基于L-M算法的射影重建技术研究

Projective reconstructing based on the Levenberg-Marquardt method

在研究已有射影重建算法的基础上,利用L M算法对基于基础矩阵的射影重建算法得到的线性结果进行优化,提高了算法的估计精度和稳定性,并在求得所有图象对应的投影矩阵后利用Bundleadjustment方法对空间结构及投影矩阵进行全局优化,取得了良好的效果。

On the basis of current projective reconstructing methods, the Levenberg-Marquardt method is used to optimize the results of the projective reconstructing based on fundamental Matrix,the precision and stability is improved. The Bundle adjustment method is also used to globally optimize the structure of scene and projective matrixes and better result is obtained.