A regional rotation algorithm of video images

Image rotation algorithm in image processing has been widely used in engineering applications, and a practical traffic video image rotation algorithm is helpful for the traffic monitoring system to analyze the video images, which would promote visualization and intelligent transportation system applications development. Based on traffic engineering practice, this paper mainly studies an image rotation algorithm of traffic video sequence, by different angles rotation and different ratios zoom of image in the region of interest and necessary interpolation operation, which can change the original irregular target region image into a rules pattern which will have the same height on the left and right sides as well as the same width on the top and bottom sides. So that the original target region uneven simple image changes into coherent and rules image, and image information is simplified widely, it's very useful for simple and efficient detection algorithm development in traffic video detection systems. Compared with the traditional image rotation algorithm, this paper uses methods of image rotation, image zoom and interpolation to regularize the image, it's closer to engineering practice and more practical in video-based traffic system, so it's innovative.