一种基于熵的距离像模板库建立方法

基于模板匹配法的目标识别,其性能好坏在很大程度上取决于模板库的质量。文中提出了一种基于最小熵添加模板库的建库方法,能够从信噪比、完备性两方面来保证模板库的质量,而且在应用中还具有易于更新维护的优点。多批实测飞机数据的实验结果表明,该方法能够获得稳定而可靠的识别。

Method of Setting up Template Base of Range Profiles Based on Entropy

The performance of a target recognition method based on template matching depends on the quality of the template database to a large extent.This paper introduces a method of sestablishing an addible template database of range profiles based on the minimum entropy rule.The new method guarantees the quality of the template base from two sides:the signal-to-noise ratio,and the completeness of the database.The new database is easy to update and maintain.Experimental result grounded on a few groups of real data demonstrates that we can get a stable and reliable recognition with the new method.