应用神经网络隐式视觉模型进行立体视觉的三维重建  

3D Reconstruction of Stereo Vision Using Neural Networks Implicit Vision Model

作　　者：袁野[1,2] 欧宗瑛[3] 田中旭[4] 

机构地区：[1]上海交通大学图象与通信研究所 [2]大连理工大学机械学院CAD&CG所大连116024 [3]大连理工大学机械学院CAD&CG所 [4]上海大学机器人中心

出　　处：《计算机辅助设计与图形学学报》2003年第3期293-296,共4页Journal of Computer-Aided Design & Computer Graphics

摘　　要：针对传统的基于精确数学模型的立体视觉方法过程繁琐的不足 ,提出一种应用BP神经网络隐式视觉模型进行三维重建的算法 该算法将多个标定平面放置在有效视场内 ,用神经网络模拟立体视觉由两个二维图像重建三维几何的过程 ,经过网络训练建模后 ,无须摄像机标定即可进行三维重建 仿真实验结果证明 ,该算法比较简单 。The classical stereo vision algorithms based on explicit model are very complicated, an algorithm of stereo vision based on BP neural networks implicit vision model is proposed. Multiple calibration planes are placed in the effective view field, neural network is used to simulate the process of generating 3D points from two 2D image points, after training the BP networks an implicit model was built, the 3D points can be reconstructed without the complicated calibration of camera. Experimental results show that the algorithm is fairly simple and accurate.

关 键 词：BP神经网络 立体视觉 三维重建 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]