Coarse-to-fine search technique to detect circles in images

Detection of patterns in images is an important high-level task in automate d manufacturing using machine vision. Straight lines, circles and ellipses are considered to be the basic building blocks of a large number of pattern s occurring in real-world images. Real-world images frequently contain nois e and occlusions resulting in discontinuous patterns in noisy images. The H ough transform can be used to detect parametric patterns, such as straight lines and circles, embedded in noisy images. The large amount of storage an d computing power required by the Hough transform presents a problem in rea l-time applications. The aim of this paper is to-propose an efficient coarse-to-fine search tech nique to reduce the storage and computing time in detecting circles in an i mage, Variable-sized images and accumulator arrays are used to-reduce the c omputing and storage requirements of the Hough transform. The accuracy and the rate of convergence of the parameters at different iterations of the al gorithm are presented The results demonstrate that the coarse-to-fine searc h strategy is very suitable for detecting circles in real-time environments having time constraints.