多尺度显著性区域提取的模板匹配

逯睿琦

doi:10.3788/OPE.20182611.2776

您当前的位置：

首页 >

文章列表页 >

多尺度显著性区域提取的模板匹配

信息科学 | 更新时间：2020-07-05

- 多尺度显著性区域提取的模板匹配
- Template matching with multi-scale saliency
- 光学精密工程 2018年26卷第11期页码：2776-2784
- 作者机构：
  
  清华大学电子工程系, 北京 100084
- 作者简介：
  
  [ "逯睿琦(1995-)，男，山西临汾人，学士，2017年于清华大学获得学士学位，主要从事数字图像处理，物体识别与检测等方面的研究。E-mail：lurq17@mails.tsinghua.edu.cn" ]
- 基金信息：
  
  国家重点基础研发计划资助项目(2016YFB0100900);国家自然科学基金资助项目(61171113);国家自然科学基金资助项目(61773231)
- DOI：10.3788/OPE.20182611.2776
  中图分类号： TP391.4
- 收稿日期：2018-06-07，
  
  录用日期：2018-6-27，
  
  纸质出版日期：2018-11-25
- 稿件说明：
移动端阅览
逯睿琦. 多尺度显著性区域提取的模板匹配[J]. 光学精密工程, 2018,26(11):2776-2784.

Rui-qi LU. Template matching with multi-scale saliency[J]. Optics and precision engineering, 2018, 26(11): 2776-2784.
逯睿琦. 多尺度显著性区域提取的模板匹配[J]. 光学精密工程, 2018,26(11):2776-2784. DOI： 10.3788/OPE.20182611.2776.

Rui-qi LU. Template matching with multi-scale saliency[J]. Optics and precision engineering, 2018, 26(11): 2776-2784. DOI： 10.3788/OPE.20182611.2776.

摘要

针对模板匹配过程中强遮挡、剧烈背景变化及物体非刚性形变等难题，本文提出了一种基于多尺度显著性区域提取的模板匹配算法。算法采用多尺度-显著性特征并行提取的方式：一方面利用空间金字塔模型将参考图像中的模板和待匹配图像中的目标区域分割成不同尺度的网格，采用可形变多相似性度量方法（Deformable Diversity Similarity，DDIS）计算不同尺度下的匹配得分；同时，算法提取模板区域的显著性区域图，形成模板区域的显著性得分；随后，利用显著性得分对不同尺度的匹配得分进行加权融合，在融合得到的匹配得分图上寻找最佳匹配区域。算法与取得目前最好结果的DDIS方法相比，AUC（Area Under Curve）指标提升2.9%。实验结果表明，显著性区域提取使匹配方法更加关注目标物体，削弱背景及遮挡物体对其影响，从而增强模板匹配方法对于背景变化及遮挡的抵抗能力。另外，空间金字塔模型能够增强模板匹配方法对于物体不同尺度下的特征提取，如物体的局部轮廓及结构特征等。二者结合有效地提高了匹配精度。

Abstract

Traditional template matching methods suffer from heavy occlusion

intense background change and non-rigid deformation. A multi-scale saliency template matching method is proposed in this article in order to deal with such conditions. The method extracted saliency and multi-scale features in parallel. On the one hand

the template and the target images were first divided into grids of different scales using spatial pyramid model. Deformable Diversity Similarity (DDIS) was calculated under such different grids. On the other hand

saliency map of the template image was calculated using saliency segmentation method. Such saliency map s are then used to weight the scores calculated by DDIS under different grids. Finally

the final score map is calculated by fusing the score maps under different grids. The method proposed achieves 2.9% AUC(Area Under Curve) improvement compared with original DDIS method. Experiments show that salient object segmentation helps the method to focus more on object than background

therefore improve the robustness to background changes and occlusion. Besides

spatial pyramid model makes the method to consider information from different scale

for example

local contours and structural features of an object. Combining these two factors raises the matching accuracy significantly.

关键词

Keywords

references

OUYANG W, TOMBARI F, MATTOCCIA S, et al.. Performance evaluation of full search equivalent pattern matching algorithms[J].IEEE Transactions on Pattern Analysis & Machine Intelligence, 2011, 34(1):127-143.

ELBOHER E, WERMAN M. Asymmetric correlation:a noise robust similarity measure for template matching[J]. IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society, 2013, 22(8):3062-3073.

HEL-OR Y, HEL-OR H, DAVID E. Matching by tone mapping:photometric invariant template matching[J].IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 36(2):317-330.

