基于FAST角点和仿射改进的随机蕨丛的单目视觉实时匹配算法

doi:10.3724/SP.J.1218.2014.00271

摘要
图/表
参考文献
相关文章 (1)

全文: PDF (975 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
罗元, 付瑜, 张毅. 基于FAST角点和仿射改进的随机蕨丛的单目视觉实时匹配算法[J]. 机器人, 2014, 36(3): 271-278.DOI: 10.3724/SP.J.1218.2014.00271. LUO Yuan, FU Yu, ZHANG Yi. A Monocular-vision Real-time Matching Algorithm Based on FAST Corners and Affine-improved Random Ferns. ROBOT, 2014, 36(3): 271-278. DOI: 10.3724/SP.J.1218.2014.00271.

摘要

针对传统配准方法在自然特征粗匹配过程中在线匹配速度较慢、识别率较低的问题，提出一种基于FAST 角点与仿射改进的适用于单目视觉实时定位的随机蕨丛匹配方案．该方案基于随机蕨半朴素贝叶斯非层次结构分类模型，突破了传统匹配方案中离线过程与在线过程的对称处理架构．采用FAST 角点作为环境自然特征提升检测速度，进而改进随机蕨丛离线过程中稳定点选取、训练片元生成的仿射策略，提升识别率、缩短离线训练时间，最后收缩随机蕨丛规模，缩短在线匹配时耗．室内、外实时匹配定位实验结果表明该方案满足单目视觉定位的实时性、识别率需求.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS

关键词 ：随机蕨丛, FAST角点, 仿射变换, 实时性, 识别率

Abstract：

For classic calibration methods, there exist problems of low online matching speed and recognition rate in rough matching of natural features. To solve the problem, a random ferns matching algorithm based on FAST (Features from Accelerated Segment Test) corners and affine-improvement is proposed to realize monocular-vision real-time localization.This algorithm is based on a random fern semi-Bayes nonhierarchic classification model, which breaks the symmetrical framework of on-line and off-line processes in classic matching methods. It adopts FAST corners as the environment natural features to accelerate online detection, improves affine strategy for stable point set selection and training fragment generation in random ferns off-line process to improve recognition rate and reduce off-line training time, and scales back the size of random ferns to reduce time consumption of online matching. Indoor and outdoor matching and localization experiments show that the proposed algorithm meets requirements of real-timeness and recognition rate for monocular-vision localization.

Key words： random ferns FAST (Features from Accelerated Segment Test) corner affine transformation real-time recognition rate

收稿日期: 2013-05-16 录用日期: 2014-02-20

ZTFLH:

TP242.2

基金资助:

科技部国际合作项目（2010DFA12160）；国家自然科学基金资助项目（51075420）

通讯作者: 李睿，lirui1860@126.com E-mail: lirui1860@126.com

作者简介: 罗元（1972- ），女，博士，教授，硕士生导师．研究领域：机器视觉，智能信号处理．
付瑜（1989- ），女，硕士生．研究领域：视觉导航．
张毅（1965- ），男，博士，教授，博士生导师．研究领域：机器人及应用，数据融合，信息无障碍技术．

链接本文:

https://robot.sia.cn/CN/10.3724/SP.J.1218.2014.00271 或 https://robot.sia.cn/CN/Y2014/V36/I3/271

[1] Lepetit V, Fua P. Keypoint recognition using randomized trees[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(9): 1465-1479.

[2] Ozuysal M, Fua P, Lepetit V. FAST keypoint recognition in ten lines of code[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, USA: IEEE, 2007: 1-8.

[3] Calonder M, Lepetit V, Fua P. Keypoint signatures for fast learning and recognition[C]//European Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2008: 58-71.

[4] Ozuysal M, Calonder M, Lepetit V, et al. FAST keypoint recognition using random ferns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(3): 448-461.

[5] Williams B, Klein G, Reid I. Real-time SLAM relocalisation[C]//International Conference on Computer Vision. Piscataway, USA: IEEE, 2007: 1-8.

[6] Oshin O, Gilbert A, Illingworth J, et al. Action recognition using randomised ferns[C]//International Conference on Computer Vision. Piscataway, USA: IEEE, 2009: 530-537.

[7] 杨明浩,王阳生.保持视觉稳定性的增强现实注册算法[J].中国图象图形学报,2010,15(12): 1842-1848.Yang M H, Wang Y S. Visual robust register algorithm for augmented reality[J]. Journal of Image and Graphics, 2010, 15(12): 1842-1848.

[8] 陈靖,王涌天,林精敦,等.基于增强现实技术的圆明园景观数字重现[J].系统仿真学报,2010,22(2): 424-428.Chen J, Wang Y T, Lin J D, et al. Digital reconstruction of Yuan Ming Yuan based on augmented reality technology[J]. Journal of System Simulation, 2010, 22(2): 424-428.

[9] 王丽婷,丁晓青,方驰.基于随机森林的人脸关键点精确定位方法[J].清华大学学报:自然科学版,2009, 49(4): 543-546.Wang L T, Ding X Q, Fang C. Accurate localization of facial feature points based on random forest classifier[J]. Journal of Tsinghua University: Science and Technology, 2009, 49(4): 543-546.

[10] Harris C, Stephens M. A combined corner and edge detector[C]//4th Alvey Vision Conference. 1988: 147-151.

[11] Rosten E, Drummond T. Machine learning for high-speed corner detection[C]///European Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2006: 430-443.

[12] Shi J, Tomasi C. Good features to track[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, USA: IEEE, 1994: 593-600.

[13] Agrawal M, Konolige K, Blas M. CenSurE: Center surround extremas for realtime feature detection and matching[C]//European Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2008: 102-115.

[13] Donoser M, Bischof H. Efficient maximally stable extremal region (MSER) tracking[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, USA: IEEE, 2006: 553-560.

[14] Lowe D. Object recognition from local scale-invariant features[C]//International Conference on Computer Vision. Piscataway, USA: IEEE, 1999: 1150-1157.

[16] Bay H, Tuytelaars T. SURF: Speeded up robust features[C]//European Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2006: 404-417.