- author = "Cynthia Beatriz {Perez Castro}",
- title = "Evolutionary Computation for Image Content Description applied to image Segmentation and Object Recognition",
- title_sp = "Computo evolutivo como enfoque en la descripcion del contenido de la imagen aplicado a la segmentacion y el reconocimiento de objetos.",
- school = "Computer Science Department, Centro de Investigacion Cientifica y de Educacion Superior de Ensenada",
- year = "2010",
- type = "Doctor en Ciencias de la Computacion",
- address = "Tijuana-Ensenada, Ensenada, Baja, Mexico",
- month = jul,
- keywords = "genetic algorithms, genetic programming, Segmentation, Statistical Descriptors, Co-ocurrence Matrix (GLCM), Local Descriptors, F-Measure, Object Recognition",
-
URL = "
http://biblioteca.cicese.mx/catalogo/tesis/ficha.php?id=18586",
-
URL = "
http://biblioteca.cicese.mx/catalogos/Ftesis/PerezCastro_thesis.pdf",
- size = "126 pages",
-
abstract = "The ability to analyse and describe the image content
from real or designed scenarios is a challenge and an
attractive task for computer vision. This work presents
two evolutionary algorithms which analyse the image
content for two particular high level tasks such as:
image segmentation and object recognition. On the one
hand, the image content is analysed using statistical
descriptors in a Gray Level Co-ocurrence Matrix (GLCM)
in order to achieve good image segmentations. On the
other hand, new local descriptor operators are proposed
using genetic programming. These operators describe the
image content in order to recognise objects localised
within indoor and outdoor scenarios presenting
different image transformations. First, we present our
EvoSeg algorithm, which uses knowledge derived from
texture analysis to identify how many homogeneous
regions exist in the scene without a priori
information. EvoSeg uses texture features derived from
the GLMC and optimises a fitness measure, based on the
minimum variance criteria, using a hierarchical GA.
Later, we include interaction within the EvoSeg
optimization process obtaining a new algorithm named
EvoSeg. This algorithm complements the chosen texture
information with direct human interaction in the
evolutionary optimisation process. Interactive
evolution helps to improve results by allowing the
algorithm to adapt using the new external information
based on user evaluation. Finally, we present
experimental results using a standard database used for
image segmentation from the USC-SIPI (Signal and Image
Processing Institute).
Second, we describe a genetic programming methodology that synthesises mathematical expressions that are used to improve a well known local descriptor algorithm. It follows the idea that object recognition in the cerebral cortex of primates makes use of features of intermediate complexity that are largely invariant to change in scale, location, and illumination. These local features have been previously designed by human experts using traditional representations that have a clear, preferably mathematically, well-founded definition. However, it is not clear that these same representations are implemented by the natural system with the same representation. Hence, the possibility to design novel operators through genetic programming represents an open research avenue where the combinatorial search of evolutionary algorithms can largely exceed the ability of human experts. Hence, we provide evidence that genetic programming is able to design new features that enhance the overall performance of the best available local descriptor. Experimental results confirm the validity of the proposed approach using a widely accept testbed and an object recognition application for indoor and outdoor scenarios using our best descriptor RDGP2.",
-
notes = "In Spanish.
Comite de tesis: Olague Caballero Gustavo; Sucar Succar Luis Enrique; Kelly Martinez Rafael de Jesus; Chernykh Andrey; Torres Rodriguez Jorge",
