改进Noble算子匹配的电子稳像法

初守艳; 席志红

doi:doi:10.3788/ope.20142201.0204

光学精密工程, 2014, 22 (1): 204, 网络出版: 2014-02-18

改进Noble算子匹配的电子稳像法

Digital image stabilization based on improved Noble feature matching

论文大纲

初守艳 ^*席志红

作者单位

哈尔滨工程大学信息与通信工程学院，黑龙江哈尔滨 150001

电子稳像全局运动全局一致性特征匹配卡尔曼滤波 digital image stabilization global motion global consistency feature matching Kalman filter

摘要

考虑全局运动估计算法对电子稳像系统准确性和实时性的影响，提出了改进Noble算子匹配的电子稳像算法。该算法通过区域预选、特征匹配、参数求解和运动补偿四步操作稳定抖动视频。首先，根据全局运动一致性的特点，采用子块间的绝对差值提前剔除部分不可靠区域，并改进传统Noble算子的全局阈值部分，根据区域梯度均值自适应调整阈值，检测各保留区域的角点，以保证图像特征的空间均匀性分布; 其次，根据特征点的邻域信息构造梯度方向描述子进行粗匹配运算，并利用最近邻次近邻比率和均值距离准则两步操作优化匹配集合; 然后，结合运动模型求解全局参数; 最后，选择卡尔曼滤波过程完成抖动图像的补偿处理。Matlab仿真结果证明: 该算法能平均提高峰值信噪比2 dB以上，当旋转角度小于5°时性能更为优异，能准确、快速地稳定视频图像。

Abstract

In consideration of the effect of global motion estimation algorithm on the real-time performance and the accuracy of a electron stabilization system, a digital image stabilization algorithm based on the improved Noble feature matching was proposed. The algorithm was composed of four steps: preselecting regions, feature matching, solving parameters and motion compensation. Firstly, the global consistency of the background pixels was used to eliminate the unreliable regions, and the adaptive threshold was added into the traditional Noble feature extraction process to ensure the features uniformly distributed in the retained image regions. Then, the neighborhood gradient information of the feature point was used to build the feature descriptors and to get the rough match sets. At the same time, the ratio of the closest neighbor to second-closest neighbor and the mean space distance criteria were both used to optimize the match sets. Furthermore, the global motion parameters were solved. Finally, the Kalman filter was used to get the true shaky components to compensate the images. Matlab experimental results indicate that the algorithm can increase the peak signal-to-noise ratio more than 2 dB, and it is especially excellent when the rotation angle is less than 5 °. In conclusion, the algorithm can stabilize video images quickly and accurately.

参考文献

[1] ELEBI A, AKBULUT O, URHAN O, et al.. Truncated gray-coded bit-plane matching based motion estimation and its hardware Architecture ［J］. IEEE Transactions on Consumer Electronics, 2009,55(3): 1530-1536.

[2] GULLU M K. Weighted constrained one-bit transform based fast block motion estimation ［J］. IEEE Transactions on Consumer Electronics, 2011, 57(2): 751-755.

[3] 张萍,魏平,于鸿洋，等.一种基于生物地理优化的快速运动估计算法［J］. 电子与信息学报,2011,33(5): 1017-1023.

ZHANG P, WEI P, YU H Y, et al.. A biogeography-based optimization algorithm for fast motion estimation ［J］. Journal of Electronics & Information Technology, 2011, 33(5): 1017-1023. (in Chinese)

[4] 谷会涛,陈书明,孙书为,等.多搜索中心的运动估计快速算法［J］.电子学报,2011,39(3): 695-699.

GU H T, CHEN SH M, SUN SH W, et al.. Fast motion estimation algorithm based on multi-search centers ［J］. Acta Electronica Sinica, 2011, 39(3): 695-699. (in Chinese)

[5] 仲训昱,朱齐丹,张智.数字图像中的快速和鲁棒运动估计研究［J］. 电子学报，2010,38(1): 251-256.

