计算机科学 ›› 2021, Vol. 48 ›› Issue (8): 134-138.doi: 10.11896/jsjkx.200600140

• 计算机图形学& 多媒体 • 上一篇    下一篇

基于改进SIFT的无人机航拍图像快速配准研究

胡育诚, 芮挺, 杨成松, 王东, 刘恂   

  1. 陆军工程大学野战工程学院 南京210007
  • 收稿日期:2020-06-22 修回日期:2020-09-27 发布日期:2021-08-10
  • 通讯作者: 芮挺(rtinguu@sohu.com)
  • 基金资助:
    国家自然科学基金(61671470);国家重点研发计划(2016YFC0802904)

Study on Aerial Image Fast Registration from UAV

HU Yu-cheng, RUI Ting, YANG Cheng-song, WANG Dong, LIU Xun   

  1. College of Field Engineering,Army Engineering University of PLA,Nanjing 210007,China
  • Received:2020-06-22 Revised:2020-09-27 Published:2021-08-10
  • About author:HU Yu-cheng,born in 1995,postgra-duate.His main research interests include image processing,pattern recognition and artificial intelligence.(361721310@qq.com)RUI Ting,born in 1972,Ph.D,professor,Master's advisor,is a member of China Computer Federation.His main research interests include image processing,pattern recognition and artificial intelligence.
  • Supported by:
    National Natural Science Foundation of China(61671470) and National Key R&D Program of China(2016YFC0802904).

摘要: 为了提高无人机航拍图像配准的实时性,通过分析无人机巡航高度相对稳定及图像缺乏高频的细节信息的特点,提出了一种改进SIFT特征点检测方法,显著提高了图像的配准速度,并构建了一个用于图像拼接的航空影像数据集进行实验验证。首先分析了SIFT(Scale Invariant Feature Transform)算法关于特征点尺度不变性的理论依据及实现方法,提出了消除冗余性能的策略;然后采用减少高斯金字塔阶数与层数以及选择在每阶的第三层图像开始检测极值点,以减小差分尺度空间规模的方法;最后在数据集上进行了与现有图像配准方法的对比实验。实验结果证明,所提方法能够获得匹配稳健、鲁棒性高的特征点,匹配耗时只有经典SIFT的1/10,该方法为无人机航拍图像快速拼接提供了技术支持。

关键词: 航拍图像, 图像配准, SIFT, 差分尺度空间, 尺度不变性

Abstract: In order to improve the real time of the UAV aerial image registration,the paper analyzes the relative stability of UAV's altitude and the lack of high-frequency details in the image,proposes an improved SIFT feature point extraction algorithm and constructs a special aerial images dataset for image mosaic for experimental verification.The paper first analyzes the theoretical basis and implementation method of scale invariance of SIFT (Scale Invariant Feature Transform),and puts forward eliminating redundant performance.The measures,such as reduction of Octave and Level of Gauss pyramid,and selecting the third Level image in each Octave to detect extreme points are taken to reduce the scale of differential scale space.Lastly,the comparable experiments based on dataset with state-of-art image mosaic methods are conducted.The experimental results show that the method proposed in this paper can extract robust feature points,and the matching time is only 1/10 of the original sift,which provides technical support for real-time image mosaic of UAV.

Key words: Aerial image, Image registration, SIFT, Differential scale space, Scale invariance

中图分类号: 

