Bio-inspired Clustering of Moving Objects

Author

Avila-Mora, Ivonne Maricela ; Castellanos-Sánchez, Claudio

Author_Institution

Lab. of Inf. Technol., Cinvestav, Ciudad Victoria, Mexico

Volume

4

fYear

2009

fDate

19-21 May 2009

Firstpage

58

Lastpage

62

Abstract

Nowadays several researches have implemented various techniques to solve the problem of clustering data. In this paper we present a visual bio-inspired approach to clustering based on the Conepvim model for visual perception of moving objects in the primary visual cortex (V1) of the human brain. This model uses the Gabor-like filters to detect motion, estimates the global speed, direction and trajectory. We have extended this model with a bio-inspired algorithm: the Self-Organization Maps (SOM) to define how many objects in motion there are in the sequence. Our approach is totally bio-inspired and it has been evaluated on natural sequences of images.

Keywords

Gabor filters; image motion analysis; image sequences; object detection; pattern clustering; self-adjusting systems; visual perception; Conepvim model; Gabor-like filters; bio-inspired clustering; global speed estimation; human brain; motion detection; moving objects; self-organization maps; trajectory; visual perception; Biological system modeling; Brain modeling; Clustering algorithms; Humanoid robots; Humans; Information technology; Laboratories; Legged locomotion; Motion estimation; Neurons; Gabor-like filters; Kohonen´s SOM; Visual bio-inspired model; clustering;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Systems, 2009. GCIS '09. WRI Global Congress on

Conference_Location

Xiamen

Print_ISBN

978-0-7695-3571-5

Type

conf

DOI

10.1109/GCIS.2009.445

Filename

5209339