Education, Science, Technology, Innovation and Life
Open Access
Sign In

Monocular Visual Odometry Based on Homogeneous SURF Feature Points

Download as PDF

DOI: 10.23977/acsat.2017.1002


Si Zengxiu, Wu Xinhua, Liu Gang

Corresponding Author

Zengxiu Si


In this paper, a monocular visual odometry method based on the uniformized SURF feature point is proposed in view of the accurate real-time location problem of the robot when moving on a flat road. By using the quadtree structure to segment the image, the feature points are evenly distributed in the image under the condition that the total number of feature points is constant. For the feature point matching, this paper builds a straight slope model based on the KNN algorithm, and uses RANSAC to quickly select the correct matching result, which further improves the accuracy of feature point matching. The experimental results show that the method of this paper has high accuracy while ensuring realtime performance.


visual odometry, SURF feature point, homogenization, feature matching

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.