  • TP391.41
[1]SHENG H,CHAO H,COOPMANS C,et al.Low-cost UAV-based thermal infrared remote sensing:Platform,calibration and applications[C]//Proceedings of 2010 IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications.IEEE,2010.
[2]WANG H F,HU Q Q,DUAN J Z.Development and applications of small airborne polarization imaging system[J].Opto-Electronic Engineering,2017(11):44-51,92.
[3]LUO L,XU Q,CHEN J,et al.UAV Image Mosaic Based on Non-Rigid Matching and Bundle Adjustment[C]//2019 IEEE International Geoscience and Remote Sensing Symposium.2019:9117-9120.
[4]WU F Q,YANG Y,PAN A N,et al.Multi-viewpoints remote sensing images registration using mixed features[J].Journal of Image and Graphics,2017,22(8):1154-1161.
[5]GILINSKY A,MANOR L Z.SIFTpack:A compact representation for efficient SIFT matching[C]//Proceedings of the 2013 IEEE International Conference on Computer Vision.IEEE,2013.
[6]BAY H,ESS A,TUYTELARRS T,et al.Speeded-Up RobustFeatures[J].Computer Vision & Image Understanding,2008,110(3):346-359.
[7]RUBLEE E,RABAUD V,KONOLIGE K,et al.ORB:An efficient alternative to SIFT or SURF[C]//2011 IEEE Internatio-nal Conference on Computer Vision.Barcelona,Spainp,2011:2564-2571.
[8]ZHEN J,QI K Y ,KAI C.SSIF T:An Improved SIFT Descriptor for Chinese Character Recognition in Complex Images[J].Information Security and Communications Privacy,2010,6:3-6.
[9]LU P,LU Q,ZHOU G L,et al.Research on Time Series Image Mosaic Method Based on Improved SIFT[J].Computer Engineering and Application,2020,56(1):196-202.
[10]MIKOLAJCZYK K,SCHMID C.A performance evalution of local descriptors[J].Computer Vision & Image Understanding,2008,11(3):404-417.
[11]YU Y,YANG N,YANG C,et al.Memristor bridge-based low pass filter for image processing[J].Journal of Systems Engineering and Electronics,2019,30(3):448-455.
[12]ZHAO A G,WANG H L,YANG X G,et al.Compressed sense SIFT descriptor mixed with geometrical feature[J].Infrared and Laser Engineering,2015,(3):1085-1091.
[13]YANG S P,CHEN J,ZHOU L,et al.Image feature matching method[J].Electronic Measurement Technology,2014,37(6):50-53.
[14]SUN S,ZENG Z.UAV image mosaic based on adaptive SIFT algorithm[C]//21st International Conference on Geoinformatics.2013:1-6.
[15]YAN C M,HAO Y F,ZHANG D,et al.An image matching method based on optimal threshold prediction under hybrid features[J].Computer Engineering and Science,2019,41(10):1803-1808.
[16]SU P F,HUANG S Q,WANG Y T ,et al.Image registration based on multi-scale corner with SIFT descriptor[J].Computer Engineering and Science,2017,39(9):1700-1705.
