Analysis of shape alignment using Euclidean and Manhattan distance metrics

Shape alignment using Euclidean distance metric is an important tool in pattern recognition and medical imaging applications due to its computational efficiency. However, Euclidian distance metric is not very robust and is sensitive to the outliers. This paper describes the use of Euclidean and Manhattan distance metric to analyze robustness and resistivity of the shape with respect to their orientation, position and size in the 2D plane. The simulated data set illustrate the properties of different distance metric for the robust shape alignment process.