Title :
Two-dimensional discriminative filters for image template detection
Author :
Mendonça, Alexandre P. ; da Silva, Eduardo A B
Author_Institution :
Departamento de Engenharia Eletrica, Instituto Militar de Engenharia, Rio de Janeiro, Brazil
fDate :
6/23/1905 12:00:00 AM
Abstract :
The image template detection is usually a very important halfway step for a computational vision algorithm. In general, the basic idea of this type of algorithm is to receive an input image and generate a set of lines, borders, edges and other well known geometrical forms as outputs. Ben-Arie and Rao (1993) have proposed a linear filter that, given a template, maximizes the energy concentration in a single sample of its output. We propose a two-dimensional generalization of this method, formulating the template matching problem as a multivariable optimization. Three different objective functions are investigated. Simulation results show that the proposed method performs well on real images
Keywords :
computer vision; image matching; optimisation; two-dimensional digital filters; 2D discriminative filters; borders; computational vision algorithm; edges; energy concentration maximization; geometrical forms; image template detection; input image; linear filter; lines; multivariable optimization; objective functions; real images; simulation results; template matching; two-dimensional discriminative filters; two-dimensional generalization method; Computer vision; Convolution; Detectors; Fourier transforms; Image edge detection; Image generation; Matched filters; Nonlinear equations; Nonlinear filters; Optimization methods;
Conference_Titel :
Image Processing, 2001. Proceedings. 2001 International Conference on
Conference_Location :
Thessaloniki
Print_ISBN :
0-7803-6725-1
DOI :
10.1109/ICIP.2001.958210
