sift points for screening based on random forest in ins/sar system

DOI

10.2991/iccset-14.2015.76How to use a DOI?

Keywords

machine learning, matching point, random forest

Abstract

The paper adopts machine learning framework to analyze matching suitability of SIFT match points in INS/SAR integrated navigation system. The paper adopts machine learning framework to analyze matching suitability of SIFT match points in INS/SAR integrated navigation system. It aims to select the matching points for the feature matching steps in the SIFT algorithm. So as to improve the proportion of the correct matching points and then reduce the time-consuming of the RANSAC algorithm. This paper first extract the adaption features of the matching points. including the nearest distance, nearest and second nearest distance, the scale and so on, and models the constructing classifier using the random forest to complete the prediction of the unknown matching points.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Hao Dou
AU  - ShaoJun Li
AU  - Xiao Sun
AU  - Tian Tian
AU  - ShuiPing Zhang
AU  - DeLie Ming
PY  - 2015/01
DA  - 2015/01
TI  - sift points for screening based on random forest in ins/sar system
BT  - Proceedings of the 2014 International Conference on Computer Science and Electronic Technology
PB  - Atlantis Press
SP  - 343
EP  - 346
SN  - 2352-538X
UR  - https://doi.org/10.2991/iccset-14.2015.76
DO  - 10.2991/iccset-14.2015.76
ID  - Dou2015/01
ER  -