二进制鲁棒不变尺度特征匹配电子稳像

崔少辉; 谢征; 王刚; 陈鹏

doi:10.3788/OPE.20152309.2715

您当前的位置：

首页 >

文章列表页 >

二进制鲁棒不变尺度特征匹配电子稳像

信息科学 | 更新时间：2020-08-12

- 二进制鲁棒不变尺度特征匹配电子稳像
- Feature matching electronic image stabilization based on binary robust invariant scalable keypionts
- 光学精密工程 2015年23卷第9期页码：2715-2723
- 作者机构：
  
  军械工程学院导弹工程系,河北石家庄,050003
- 作者简介：
- 基金信息：
  
  军内科技创新计划资助项目(装司字[2012]665)()
- DOI：10.3788/OPE.20152309.2715
  中图分类号： TP391.4
- 收稿日期：2015-04-29，
  
  修回日期：2015-06-25，
  
  纸质出版日期：2015-09-25
- 稿件说明：
移动端阅览
崔少辉, 谢征, 王刚等. 二进制鲁棒不变尺度特征匹配电子稳像[J]. 光学精密工程, 2015,23(9): 2715-2723

CUI Shao-hui, XIE Zheng, WANG Gang etc. Feature matching electronic image stabilization based on binary robust invariant scalable keypionts[J]. Editorial Office of Optics and Precision Engineering, 2015,23(9): 2715-2723
崔少辉, 谢征, 王刚等. 二进制鲁棒不变尺度特征匹配电子稳像[J]. 光学精密工程, 2015,23(9): 2715-2723 DOI： 10.3788/OPE.20152309.2715.

CUI Shao-hui, XIE Zheng, WANG Gang etc. Feature matching electronic image stabilization based on binary robust invariant scalable keypionts[J]. Editorial Office of Optics and Precision Engineering, 2015,23(9): 2715-2723 DOI： 10.3788/OPE.20152309.2715.

摘要

针对捷联图像制导系统中视频序列受弹体姿态变化和抖动而出现的不稳定现象

提出了一种基于二进制鲁棒不变尺度特征关键点(BRISK)的高精度快速鲁棒电子稳像算法。首先

用BRISK算法提取图像BRISK特征描述符;为了保证匹配精度和速度

采用引导互匹配策略实现BRISK特征点跟踪匹配;然后

利用改进的随机抽样一致方法(RANSAC)剔除误匹配点对;最后

利用最小二乘算法(LSA)估计全局运动参数进而实现稳像。对标准图片的匹配测试和实拍视频稳像的结果表明:结合BRISK算法的电子稳像技术

运行时间小于30 ms

定位精度达到0.1 pixel

对光照变化、噪声以及复杂环境遮挡具有较强的鲁棒性

能快速有效地补偿捷联制导图像的复杂随机抖动并提高视频质量。

Abstract

To overcome the instability of video sequences caused by attitude changes and jitters of a projectile in the strapdown image guidance system

a robust high-precision fast electronic image stabilization algorithm based on Binary Robust Invariant Salable Keypoints(BRISK) was proposed. Firstly

the BRISK algorithm was used to extract BRISK feature descriptors of the image. To improve the matching accuracy and speeds

the mutual matching strategy was adopted to realize BRISK matching feature point tracking. Then

the improved Random Sample Consensus (RANSAC) method was used to eliminate the false matching points.Finally

the Least Square Algorithm (LSA) was taken to estimate the global motion parameters to realize image stabilization. The standard image matching test and real video shoot image results show that the strapdown image guidance stabilization technology based on BRISK algorithm can provide short processing time less than 30 ms and high positioning accuracy of 0.1 pixel. Moreover

the technology has stronger robustness to changed illumination

noise and complex environment barrier and can quickly and effectively compensate complex random jitters and improve the quality of video images.

关键词

Keywords

references

张跃,储海荣. 全捷联图像导引头制导的技术特点[J]. 光学精密工程,2014,22(10):2825-2831. ZHANG Y,CHU H R. Technical characteristics of strapdown image seeker guidance [J]. Opt. Precision Eng., 2014, 22(10):2825-2831. (in Chinese)

王志民,徐晓刚. 电子稳像技术综述[J]. 中国图像图形学报,2010,15(3):470-479. WANG ZH M, XU X G. A survey on electronic image stabilization [J].Journal of Image and Graphics, 2010, 15(3):470-479. (in Chinese)

张永祥,卢岩,栾中,等. 视频图像运动补偿系统的设计与实现[J]. 液晶与显示,2013,28(3):424-428. ZHANG Y X, LU Y, LUAN ZH, et al.. Design and implementation of motion compensation system based on FPGA[J]. Chinese Journal of Liquid Crystals and Displays, 2013, 28(3):424-428. (in Chinese)

TIAN S Q, ZHAO P, WANG N H,et al.. Aims at moving objects' improvement based on gray projection of algorithm of the electronic image stabilization [C]. Proceedings of 2010 3rd International Congress on Image and Signal Processing, 2010:2483-2487.

