Detecting Moving Objects with an Omnidirectional Camera and Subtraction Whose Background Image is Renewed

2 Detecting moving objects When we use a PTZ camera and background subtraction to detect moving objects, we must synchronize the PTZ of the camera and the PTZ of the background image. This paper describes a method a new method of detecting moving objects with an omnidirectional camera. The proposed method uses a fixed omnidirectional camera which can acquire images surrounding the camera, and background subtraction based on updating background image. When we subtract the background image from the input image, we consider oscillation of intensity. Therefore, the proposed method can robustly detect moving objects under the condition that there were fluorescent lights and windows which outside light came into. In the experiment, the system can detect moving objects and show the images of the objects in the room. 2.1 Outline of the method The proposed method uses an omnidirectional camera and background subtraction. The fixed omnidirectional camera acquires the omnidirectional images surrounding the camera. At first, we estimate the background image and amplitude image from the first several frames of the input omnidirectional images. After the background image and amplitude image are estimated, we detect moving objects from the input images, the background image and the amplitude image, updating the background image and the amplitude image. Next, we calculate the fan-shaped bounding-boxes which enclose the detected moving objects, generate the common perspective images of the moving objects, and show the user the common perspective images. This section is structured as follows. First, we describe the omnidirectional camera: HyperOmni Vision. Next, we describe the method of detecting moving objects from the omnidirectional input images. At last, we describe the method of calculating the fan-shaped bounding-boxes which enclose the detected moving objects.