It is illegal to pass a stopped school bus when the vehicle's stop-arm is extended and the red lights are flashing. Public opinion on this issue is very clear. A random phone survey of the public conducted by the National Highway Traffic Safety Administration (NHTSA) reported that more than 90 percent of respondents rated "passing a school bus that has its red lights flashing and the stop arm in full view" as a somewhat or extremely dangerous driving behavior (Boyle, Dienstfrey, and Sothoron, 1998). Despite this public opinion, there is evidence that the number of vehicles that illegally pass school buses each day is substantial. Based on data collected throughout the state of Illinois, the Illinois Department of Transportation (1996) estimated that more than 10,000 vehicles illegally pass school buses every day. Similar findings were reported in Florida (Center for Urban Transportation Research, 1996).

To address this problem, NHTSA sponsored a research effort aimed at developing an automated system for detecting and recording the license plates of vehicles as well as their drivers who illegally pass school buses. The overall objective of this research was to develop a prototype system that would automatically detect and record vehicles that illegally pass school buses (i.e., bus' stop-arm is extended and lights are flashing). Based on the results of technical, administrative, and legal feasibility analyses, system specifications were developed and a prototype unit was built. The prototype system was then field-tested in a variety of real-world conditions in both a controlled setting and on an actual school bus route. The results of the field test proved the prototype system to be comparable with other automated enforcement systems. Testing showed that recorded images were more identifiable when the violation occurred in the lane next to the school bus. In addition, frontal facial recordings were found to be 1.5 times more useful then profile recordings. It must be stressed that the purpose of the field test was to gather data that could be used in support of design recommendations and changes for the next generation of the system.