Road safety is an imperative issue to be considered towards the aim of decreasing road accidents and fatalities thereof. The nature of traffic in India is identified to be highly heterogenic. Considering this an Intelligent Driver Assistance System is proposed in this paper. The said system, comprising of three phases: Image Acquisition, Image Processing and Reporting, employs a sequence of efficient image processing algorithms. The three modules of the Image Processing Phase work towards different aims to provide complete driver assistance in totality. These modules are: Prediction of crash; Least Congestion Route Suggestion and Lane Departure Detection. The proposed system makes use of onboard sensors for image acquisition. Canny edge detection algorithm is employed on the greyscale converted images to yield better results than other first order derivative filters. In the Crash prediction module, the system engages Kalman filters for trajectory analysis of the equipped vehicle and the surrounding vehicles in defined proximity. The proposed system is superior to its counterparts by virtue of its integrated modules providing comprehensive driver assistance.