Abstract
In this paper we study the problem of feature point matching via a technique well known in geometry, leading to a new gradient algorithm in machine vision. The procedure does not require one to solve explicit correspondences between feature points but relies on a specific form of "spatial averaging." In this sense the estimation procedure is robust. In this paper we describe the algorithm and report simulation results.
Original language | English |
---|---|
Pages (from-to) | 1671-1676 |
Number of pages | 6 |
Journal | IEEE Transactions on Automatic Control |
Volume | 41 |
Issue number | 11 |
DOIs | |
State | Published - 1996 |