subsequently, this marker is inserted into the file by means of Information Fusion through wavelet and wavelet packets techniques. Finally, we demonstrate the correctness of the approach in unequivocally establishing the legitimate owner of the data and the exact ordered sequence of users who have accessed the data.

CANTORIAN SPACE IN NATURE AND DYNAMICAL SYSTEMS

In this paper we consider the motion of point-like massive systems on a Cantorian path. The results is that an harmonic material support with an external force is equivalent, under some assumptions, to a classical or quantum harmonic system on a continuous support. The idea that we want stress in this paper is that a Cantorian space could explain some relevant stochastic and quantum processes, if the space acts as an harmonic oscillating support, such as happen in Nature. This means that a quantum process, in some cases, could be explained as a classical one, but on a non continuous and fractal support. We consider the validity of this point of view, that in principle could be more realistic, since it describe the real nature of the matter and space, which does not only exist in Euclidean space or curved one, but in a Cantorian one.