两阶段旋转、平移亚像素数字稳像算法

提出一种稳健的两阶段层次匹配稳像算法, 在频域结合相位相关和谱对消技术对抖动图像序列间的平移和旋转矢量获得亚像素估计精度。算法第一阶段首先在对数极坐标表示下, 利用相位相关法得到图像帧间的旋转角度和整数像素平移参数; 算法第二阶段运用谱对消技术确定亚像素平移参数。在建立参数运动模型的基础上,通过自适应均值滤波确定各帧修正矢量。实验结果表明, 该方法具有亚像素级的稳像精度。     

一种改进的代表点匹配算法在稳像技术中的应用

提出一种用于电子稳像技术改进的代表点匹配算法。通过合理的划分区域,保证在区域内各点运动的一致性,利用代表点匹配算法确定各区域内的运动矢量,并把其代入变换模型可得关于旋转、平移和变焦等信息的线性方程组,求其方程的解得全局运动矢量。利用该方法的特点是,具有速度快,并能求出平移运动和旋转运动参数,实现图像运动的平移和旋转补偿。实验表明,改进的稳像算法对于提高动态图像的稳定性具有较好的效果。     

基于霍夫变换和相位相关的电子稳像算法研究

提出了一种新的基于霍夫变换和相位相关的电子稳像算法。首先根据1D相位相关技术,通过搜索角度脊来得到旋转参数;将待处理图像去旋转后,再采用2D相位相关来得到平移参数;在建立参数运动模型的基础上,通过自适应均值滤波确定各帧修正矢量。提出的方法可以充分利用FFT的高效运算,实验结果表明采用本方法能准确有效地处理具有旋转和平移的电子稳像问题。     

基于角点匹配的电子稳像算法

提出了一种基于角点匹配的电子稳像算法。分析了Harris角点检测的原理。获得角点后,根据相邻帧对应角点的坐标分布,建立当前帧与参考帧的映射关系,并运用仿射变换模型,以最小二乘解的形式获得帧间全局运动估计矢量。最后,采用Kalman滤波器对运动估计矢量作低通滤波,平滑运动参数,获得运动补偿矢量,实现视频序列的实时稳像。实验表明该算法较好地去除了视频序列的高频抖动,同时保留了摄像机的主动运动,稳像后视频序列的峰值信噪比明显提高。     

运动背景下的帧间稳像技术

针对运动背景下帧间稳像技术,提出了一种带运动矢量修正的灰度投影运动估计算法。采用该算法分别对当前帧和参考帧的行、列计算灰度投影序列;将当前帧投影序列局部分块,分别将每一分块与参考帧行、列投影曲线进行互相关计算,得到基于局部投影的行、列运动矢量集合;以分块区域的相关置信度为权系数衡量参数,计算每一分块的像素位移权值,从而计算某一方向帧间的加权运动矢量。实验结果表明:该方法可以使运动目标造成的影响只作用于其中若干个局部分块,而其他分块不受此影响,尽可能保证稳像的准确性。采用该方法稳像后的图像与参考帧图像的均方根误差(RMSE)值明显下降,与参考帧图像更加吻合。     

基于特征匹配和校验的鲁棒实时电子稳像

提出了一种基于特征点匹配和校验的鲁棒实时电子稳像算法.首先利用Kanade-Lucas-Tomasi角点检测器提取参考帧和当前帧的特征点,并用绝对误差和准则进行特征点匹配|在校验阶段,提出一种能够有效剔除前景运动物体特征点和错误匹配点的空间位置不变准则|最后,在相似运动模型下,利用最小二乘法求解全局运动矢量进行运动补偿.实验证明:该算法满足实时性要求,对视频的平移、旋转、缩放运动都有较好的稳像效果,并对运动物体具有鲁棒性.     

运动背景下的帧间稳像技术

针对运动背景下帧间稳像技术,提出了一种带运动矢量修正的灰度投影运动估计算法。采用该算法分别对当前帧和参考帧的行、列计算灰度投影序列;将当前帧投影序列局部分块,分别将每一分块与参考帧行、列投影曲线进行互相关计算,得到基于局部投影的行、列运动矢量集合;以分块区域的相关置信度为权系数衡量参数,计算每一分块的像素位移权值,从而计算某一方向帧间的加权运动矢量。实验结果表明：该方法可以使运动目标造成的影响只作用于其中若干个局部分块,而其他分块不受此影响,尽可能保证稳像的准确性。采用该方法稳像后的图像与参考帧图像的均方根误差（RMSE）值明显下降,与参考帧图像更加吻合。     

