FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDS
In this paper, a new family of shape descriptors called Free Shape Contexts (FSC) is introduced to generalize the existing 3D Shape Contexts. The FSC introduces more degrees of freedom than its predecessor by allowing the level of complexity to vary between its parts. Also, each part of the FSC has...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2015-08-01
|
Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Online Access: | http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-3-W5/41/2015/isprsannals-II-3-W5-41-2015.pdf |
id |
doaj-a45e5715b631423daba9605316ee27b3 |
---|---|
record_format |
Article |
spelling |
doaj-a45e5715b631423daba9605316ee27b32020-11-25T01:00:59ZengCopernicus PublicationsISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences2194-90422194-90502015-08-01II-3-W5414810.5194/isprsannals-II-3-W5-41-2015FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDSP. Polewski0W. Yao1M. Heurich2P. Krzystek3U. Stilla4Dept. of Geoinformatics, Munich University of Applied Sciences, 80333 Munich, GermanyDept. of Geoinformatics, Munich University of Applied Sciences, 80333 Munich, GermanyDept. for Research and Documentation, Bavarian Forest National Park, 94481 Grafenau, GermanyDept. of Geoinformatics, Munich University of Applied Sciences, 80333 Munich, GermanyPhotogrammetry and Remote Sensing, Technische Universität München, 80333 Munich, GermanyIn this paper, a new family of shape descriptors called Free Shape Contexts (FSC) is introduced to generalize the existing 3D Shape Contexts. The FSC introduces more degrees of freedom than its predecessor by allowing the level of complexity to vary between its parts. Also, each part of the FSC has an associated activity state which controls whether the part can contribute a feature value. We describe a method of evolving the FSC parameters for the purpose of creating highly discriminative features suitable for detecting specific objects in sparse point clouds. The evolutionary process is built on a genetic algorithm (GA) which optimizes the parameters with respect to cross-validated overall classification accuracy. The GA manipulates both the structure of the FSC and the activity flags, allowing it to perform an implicit feature selection alongside the structure optimization by turning off segments which do not augment the discriminative capabilities. We apply the proposed descriptor to the problem of detecting single standing dead tree trunks from ALS point clouds. The experiment, carried out on a set of 285 objects, reveals that an FSC optimized through a GA with manually tuned recombination parameters is able to attain a classification accuracy of 84.2%, yielding an increase of 4.2 pp compared to features derived from eigenvalues of the 3D covariance matrix. Also, we address the issue of automatically tuning the GA recombination metaparameters. For this purpose, a fuzzy logic controller (FLC) which dynamically adjusts the magnitude of the recombination effects is co-evolved with the FSC parameters in a two-tier evolution scheme. We find that it is possible to obtain an FLC which retains the classification accuracy of the manually tuned variant, thereby limiting the need for guessing the appropriate meta-parameter values.http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-3-W5/41/2015/isprsannals-II-3-W5-41-2015.pdf |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
P. Polewski W. Yao M. Heurich P. Krzystek U. Stilla |
spellingShingle |
P. Polewski W. Yao M. Heurich P. Krzystek U. Stilla FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDS ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
author_facet |
P. Polewski W. Yao M. Heurich P. Krzystek U. Stilla |
author_sort |
P. Polewski |
title |
FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDS |
title_short |
FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDS |
title_full |
FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDS |
title_fullStr |
FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDS |
title_full_unstemmed |
FREE SHAPE CONTEXT DESCRIPTORS OPTIMIZED WITH GENETIC ALGORITHM FOR THE DETECTION OF DEAD TREE TRUNKS IN ALS POINT CLOUDS |
title_sort |
free shape context descriptors optimized with genetic algorithm for the detection of dead tree trunks in als point clouds |
publisher |
Copernicus Publications |
series |
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
issn |
2194-9042 2194-9050 |
publishDate |
2015-08-01 |
description |
In this paper, a new family of shape descriptors called Free Shape Contexts (FSC) is introduced to generalize the existing 3D Shape
Contexts. The FSC introduces more degrees of freedom than its predecessor by allowing the level of complexity to vary between its
parts. Also, each part of the FSC has an associated activity state which controls whether the part can contribute a feature value. We
describe a method of evolving the FSC parameters for the purpose of creating highly discriminative features suitable for detecting
specific objects in sparse point clouds. The evolutionary process is built on a genetic algorithm (GA) which optimizes the parameters
with respect to cross-validated overall classification accuracy. The GA manipulates both the structure of the FSC and the activity flags,
allowing it to perform an implicit feature selection alongside the structure optimization by turning off segments which do not augment
the discriminative capabilities. We apply the proposed descriptor to the problem of detecting single standing dead tree trunks from
ALS point clouds. The experiment, carried out on a set of 285 objects, reveals that an FSC optimized through a GA with manually
tuned recombination parameters is able to attain a classification accuracy of 84.2%, yielding an increase of 4.2 pp compared to features
derived from eigenvalues of the 3D covariance matrix. Also, we address the issue of automatically tuning the GA recombination metaparameters.
For this purpose, a fuzzy logic controller (FLC) which dynamically adjusts the magnitude of the recombination effects
is co-evolved with the FSC parameters in a two-tier evolution scheme. We find that it is possible to obtain an FLC which retains the
classification accuracy of the manually tuned variant, thereby limiting the need for guessing the appropriate meta-parameter values. |
url |
http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-3-W5/41/2015/isprsannals-II-3-W5-41-2015.pdf |
work_keys_str_mv |
AT ppolewski freeshapecontextdescriptorsoptimizedwithgeneticalgorithmforthedetectionofdeadtreetrunksinalspointclouds AT wyao freeshapecontextdescriptorsoptimizedwithgeneticalgorithmforthedetectionofdeadtreetrunksinalspointclouds AT mheurich freeshapecontextdescriptorsoptimizedwithgeneticalgorithmforthedetectionofdeadtreetrunksinalspointclouds AT pkrzystek freeshapecontextdescriptorsoptimizedwithgeneticalgorithmforthedetectionofdeadtreetrunksinalspointclouds AT ustilla freeshapecontextdescriptorsoptimizedwithgeneticalgorithmforthedetectionofdeadtreetrunksinalspointclouds |
_version_ |
1725211544564793344 |