A WEIGHTED ITERATIVE METHOD FOR ROBUST SELF-CALIBRATION  

作　　者：Liu Shigang Wu Chengke Tang Li Jia Jing 

机构地区：[1]National Key Laboratory of ISN, Xidian University, Xi＇an 710071, China

出　　处：《Journal of Electronics(China)》2006年第5期713-717,共5页电子科学学刊（英文版）

基　　金：Supported by the National Natural Science Foundation of China (No.60473119 and No.60372043).

摘　　要：A robust self-calibration method is presented, which can efficiently discard the outliers based on a Weighted Iteration Method (WIM). The method is an iterative process in which the projective reconstruction is obtained based on the weights of all the points, whereas the weights are defined in inverse proportion to the re- ciprocal of the re-projective errors. The weights of outliers trend to zero after several iterations, and the accu- rate projective reconstruction is determined. The location of the absolute conic and the camera intrinsic pa- rameters are obtained after the projective reconstruction. The theory and experiments with both simulate and real data demonstrate that the proposed method is very efficient and robust.A robust self-calibration method is presented, which can efficiently discard the outliers based on a Weighted Iteration Method （WIM）. The method is an iterative process in which the projective reconstruction is obtained based on the weights of all the points, whereas the weights are defined in inverse proportion to the reciprocal of the re-projective errors. The weights of outliers trend to zero after several iterations, and the accurate projective reconstruction is determined. The location of the absolute conic and the camera intrinsic parameters are obtained after the projective reconstruction. The theory and experiments with both simulate and real data demonstrate that the proposed method is very efficient and robust.

关 键 词：Weight iteration Self calibration Projective reconstruction 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]