This paper proposes a novel vehicle candidate generation algorithm for vehicle detection in on-road environments. The proposed algorithm uses shadow and edge information of original image collected from a camera in the vehicle. The bottom of vehicle has a shadow which is darker than the road. The proposed algorithm extracts the shadow region darker than the road, and creates an edge image using the Canny edge detector. It performs AND operation between the shadow image and the edge image. We can effectively generate the vehicle candidates from the result image. The experimental results show that the proposed algorithm provides fast and accurate vehicle candidate generation. And the number of candidates is reasonable for the vehicle verification.