一种面向弱纹理图像的特征点描述子

doi:10.13190/j.jbupt.2021-072

北京邮电大学学报 ›› 2021, Vol. 44 ›› Issue (6): 13-19.doi: 10.13190/j.jbupt.2021-072

一种面向弱纹理图像的特征点描述子

程鹏飞¹, 周修庄², 唐玲², 魏世民¹, 高欢¹

1. 北京邮电大学现代邮政学院, 北京 100876;
2. 北京邮电大学人工智能学院, 北京 100876

收稿日期:2021-04-28 出版日期:2021-12-28 发布日期:2021-12-28
通讯作者: 唐玲(1986—),女,高级工程师,硕士生导师,E-mail:tangling@bupt.edu.cn. E-mail:tangling@bupt.edu.cn
作者简介:程鹏飞(1995—),男,硕士生.
基金资助:
国家自然科学基金项目(61972046);北京市自然科学基金项目(4202051)

A Feature Point Descriptor for Texture-Less Images

CHENG Peng-fei¹, ZHOU Xiu-zhuang², TANG Ling², WEI Shi-min¹, GAO Huan¹

1. School of Modern Post, Beijing University of Posts and Telecommunications, Beijing 100876, China;
2. School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China

Received:2021-04-28 Online:2021-12-28 Published:2021-12-28

摘要/Abstract

摘要： 现有的大多数特征点提取算法适用于处理纹理丰富的图像,而对于弱纹理图像则无法提取有效的特征点. 对此,提出了多邻域结构张量特征(MNSTF)算法. 基于一系列固定的邻域和图像结构张量,通过表达局部图像的结构和纹理信息,解决了弱纹理和无纹理场景下特征点提取和匹配等相关问题;同时,通过计算邻域之间的相对方向,实现了MNSTF算法特征描述子的旋转不变性. 实验结果表明,MNSTF算法在经过旋转的弱纹理图像测试集上的特征点匹配准确率达到了99.9%以上,验证了其良好的适用性、旋转不变性和鲁棒性.

关键词: 特征点描述子, 多邻域, 结构张量, 弱纹理图像

Abstract: Most of the existing feature point extraction algorithms are suitable for processing images with rich texture, but they cannot extract effective feature points for texture-less images. To solve this problem, a new feature point descriptor, termed as multi-neighborhood structure tensor features (MNSTF) is proposed. The algorithm is based on a series of fixed neighborhoods and image structure tensors. By expressing the structure and texture information of local images, it solves the problems such as feature point extraction and matching in texture-less and non-texture scenes in existing feature point extraction algorithms. At the same time, by calculating the relative direction between neighborhoods, the rotation invariance of the MNSTF feature descriptor. The experimental results show that the accuracy of feature point matching of the MNSTF algorithm on the rotated texture-less image test set is over 99.9%, which verifies the good applicability, rotation invariance and robustness of the MNSTF feature descriptor.

Key words: feature point descriptor, multi-neighborhood, structure tensor, texture-less images

中图分类号:

TP391.4

程鹏飞, 周修庄, 唐玲, 魏世民, 高欢. 一种面向弱纹理图像的特征点描述子[J]. 北京邮电大学学报, 2021, 44(6): 13-19.

CHENG Peng-fei, ZHOU Xiu-zhuang, TANG Ling, WEI Shi-min, GAO Huan. A Feature Point Descriptor for Texture-Less Images[J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(6): 13-19.

参考文献

[1] Lowe D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision, 2004, 60(2):91-110.
[2] Bay H, Tuytelaars T, van Gool L.SURF:speeded up robust features[M]//Computer Vision-ECCV 2006.Berlin:Springer Berlin Heidelberg, 2006:404-417.
[3] Calonder M, Lepetit V, Strecha C, et al.BRIEF:binary robust independent elementary features[M]//Computer Vision-ECCV 2010.Berlin:Springer Berlin Heidelberg, 2010:778-792.
[4] Rublee E, Rabaud V, Konolige K, et al.ORB:an efficient alternative to SIFT or SURF[C]//2011 International Conference on Computer Vision.Barcelona:IEEE Press, 2011:2564-2571.
[5] Rosten E, Drummond T.Machine learning for high-speed corner detection[M]//Computer Vision-ECCV 2006.Berlin:Springer Berlin Heidelberg, 2006:430-443.
[6] Tombari F, Franchi A, Di L.BOLD features to detect texture-less objects[C]//2013 IEEE International Conference on Computer Vision.Sydney:IEEE Press, 2013:1265-1272.
[7] Chan J, Lee J A, Qian Kemao.BORDER:an oriented rectangles approach to texture-less object recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas:IEEE Press, 2016:2855-2863.
[8] Hinterstoisser S, Holzer S, Cagniart C, et al.Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes[C]//2011 International Confe-rence on Computer Vision.Barcelona:IEEE Press, 2011:858-865.
[9] Hinterstoisser S, Cagniart C, Ilic S, et al.Gradient response maps for real-time detection of textureless objects[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(5):876-888.
[10] Cvetković S, Grujić N, Ilić S, et al.Detection of texture-less objects by line-based approach[J].Facta Universitatis, Series:Automatic Control and Robotics, 2020, 18(2):79-94.
[11] Chan J, Lee J A, Qian Kemao.BIND:binary integrated net descriptors for texture-less object recognition[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Honolulu:IEEE Press, 2017:3020-3028.
[12] Yu Hancheng, Qin Haibao, Peng Maoting.A fast approach to texture-less object detection based on orientation compressing map and discriminative regional weight[J].Algorithms, 2018, 11(12):201-217.
[13] Fang Wei, Ding Yewen, Zhang Feihong, et al.DOG:a new background removal for object recognition from images[J].Neurocomputing, 2019, 361:85-91.
[14] Kaskman R, Shugurov I, Zakharov S, et al.6 DoF pose estimation of textureless objects from multiple RGB frames[C]//Computer Vision-ECCV 2020 Workshops.[S.l.]:Springer, 2020:612-630.
[15] Jiang Ping, Ishihara Y, Sugiyama N, et al.Depth image-based deep learning of grasp planning for textureless planar-faced objects in vision-guided robotic Bin-picking[J].Sensors, 2020, 20(3):706-729.
[16] Weickert J.Anisotropic diffusion in image processing[M].Stuttgart:Teubner, 1998:56-57.
[17] Kuhne G, Weickert J, Schuster O, et al.A tensor-driven active contour model for moving object segmentation[C]//Proceedings 2001 International Conference on Image Processing.Thessaloniki:IEEE Press, 2001:73-76.
[18] Wang Zhou, Bovik A C.Modern image quality assessment[J].Synthesis Lectures on Image, Video, and Multimedia Processing, 2006, 2(1):1-156.
[19] Zhang Weichuan, Sun Changming.Corner detection using multi-directional structure tensor with multiple scales[J].International Journal of Computer Vision, 2020, 128(2):438-459.
[20] Li Shimiao.A review of feature detection and match algorithms for localization and mapping[J].IOP Conference Series:Materials Science and Engineering, 2017, 231:012003.
[21] Dong J, Soatto S.Domain-size pooling in local descriptors:DSP-SIFT[C]//Proceedings of the IEEE Confe-rence on Computer Vision and Pattern Recognition.Baston:IEEE Press, 2015:5097-5106.

一种面向弱纹理图像的特征点描述子

A Feature Point Descriptor for Texture-Less Images

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价