KORMAN S, REICHMAN D, TSUR G, et al .. FasT-match: fast affine template matching[C]. IEEE Conference on Computer Vision and Pattern Recognition . IEEE Computer Society , 2013: 2331-2338. https://ieeexplore.ieee.org/document/6619146

TIAN Y, NARASIMHAN S G. Globally optimal estimation of nonrigid image distortion[J]. International Journal of Computer Vision, 2012, 98(3):279-302.

DEKEL T, ORON S, RUBINSTEIN M, et al .. Best-buddies similarity for robust template matching[C]. Computer Vision and Pattern Recognition . IEEE , 2015: 2021-2029. https://ieeexplore.ieee.org/document/7298813

ORON S, DEKEL T, XUE T, et al.. Best-buddies similarity-robust template matching using mutual nearest neighbors[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1-14.

TALMI I, MECHREZ R, ZELNIK-MANOR L. Template matching with deformable diversity similarity[C] . IEEE Conference on Computer Vision and Pattern Recognition . IEEE , 2017: 1311-1319. https://arxiv.org/abs/1612.02190v2

邵枫, 姜求平, 蒋刚毅, 等.基于显著性分析的立体图像视觉舒适度预测[J].光学精密工程, 2014, 22(6):1631-1638.

SHAO F, JIANG Q P, JIANG G Y, et al.. Prediction of visual discomfort of stereoscopic images based on saliency analysis[J]. Opt. Precision Eng., 2014, 22(6):1631-1638.(in Chinese)

JIANG H, WANG J, YUAN Z, et al .. Salient object detection: a discriminative regional feature integration approach[C]. Computer Vision and Pattern Recognition . IEEE , 2013: 2083-2090. https://link.springer.com/article/10.1007%2Fs11263-016-0977-3

LAZEBNIK S, SCHMID C, PONCE J. Beyond bags of features: spatial pyramid matching for recognizing natural scene categories[C]. Computer Vision and Pattern Recognition , 2006 IEEE Computer Society Conference on . IEEE , 2006: 2169-2178. https://hal.inria.fr/inria-00548585

GRAUMAN K, DARRELL T. Discriminative classification with sets of image features[C]. International Conference on Computer Vision , 2005.

WALLRAVEN C, CAPUTO B, GRAF A. Recognition with local features: the kernel recipe[C]. IEEE International Conference on Computer Vision . IEEE Computer Society , 2003: 257. https://ieeexplore.ieee.org/document/1238351

WILLAMOWSKI J, ARREGUI D, CSURKA G, et al.. Categorizing nine visual classes using local appearance descriptors[J]. Icpr Workshop on Learning for Adaptable Visual Systems, 2004.

ZHANG J, MARSZALEK M, LAZEBNIK S, et al .. Local features and kernels for classifcation of texture and object categories: An in-depth study[R]. Technical Report RR -5737, INRIA Rhône-Alpes , 2005.

HE K, ZHANG X, REN S, et al.. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2014, 37(9):1904-1916.

秦翰林, 周慧鑫, 刘群昌, 等.采用多尺度隐式马尔可夫模型的红外图像背景抑制[J].光学精密工程, 2011, 19(8):1950-1956.

QIN H L, ZHOU H X, LIU Q CH, et al.. Suppression of infrared image background by multiscale hidden Markov model[J]. Opt. Precision Eng., 2011, 19(8):1950-1956.(in Chinese)

王玮, 黄非非, 李见为, 等.使用多尺度LBP特征描述与识别人脸[J].光学精密工程, 2008, 16(4):696-705.

WANG W, HUANG F F, LI J W, et al.. Face description and recognition using multi-scale LBP feature[J]. Opt. Precision Eng., 2008, 16(4):696-705.(in Chinese)

WU Y, LIM J, YANG M H. Online object tracking: a benchmark[C]. Computer Vision and Pattern Recognition . IEEE , 2013: 2411-2418. https://ieeexplore.ieee.org/document/6619156

SIMAKOV D, CASPI Y, SHECHTMAN E, et al .. Summarizing visual data using bidirectional similarity[C]. Computer Vision and Pattern Recognition , 2008. CVPR 2008. IEEE Conference on . IEEE , 2008: 1-8. https://ieeexplore.ieee.org/document/4587842

浏览量

194

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向机翼线缆支架的装配符合性视觉检测

采用多特征共生矩阵的模板匹配

并行压缩成像系统的压缩域小目标检测

基于双重模型的微光像增强器分辨力

融合距离信息的红外自适应相关跟踪