Constrained Tree Search for Feature Correspondence

Author: Sei-Wang Chen(Department of Information and Computer Education, National Taiwan Normal University)

Abstract：

Constrained search is a useful technique for dealing with feature corre-spondence problems frequently encountered in machine vision. In this paper, an indexing scheme, inspired by the Dewey decimal notation, is proposed for simplifying the implementation of the search algorithm. This scheme also enables the algorithm to be easily realized in hardware in view that the algor-ithm characterized by the scheme will behave like a digital counter. Through analyzing the complexity of the search algorithm, a process, that can be used to select a robust subset of features and constraints for the purpose of object recognition, is presented. Experiments demonstrate the feasibilities of both the algorithm and the process.

《Full Text》