One of the important functionalities of Advanced Driver Assistance System (ADAS) is the detection of lanes on the roads. This functionality is used to monitor other vehicles that are moving too close to the car or any obstacles that might disrupt the motion and either alerts the driver or takes full charge of the car. This system is built for driving automation ranging from Level 2 to Level 5 based on The National Highway Traffic Safety Administration (NHTSA). Many manufacturers are actively finding solutions for safety problems involving ADAS in semi-automated or fully automated cars as autonomous vehicles are significantly growing more popular. Lane departure detection is a basic feature of the majority of the current ADASs. There is one main significant challenge due to challenging road conditions (e.g. dotted lane, solid lane, curvy lane) and environments such as heavy rain, night vision and obstacles like trees, cones. Lane detection algorithms can be roughly classified into two categories: visual-based lane detection and machine learning-based lane detection. In this project it is intended to review and investigate few approaches in detection lanes in night time imagery. Two lane detection approach using standard Hough Transform and probabilistic Hough transform with Canny detector to detect the edges is evaluated and compared its accuracy in terms of DICE coefficients.