Pavement monitoring and maintenance have become an essential task for maintaining a safe national transportation infrastructure. This has motivated research in automation of this process as it promises to result in great monetary savings, and can lead to more frequent inspection cycles. A new approach that is fully automated to detect road surface defects from large amounts of high-resolution grayscale images is presented. The images are collected with a vehicle-mounted rear-facing 5MP video camera complemented by GPS based positioning information. The algorithm starts by correcting the images for radial and angular distortion to get a bird’s-eye view image. This results in images with known dimensions (consistent in width per pixel) which allow data to be accurately placed on geo-referenced maps. Each of the pixels in the image is labeled as crack or non-crack using multinomial logistic regression model. The parameter values for the model are determined using maximum posterior estimate (MAP). The data used for testing and training are disjoint sets of images collected from the streets of Boston.