Towards the introduction of human perception in a natural scene classification system

G. Nathalie1, L.B. Herve1, H. Jeanny1, G.-D. Anne2
1LIS, Grenoble, France
2CLIPS, Grenoble, France

Tóm tắt

We develop a method to optimize a machine-based semantic categorization of natural images according to human perception. First, the categories are determined through a psychophysical experiment. The similarity matrices obtained from the human responses are analyzed by a multidimensional scaling technique called curvilinear component analysis (CCA). The same is done with an automatic image indexing system based on similarities between the outputs of Gabor filters applied to the images. Then we show that, by using the human categorization to balance the filter outputs, the system's performance may be significantly improved.

Từ khóa

#Humans #Layout #Image databases #Gabor filters #Indexing #Image retrieval #Psychology #Multidimensional systems #Multimedia databases #Navigation

Tài liệu tham khảo

rogowitz, 1998, Perceptual image similarity experiments, Proc SPIE Human Vision Electron Imaging, 3299 10.1109/72.554199 10.1109/ICIP.2001.958942 10.1109/34.790428 hérault, 2001, « De la rétine biologique aux circuit neuromorphiques », Trait IC2 Les Systemes de Vision jones, 1987, « An evaluation of the two-dimensional Gabor filter model of simple receptive fields in cat striate cortex, J Neurophysiol, 58, 1233, 10.1152/jn.1987.58.6.1233 de valois, 1988, « Spatial Vision » 10.1016/S0031-3203(98)00028-4 10.1016/S0167-8655(00)00074-X del bimbo, 1999, Visual Information Retrieval 10.1109/CAIVD.1998.646032 10.1109/ICPR.1994.576325 10.1109/IVL.1997.629723 10.1016/S0031-3203(98)00079-X johansson, 2000, A Survey on: Contents Based Search in Image Databases 10.1109/ICPR.1996.546971 hérault, 1997, Scene Categorisation by Curvilinear Component Analysis of Low Frequecy Spectra, ESANN'97