您现在的位置: 首页> 研究主题> 全局运动估计

全局运动估计

全局运动估计的相关文献在1999年到2022年内共计175篇,主要集中在自动化技术、计算机技术、无线电电子学、电信技术、系统科学 等领域,其中期刊论文123篇、会议论文9篇、专利文献132967篇;相关期刊78种,包括军民两用技术与产品、中国图象图形学报、光学精密工程等; 相关会议9种,包括2009年全国模式识别学术会议暨首届中日韩模式识别学术研讨会、第五届和谐人机环境联合学术会议(HHME2009)、第18届全国多媒体学术会议(NCMT2009)、第4届全国普适计算学术会议(PCC2009)、第5届全国人机交互学术会议(CHCI2009)、2008年全国博士生学术会议(光学测试新理论、新技术)等;全局运动估计的相关文献由351位作者贡献,包括覃团发、郑嘉利、李波等。

全局运动估计—发文量

期刊论文>

论文:123 占比:0.09%

会议论文>

论文:9 占比:0.01%

专利文献>

论文:132967 占比:99.90%

总计:133099篇

全局运动估计—发文趋势图

全局运动估计

-研究学者

  • 覃团发
  • 郑嘉利
  • 李波
  • 姚军
  • 杜登崇
  • 林福辉
  • 罗小伟
  • 蒋晓瑜
  • K·帕努索庞
  • 于白
  • 期刊论文
  • 会议论文
  • 专利文献

搜索

排序:

