文摘
|
提出了一种改进的尺度不变特征点的图像配准方法。该方法在SURF算法的基础上使用图像的熵来对匹配图像作特征检测区域选择,建立特征点筛选机制来从初步的特征检测中得到最显著的特征点,以控制特征点的数目来减少后继的计算量和算法性能。同时改进了SURF的特征描述方法的计算过程,提出了一种改进的特征描述方法。实验表明,该方法在提高算法性能的同时,明显改进了特征点的匹配率。 |
其他语种文摘
|
An improved scale invariant feature extraction method based on speed-up robust feature(SURF) algorithm was proposed.The SURF algorithm was a time-consuming computation,using fast-Hessian detector,which was a differential operator,to get the interest points.Because the statistical property of the image(like image entropy) could present the information abundance,the image could be divided into blocks,to estimate the interest point distribution using the image entropy.The block which had higher image entropy had more interest points.By selecting the image block with higher image entropy to extract the SURF interest point,the computation time can be reduced.Then the SURF feature point orientation computation method was and refined the feature descriptor was improved.The experimental results show that the algorithm is faster than SURF,and has better correct match rate. |
来源
|
红外与激光工程
,2012,41(2):537-542 【核心库】
|
关键词
|
特征点
;
SURF
;
区域选择
;
图像熵
;
图像配准
|
地址
|
1.
中国科学院研究生院, 中国科学院光电信息处理重点实验室;;辽宁省图像理解与视觉计算重点实验室, 北京, 100049
2.
中国科学院沈阳自动化研究所, 中国科学院光电信息处理重点实验室;;辽宁省图像理解与视觉计算重点实验室, 辽宁, 沈阳, 100049
|
语种
|
中文 |
ISSN
|
1007-2276 |
学科
|
自动化技术、计算机技术 |
文献收藏号
|
CSCD:4495396
|
|
1.
丁琳. 基于不变特征的CCD影像与红外影像的自动配准.
红外与激光工程,2011,40(2):350-354
|
被引
4
次
|
|
|
|
2.
邸男. 采用SIFT的末制导图像匹配技术.
红外与激光工程,2011,40(8):1589-1593
|
被引
6
次
|
|
|
|
3.
Lowe David G. Distinctive Image features from scaleinvariant keypoints.
International Journal of Computer Vision,2004,2(60):91-110
|
被引
4559
次
|
|
|
|
4.
Yan Ke. PCA-SIFT: a more distinctive representation for local image descriptors.
Proc Conf Computer Vision and Pattern Recognition,2004:511-517
|
被引
2
次
|
|
|
|
5.
Mikolajczyk K. A performance evaluation of local descriptors.
Proceedings of Computer Vision and Pattern Recognition,2003
|
被引
4
次
|
|
|
|
6.
Herbert Bay. Speeded-up robust feature.
Computer Vision and Image Understanding,2008,110:346-359
|
被引
900
次
|
|
|
|
7.
Simard P.
Boxlets: a fast convolution algorithm for signal processing and neural networks,1998
|
被引
1
次
|
|
|
|
8.
朱庆. 具有良好重复率与信息量的立体影像点特征提取方法.
电子学报,2006,34(2):205-209
|
被引
6
次
|
|
|
|
9.
.
http://www.robots.ox.ac.uk/~vgg/research/affine
|
被引
1
次
|
|
|
|
10.
.
http://web.engr.oregonstate.edu/~hes/
|
被引
1
次
|
|
|