Particle Filter with Temporal Smoothing for Mobile Robot Vision-Based Localization

2009 ◽  
pp. 167-180 ◽  
Author(s):  
Walter Nisticò ◽  
Matthias Hebbel
Keyword(s):  
Mobile Robot ◽  
Particle Filter ◽  
Robot Vision ◽  
Temporal Smoothing ◽  
Mobile Robot Vision ◽  
Vision Based Localization

Improved Hough Transform Applying in Mobile Robot Vision Navigation

2012 ◽  
Vol 232 ◽  
pp. 408-413
Author(s):  
Yin Ping Jiang ◽  
Xian Xian Zhang ◽  
Xiao Peng Fu
Keyword(s):  
Mobile Robot ◽  
Hough Transform ◽  
Robot Vision ◽  
Geometric Figure ◽  
Vision Navigation ◽  
Line Extraction ◽  
Local Threshold ◽  
Global Threshold ◽  
Image Quality Reduction ◽  
Mobile Robot Vision

This paper mainly discusses that in mobile robot vision navigation system, by using the improved Hough transform, we can improve the accuracy of line extraction and therefore avoid the image quality reduction caused by noise points. Considering the limitations of the standard Hough transform, we come up with a method with which we will accumulates the H (ρ, θ) through distributing the increment value, set a global threshold to shun the pointless measurements, eliminate the false lines by comparing θ difference between tow arbitrary lines, find the peaks by using rectangle window, and set a local threshold to eliminate false peaks. In this way, we can gain a method superior to the standard Hough transform which works better in extracting lines in application. The experiments show that this method can not only extract line features of geometric figure effectively in brief background, but also eliminate the iterative lines efficiently.

Still Corresponding Points Extraction Using a Moving Monocular Camera with a Motion Sensor

2015 ◽  
Vol 19 (2) ◽  
pp. 319-329
Author(s):  
Toshihiro Akamatsu ◽  
◽  
Fangyan Dong ◽  
Kaoru Hirota
Keyword(s):  
Computer Graphics ◽  
Mobile Robot ◽  
Moving Objects ◽  
Robot Vision ◽  
Motion Sensor ◽  
3D Measurement ◽  
Sensor Noise ◽  
Monocular Camera ◽  
Mobile Robot Vision ◽  
Extraction Experiment

The method of extracting still corresponding points proposed in this paper uses a moving monocular camera connected to a 6-axis motion sensor. It classifies corresponding points between two consecutive frames containing still/moving objects and chooses corresponding points appropriate for 3D measurement. Experiments are done extracting still corresponding points with 2 scenes from original computer graphics images. Results for scene 1 show that accuracy is 0.98, precision 0.96, and recall 1.00. Robustness against sensor noise is confirmed. Extraction experiment results with real scenes show that accuracy is 0.86, precision 0.88, and recall 0.94. We plan to include the proposed method in 3D measurement with real images containing still/moving objects and to apply it to obstacles avoidance for vehicles and to mobile robot vision systems.

Sign in / Sign up

Export Citation Format

Share Document