Improving Fitness Function and Optimising Training
Data in GP for Object Detection
M. Zhang, and M. Lett. CS-TR-06-8. Computer Science, Victoria University of Wellington, New Zealand, (2006)
Abstract
the refinement of a fitness function and the
optimisation of training data in genetic programming
(GP) for object detection particularly object
localisation problems. The fitness function uses the
weighted F-measure of a genetic program and considers
the localisation fitness values of the detected object
locations. To investigate the training data with this
fitness function, we categorise the training data into
four types: exact centre, close to centre, include
centre, and background. The approach is examined and
compared with an existing fitness function on three
object detection problems of increasing difficulty. The
results suggest that the new fitness function
outperforms the old one by producing far fewer false
alarms and spending much less training time and that
the first two types of the training examples contain
most of the useful information for object detection.
The results also suggest that the complete background
type of data can be removed from the training set.
%0 Report
%1 vuw-CS-TR-06-8
%A Zhang, Mengjie
%A Lett, Malcolm
%C New Zealand
%D 2006
%K Fitness algorithms, classification, detection, examples, function, genetic localisation object programming, recognition, training
%N CS-TR-06-8
%T Improving Fitness Function and Optimising Training
Data in GP for Object Detection
%U http://www.mcs.vuw.ac.nz/comp/Publications/CS-TR-06-8.abs.html
%X the refinement of a fitness function and the
optimisation of training data in genetic programming
(GP) for object detection particularly object
localisation problems. The fitness function uses the
weighted F-measure of a genetic program and considers
the localisation fitness values of the detected object
locations. To investigate the training data with this
fitness function, we categorise the training data into
four types: exact centre, close to centre, include
centre, and background. The approach is examined and
compared with an existing fitness function on three
object detection problems of increasing difficulty. The
results suggest that the new fitness function
outperforms the old one by producing far fewer false
alarms and spending much less training time and that
the first two types of the training examples contain
most of the useful information for object detection.
The results also suggest that the complete background
type of data can be removed from the training set.
@techreport{vuw-CS-TR-06-8,
abstract = {the refinement of a fitness function and the
optimisation of training data in genetic programming
(GP) for object detection particularly object
localisation problems. The fitness function uses the
weighted F-measure of a genetic program and considers
the localisation fitness values of the detected object
locations. To investigate the training data with this
fitness function, we categorise the training data into
four types: exact centre, close to centre, include
centre, and background. The approach is examined and
compared with an existing fitness function on three
object detection problems of increasing difficulty. The
results suggest that the new fitness function
outperforms the old one by producing far fewer false
alarms and spending much less training time and that
the first two types of the training examples contain
most of the useful information for object detection.
The results also suggest that the complete background
type of data can be removed from the training set.},
added-at = {2008-06-19T17:46:40.000+0200},
address = {New Zealand},
author = {Zhang, Mengjie and Lett, Malcolm},
biburl = {https://www.bibsonomy.org/bibtex/2c781de5e447789ffa3b2ced6e0f1e927/brazovayeye},
institution = {Computer Science, Victoria University of Wellington},
interhash = {129b1a8168600d2a9b636cde698599df},
intrahash = {c781de5e447789ffa3b2ced6e0f1e927},
keywords = {Fitness algorithms, classification, detection, examples, function, genetic localisation object programming, recognition, training},
number = {CS-TR-06-8},
timestamp = {2008-06-19T17:55:37.000+0200},
title = {Improving Fitness Function and Optimising Training
Data in {GP} for Object Detection},
url = {http://www.mcs.vuw.ac.nz/comp/Publications/CS-TR-06-8.abs.html},
year = 2006
}