三维重构系统的极线矫正算法研究

Algorithm for Rectification in 3-D Reconstruction System

在设计三维重构系统的过程中发现,极线矫正算法的优劣将直接关系到三维重构结果的准确性和精确度。因此,文中采用了先进行控制点变换再进行灰度值插值的算法来达到极线矫正的目的。该方法通过控制点变换来定义空间变换本身,用其来描述每个像素如何从初始位置"移动"到终止位置,再利用灰度值插值算法产生一个光滑映射,并将像素点的灰度值改变,从而达到矫正的目的。文中不仅对矫正前后左右目摄像头采集到的图像进行了对比,还对右图矫正前后的灰度波形进行了对比,矫正后的图像整体向左平移了150个像素点。

In designing the 3-D reconstruction system,it is found that the accuracy and precision of 3D reconstruction results is subject directly to the accuracy of the rectification algorithm. The control point transformation is combined with the gray value interpolation algorithm to rectify the plume. This method defines the space transform itself by a control point transformation,which is used to describe how each pixel ＂ mobile＂ from the initial position to the end position. Then,a smooth mapping is produced by grey value interpolation algorithm and the gray levels of pixels are changed to achieve rectification. Finally,the images before and after the rectification and their gray values are compared,showing that the whole image has a 150 px left shift.