Full text: Proceedings of the Symposium on Global and Environmental Monitoring (Part 1)

724 
RANGE IMAGE SEGMENTATION AND OBJECT RECOGNITION 
USING MARKOV RANDOM FIELD MODEL 
N.N. Pasechny, V.A. Stephanov, V.M. Lisitsyn 
ABSTRACT 
The problem of remote sensing and automatic object recognition using three-dimensional range data is 
considered. It is assumed that range image is obtained by a laser radar which determines the distance from 
the sensor to the object in every pixel. Each object to be recognized is a solid of arbitrary shape and composed 
with a small number of primitive surfaces. The sensed range image is then divided into homogenous regions 
which approximate different objects surfaces. The range image digital processing is based on a two-level 
hierarchical Markov random field model and sequential segmentation of the range data within a formal 
statistical estimation framework. It is shown that the range image segmentation algorithm proposed can be 
realised in the sistolic array processor. Theoretical and experimental characteristics of proposed segmentation 
algorithm are given with discussion of practical applications.
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.