Paper Title
Detection of Incomplete Elliptical Shapes in Thinned Odia Character Image using Iterative Randomized Hough Transform

Abstract
In this paper, detection of incomplete elliptical strokes by using the Iterative Randomized Hough Transform (IRTH) method has been experimented. Most of the characters in Odia script is having a circular or elliptical structure on its top, bottom, left or right side region. For extracting such structure, the IRHT method is implemented. IRHT iteratively applies the Randomized Hough transform (RHT) to a region of interest in the image space. Region of interest is determined from the latest estimation of ellipse parameter. During the iteration process the noise pixels are gradually excluded from the region of interest. When IRHT method is implemented on Odia character image, its approximation of detecting ellipse is more than 75% of accuracy. Results demonstrate that the IRHT is robust and efficient ellipse detection method for thinned Odia character image. Keywords - Ellipse Detection, Randomized Hough transform, OpenCV, Odia Script, OCR