ABSTRACT
This chapter discusses the problem of developing low-complexity computer vision algorithms targeted to wireless multimedia sensor networks (WMSNs) devices for enabling pervasive intelligent transportation systems. The WMSNs development has been mainly fostered by a new generation of low-power and high performance microcontrollers able to speed up the processing capabilities of a single wireless node, as well as the development of new microcameras and microphones imported from the mobile phones industry. The chapter provides an overview of the most popular embedded vision platforms in the WMSNs domain, as well as a description of their main hardware and image processing characteristics. Wireless transceivers are not provided on board and communications can be enabled through mote systems. The performance of vision-based algorithms targeted at WMSN devices mainly depends on the computational capabilities of the whole smart camera system. The effectiveness of the proposed algorithm can be seen as its ability in reflecting the real behavioral trend of monitored parking spaces.
