Hough transform and uncertainty handling. Application to circular object detection in ultrasound medical images

The Hough Transform (HT) has been shown to be an effective method to detect parameterized curves in images by mapping image edge points into manifolds in the parameter space and then finding "peaks" in the parameter space (P.V.C. Hough, 1962; R.O. Duda and P.E. Hart, 1972). In this paper, the application of the HT to ultrasound images of the peri-esophageal region for the localization of the aorta is discussed. The proposed algorithm has the important feature of integrating a priori knowledge as well as two types of uncertainties through the accumulation process. The first uncertainty type is related to the imprecise estimation of the gradient direction. The second, concerns the ambiguity due to the lack of a precise aorta radius determination.