Extraction of correspondences in color coded pattern for the 3D reconstruction using structured light

This paper presents a methodology to extract correspondences on images of color patterns based on the De Bruijn sequence projected on static objects and surfaces with no specular highlights, decreasing in great number the problems with occlusions which appear with this pattern. The methodology aims...

Full description

Saved in:
Bibliographic Details
Published in2014 XIX Symposium on Image, Signal Processing and Artificial Vision pp. 1 - 5
Main Authors Ramirez, Luisa F., Gutierrez, Andres F., Ocampo, Jose L., Madrigal, Carlos A., Branch, John W., Restrepo, Alejandro
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2014
Subjects
3D reconstruction
Arrays
color coded pattern
Computer vision
De Bruijn sequence
Filtering
Image color analysis
Image edge detection
Image segmentation
Reflectivity
structured light
Three-dimensional displays
Online AccessGet full text

Cover

More Information
Summary:This paper presents a methodology to extract correspondences on images of color patterns based on the De Bruijn sequence projected on static objects and surfaces with no specular highlights, decreasing in great number the problems with occlusions which appear with this pattern. The methodology aims to capture and process the image, the detection of changes in intensity between stripes of the pattern for each color channel R, G, B, a selective contextual algorithm and finally the comparison of neighbors to determine the similar regions. Experimental tests demonstrate that the proposed methodology identifies large number of correspondences in the images with a low error rate.
ISSN:2329-6232
DOI:10.1109/STSIVA.2014.7010164