首页 > 论文 > 激光与光电子学进展 > 56卷 > 14期(pp:141008--1)

基于卷积神经网络的SIFT特征描述子降维方法

Convolutional Neural Network-Based Dimensionality Reduction Method for Image Feature Descriptors Extracted Using Scale-Invariant Feature Transform

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

针对128维尺度不变特征变换(SIFT)特征描述子进行图像局部特征点提取时匹配时间过长,以及三维重建进行特征点配准时的应用局限性,结合深度学习方法,提出一种基于卷积神经网络的SIFT特征描述子降维方法。该方法利用卷积神经网络强大的学习能力实现了SIFT特征描述子降维,同时保留了良好的仿射变换不变性。实验结果表明,经过训练的卷积神经网络将SIFT特征描述子降至32维时,新的特征描述子在旋转、尺度、视点以及光照等仿射变换下均具有良好的匹配效果,匹配效率比传统SIFT特征描述子效率提高了5倍。

Abstract

Since local feature descriptors extracted from an image using the traditional scale-invariant feature transform (SIFT) method are 128-dimensional vectors, the matching time is too long, which limits their applicability in some cases such as feature point matching based on the three-dimensional reconstruction. To tackle this problem, a SIFT feature descriptor dimensionality reduction method based on a convolutional neural network is proposed. The powerful learning ability of the convolutional neural network is used to realize the dimensionality reduction of SIFT feature descriptors while maintaining their good affine transformation invariance. The experimental results demonstrate that the new feature descriptors obtained using the proposed method generalize well against affine transformations, such as rotation, scale, viewpoint, and illumination, after reducing their dimensionality to 32. Furthermore, the matching speed of the feature descriptors obtained using the proposed method is nearly five times faster than that of the SIFT feature descriptors.

Newport宣传-MKS新实验室计划
补充资料

DOI:10.3788/LOP56.141008

所属栏目:图像处理

基金项目:国家自然科学基金、国家973计划;

收稿日期:2019-01-14

修改稿日期:2019-02-21

网络出版日期:2019-07-01

作者单位    点击查看

周宏浩:中国科学技术大学环境科学与光电技术学院, 安徽 合肥 230031中国科学院通用光学与表征技术重点实验室, 安徽 合肥 230031
易维宁:中国科学院通用光学与表征技术重点实验室, 安徽 合肥 230031
杜丽丽:中国科学院通用光学与表征技术重点实验室, 安徽 合肥 230031
乔延利:中国科学技术大学环境科学与光电技术学院, 安徽 合肥 230031中国科学院通用光学与表征技术重点实验室, 安徽 合肥 230031

联系人作者:杜丽丽(ylqiao@aiofm.ac.cn); 乔延利(lilydu@aiofm.ac.cn);

备注:国家自然科学基金、国家973计划;

【1】Wu C C. Towards linear-time incremental structure from motion. [C]∥2013 International Conference on 3D Vision, June 29-July 1, 2013, Seattle, WA, USA. New York: IEEE. 127-134(2013).

【2】Guo M, Hu L L and Li J. Local point cloud reconstruction of ceramic-bowl-surface defect based on multi-image sequences. Acta Optica Sinica. 37(12), (2017).
郭萌, 胡辽林, 李捷. 基于多幅图像的陶瓷碗表面缺陷的局部点云重建. 光学学报. 37(12), (2017).

【3】Lowe D G. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision. 60(2), 91-110(2004).

【4】Bay H, Ess A, Tuytelaars T et al. Speeded-up robust features (SURF). Computer Vision and Image Understanding. 110(3), 346-359(2008).

【5】Ke Y and Sukthankar R. PCA-SIFT: a more distinctive representation for local image descriptors. [C]∥2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 27-July 2, 2004, Washington, DC, USA. New York: IEEE. 8161522, (2004).

【6】Leutenegger S, Chli M and Siegwart R Y. BRISK: binary robust invariant scalable keypoints. [C]∥2011 International Conference on Computer Vision, November 6-13, 2011, Barcelona, Spain. New York: IEEE. 2548-2555(2011).

