| 基于Gauss-Newton法的L1数据拟合 |
| L1 Data Fitting Based on Gauss-Newton Method |
| |
| HTML View Full Text View/Add Comment Download reader |
| 中文摘要: |
| 在数据拟合中，当个别数据存在较大测量误差时，采用Ｌ１准则对数据进行拟合要优于最小二乘准则。本文通过建立Ｌ１准则与最小二乘准则之间的关系，给出了一种基于Ｇａｕｓｓ－Ｎｅｗｔｏｎ法的Ｌ１数据拟合自满。计算结果表明，该方法易于理解，便于编程，具有一定的理论意义和现实意义。 |
| 英文摘要: |
| It is a good choose to apply L\-1 Criterion other than least square criterion in data fitting when one or two large errors exist in the measuring data.This paper introduces a new algorithm for L\-1 data fitting based on Gauss\|Newton method through establishing the connection between L\-1 Criterion and least square criterion.Experimental results show that the method can be realised easily. |
| 田社平 颜德田 丁国清 |
| 上海交通大学精密工程及智能微系统研究所,上海 |
| 中文关键词: L1数据拟合 Gauss－Newton法 测量 数据处理 |
| 英文关键词:L\-1 Data fitting,Gauss\|Newton method,Measurement |
| 基金项目: |
| DOI： |
| Close |
