Landmark-Based Shape Context for Handwritten Digit Recognition

Abstract

This chapter introduces landmark-based shape context. Standard shape context computation samples at regular interval on the contour of an object. Corner points of an object being landmarks on the contour; set of corner points is a good descriptor of shape. Landmark-based shape context is computed by sampling a reduced set of points based on such landmarks. In this chapter, an approach to recognizing handwritten digits based on shape similarity computed through landmark-based shape context is presented. Shape context based object recognition being an iterative process; the reduction in the number of sample points provides a basis for faster recognition.