[17]LIU X,WANG F,WEN R S.Research of UAV Visual Navigation Based on SIFT Algorithm[J].Radio Engineering,2017,47(5):19-22.
[18]LOWE D G.Object Recognition from Local Scale-Invariant Features[C]//Proceedings of the Seventh IEEE International Conference on Computer Vision.1999:1150-1157.
[19]LOWED G.Distinctive Image Features from Scale-invariantkeypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[20]HE J,LI Y S,LU H,et al.Research of UAV Aerial Image Mosaic Based on SIFT[J].Opto-Electronic Engineering,2011,38(2):122-126.
[1] 高玉潼, 雷为民, 原玥. 复杂环境下基于聚类分析的人脸目标识别[J]. 计算机科学, 2020, 47(7): 111-117.
[2] 贺超雷,毕秀丽,肖斌. 一种基于Zernike矩的局部特征检测方法[J]. 计算机科学, 2020, 47(2): 135-142.
[3] 焦扬, 杨传颖, 石宝. 基于SVM相关反馈的鞋印图像检索算法[J]. 计算机科学, 2020, 47(11A): 244-247.
[4] 杨思燕,贺国旗,刘如意. 基于SIFT算法的大场景视频拼接算法及优化[J]. 计算机科学, 2019, 46(7): 286-291.
[5] 邢文博, 杜志淳. 数字图像复制粘贴篡改取证[J]. 计算机科学, 2019, 46(6A): 380-384.
[6] 邵进达, 杨帅, 程琳. 改进SIFT算法结合两级特征匹配的无人机图像匹配算法[J]. 计算机科学, 2019, 46(6): 316-321.
[7] 孙雪强, 黄旻, 张桂峰, 赵宝玮, 丛麟骁. 基于改进SIFT的多光谱图像匹配算法[J]. 计算机科学, 2019, 46(4): 280-284.
[8] 刘朝霞,邵峰,景雨,祁瑞华. 基于视觉约束能量最小化的特征点匹配算法[J]. 计算机科学, 2018, 45(5): 228-231.
[9] 龚安,费凡,郑君. 基于卷积神经网络的多人行为识别方法[J]. 计算机科学, 2018, 45(2): 306-311.
[10] 楼浩锋, 张端. 高斯过程下的CMA-ES在医学图像配准中的应用[J]. 计算机科学, 2018, 45(11A): 234-237.
[11] 厉丹,肖理庆,田隽,孙金萍. 基于改进相位相关与特征点配准的多图拼接算法[J]. 计算机科学, 2018, 45(1): 313-319.
[12] 郑伟,蒋晨娇,刘帅奇,赵杰. 改进的鸡群优化算法及其在DTI-FA图像配准中的应用[J]. 计算机科学, 2018, 45(1): 285-291.
[13] 刘川熙,赵汝进,刘恩海,洪裕珍. 基于RANSAC的SIFT匹配阈值自适应估计[J]. 计算机科学, 2017, 44(Z6): 157-160.
[14] 王怡,徐文迪,余慧斌,郑河荣,潘翔. 显著性特征约束的交互式协同分割[J]. 计算机科学, 2017, 44(Z11): 269-272.
[15] 吴鹏,于秋则,闵顺新. 一种快速鲁棒的SAR图像匹配算法[J]. 计算机科学, 2017, 44(7): 283-288.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 夏庆勋,庄毅. 一种基于局部性原理的远程验证机制[J]. 计算机科学, 2018, 45(4): 148 -151 .
[2] 贾伟,华庆一,张敏军,陈锐,姬翔,王博. 基于改进粒子群优化的移动界面模式聚类算法[J]. 计算机科学, 2018, 45(4): 220 -226 .
[3] 赵慧赟,潘志松. 基于shapelets学习的多元时间序列分类[J]. 计算机科学, 2018, 45(5): 180 -184 .
[4] 耿焕同,丁洋洋,周利发,韩伟民. 一种基于自适应选择策略的改进型MOEA/D算法[J]. 计算机科学, 2018, 45(5): 201 -207 .
[5] 李小薪,吴克宋,齐盼盼,周旋,刘志勇. 局部球面规范化嵌入:PCANet的一种改进方案[J]. 计算机科学, 2018, 45(5): 238 -242 .
[6] 王洋洋, 韦皓诚, 柴云鹏. 基于SSD-SMR混合存储的LSM树键值存储系统的性能优化[J]. 计算机科学, 2018, 45(7): 61 -65 .
[7] 郭炳, 郑文萍, 韩素青. 一种基于突变基因网络的癌症驱动通路识别算法[J]. 计算机科学, 2018, 45(7): 230 -236 .
[8] 倪园慧,陈巍文,王磊,邱柯妮. 面向MLC STT-RAM的寄存器分配策略优化研究[J]. 计算机科学, 2018, 45(6A): 562 -567 .
[9] 王伟, 杨本朝, 李光松, 斯雪明. 异构冗余系统的安全性分析[J]. 计算机科学, 2018, 45(9): 183 -186 .
[10] 张志禹, 刘思媛. 一种基于Curv-SAE特征融合的人脸降维和识别方法[J]. 计算机科学, 2018, 45(10): 267 -271 .