.png){kind=icon}
.png){kind=icon}
Return
Detecting background and foreground with a laser array system
.png){kind=icon}
PRE
AI
DOI：10.1016/j.measurement.2014.12.004.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Ranging Laser Sensors are typically used as method for extracting tridimensional information on the environment. This information takes the sensor's location as reference, obtaining a points cloud which corresponds to the relative distances of the surrounding surfaces, which are related to the objects around the reference. However, in outdoor scenarios, a great diversity of non-controllable events makes difficult to analyze the dynamic objects. Consequently, an efficient criterion to discriminate which information corresponds to fixed zones and which information corresponds to dynamic zones is complicated to define, due to a wide variety of involved situations. This work presents a stochastic approach for modeling a tridimensional environment. This approach is used to estimate the foreground and background using information obtained from a Laser Imaging Detection and Ranging (LiDAR) sensor. It represents an extension of the background subtraction approach, using a Mixture of Gaussians (MoG) on the image sequences. The environment is visualized using a laser array system, which acquires the distance of every surrounding surface in a radius of 120 m. The obtained data are set into an array of parametric form and spherical discretized notation to model every interval as a Mixture of Gaussians. The Expectation Maximization (EM) technique is used to estimate the parameters of every MoG. Finally, this new algorithm allows characterizing objects according to the distance between their surfaces and the laser array system. Furthermore, as a consequence of the technique itself, it also eliminates noise and obtains a trust level of the implemented method. (C) 2014 Elsevier Ltd. All rights reserved.
Keywords:
Background estimation
Laser
LiDAR
Mixture of Gaussians (MoG)
Expectation Maximization (EM)
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
5.6
Papers:
2.0W
Citations:
5.4W
View more organizations