2.2.1. Preprocessing for Denoising

Various preprocessing methods have been used to enhance the performance or efficiency of the experiment in most technical research and process fields as well as image processing fields. In our system, the input image data is converted into a grayscale image, and a median filter is applied to remove noise.

Since only motion information is needed from the input image, color information is not required. Therefore, the input color image is converted into a grayscale image to reduce the dimension of the data. Then, the transformed grayscale image is filtered by a median filter to remove noise. Median filters are often used to remove signal noise. Unlike other smoothing filters, it also preserves the boundary values well in the noise removal process. In the case of the CCTV image, there is much noise due to sunlight or other illumination. Besides, there are many cases where the image quality is low, so it is necessary to enhance the edge. Using the median filter, we can remove noise and enhance the edge.
