To prevent vehicles running off road, road edge detection is a fundamental function. Current work on road edge detection has not exhaustively tackled all weather and illumination conditions. In this paper, we first sort roads under various illuminations. Then, data mining is applied to a large driving video set to learn the qualitative and statistical distribution of road edge appearances. The corresponding features are selected for robust road edge detection. To visualize road appearance and evaluate the accuracy of detected road, a compact road profile image is generated from video data. Through the examination of all weather and illumination conditions, our road detection methods can locate road edges in good weather, reduce errors in dark illuminations, and report road invisibility in poor illumination.