In this chapter, we present an innovative technique of line sensor based image capturing and processing in order to detect moving objects such as vehicles. Line Sensor techniques, when used in MWSN, may achieve faster processing results with much less storage and bandwidth requirements while conserving node energy. Line Sensor based processing algorithms provide novel ways for object counting, classification and speed measurement. This solution presents itself as an ideal low-cost candidate for Intelligent Transport Systems (ITS) to monitor and control urban traffic.
WSN is a promising technology for distributed sensing and computation. Although constrained by limited computational capabilities, sensor networks are cost-effective and have good scaling virtues. The use of a low-power radio communication protocol such as the IEEE 802.15.4 allows to eliminate the cost of cabling the sensors, and allows more flexibility in the deployment. So far, WSNs have been mainly used in applications with low-frequency sampling and little computational complexity, such as environmental monitoring in agriculture, monitoring and control of the temperature and light in home automation, etc.
Recently, researchers and engineers started to investigate the use of WSNs to support more demanding applications, including process control, industrial automation, video surveillance, and multimedia streaming. Some early attempts have been made to use WSN technology for low-quality video streaming (Kulkarni, Ganesan, Shenoy & Lu, 2005). The success of such attempts have created a new challenging area of research, Multimedia Wireless Sensor Networks (MWSN) (Akyildiz, Melodia & Chowdhury, 2007).
MWSN technology is a good candidate for use in pervasive contexts like info-mobility. The idea is to use a set of inexpensive sensors nodes equipped with low-cost cameras in city streets to monitor traffic flows, number of cars in parking lots, etc.. The information are collected by the sensors and sent through wireless communication to a concentrator node that aggregates the incoming data (sensor fusion). The concentrator, which is in charge of collecting data related to a specific city area, is then connected to a wide area network, and sends aggregated data to higher levels of the information system hierarchy, which monitor and control urban mobility.
The architecture of such a system is shown in Figure 1. The backbone network can be wired (e.g. ADSL or Ethernet) or wireless (e.g. WiFi, WiMax, GPRS or UMTS), depending on the specific needs. In this proposal, we are interested in the lower level of the hierarchy that includes the sensor nodes and the concentrator.
General System Architecture
Examples of applications are:
Counting the number of cars passing on city streets. This information can help in estimating the traffic flow entering in (or exiting from) a city area, and take appropriate actions to prevent congestion. While there are other methods to count the number of cars passing in a street (by magnetic sensors, photoelectric cells, etc.), we envision that the one based on WSN cameras will be more flexible, easier to install and more cost-effective.
Identifying the occupancy level of a parking lot in a open area. This information can help to provide appropriate advices to drivers and guide them to the free spots. Also in this case there are many working examples of parking lots equipped with sensor systems. However, these systems are more difficult to be installed in open environments, due to high cabling cost. The use of WSN will help to lower the cost of installation and maintenance.
Counting number of people entering a building or in a room, or detect intrusion into limited access areas. Such information can be exploited for marketing purposes (e.g. how many people pass by a corridor) or scheduling personnel (i.e. how many employees are needed at a particular time of the day). A typical solution is an embedded unit with a single counter and camera integrated. The use of WSN could increase the system robustness, especially in challenging scenarios where the error rate goes up to 30%.
Counting people outdoor is a state-of-the-art challenge and no product is doing that reliably these days. The information is useful for public authorities (e.g. town-hall, police) that are interested in to distributing police agents and schedule their personnel. The use of WSN in outdoor counting will help lower main technical problems associated to this scenario, like harsh illumination conditions, weather, occlusions due to crowds, etc.
Monitoring industrial processes. Video cameras are being used in industrial processes to identify defects in products, or to monitor a hazardous area. Using less expensive wireless cameras would simplify this tasks and allow more cameras to be installed with less effort and inferior cost with respect to current monitoring systems. Also, a real-time architecture for WSN would allow to control the manufacturing process, and de-localize control algorithms, to make them closer to the sensors, thus reducing the overall cost.