基于CUDA 和卡尔曼预测的实时电子稳像方法

吉林大学学报(信息科学版) ›› 2015, Vol. 33 ›› Issue (1): 45-51.

基于CUDA 和卡尔曼预测的实时电子稳像方法

朱振伍, 何凯, 王新磊

天津大学电子信息工程学院, 天津300072

收稿日期:2014-07-18 出版日期:2015-01-24 发布日期:2015-03-20
作者简介:朱振伍(1987—), 男, 天津人, 天津大学硕士研究生, 主要从事数字图像与视频处理研究, (Tel)86-13820468876 (E-mail)763721291@ qq. com; 何凯(1972—), 男, 沈阳人, 天津大学副教授, 硕士生导师, 主要从事数字图像处理研究, (Tel)86-15510810452(E-mail)hekai626@163. com。
基金资助:
国家自然科学基金资助项目(61271326)

Real-Time Electronic Image Stabilizing Approach Based on CUDA and Kalman Predictor

ZHU Zhenwu, HE Kai, WANG Xinlei

School of Electronic Information Engineering, Tianjin University, Tianjin 300072, China

Received:2014-07-18 Online:2015-01-24 Published:2015-03-20

摘要/Abstract

摘要：

针对传统电子稳像方法无法实现视频的实时处理的问题, 提出以SURF(Speed Up Robust Features)配准算法为基础, 基于CUDA(Compute Unified Device Architecture)编程实现算法的加速, 并利用卡尔曼预测器进行实时预测。算法利用CUDA 并行编程实现帧间特征点的提取和配准, 获得帧间运动矢量; 利用卡尔曼预测器获得稳定后的运动矢量, 实现对当前帧的运动矢量的补偿, 以达到实时稳像的目的。仿真实验结果表明, 该方法可有效去除视频帧间的抖动, 稳像效果良好, 实现了视频的实时处理。

关键词: 电子稳像, CUDA 编程, SURF 图像配准, 卡尔曼预测

Abstract:

The traditional electronic image stabilization approaches cannot be realized in real time. For this reason, we proposed a novel method based on SURF ( Speed Up Robust Features) registration algorithm. Additionally, CUDA(Compute Unified Device Architecture) programming and Kalman predictor are used for algorithm acceleration and real-time prediction. The proposed method extracts, selects and registers feature points based on CUDA parallel programming, and then obtains the motion vector between video frames. The Kalman predictor is used to predict the motion vector of the current frame, and then to realize the motion compensation as well as the real-time image stabilization. The simulation results show that the proposed algorithm can effectively remove the shake between frames, obtain perfect image stabilization effect, and deal with the question of real time process.

Key words: electronic image stabilizing, compute unified device architecture (CUDA) programming, speed up robust features (SURF) matching, Kalman predictor

中图分类号:

TP391. 41

朱振伍, 何凯, 王新磊. 基于CUDA 和卡尔曼预测的实时电子稳像方法[J]. 吉林大学学报(信息科学版), 2015, 33(1): 45-51.

ZHU Zhenwu, HE Kai, WANG Xinlei. Real-Time Electronic Image Stabilizing Approach Based on CUDA and Kalman Predictor[J]. Journal of Jilin University(Information Science Ed, 2015, 33(1): 45-51.