Parametric Object Detection Using Estimation of Distribution Algorithms
Iván Cruz-Aceves, Jesús Guerrero-Turrubiates, Juan Manuel Sierra-Hernandez · 2017
This paper presents a novel strategy for the detection of parametric objects by using Estimation of Distribution Algorithms (EDAs). This strategy is used to detect lines, circles, and parabolas on synthetic and retinal fundus images. In the first stage, the detection problem is addressed as an optimization task by applying the trigonometric equations for each object. Moreover, a comparative analysis between EDAs and two variants of the Hough Transform is carried out in terms of computational time. The experimental results using the proposed method obtained the lowest average execution time ( 0.008 https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315154152/905dd614-092a-44c9-95b3-cae749202c91/content/inline-math4_1.tif"/> seconds) since it avoids the exhaustive search of the Hough Transform algorithm. In addition, the experimental results reveal that the proposed strategy based on EDAs to detect parametric objects reduced the computational time on about 94% on synthetic images and 90% on retinal fundus images.