Urban target recognition at intersections using multimodality wireless sensor networks is a very promising system for the reduction of accidents by detecting unusual events in real time. To provide an alarm signal about an incoming car to a pedestrian using sound or to notify a driver about a pedestrian using infrastructure-to-vehicle communication, the deployed sensor system collects the sensed data from multimodality sensor nodes, performs data fusion, and conducts reactions to avoid the imminent accident. To facilitate the application, we design and implement an application-aware event-oriented MAC protocol (App-MAC) to support prioritized event delivery, provide inter-event fairness, improve the performance of channel utilization, and reduce energy consumption. App-MAC leverages the advantages of contention- and reservation-based medium access control (MAC) protocols to coordinate channel access and propose channel contention and reservation algorithms to adaptively allocate the time slots according to application requirements and current events status. To evaluate App-MAC, we have conducted simulations through the TOSSIM simulator and empirical studies using Berkeley TelosB motes with synthesized target recognition events and compared App-MAC with three state-of-the-art MAC protocols, i.e., sensor MAC (S-MAC), time-division multiple access (TDMA), and traffic-adaptive medium access (TRAMA), in terms of the proposed performance metrics, namely, averageeventdeliverylatency,eventfairnessindex,channelutilizationefficiency, and energyconsumptionefficiency. We found that App-MAC tremendously outperforms the other protocols in this application scenario.