A decentralized hybrid traffic intersection control for urban areas is presented. First, an isolated intersection control problem is analyzed. Based on the different traffic conditions in an isolated intersection, a simple hybrid controller is proposed. Then this controller is extended to the network level, i.e., the effects of the control between multi-intersections are considered. This extension is concluded as a decentralized hybrid intersection controller, in which a finite state machine contains five states to represent the different traffic conditions. A simple simulation example is given and the simulation results show that this decentralized hybrid intersection control method can improve the performance of traffic networks.