GPS correction with the sky-subimage recognition

In the present time almost every new car has been introduced by the manufacturing companies with the facilities of GPS and vehicle event data recorder (VEDR). The function of GPS satellite navigation is often used as a vehicle navigation purposes, and VEDR is used for recording videos during driving. When a car drives through tunnel, underpass, or under the viaduct, etc., GPS will not be able to receive the satellite signals with shading, and it is resulted in the error GPS satellite navigation. In order to resolve this problem, a sky-image recognition technique for the video of VEDR is proposed in this paper to complete this task. Since the gray-level range of color sky image is too larger to be separated from original image, the color input image is converted to gray scale image in the proposed algorithm first. Then, based on the entropy of image histogram within sky subimage, the road conditions of tunnel, underpass, viaduct, and T-junction are distinguished in the proposed algorithm. Some different road conditional images are collected and experimented with our proposed algorithm. Our experimental results show our proposed algorithm succeed in recognition of the road condition in the daytime and computational inexpensiveness.