Research On Texture Mapping Method For Aircraft 3D Geometric Model

Texture Mapping have aroused general interest in computer graphics.Aiming at the application of the topic in civil aviation,aircraft 3D geometric model texture mapping method is researched,and the camera pose estimation method is studied.At the same time,a new method for texture mapping based on geometric relationship and energy information is proposed to realize the multi-view browsing of the realistic three-dimensional model.The main work of the thesis is as follows:Firstly,aiming at the huge and time-consuming calculation in image matching for camera pose estimation,a mismatch culling method based on Gaussian mixture model is proposed.The feature points are detected and initially matched by ORB algorithm,and the feature point sets are placed in Reproducing Kernel Hilbert Space.The Gaussian mixture model and smooth vector field are used to select the correct correspondences.Experiments show that the algorithm has a good classification effect,which is more conducive to estimate the camera pose in real time.Secondly,a hybrid model motion estimation method is adopted to obtain the camera pose.Based on the matched feature point sets,the 3D-3D and 3D-2D model estimation method are utilized to obtain the initial camera position.The accuracy of pose estimation is improved by using nonlinear method.The experiments prove that the camera pose trajectory obtained by this method has a high degree of fitting with the actual camera pose trajectory,which proves the effectiveness of the algorithm.Finally,aiming at the texture mapping process of aircraft 3D geometric model,the attitude of the camera relative to the geometric model is indirectly obtained by aligning the geometric model with the image reconstruction model.The texture mapping is performed from the geometric relationship and energy information of the texture images to build a 3D realistic aircraft model.Then a 3D browser is designed to observe the different views and states of the aircraft model.Experimental results show that the method is effective in texture mapping of 3D geometric models.