Models, Algorithms and Architectures for Video Analysis in Real-time

This book is a revised version of my Phd thesis on computer engineering. It has been divided in three main parts. The first part will describe the study of embedded, special-purpose systems as a solution to real-time video analysis. We will first describe the development of a Real Time Convolver (RTC) with a parallelized systolic architecture. The system has been improved by functionally partitioning it onto a multi-FPGA device. Moreover, the FPGA solution is proposed for an UTC (Urban Traffic Control) system called VTTS (Vehicular Traffic Tracking System). The second part will focus on the multimedia cache research. In this part, a cache tuned to multimedia and image processing application has been studied and has allowed the development of novel prefetching techniques able to improve the overall performance of the cache of up to 140%!!! The last part is, indeed, related to video surveillance and, in particular, motion detection (Sakbot system) and shadow detection algorithms. Lastly, preliminary results of high-speed video analysis with the aim of computing the angle that the steering wheel of a Formula 1 s car does will be presented in the last chapter of this part.

Autorentext

Andrea Prati (MsC 1998, PhD 2002) is currently an Assistant Professor at Univ. of Modena and Reggio Emilia, Italy. His research interests range from robotic vision, to performance analysis for multimedia, to video-surveillance. He is author of more than 90 papers, and has been invited speaker, organizer of workshops and journal's special issues.