An Improved PCA-SIFT Algorithm for Image Mosaics

YANG Bing-kun; CHENG Shu-ying; ZHENG Qian-ying

Chinese

您当前的位置：

首页 >

文章列表页 >

An Improved PCA-SIFT Algorithm for Image Mosaics

更新时间：2026-09-04

- An Improved PCA-SIFT Algorithm for Image Mosaics
- Vol. 35, Issue 12, Pages: 70-75(2018)
- 作者机构：
  
  福州大学物理与信息工程学院,福建,福州,350108
- 作者简介：
- 基金信息：
- DOI：
  CLC：
- Published：2018
- 稿件说明：
移动端阅览
杨炳坤, 程树英, 郑茜颖. 一种面向图像拼接的改进PCA-SIFT算法[J]. 微电子学与计算机, 2018, 35(12): 70-75.

YANG Bing-kun, CHENG Shu-ying, ZHENG Qian-ying. An Improved PCA-SIFT Algorithm for Image Mosaics[J]. 2018, 35(12): 70-75.
杨炳坤, 程树英, 郑茜颖. 一种面向图像拼接的改进PCA-SIFT算法[J]. 微电子学与计算机, 2018, 35(12): 70-75. DOI：

YANG Bing-kun, CHENG Shu-ying, ZHENG Qian-ying. An Improved PCA-SIFT Algorithm for Image Mosaics[J]. 2018, 35(12): 70-75. DOI：

摘要

针对图像拼接中尺度不变特征变换(SIFT)算法没有充分考虑特征点的分布情况且计算复杂、耗时较长等问题

提出了一种基于改进的PCA-SIFT算法.该算法首先在空间极值点检测阶段引入改进的非极大值抑制法对初始特征点进行优选

得到分布更加均匀的特征点集; 然后在构建描述符阶段基于圆形领域提取64维SIFT描述符

并使用主成分分析(PCA)法对描述符进一步降维

减少描述符的数据复杂度; 最后在特征匹配阶段引入基于K-D树的BBF搜索策略

采用随机抽样一致性(RANSAC)算法剔除误匹配点

从而提高了匹配速度与匹配精度.在10组图像拼接实验中

本文算法的拼接速度是传统SIFT算法的1.6~2.2倍.实验结果表明

本文算法具有较高的精度、较好的鲁棒性

较强的实时性.

Abstract

Aiming at the problem that the SIFT algorithm does not fully consider the distribution of feature points in the image splicing and the calculation is complex and takes a long time

an improved PCA-SIFT algorithm is proposed. The algorithm firstly introduces an improved non-maximum suppression method to optimize the initial feature points so as to obtain a more even distribution of feature point sets. Then the 64-dimension SIFT descriptor is extracted based on the circular neighborhood

and the descriptor is further reduced using PCA to reduce the data complexity of the descriptor. Finally

the BBF search strategy based on K-D tree was introduced. The RANSAC was used to eliminate the false matching points

which improved the matching speed and matching accuracy. In the 10 sets of image stitching experiments

the stitching speed of this algorithm is 1.6~2.2 times that of the traditional SIFT algorithm. Experimental results show that the proposed algorithm has higher accuracy

better robustness

and stronger real-time performance.

关键词

Keywords

references

Mikolajczyk K, Schmid C. A performance evaluation of local descriptors[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(10): 1615-1630. DOI: 1615

Luan H, Law N F. A novel dual-threshold SIFT-based copy-move forgery detection[C]// Signal and Information Processing Association Summit and Conference. Kuala Lumpur, Malaysia, IEEE, 2017: 1-5.

Dong Q, Liu J. Seamline determination based on PKGC segmentation for remote sensing image mosaicking[J]. Sensors, 2017, 17(8):1721. DOI: 1721

Lowe D G. Distinctive image features from scale invariant key points[J]. International Journal of Computer Vision, 2004, 60(2): 91-110. DOI: 91

张勇, 王志锋, 马文.基于改进SIFT特征点匹配的图像拼接算法研究[J].微电子学与计算机, 2016, 33(3):60-64. DOI: 60

翟优, 曾峦, 熊伟.不同局部邻域划分加速鲁棒特征描述符的性能分析[J].光学精密工程, 2013, 21(9):2395-2404. DOI: 2395

蒋波, 翟旭平.基于PCA-SIFT特征匹配的图像拼接算法[J].计算机应用, 2016, 36(2):143-145. DOI: 143

Yu Qing H E, Xue W, Wang S Y, et al. Modified SIFT descriptor and key-point matching for fast and robust image mosaic[J].北京理工大学学报(英文版), 2016, 25(4):562-570. DOI: 562

杨恒, 王庆.一种新的局部不变特征检测和描述算法[J].计算机学报, 2010, 33(5):935-944. DOI: 935

Zhang Z, Kuang L, Han X, et al. An improved RANSAC algorithm for image mosaic[J]. Computer Measurement % Control, 2014, 22(6): 1856-1858. DOI: 1856

李玉峰, 李广泽, 谷绍湖, 等.基于区域分块与尺度不变特征变换的图像拼接算法[J].光学精密工程, 2016, 24(5): 1197-1205. DOI: 1197

蒋晓玲, 章慧.基于多特征融合的激光医学图像拼接[J].激光杂志, 2015, 36(12):55-58. DOI: 55

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A review of large model development: technological evolution from BERT to Deepseek

Research on assembly action recognition based on MSS-YOLO-MPHands fusion architecture

Design of ultra-low power real-time clock circuits

Lightweight countermeasures against Spectre attacks on the branch prediction unit

Aerial vehicle detection algorithm based on multi-scale fusion and small-object enhancement

Related Author

No data

Related Institution

No data

AI问答

Postal code：100079
Tel：（010）53879206 Email：tmw@bjxintong.com.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09082226号-64 京公网安备11010602201714号
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