附加多视线特征约束的城市场景网格模型优化  

Refinement of urban mesh model with multi-view line feature constraints

作　　者：俞思帆 闫利[1] 谢洪[1] 陈长军[1] YU Sifan;YAN Li;XIE Hong;CHEN Changjun(School of Geodesy and Geomatics,Wuhan University,Wuhan 430079,China)

机构地区：[1]武汉大学测绘学院,武汉430079

出　　处：《测绘科学》2022年第2期102-109,共8页Science of Surveying and Mapping

基　　金：国家重点研发计划项目(2017YFC0803801,2017YFC0803802)。

摘　　要：基于多视立体几何技术重建的城市场景密集点云难免会包含噪声,尤其是在建筑边缘区域,这会导致三维模型在建筑边缘处发生扭曲,缺乏应有的平直锐利性。针对这一问题,该文提出一种三维网格模型优化方法:首先,从多视影像中提取二维线特征,定位建筑模型边缘区域,并构造多视线特征约束;然后,以网格顶点的影像灰度一致性约束和多视线特征约束作为数据项,以顶点自身曲率作为平滑项,构造能量函数;最后通过梯度下降法求解,逐步调整网格顶点位置。实验结果表明,该文所提算法在标准数据集上有效提高了网格模型的精确度和完整度,同时也恢复了模型边缘处平直锐利的特性,说明该文所提算法能有效地优化城市场景网格模型顶点位置,提高模型质量。The dense point cloud of urban scene reconstruction based on multi view stereo geometry technology inevitably contains noise,especially in the building edge area,which will lead to the distortion of 3 D model at the building edge and lack of straightness and sharpness.To solve this problem,a 3 D mesh model optimization method is proposed:firstly,the 2 D line feature information is extracted from the multi view image,the edge area of the building model is located,and the multi-view line feature constraint is constructed;secondly,the grid vertex image gray consistency constraint and multi-view line feature constraint are taken as the data items,and the curvature of the mesh itself is taken as the smoothing term to form the energy function;finally,the energy function is formed The gradient descent method is used to solve the problem,and the vertex position is adjusted step by step.The experimental results showed that the proposed algorithm could effectively improve the accuracy and integrity of the mesh model on the standard dataset,and also restore the straight and sharp edge of the model,which showed that the proposed algorithm could effectively optimize the vertex position of the urban scene mesh model and improve the quality of the model.

关 键 词：三维重建 网格优化 线特征 能量函数 梯度下降 

分 类 号：P237[天文地球—摄影测量与遥感]