离散不变矩算法及其在目标识别中的应用
DISCRETE MOMENT INVARIANT ALGORITHM AND ITS APPLICATION ON TARGET RECOGNITION
-
摘要: 该文分析了在离散状态下不变矩特征受比例因子变化的影响,提出了一种修正归一化中心矩的方法,保证了不变矩特征具有平移、旋转和比例因子的不变性,并将此用于目标识别,最后通过仿真实验,比较了修正前后的识别率,表明本文采用的方法是行之有效的方法。
-
关键词:
- 目标识别; 矩; 不变矩
Abstract: The paper analyses the influence of scale factor on discrete moment invariant characteristics. A modified attributive center characteristics with scale, translation and rotating invariances are proposed. The moment invariant characteristics are used in target recognition. Finally through simulation, its recognition rate with that before modification is compared. The results show this way is feasible. -
M.K.Hu,Visual pattern recognition by moment invariant,IEEE Trans.on Information Theory,1962,8(2),179-187.[2]Y.R.Wong,Scene matching with invariant moments,Computer Graphics and Image Processing,1978,8(1),16-24.[3]丁明跃等,不变矩算法研究,数据采集与处理,1992,7(1),1-9.[4]R.J.Prokop,A.P.Reeves,A survey of moment-based techniques for unoccluded object representation and recognition,Graphical,Models and Processing,1992,54(5),438-460.[5]沈请等,模式识别导论,长沙,国防科技大学出版社,1991年,第1版,110-112.[6]杜亚娟,潘泉,张洪才,图像多灰度非线性模糊增强算法研究,数据采集与处理,1999,M(2),140-143. -
计量
- 文章访问数: 2035
- HTML全文浏览量: 103
- PDF下载量: 1440
- 被引次数: 0
下载:
