Theia: Multispectral Image Analysis and Archaeological Survey Vito Roberto 1,2 and Massimiliano Hofer 1 1 Dipartimento di Matematica e Informatica 2 Norbert Wiener Center University of Udine, Italy {vito.roberto,massimiliano.hofer}@dimi.uniud.it Abstract. Theia is a software framework for multispectral image analy- sis. The design is grounded on the object-oriented approach and a model combining parallel computation with selective data processing. Multi- spectral images from the archaeological site of Aquileia, Italy, have been used as the experimental testbed in order to assess the effectiveness and performance of the system; satisfactory results are reported, and are quite promising towards the use of the framework as a dynamic, inter- active interface to real-time data exploration and processing. Keywords: Multispectral, Hyperspectral, Image Processing, Interac- tive, Visualization, Object Oriented Design, Cultural Heritage, Archae- ological survey, Remote sensing. 1 Introduction The analysis of multispectral images is a central issue in a number of research and management tasks, such as environmental planning, medical diagnosis, ar- chaeological survey and surveillance for both military and civilian applications. Problems arise from the data acquisition by heterogeneous sensory systems; the massive data sets to be handled; the need for efficient algorithms for data visu- alization, filtering and mining. As a consequence, a number of challenging topics are to be faced by the designers. Proprietary frameworks are available like ENVI R  [2], or freeware like ISIS [3], which are the outcome of long-term research projects, and so are mature enough to address a large number of applications. More recent developments include Opticks [4] and Next ESA SAR Toolbox (NEST) [5]. However, most computers now process gigabytes of data in a second and load large data sets in the RAM; uniprocessor systems are being rapidly replaced by CMP, SMT and SMP architectures even in low-end hardware. Novel software systems should be put into operation to take advantage from this state of affairs [10]: new solutions can be explored towards real-time, dynamic processing of huge amounts of information, to be driven by fully interactive user interfaces. On the other hand, the fast growth and differentiation of applications sug- gest to adopt well-posed criteria of software design – inspired by modularity P. Foggia, C. Sansone, and M. Vento (Eds.): ICIAP 2009, LNCS 5716, pp. 337–345, 2009. c  Springer-Verlag Berlin Heidelberg 2009