Research On National Stylization For Image

Digital Image National Stylization is to generate rendering images, with embroidery style. This style of processing can be applied to animation, video games and other entertainment areas, bringing the user experience of national art when reducing cognitive fatigue. Digital Image National Stylization belongs to non-photorealistic rendering (NPR). Its scope is the combination of computer technology and painting. The aim of NPR is to express the art of image and to simulate various painting styles, not to render objects realistically. Today a variety of styles have been simulated successfully, such as relief, sketch, et al. However, ethnic-style analog is rarely, so simulation for ethnic-style is significant. It is not only the study of computer science, but also to develop national art. This paper proposed the algorithm of the national style.Cross stitch is using interweave method with embroidery thread on gingham and it is our common homemade crafts. Embroidery is sewing on the fabric with kinds of images; it includes Su embroidery, Xiang embroidery, Shu embroidery and Yue embroidery. There are wrong embroidery and crewel embroidery, etc. Pyrography uses a hot iron to draw on bamboo. It is the creator’s thoughts and emotions. To simulate the three ethnic styles is not only to bring new opportunity for our outstanding culture heritage, but also to bring new guideline for the expansion of NPR technology. Currently there are already some simulation technology exists, but users need understand the basic theory of art, the excellent results can be simulated.For these three ethnic styles, we present excellent effect algorithms to automatically convert the input image into image with cross-stitch, embroidery and pyrogaphy effect, without background experience about painting. Cross stitch technique is relatively simple, we set small areas as the roving, take the eve-value as the color. We can connect the cross. The key to simulate embroidery is to simulate the embroidery. We propose the embroidery difference model, it can form a single strand embroidery thread. For black and white line pattern, this paper proposes a method based on learning the art style. Firstly, processing the image into relief to form a visual sense of convex surface. Fusing with the original image to highlight the main pattern. Finally background color can be given to get good visual sense of embroidery. Pyrography has good edge characteristics, and its color is based on brown. We need edge detection. However, edges which are extracted by the mature log or canny operator are stiff and cannot keep the details. This paper proposes a hybrid model of the color dodge. Our algorithm cannot only extract the edge information of the image, and can be well preserved details, and can make the image more vivid, and this is one of the innovations of this paper. Our methods were all described on computer, and achieved good results, and it has important value for future related research.