遆晓光,靳万鑫,余颖. 大运动前景和旋转抖动视频的快速数字稳定[J]. 光学精密工程. 2014,22(1):177-184. DI X G,JIN W X,YU Y. Digital video stabilization for large-scale moving foreground object and rotation jitter [J]. Opt. Precision Eng., 2014, 22(1):177-184. (in Chinese)

WEN D H, JIANG Y S, ZHANG Y D, et al.. Modified block-matching 3-D filter in Laplacian pyramid domain for speckle reduction [J]. Optics Communications,2014, 332(2):150-154.

AOYAMA S, ITO K, AOKI T. A finger knuckle print recognition algorithm using phase-based local block matching [J]. Information Sciences, 2014, 268(12):53-64.

ALAN A, HARIN S. Stego quality enhancement by message size reduction and fibonacci bit-plane mapping [J]. Lecture Notes in Computer Science, 2014, 8893(1):151-166.

AGRAWAL P,KHURSHID A. DWT and GA-PSO based novel watermarking for videos using audio watermark [J].Lecture Notes in Computer Science,2014,8795(1):212-220.

SCHULZE C, LIWICKI M. Optimized SIFT feature matching for image retrieval [J]. Lecture Notes in Computer Science, 2014, 1(1):102-115.

HERBERT B, ANDREAS E, TINNE T, et al..Speeded Up Robust Feature(SURF) [J].Computer Vision and Image Understanding(CVIU), 2008,110(3):346-359.

张从鹏,魏学光. 基于Harris角点的矩形检测[J]. 光学精密工程, 2014, 22(8):2259-2266. ZHANG C P, WEI X G. Rectangle detection based on Harris corner[J]. Opt. Precision Eng., 2014, 22(8):2259-2266. (in Chinese)

BIADGIE Y, SOHN K A. Feature detector using adaptive accelerated segment test [C]. International Conference on Information Science & Applications (ICISA), 2014:1537-1541.

FOTEINI P, NIKIFOROS M, OLGA P. A study on the use of a binary local descriptor and color extensions of local descriptors for video concept detection[J]. Lecture Notes in Computer Science, 2015, 8935(1):282-293.

LEUTENEGGER S, CHLI M, SIEGWART R Y. BRISK:Binary robust invariant scalable keypoints [C]. Proceedings of the 2011 IEEE International Conference on Computer Vision (ICCV), 2012:1589-1596.

林森,苑玮琦,宋辉. 二进制鲁棒不变尺度特征在非接触掌纹识别中的应用[J]. 仪器仪表学报. 2013,34(12):2785-2792. LIN S, YUAN W Q, SONG H. Application of binary robust invariant scalable keypoints in non-contact palmprint recognition [J].Chinese Journal of Scientific Instrument, 2013, 34(12):2785-2792. (in Chinese)

许允喜,蒋云良, 陈方. 惯性组合导航系统中基于BRISK的快速景象匹配算法[J]. 光电子·激光,2013,23(8):1589-1596. XU Y X, JIANG Y L, CHEN F. Rapid scene matching algorithm based on BRISK for inertial integrated navigation system [J]. Journal of Optoelectronics Laser, 2013, 23(8):1589-1596. (in Chinese)

张勇,王新赛,李明明,等. 鲁棒的车载红外视频电子稳像算法[J]. 强激光与粒子束,2013,25(4):853-857 ZHANG Y, WANG X S, LI M M, et al.. Robust infrared video electronic stabilization algorithm for vehicle [J]. High Power Laser and Particle Beams, 2013, 25(4):853-857. (in Chinese)

李刚,王赫. 机载光电平台实时图像消旋[J]. 液晶与显示,2014,29(2):304-309. LI G, WANG H. Real-time image rotation-elimination for airborne photoelectric platform [J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(2):304-309. (in Chinese)

陈付幸,王润生. 基于预检验的快速随机抽样一致性算法[J]. 软件学报,2005,16(8):1431-1437. CHEN F X, WANG R SH. Fast RANSAC with preview model parameters evaluation [J]. Journal of Software, 2005, 16(8):1431-1437. (in Chinese)

浏览量

327

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

弱纹理环境下融合点线特征的双目视觉同步定位与建图

针对目标遮挡的自适应特征匹配网络

基于自适应惯导辅助特征匹配的视觉SLAM算法

融合二维图像和三维点云的相机位姿估计

采用轮廓特征匹配的红外-可见光视频自动配准