Institute of Transportation, MOTC

　　In this paper a novel parking guidance system based on radio frequency and computer vision is presented. The goal of this paper is to present a wireless guidance sensor network that combines with a visual-based surveillance system that can allow easy, inexpensive monitoring of the vehicle berth in a car park. The wireless guidance sensor nodes are small in size, low cost, easily installed, and use ZigBee wireless communication technology to direct drivers to available spaces close to them. In addition, we present an unsupervised vision-based system for parking lot occupancy detection. The proposed method exhibits low computation complexity and uses just a few processes. The method is based on four main processing stages. In the first section, a raw image acquired from a camera system is preprocessed to mark the regions of the parking spaces. In the second step, the raw image is converted to a gray image. In the following step, car characteristics in the gray image are detected via the Harris corner detection algorithm. Finally, a parking place status is evaluated. Experimental results from both artificial and real scenes show promising results even in quite challenging conditions. Nonetheless a scope for improvement of the presented method is given.