基于灰度投影法运动估计的成像CCD平移补偿法

为了提高成像质量,得到高分辨率的图像,通过对灰度投影法进行分析,提出一种基于灰度投影法运动估计的成像CCD平移补偿法,给出其系统结构图。该方法采取只平移成像CCD,探测CCD固定,即在成像CCD补偿位移量的过程中,探测CCD不补偿位移,探测CCD在成像开始的第一帧为全局参考帧,其他帧与之比较来获得位移量。验证了分辨率标板在各种运动情况下系统的补偿效果,对拍摄的实物图像进行恢复,结果证明：该补偿法对慢速运动图像和随机振动图像具有较高的恢复能力。     

基于多帧参考的整帧丢失错误掩盖算法

为了解决视频传输中整帧图像丢失的误差扩散问题,提出了一种基于多参考帧运动矢量外推的整帧丢失错误掩盖算法.通过对多个参考帧的运动矢量外推得到丢失块的候选运动矢量集;采用多参考帧边界匹配准则进行丢失帧和其后续帧的误差估计,并选择丢失块的最优运动矢量;最后采用自适应的重叠块运动补偿方法重建丢失帧.实验结果表明,该算法在恢复整帧图像的同时,有效地降低了视频序列中整帧图像丢失的误差扩散影响,与已有算法比较,该算法恢复的视频序列平均峰值信噪比提高了0.5～1dB,具有更好的错误掩盖效果.     

基于多帧参考的整帧丢失错误掩盖算法

为了解决视频传输中整帧图像丢失的误差扩散问题,提出了一种基于多参考帧运动矢量外推的整帧丢失错误掩盖算法.通过对多个参考帧的运动矢量外推得到丢失块的候选运动矢量集|采用多参考帧边界匹配准则进行丢失帧和其后续帧的误差估计,并选择丢失块的最优运动矢量|最后采用自适应的重叠块运动补偿方法重建丢失帧.实验结果表明,该算法在恢复整帧图像的同时,有效地降低了视频序列中整帧图像丢失的误差扩散影响,与已有算法比较,该算法恢复的视频序列平均峰值信噪比提高了0.5~1 dB,具有更好的错误掩盖效果.     

一种新型不变矩在图像识别中的应用

给出了仿射变换的定义,并在此基础上建立了一种新的仿射变换途径——多尺度自卷积变换。构造了一组新的仿射不变矩(MSA矩),分析了其性质,给出了它的明确表达式及其具体实现方法。分别应用最小距离法、k-近邻法和相似性测度三种识别方法进行了仿真实验。结果表明,MSA矩对平移、比例、旋转、拉伸和扭曲等仿射变换均具有很好的不变性,与典型的Flusser矩相比,稳健性好、识别率高、抗噪声干扰能力强,具有一定的应用价值。     

Thomas precession for spin and for a rod

In this paper we obtain the differential equations which describe the rotating rod and precession of the spin of a gyroscope moving along a curved trajectory. Several examples of such motion are considered. The obtained equations differ from the traditional Thomas expression, interpreted as a rotation of the noninertial frame relative to the fixed one. The cause of this disagreement is the fact that, in general, the axes of the moving frame are not orthogonal for the fixed observers. When the velocity changes, the axes’ direction changes, due to both Wigner rotation and Lorentz contraction. In the present paper we take into account both of these factors. It is shown that the vectors representing various physical quantities transform in different ways in the moving reference frame. Thus, the kinematic equations describing the motion of these quantities in a fixed frame are different as well.     

一种鲁棒的目标跟踪方法

针对传统特征光流场跟踪方法中由于误差积累和错误匹配而导致的特征点丢失问题,基于一种新的Harris-SIFT特征点表示方法,提出基于预测帧与关键帧的算法框架,实现了光流场运动估计与局部特征识别相结合的目标跟踪方法.预测帧利用塔式分解和递归算法计算特征点的光流场运动矢量,使用运动矢量直方图获取目标的运动矢量,并剔除误匹配点;当特征点数量小于5个时,关键帧使用Harris-SIFT特征点进行局部特征匹配,利用仿射模型对目标精确定位及姿态修正.实验结果表明,本方法对视频序列中的纹理特征目标跟踪的鲁棒性较好,在背景复杂、目标遮挡或暂时丢失情况下,仍可以继续完成目标的可靠跟踪.     

一种鲁棒的目标跟踪方法

针对传统特征光流场跟踪方法中由于误差积累和错误匹配而导致的特征点丢失问题,基于一种新的Harris-SIFT特征点表示方法,提出基于预测帧与关键帧的算法框架,实现了光流场运动估计与局部特征识别相结合的目标跟踪方法.预测帧利用塔式分解和递归算法计算特征点的光流场运动矢量,使用运动矢量直方图获取目标的运动矢量,并剔除误匹配点;当特征点数量小于5个时,关键帧使用Harris-SIFT特征点进行局部特征匹配,利用仿射模型对目标精确定位及姿态修正.实验结果表明,本方法对视频序列中的纹理特征目标跟踪的鲁棒性较好,在背景复杂、目标遮挡或暂时丢失情况下,仍可以继续完成目标的可靠跟踪.     