ZHONG X Y, ZHU Q D, ZHANG ZH. Study of fast and robust motion estimation in the digital image stabilization ［J］. Acta Electronica Sinica, 2010,38 (1): 251-256. (in Chinese)

[6] 邱家涛,李玉山,初秀琴,等.稳定运动物体视频的特征方法［J］. 光学精密工程,2012,20(10): 2300-2307.

QIU J T, LI Y SH, CHU X Q, et al.. Feature-based approach for stabilizing videos with moving objects［J］. Opt .Precision Eng., 2012, 20(10): 2300-2307. (in Chinese)

[7] 张跃飞.车载摄像机数字稳像技术研究［D］.成都: 电子科技大学,2011.

ZHANG Y F. Research on digital image stabilization for in-car video camera ［D］.Chengdu: University of Electronic Science and Technology of China,2011. (in Chinese)

[8] 王山虎,尤红建,付琨,等.基于大尺度双边SIFT的SAR图像同名点自动提取方法［J］. 电子与信息学报,2012,34(2): 287-293.

WANG SH H, YOU H J, FU K, et al.. An automatic method for finding matches in SAR images based on coarser scale bilateral filtering SIFT［J］. Journal of Electronics & Information Technology, 2012, 34(2): 287-293. (in Chinese)

[9] 孙韬,方俊永,刘学,等.基于SIFT特征匹配的全自动辐射归一化算法［J］. 红外与毫米波学报,2012,31(4): 355-359.

SUN T, FANG J Y, LIU X, et al. Automatic relative radiometric normalization method based on SIFT feature matching ［J］. Journal of Infrared and Millimeter Waves, 2012, 31(4): 355-359. (in Chinese)

[10] 张小洪,李博,杨丹,等.一种新的Harris多尺度角点检测［J］. 电子与信息学报,2007,29(7): 1735-1738.

ZHANG X H, LI B,YANG D, et al.. A novel Harris multi-scale corner detection algorithm［J］. Journal of Electronics & Information Technology, 2007, 29(7): 1735-1738. (in Chinese)

[11] 朱娟娟,郭宝龙.复杂场景中基于变块差分的运动目标检测［J］. 光学精密工程,2011,19(1): 183-191.

ZHU J J, GUO B L. Moving object detection based on variant block difference in complex scenes［J］. Opt .Precision Eng., 2011, 19(1): 183-191. (in Chinese)

[12] 马丽红,任淼,谭幸均,等.基于局部熵和方差调整的Noble角点检测算法改进［J］. 华南理工大学学报: 自然科学版, 2011,39(2): 51-59.

MA L H, REN M, TAN X J, et al.. Modified Noble corner detection algorithm based on fine tuning of local entropy and variance［J］. Journal of South China University of Technology: Natural Science Edition, 2011, 39(2): 51-59. (in Chinese)

[13] 赵亚湘,樊晓平,刘少强,等.基于运动矢量多级分析的视频全局运动估计［J］.电子与信息学报,2012,34(7): 1538-1542.

ZHAO Y X, FAN X P, LIU SH Q, et al.. Global motion estimation based on the multi-stage analysis of motion vectors［J］. Journal of Electronics & Information Technology, 2012, 34(7): 1538-1542. (in Chinese)

[14] XI Z H, CHU S Y. Two-step global motion estimation algorithm based on the robustness region ［J］. Journal of Computational Information Systems,2013, 9(3): 1019-1026.

[15] 张坤,许廷发,王平,等.高精度实时全帧频SURF电子稳像方法［J］. 光学精密工程,2011,19 (8): 1964-1972.

ZHANG K, XU T F, WANG P, et al.. Real-time full-frame digital image stabilization system by SURF［J］. Opt.Precision Eng., 2011, 19 (8): 1964-1972. (in Chinese)

初守艳, 席志红. 改进Noble算子匹配的电子稳像法[J]. 光学精密工程, 2014, 22(1): 204. CHU Shou-yan, XI Zhi-hong. Digital image stabilization based on improved Noble feature matching[J]. Optics and Precision Engineering, 2014, 22(1): 204.

改进Noble算子匹配的电子稳像法

关于本站 Cookie 的使用提示

全站搜索

改进Noble算子匹配的电子稳像法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索