Ellipse detection based on principal component analysis

A method of ellipse detection and location based on principal components analysis is proposed. According to the fact that major axis of an ellipse is the direction of principal components, the data points are projected on the coordinate frame of principal axis using the principal components analysis. A new evaluation method for elliptical profile error is proposed, and the ellipse objects are recognized according to the elliptical profile error of the data points after transform. The ellipse center coordinates are optimized by Levenberg-Marquardt algorithm, and the ones of the ellipse before transform can be obtained by the anti-transform. The method simplifies the process of ellipse detection by transforming arbitrary ellipse to standard one, and the location precision is less than 0.05 pixels when the noise variance is 0.05.