An Image-Based Object Detection Method Using Two Cameras

To develop a quick detection algorithm for the image-based autonomous vehicle (AV) or computer assisted driving system, this paper proposes a real-time object detection method using two coordinated cameras. The mapping between two simultaneous images from the two cameras and coordination of the virtual sensitive areas can be resolved using parameters like the camera shot angle, the camera focal length, and the virtual square which is the interested area. This calculation need minimum time and further object detection can be performed within the virtual area, which can considerably reduced the image size and accelerate the object detection. The method can be used for tailgating prevention or fast lane detection.