The number of vehicles is increasing rapidly in urban areas and metropolitan cities with in crease in population. To manage such a huge amount of traffic in an efficient way is the need of the hour. The conventional traffic signalling system are not adaptable to change in the amount of traffic. This paper solves the problem by making traffic signals smart enough to distinguish between the lanes which have different densities of traffic present in them. This paper makes use of image processing with the help of stored data set of respective lane images and a microprocessor to compute an output.