计算机工程
計算機工程
계산궤공정
COMPUTER ENGINEERING
2015年
4期
263-266,272
,共5页
超分辨率重建%粒子滤波%运动估计%匹配精度%无迹卡尔曼滤波%权值
超分辨率重建%粒子濾波%運動估計%匹配精度%無跡卡爾曼濾波%權值
초분변솔중건%입자려파%운동고계%필배정도%무적잡이만려파%권치
super resolution reconstruction%particle filtering%motion estimation%matching accuracy%Unscented Kalman Filtering( UKF)%weight
视频超分辨率重建的一个必要步骤是视频运动估计，相对其他图像匹配算法，基于特征点的视频匹配算法具有更高的鲁棒性，但精确度受特征点的定位、选取和匹配误差的影响较大。为此，提出将粒子滤波应用到视频超分辨率的运动估计问题中，用粒子滤波算法来修正匹配误差，并针对粒子滤波中的粒子匮乏问题改进基本粒子滤波算法。实验结果表明，该算法比其他经典滤波算法估计精度有了较大提高，且在超分辨率重建中能更精确地进行运动估计，匹配精度和稳定性能都有所改善。
視頻超分辨率重建的一箇必要步驟是視頻運動估計，相對其他圖像匹配算法，基于特徵點的視頻匹配算法具有更高的魯棒性，但精確度受特徵點的定位、選取和匹配誤差的影響較大。為此，提齣將粒子濾波應用到視頻超分辨率的運動估計問題中，用粒子濾波算法來脩正匹配誤差，併針對粒子濾波中的粒子匱乏問題改進基本粒子濾波算法。實驗結果錶明，該算法比其他經典濾波算法估計精度有瞭較大提高，且在超分辨率重建中能更精確地進行運動估計，匹配精度和穩定性能都有所改善。
시빈초분변솔중건적일개필요보취시시빈운동고계，상대기타도상필배산법，기우특정점적시빈필배산법구유경고적로봉성，단정학도수특정점적정위、선취화필배오차적영향교대。위차，제출장입자려파응용도시빈초분변솔적운동고계문제중，용입자려파산법래수정필배오차，병침대입자려파중적입자궤핍문제개진기본입자려파산법。실험결과표명，해산법비기타경전려파산법고계정도유료교대제고，차재초분변솔중건중능경정학지진행운동고계，필배정도화은정성능도유소개선。
In the super resolution reconstruction, a key step is the video motion estimation. Compared with other methods,matching algorithm based on features of video has higher robustness. However, the accuracy of this kind of methods is affected by the position and selection of feature points. To overcome this problem,this paper introduces the particle filtering into the motion estimation to reduce the matching error. The main disadvantage of the particle filtering is particle degeneracy. In this paper, an extended Kalman filtering is used to general the proposal distribution, and an Unscented Kalman Filtering( UKF) is used to refine particles. Experimental results show that,compared with other eight classic filtering algorithms, the proposed algorithm has much better performance, and for the super resolution reconstruction issue,the proposed algorithm can estimate the motion more accurately.
