Anais do X SIBGRAPI (1997) #ART21


An Iterative Algorithm for Segmenting Lanes in Gel Electrophoresis Images

Alexei M. C. Machado, Departamento de Ciência da Computação, UFMG
Mario F. M. Campos, Departamento de Ciência da Computação, UFMG
Ari M. Siqueira, Departamento de Bioquímica, UFMG
Osvaldo S. F. de Carvalho, Departamento de Ciência da Computação, UFMG

Abstract. In this paper we discuss the application of spatial-domain filters for solving the problem of automatic lane detection on gel electrophoresis computer images. The problem can be posed as the determination of the number, location and orientation of lanes on the image, based on the analysis of their gray-level intensities. A novel iterative filtering algorithm is proposed based on the periodicity of image projection pattern. The algorithm is compared with clustering and frequency-domain filtering methods in terms of their effectiveness and efficiency from which a trade-off solution is presented.

[ full version | contact authors ]