年份

    • 冯欣; 张洁; 张杰; 蒋友妮; 杨武
    • 摘要: 视频内容的分析与理解往往基于对视频中目标对象的空间、运动特征进行感知。然而,在实际拍摄的视频中,目标对象的真实运动轨迹往往受到同时存在的相机全局运动影响。这种由相机自运动带来的全局运动在当前流行的自媒体视频中十分普遍。为了消除全局运动对视频中对象的真实运动轨迹的影响,提出了一种基于L1范数最小化的全局运动参数估计算法,并在此基础上实现了视频的全局运动补偿,得到了前景对象的真实运动轨迹。实验结果表明该算法能准确有效地去除全局运动的影响并准确恢复出运动对象的真实运动轨迹。
    • 方勇
    • 摘要: 本文结合运动与边缘信息提出了一种视频对象提取方法,首先在时间域估计图像帧的运动场,再基于六参数仿射运动模型使用全局运动估计方法获得全局运动区域和局部运动区域;对局部运动区域再进行多阶段仿射运动一致性分割,获得独立运动区域。在空间域方面,使用Canny算子和运动检测提取运动边缘,并对边缘进行优化,然后进行高斯扩展,形成光滑的吸引力场。在时空联合分析时,对独立运动区域和跟踪目标进行识别,区分出跟踪目标和新目标,并使用曲线演化的方法进行边界校正。
    • 沈秋; 李小凡; 严小乐; 孔繁锵
    • 摘要: 为克服无人机视频应用中计算能力和传输带宽的限制,提出一种视频分层表达方法,实现快速、高质量、灵活的视频编码.将无人机视频分为背景层和目标层,并分别采用基于全局运动估计和基于局部块匹配的方法进行压缩,压缩后的码流可以根据实际的网络情况和应用需求进行单独传输或组合传输.实验结果表明:在极低码率如50 kbit/s时,H.264的PSNR低于28 dB,且图像主观质量超出可接受范围,而采用分层表达中的背景层压缩,可以在保证PSNR在28 dB以上,且图像细节清晰、主观质量较高;在码率较高时,采用背景层和目标层结合,在同等码率条件下,目标的细节比H.264更清晰;整个压缩过程消耗的时间只有H.264的18%.本文方法在不降低视频质量的同时提高了压缩效率,且码流具备灵活性,适用于低延时、低带宽、复杂的无人机应用.%To overcome the limitations of compute capability and bandwidth in UAV (Unmanned Aerial Vehicle) video,this paper proposes a novel video compression algorithm based on hierarchical description.The video is split into background layer and object layer.Background layer is encoded with global motion estimation,while object layer is encoded with local block matching.The experimental results demonstrate that,when bandwidth is narrow e.g.50 kbit/s,using background layer encoding can achieve higher PSNR than H.264,and the subjective quality is much more acceptable;in most cases,combining background and object layer can give more details than H.264 with the same bitrate;the consuming time is only 18% of H.264's.Additionally,alternative bitstream can be extracted for different bandwidth.Consequently,the proposed algorithm is suitable for low latency and complexity applications in UAV video due to its good performance,high efficiency and flexibility.
    • 王朋月; 毛征; 张晨; 张广申
    • 摘要: 为提高全局运动估计的准确性,提出一种基于定向旋转不变性(oriented FAST and rotated BRIEF,ORB)特征匹配的全局运动估计算法.首先对差分图像进行分块,通过设定阈值来筛选出有效的背景运动模块,然后经ORB特征匹配后,代入六参数运动模型,通过最小二乘法求解出估计参数,最后对当前帧做补偿,比较补偿前后的差分图像中的目标轮廓的清晰度.实验结果表明:该算法能有效去除局部运动的影响,在保证全局运动估计准确性的同时,也确保该算法的有效性.%In order to improve the accuracy of global motion estimation, this paper proposes a global motion estimation method based on oriented FAST and rotated BRIEF (ORB) features. Firstly, the difference image is divided into blocks, and the effective background motion module is selected by setting the threshold. Then, the selected background modules are matched by the ORB feature, and substituted into the six-parameter motion model. Finally, the estimation parameters are solved by the least squares method, which are used to compensate the current frame. Compare the sharpness of the target contour of difference image before and after compensation. The experimental results demonstrate that the proposed algorithm can effectively remove the influence of the local motion and ensure the accuracy of the global motion estimation, while ensuring the effectiveness of the algorithm.
    • 卜丽静; 郑新杰; 肖一鸣; 张正鹏
    • 摘要: According to the super-resolution reconstruction of Jilin-1 satellite and on the basis of an analysis of different characteristics of satellite video and common video as well as remote sensing image,the authors studied the motion estimation applicability of the Vandewalle method and the method of LK ( Lucas-Kanade ) optical flow in pyramid under the situation of imaging scene containing moving objects. At the same time, according to the median shift and add( MSA) method, the new median shift and add( NMSA) method was proposed to tackle the problem that the edge information is not clear because of the lack of complementary information between the frames in the video satellite image reconstruction. First, a resolution grid is established based on the multiple of reconstruction, and the two grids are unified into a unified space. Then, the low-resolution pixel values that participate in estimation are determined. With the pixel of the image reconstruction grid to be determined as the center, the allowed error is used to determine the reconstruction pixel values. The experiments using the data of Jilin -1 satellite prove the effectiveness of the method proposed in this paper.%针对吉林一号(Jilin-1)卫星视频影像的超分辨率重建问题,在分析卫星视频与普通视频及遥感影像不同特性的基础上,研究了在成像场景内含有运动物体情况下,Vandewalle方法和金字塔LK(Lucas-Kanade)光流法运动估计方法的适用性.同时,针对目前中值平移并添加(median shift and add,MSA)方法在视频卫星影像重建中因帧间缺少互补信息而引起的边缘信息不清晰问题,提出了新的中值平移并添加(new median shift and add,NMSA)方法.首先,根据重建的倍数建立重建前、后的空间分辨率格网,并将2个格网统一到1个空间内;然后,以待确定的重建后影像格网像元为中心,确定参与估计的低空间分辨率像元值,利用容许误差确定重建后的像元值.通过利用吉林一号(Jilin-1)卫星数据进行实验,验证了NMSA方法的有效性.
