In this paper a method is proposed to discriminate
natural and manmade scenes of similar depth. Increase in
image depth leads to increase in roughness in manmade
scenes; on the contrary natural scenes exhibit smooth behavior
at higher image depth. This particular arrangement of pixels
in scene structure can be well explained by local texture
information in a pixel and its neighborhood. Our proposed
method analyses local texture information of a scene image
using texture unit matrix. For final classification we have
used unsupervised learning using Self Organizing Map
(SOM). This technique is useful for online classification due
to very less computational complexity.
%0 Generic
%1 rath2013texture
%A Rath, Dr. N.P.
%A Mishra, Bandana
%B 2013 Mobile Communication - II
%D 2013
%E Shankaranarayanan, Dr.
%I ACEEE (A Computer division of IDES)
%K (SOM) Image-depthTexture Map Organizing Self Texture image matrix scene unit
%T Texture Unit based Approach to Discriminate Manmade Scenes from Natural Scenes
%U http://searchdl.org/public/book_series/LSCS/3/78.pdf
%X In this paper a method is proposed to discriminate
natural and manmade scenes of similar depth. Increase in
image depth leads to increase in roughness in manmade
scenes; on the contrary natural scenes exhibit smooth behavior
at higher image depth. This particular arrangement of pixels
in scene structure can be well explained by local texture
information in a pixel and its neighborhood. Our proposed
method analyses local texture information of a scene image
using texture unit matrix. For final classification we have
used unsupervised learning using Self Organizing Map
(SOM). This technique is useful for online classification due
to very less computational complexity.
@conference{rath2013texture,
abstract = { In this paper a method is proposed to discriminate
natural and manmade scenes of similar depth. Increase in
image depth leads to increase in roughness in manmade
scenes; on the contrary natural scenes exhibit smooth behavior
at higher image depth. This particular arrangement of pixels
in scene structure can be well explained by local texture
information in a pixel and its neighborhood. Our proposed
method analyses local texture information of a scene image
using texture unit matrix. For final classification we have
used unsupervised learning using Self Organizing Map
(SOM). This technique is useful for online classification due
to very less computational complexity.},
added-at = {2014-02-05T06:23:28.000+0100},
author = {Rath, Dr. N.P. and Mishra, Bandana},
biburl = {https://www.bibsonomy.org/bibtex/280ca10251bc8e4e8d90627bef83355ea/idescitation},
booktitle = {2013 Mobile Communication - II},
editor = {Shankaranarayanan, Dr.},
interhash = {79bcbb7c28cd90343b18756dfebb8073},
intrahash = {80ca10251bc8e4e8d90627bef83355ea},
keywords = {(SOM) Image-depthTexture Map Organizing Self Texture image matrix scene unit},
organization = {Institute of Doctors Engineers and Scientists},
publisher = {ACEEE (A Computer division of IDES)},
timestamp = {2014-02-05T06:23:28.000+0100},
title = {Texture Unit based Approach to Discriminate Manmade Scenes from Natural Scenes},
url = {http://searchdl.org/public/book_series/LSCS/3/78.pdf},
year = 2013
}