遥感信息
遙感信息
요감신식
2015年
3期
99-103
,共5页
多源遥感影像%影像配准%点特征%互信息%配准精度
多源遙感影像%影像配準%點特徵%互信息%配準精度
다원요감영상%영상배준%점특정%호신식%배준정도
multi-source remote sensing image%image registration%point feature%mutual information%registration accuracy
针对多源遥感影像间几何形变和辐射差异所引起的配准困难问题，该文提出一种利用点特征和互信息的影像配准方法。该方法首先利用 Harris 算子，采用分块的提取策略，在参考影像上检测出大量、分布均匀的特征点，然后以互信息作为相似性，并采用双向匹配的策略在输入影像上进行同名点识别，同时进行粗差剔除，最后根据获取的同名点建立不规则三角网，对输入影像进行逐三角网纠正，最终实现影像的配准。通过选择两种不同类型的多源遥感影像进行试验，结果表明该方法具有较强的适应性并且配准精度较高。
針對多源遙感影像間幾何形變和輻射差異所引起的配準睏難問題，該文提齣一種利用點特徵和互信息的影像配準方法。該方法首先利用 Harris 算子，採用分塊的提取策略，在參攷影像上檢測齣大量、分佈均勻的特徵點，然後以互信息作為相似性，併採用雙嚮匹配的策略在輸入影像上進行同名點識彆，同時進行粗差剔除，最後根據穫取的同名點建立不規則三角網，對輸入影像進行逐三角網糾正，最終實現影像的配準。通過選擇兩種不同類型的多源遙感影像進行試驗，結果錶明該方法具有較彊的適應性併且配準精度較高。
침대다원요감영상간궤하형변화복사차이소인기적배준곤난문제，해문제출일충이용점특정화호신식적영상배준방법。해방법수선이용 Harris 산자，채용분괴적제취책략，재삼고영상상검측출대량、분포균균적특정점，연후이호신식작위상사성，병채용쌍향필배적책략재수입영상상진행동명점식별，동시진행조차척제，최후근거획취적동명점건립불규칙삼각망，대수입영상진행축삼각망규정，최종실현영상적배준。통과선택량충불동류형적다원요감영상진행시험，결과표명해방법구유교강적괄응성병차배준정도교고。
In order to address the problem of the registration of multi-source remote sensing images with geometric distortions and intensity differences,this paper proposes a registration method using point features and mutual information for such images.The Harris detector is firstly applied to extract a large amount of evenly distributed feature points in the reference image using a block-based scheme.Then,mutual information is employed as the similarity metric to detect the correspondences using a bidirectional matching technique,and the correspondences with large errors are removed.Finally,image registration is achieved through a piecewise linear transform,which constructs the triangulated irregular networks with the correspondences, and rectifies the input image on each triangle region one by one.The proposed method is evaluated with two different types of multi-source remote sensing images.The experimental results demonstrate that the proposed method is suitable for multi-source remote sensing images,and achieves high registration accuracy.
