基于多幅图像的三维形状恢复新算法研究  

Study on New Algorithm of Shape Reconstruction Based on Multi-images

作　　者：苏秋萍[1] 司存瑞[1] 

机构地区：[1]陕西教育学院现代教育技术中心,陕西西安710061

出　　处：《航空计算技术》2007年第4期17-19,共3页Aeronautical Computing Technique

基　　金：陕西省教育厅专项科研计划项目(04KJ236)

摘　　要：提出了基于多幅图像的由明暗恢复三维形状的新算法。传统的SFS是基于一幅图像,而反射图方程是含有两个自由度的PDE,因此SFS是一个病态问题。本文提出了一种基于三幅图像的SFS新算法,克服了SFS的病态性。首先在图像的Lambertian反射模型的基础上,由三幅图像的逐个象素点的图像灰度值,建立物体表面方向的一次代数方程组,得到表示物体表面方向的唯一梯度向量。然后经过梯度恢复高度的算法,得到物体表面的形状。最后给出的仿真实例表明算法的有效性。New shape from shading algorithm based on multi - images is proposed in this paper. Traditional shape from shading is based on single image, but the reflectance map equation is a nonlinear partial differential equation （PDE） about two freedom degrees, so the SFS is an ill-posed problem. The proposed new SFS method uses three images to avoid this problem of SFS. The image is formulated under condition that the surface is Lambertian reflectance. For each pixel of the image, three reflectance map equations derive a set of linear equations about orientation of surface that determine the orientation uniquely. Then using an integration method from orientation, the height function of the surface is gained. Experimental simulation shows the proposed method is accurate.

关 键 词：形状重构 多幅图像 Lambertian反射模型 反射图方程 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]