基于倾斜摄影的三维模型单体化方法研究

利用三角面片的可分割性质，解决倾斜摄影数据的三维模型可分离单体化问题。该方法首先绘制切割多边形，确定单体化模型边界，然后使用包围盒方法进行求交检测，获得需要切割的三角面片。之后计算求交检测所得三角面片与切割多边形的交点，然后详细分析切割多边形与三角面片相交时的不同相交情形，并针对不同的相交情形采用不同的切割方法。最后，对相交区域三角面片进行裁切并重构，再纹理重构并实现多细节分层后得到分离的单体化模型。实验结果表明，该方法可有效实现倾斜摄影三维模型的分离单体化。

Study of method for achieving 3D model to be single based on oblique photogrammetry image

The splitting property of triangle patches are used to solve the problem of achieving the separate 3D building model to be single based on oblique photogrammetry image. Firstly, this paper draws a cutting polygon to determine the model boundary. Then bounding box is used to make the intersection detection to obtain the triangle patches which need to be cut. And the intersection points of triangles are found which may intersect with cutting polygon. Each of intersecting cases is detailed introduced and analyzed. All these intersection should be cut according to their actual conditions respectively. The cut triangular patches should be reconstructed. Through reconstructing the texture and levels of detail can achieve the separate 3D building model to be single. The experimental results show that this method can effectively achieve 3D model to be single.