【7】Rublee E, Rabaud V, Konolige K et al. ORB: an efficient alternative to SIFT or SURF. [C]∥2011 International Conference on Computer Vision, November 6-13, 2011, Barcelona, Spain. New York: IEEE. 2564-2571(2011).

【8】Rosten E and Drummond T. Machine learning for high-speed corner detection. ∥Leonardis A, Bischof H, Pinz A. Lecture notes in computer science. Berlin, Heidelberg: Springer. 3951, 430-443(2006).

【9】Calonder M, Lepetit V, Strecha C et al. BRIEF: binary robust independent elementary features. ∥Daniilidis K, Maragos P, Paragios N. Lecture notes in computer science. Berlin, Heidelberg: Springer. 6314, 778-792(2010).

【10】Alahi A, Ortiz R and Vandergheynst P. FREAK: fast retina keypoint. [C]∥2012 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), June 16-21, 2012, Providence, RI, USA. New York: IEEE. 510-517(2012).

【11】Xu P F, Jiang Z L, Zhao Y et al. Feature point extraction and matching algorithm of smooth surfaces without chromatic aberration. Acta Optica Sinica. 38(4), (2018).
许鹏飞, 姜兆亮, 赵阳 等. 无色差光滑曲面特征点的提取及匹配算法. 光学学报. 38(4), (2018).

【12】Chen S, Yang T and Sun S Y. Feature point matching algorithm for fusion of color invariants and SURB detection. Laser & Optoelectronics Progress. 55(5), (2018).
陈树, 杨天, 孙顺远. 融合彩色不变量和SURB检测的特征点匹配算法. 激光与光电子学进展. 55(5), (2018).

【13】Huang H, He K, Zheng X L et al. Spatial-spectral feature extraction of hyperspectral image based on deep learning. Laser & Optoelectronics Progress. 54(10), (2017).
黄鸿, 何凯, 郑新磊 等. 基于深度学习的高光谱图像空-谱联合特征提取. 激光与光电子学进展. 54(10), (2017).

【14】Shu C X, He Y T and Sun Q K. Point cloud registration based on convolutional neural network. Laser & Optoelectronics Progress. 54(3), (2017).
舒程珣, 何云涛, 孙庆科. 基于卷积神经网络的点云配准方法. 激光与光电子学进展. 54(3), (2017).

【15】Verdie Y, Yi K M, Fua P et al. TILDE: a temporally invariant learned detector. [C]∥2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 7-12, 2015, Boston, MA, USA. New York: IEEE. 5279-5288(2015).

【16】Simo-Serra E, Trulls E, Ferraz L et al. Discriminative learning of deep convolutional feature point descriptors. [C]∥2015 IEEE International Conference on Computer Vision (ICCV), December 7-13, 2015, Santiago, Chile. New York: IEEE. 118-126(2015).

【17】Yi K M, Verdie Y, Fua P et al. Learning to assign orientations to feature points. [C]∥2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), June 27-30, 2016, Las Vegas, NV, USA. New York: IEEE. 107-116(2016).

【18】Yi K M, Trulls E, Lepetit V et al. LIFT: learned invariant feature transform. ∥Leibe B, Matas J, Sebe N, et al. Lecture notes in computer science. Cham: Springer. 9910, 467-483(2016).

【19】Sch ps T, Sch nberger J L, Galliani S et al. A multi-view stereo benchmark with high-resolution images and multi-camera videos. [C]∥2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), July 21-26, 2017, Honolulu, HI, USA. New York: IEEE. 2538-2547(2017).

【20】Mikolajczyk K and Schmid C. A performance evaluation of local descriptors. IEEE Transactions on Pattern Analysis and Machine Intelligence. 27(10), 1615-1630(2005).

引用该论文

Zhou Honghao,Yi Weining,Du Lili,Qiao Yanli. Convolutional Neural Network-Based Dimensionality Reduction Method for Image Feature Descriptors Extracted Using Scale-Invariant Feature Transform[J]. Laser & Optoelectronics Progress, 2019, 56(14): 141008

周宏浩,易维宁,杜丽丽,乔延利. 基于卷积神经网络的SIFT特征描述子降维方法[J]. 激光与光电子学进展, 2019, 56(14): 141008

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF