Traffic delay estimation is one of the main concerns for big and smart cities where traffic congestion problem should be solved. In this sense, mechanisms for rapid and accurate traffic delay estimation that allow evaluation of road network performance are crucial. Recently, in Intelligent Transportation Systems (ITS), cooperative vehicular communications have become an attractive traffic surveillance method. The main drawback of the cooperative vehicular communications is that it requires minimum number of vehicles to be equipped with communication capability, known as penetration rate (PR), so that it can offer a good traffic surveillance. In this paper, a novel approach specially designed to estimate traffic delay with low PR for an isolated intersection. In particular, arrival and departure patterns for each incoming edge is estimated using fuzzy logic system. Based on average speed provided by vehicular communication, arrival and leaving time are defined and estimated. The estimated arrival and leaving time are used to determine the delay for each incoming edge with suitable adaptation mechanism. Averaging the delay for all the incoming edges give an indication for the traffic delay of the intersection as a whole. The proposed approach is realized and evaluated using COLOMBO framework through simulation. The simulation results reveal that the approach proposed can be used for traffic delay estimation with good accuracy even under low PR.