基于特征点跟踪的数字稳像算法

针对图像平移和旋转的数字电子稳像问题,提出了一种新的稳像算法。采用K-R法(Kitchen-Rosenfeld)提取图像特征,根据Similarity变换模型进行运动估计,将图像平移和旋转模型转化为仅绕中心点旋转的单一旋转模型,降低了模型的复杂度和计算量。实验证明,这种算法具有很好的稳像效果。     

Landau electron in a rotating environment: A general factorization of time evolution

For the Landau problem with a rotating magnetic field and a confining potential in the (changing) direction of the field, we derive a general factorization of the time evolution operator that includes the adiabatic factorization as a special case. The confining potential is assumed to be of a general form and it can correspond to nonlinear Heisenberg equations of motion. The rotation operator associated with the solid angle Berry phase is used to transform the problem to a rotating reference frame. In the rotating reference frame, we derive a natural factorization of the time evolution operator by recognizing the crucial role played by a gauge transformation. The major complexity of the problem arises from the coupling between motion in the direction of the magnetic field and motion perpendicular to the field. In the factorization, this complexity is consolidated into a single operator which approaches the identity operator when the potential confines the particle sufficiently close to a rotating plane perpendicular to the magnetic field. The structure of this operator is clarified by deriving an expression for its generating Hamiltonian. The adiabatic limit and non-adiabatic effects follow as consequences of the general factorization which are clarified using the magnetic translation concept.     

当动系为平面运动时的加速度合成定理

本文给出动系是平面运动时的加速度合成定理。它包含了动系是平动和定轴转动两种情况。并给出数值算例。     

Improving motion estimation by accounting for local image distortion

Cardiac elastography is a useful diagnostic technique for detection of heart function abnormalities, based on analysis of echocardiograms. The analysis of the regional heart motion allows assessing the extent of myocardial ischemia and infarction. In this paper, a new two-stage algorithm for cardiac motion estimation is proposed, where the data is taken from a sequence of 2D echocardiograms. The method combines the advantages of block-matching and optical flow techniques. The first stage employs a standard block-matching algorithm (sum of absolute differences) to provide a displacement estimate with accuracy of up to one pixel. At the second stage, this estimate is corrected by estimating the parameters of a local image transform within a test window. The parameters of the image transform are estimated in the least-square sense. In order to account for typical heart motions, like contraction/expansion, translation and rotation, a local affine model is assumed within the test window. The accuracy of the new algorithm is evaluated using a sequence of 500 grayscale B-mode images, which are generated as distorted, but known copies of an original ROI, taken from a real echocardiogram. The accuracy of the motion estimation is expressed in terms of errors: maximum absolute error, root-mean-square error, average error and standard deviation. The errors of the proposed algorithm are compared with these of the known block-matching technique with cross-correlation and interpolation in the sub-pixel space. Statistical analysis of the errors shows that the proposed algorithm provides more accurate estimates of the heart motion than the cross-correlation technique with interpolation in the sub-pixel space.     

Improving motion estimation by accounting for local image distortion

Cardiac elastography is a useful diagnostic technique for detection of heart function abnormalities, based on analysis of echocardiograms. The analysis of the regional heart motion allows assessing the extent of myocardial ischemia and infarction. In this paper, a new two-stage algorithm for cardiac motion estimation is proposed, where the data is taken from a sequence of 2D echocardiograms. The method combines the advantages of block-matching and optical flow techniques. The first stage employs a standard block-matching algorithm (sum of absolute differences) to provide a displacement estimate with accuracy of up to one pixel. At the second stage, this estimate is corrected by estimating the parameters of a local image transform within a test window. The parameters of the image transform are estimated in the least-square sense. In order to account for typical heart motions, like contraction/expansion, translation and rotation, a local affine model is assumed within the test window. The accuracy of the new algorithm is evaluated using a sequence of 500 grayscale B-mode images, which are generated as distorted, but known copies of an original ROI, taken from a real echocardiogram. The accuracy of the motion estimation is expressed in terms of errors: maximum absolute error, root-mean-square error, average error and standard deviation. The errors of the proposed algorithm are compared with these of the known block-matching technique with cross-correlation and interpolation in the sub-pixel space. Statistical analysis of the errors shows that the proposed algorithm provides more accurate estimates of the heart motion than the cross-correlation technique with interpolation in the sub-pixel space.