Design and development of image defogging system

The proposed work presents a prototype of novel method that comprises of a complete embedded system for the process of restoring the foggy images. Foggy conditions are considered to be a major source of accidents for vehicles. Use of a novel `Mean Channel Prior (MCP)' algorithm for defogging gave more accurate results as compared with the earlier existing methodologies. The prototype of proposed system works with use of raspberry pi, camera and display screen. Use of low cost, low power processing core raspberry pi made it low power, portable and stand alone system. Results of MCP image defogging algorithm are compared with Dark Channel Prior (DCP), Median channel Prior (MCP) and Improved single image dehazing using dark channel prior (ISIDDCP) on the basis of two parameters: Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR).