A Wireless Sensor Networks (WSNs) is a dynamic wireless network which consists of a network of sensor nodes, in which each node communicate and has to rely on others to relay its data packets. Since the sensor nodes are normally constrained by battery and computing resources, therefore some nodes may choose, not to cooperate by refusing to do so while still using the network to forward their packets. Greedy nodes avoid themselves from being asked to forward data packets and hence conserve the resources for their own use. The resources in WSN are limited like energy and bandwidth which motivate nodes to reduce their energy consumption. Detection these malicious nodes is a real challenge in WSNs. In this paper we propose a SelfCentered Friendship tree (SCFT) algorithm to detect and remove greedy nodes from the network. In this paper, we focus on the detection phase and tried to improve the rate of packet loss due to existence of greedy nodes. Simulation results that this algorithm is highly effective and can reliably detect and remove greedy nodes.