    • 刘刚; 凌强; 李峰; 王嵩
    • 摘要: 针对因交通监控设备抖动而引起的视频图像序列不稳定问题,提出了一种在设备内的DSP上直接工作的快速稳像方法.该方法采用Harris算子提取参考帧图像的特征点,利用金字塔LK光流在当前帧跟踪参考帧的特征点,从而获得运动矢量.通过RANSAC剔除匹配错误或前景物体的特征点对,筛选出全局运动矢量.采用描述复杂运动的单应性模型估算全局运动参数,用单纯形法和模板相似性微调参数优化整体效果.最后进行全局运动补偿,实现图像的稳定.依赖TI公司C674X系列DSP的优化方法,实现了对720×480交通视频精确、实时的图像稳定处理.
    • 李颖娟; 纪明; 贺峻峰; 马爱秋; 杨萌
    • 摘要: 为了实现亚像素级图像快速全局运动估计,提出了一种基于相位相关的快速亚像素全局运动估计方法。首先对待计算的两幅图像进行下采样并计算整像素级的全局运动矢量,然后选取两幅图像重叠的部分进行插值,再计算亚像素级的全局运动矢量。最后,将整像素级和亚像素级的计算结果进行加权计算,得出亚像素级的全局运动位移。实验结果表明,文中算法对噪声影响、光照变化和局部遮挡具有较好的鲁棒性,同时能够有效地提高运动估计的精确性和计算效率。%Proposed in this paper is a fast sub-pixel algorithm based on phase correlation and image down-sampling for sub-pixel global motion estimation. First,down-sampling of the two images, then the pixel-level global motion vector of two images calculated. Then chooses the repeated area of the two images and interpolated them,the global motion estimation of sub-pixel-level is achieved. Finally,weighted calculate the results of pixel -level and sub -pixel -level,the global motion displacement of sub-pixel-level of the two images will be calculated out. Experimental results show that the proposed algorithm has good robustness to noise,illumination and partial occlusion,and can effectively improve the accuracy and efficiency of motion estimation.
    • 黄文娟; 王敬东; 薛重飞; 王崟
    • 摘要: Reasonable selection of reference frames in the electronic image stabilization is the basis of global motion estimation and motion compensation. In order to solve the problem of frame jump caused by the traditional reference frame selection strategy, a reference frame update strategy based on the cumulative transformation is presented in this paper. By the cumulative transformation of the adjacent frames, the cumulative global motion parameter of the current frame with respect to the first frame is obtained, and the compensation standard is unified. Aiming at the problem of cumulative error in the process, the change of the reference frame is adjusted by setting the matching threshold, and the number of parameters is reduced. The experimental results show that this improved algorithm can effectively solve the problem of inter frame jump, alleviate the error accumulation process, and output a smooth video.%在电子稳像中合理的选择参考帧是全局运动估计和运动补偿的基础。为了解决传统参考帧选择策略带来的帧间跳变的缺陷,给出一种基于累积变换的参考帧更新策略,通过相邻帧间运动参数的累积变换得到当前帧相对于第一帧的累积全局运动参数,统一了补偿基准,并针对过程中产生的累积误差问题,通过设定匹配阈值调整参考帧更换条件,减少参数的传递次数。实验结果表明,改进算法能够有效解决帧间跳变问题,缓解误差的累积进程,输出平稳的结果视频。
    • 肖春明
    • 摘要: To resolve the indistinct background problem caused by dynamic camera behavior,and in combination with the global motion estimation algorithm of angular point distance feature,the maximum interclass variance method is introduced into the RANSAC estimation algorithm to improve the common RANSAC algorithm. The moving object detection algorithm based on difference multiply principle is used to detect and locate the moving object in object detection process. Aiming at the object de⁃tection of follow⁃up frame sequence of the obtained object location,the moving target detection algorithm based on angular point marching of improved singular value decomposition is proposed. The experimental results demonstrate that the proposed algo⁃rithm has good background motion compensation effect in dynamic background,and can accurately detect the moving target of the follow⁃up frame sequence in the subsequent measured dynamic background measurement. The algorithm has good robustness.%由于动态摄像机行为造成了背景模糊等问题,结合角点矩特征的全局运动估计算法,将最大类间方差法引入RANSAC估计算法,对常用的RANSAC算法进行改进。在目标检测过程中,利用基于差分相乘原理的运动目标检测算法实现了运动目标的检测定位。针对获取到目标位置的后续帧序列的目标检测,提出了基于改进的奇异值分解的角点匹配的运动目标检测算法。实验表明,提出的算法对于背景发生变化的场景,能达到较好的背景运动补偿效果,在随后测量的动态背景状态下,可以准确检测后续帧序列的运动目标,并且具有良好的鲁棒性。
    • 徐济惠; 祝晓东; 刘翠娟; 郭国文
    • 摘要: 研究了将粒子群优化算法同万有引力算法相结合进行边缘检测,利用万有引力原理进行启发函数的计算,指导蚁群运动趋向,快速检测出边缘线,作为图像分析的预处理结果;另外根据视频图像中局部运动集中在图像中部概率较大的特性,提出对图像的几个小区域数据进行全局运动估计,这样大幅度减少了运算量,实验中通过采样4个六分之一原图边长的矩形区域进行全局估计,实验证实了在精度相同的情况下,运算速度提高了九倍左右.
  • 查看更多

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号