Array Processing and Fast Optimization Algorithms for Distorted Circular Contour Retrieval
- PDF / 2,939,943 Bytes
- 13 Pages / 600.03 x 792 pts Page_size
- 103 Downloads / 176 Views
Research Article Array Processing and Fast Optimization Algorithms for Distorted Circular Contour Retrieval Julien Marot and Salah Bourennane GSM, Institut Fresnel, CNRS-UMR 6133, Ecole Centrale Marseille, Universit´e Aix-Marseille III, D.U. de Saint J´erˆome, 13397 Marseille Cedex 20, France Received 19 July 2006; Revised 20 December 2006; Accepted 17 February 2007 Recommended by Wilfried Philips A specific formalism for virtual signal generation permits to transpose an image processing problem to an array processing problem. The existing method for straight-line characterization relies on the estimation of orientations and offsets of expected lines. This estimation is performed thanks to a subspace-based algorithm called subspace-based line detection (SLIDE). In this paper, we propose to retrieve circular and nearly circular contours in images. We estimate the radius of circles and we extend the estimation of circles to the retrieval of circular-like distorted contours. For this purpose we develop a new model for virtual signal generation; we simulate a circular antenna, so that a high-resolution method can be employed for radius estimation. An optimization method permits to extend circle fitting to the segmentation of objects which have any shape. We evaluate the performances of the proposed methods, on hand-made and real-world images, and we compare them with generalized Hough transform (GHT) and gradient vector flow (GVF). Copyright © 2007 J. Marot and S. Bourennane. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1.
INTRODUCTION
Circular features in digital images are sought very often in digital image processing. An image containing one or several contours is composed of black pixels with value “1” which represent the contours, over a white background with pixel value “0.” Circle fitting, in particular, is faced in several application fields such as quality inspection for food industry and mechanical parts, fitting particle trajectories [1, 2]. Circle fitting also has applications in microwave engineering, and ball detection in robotic vision systems [3]. Several methods have been proposed for solving this problem using, among others, the generalized Hough transform (GHT) [4, 5], array processing methods [6, 7], contour-based snakes methods [8, 9]. The formalism proposed by Aghajan [6] permits to detect circular or elliptic contours. The coordinates of the center of a circle are estimated by an array processing method [6] that works on virtual signals generated from the image. Each row or column of the image is associated with a sensor of a linear antenna. In this paper, we propose a new approach which employs a circular antenna for the estimation of the radius of a circle, and we propose to adapt an optimization method to retrieve
the distortions between any nearly-circular contour and a circle. We adopted a similar strategy in [10], in the
Data Loading...
