Bas-relief Modeling Based On Digital Images

The digital modeling of traditional art bas-relief is a challenging research direction in the field of computer 3D modeling.At present,there are two main types of bas-relief digital modeling technology.One is to compress the three-dimensional model of the object to form a bas-relief model.The other is to restore the height value of the object from the digital image to obtain a bas-relief model.The premise of bas-relief modeling technology based on 3D model is that the 3D model of the object already exists.However,in practical applications,many objects that need to be made into bas-relief are difficult to obtain their original 3D model.The bas-relief modeling technology based on digital image can recover the three-dimensional model of the object from the image,enrich the object of bas-relief modeling,expand the creation space of bas-relief,and improve the production efficiency of bas-relief.Taking a single ordinary digital image as the research object,this thesis focuses on the key techniques of image contour extraction,region division,height value restoration and surface 3D reconstruction,in order to quickly construct a bas-relief model with three-dimensional stereoscopic and planar sense.Mainly carried out the following work:1.Outline extraction of digital images.The flow-based Gaussian difference algorithm is improved and optimized to reduce the complexity of the algorithm and realize the extraction of high-quality image contours,which lays a foundation for image region division.2.Area division of digital images.The traditional connected domain calibration algorithm is improved,and the algorithm is optimized under the premise of improving the stability and efficiency.Using the known image contour,the improved connected domain calibration algorithm is used to divide the digital image into regions,and the appropriate image region is obtained to provide support for height value recovery.3.The height value of the digital image is restored.Analyze the principle of calculating the pixel height value by the shading recovery shape(SFS)algorithm,and improve the SFS algorithm to obtain more realistic 3D point cloud data of the object surface.4.3D reconstruction of point cloud data.Using the triangulation algorithm,the point cloud data generated by the restoration is reconstructed in three dimensions to obtain a bas-relief model of the digital image.5.Experimental verification and evaluation: Analytic hierarchy process is used to analyze and evaluate the traditional SFS algorithm and the bas-relief model constructed by the algorithm.It is clear that the results obtained by the algorithm are more realistic than the bas-relief formed by the traditional SFS algorithm.