Savitzky-Golay滤波和L1范数优化相结合的视频稳像算法

近年来,越来越多的人利用手持设备拍摄视频表达自己或社会问题等,这类视频常存在抖动现象,如何消除视频序列中的随机抖动成为近年来的研究热点。本文面向视频去抖动应用场景,提出了一种Savitzky-Golay滤波和L1范数优化相结合的视频稳像算法。首先基于帧间配对特征点进行运动估计;然后利用Savitzky-Golay滤波算法进行路径平滑处理,去除运动路径中的高频抖动分量,避免了噪声对后续生成稳定路径的影响;进一步,对平滑后的有效路径信息进行基于优化的L1范数优化算法拟合,得到最终的稳定路径;最后对原始视频图像序列进行运动补偿,生成稳定的图像序列。对比实验结果表明,本文方法稳像效果优于原始L1范数优化算法,算法速度快于目前效果较好的二维稳像算法。      

快速平滑点特征轨迹电子稳像

为了提高机载成像系统输出视频的图像质量,提出了一种快速平滑点特征轨迹的稳像算法。以消除全局运动估计的帧间匹配累积全局运动、实现长时快速稳像为目的,建立有别于传统实时稳像模式的系统框架。首先采用SURF算法从原始的抖动视频中提取不稳定的特征点;其次利用Delaunay三角剖分算法判断特征点的邻接性,生成点特征轨迹;再次采用Kalman 滤波器对不稳视频中得到的点特征轨迹进行滤波处理,得到平滑的点特征轨迹;最后由原始点特征轨迹和平滑点特征轨迹估算出直接需要补偿的全局运动矢量。实验结果表明:该方法不仅能够实时处理失稳航摄视频,有效改善机载成像系统的图像质量,而且能够估计出相互独立的帧间全局运动矢量,可以应用于需要长时间稳像的场合。     

Adaptive postprocessing algorithms for low bit rate video signals

A new adaptive postprocessing algorithm to enhance the quality of a noisy video sequence is presented. The algorithm recognizes that the visibility of noise depends on local signal characteristics. It therefore classifies the video signal into different classes and uses separate nonlinear filters matched to each class. The most general version of the algorithm employs motion-compensated frame averaging to improve picture quality in a first stage. A classification algorithm subsequently divides subblocks of pixels in the averaged frame into four classes: edge, smooth, nonsmooth with motion and nonsmooth without motion. Spatial algorithms that perform multilevel median filtering, double median filtering, and median filtering are used for pixels belonging to edge, smooth, and nonsmooth with motion categories. Pixels in the nonsmooth, unmoving category are left unfiltered to preserve corresponding image texture. In a simpler version of this four-class system, the motion cues and motion-compensated frame averaging are eliminated, and the purely spatial filtering is based on a three-class algorithm. When used at the output of a 3-D subband coder at 384 kbps, the spatial postfilter was shown to provide a consistent gain in subjectively evaluated picture quality. Twenty-five viewers participated in an experiment involving three coded sequences. In a pairwise comparison of postfiltered and unfiltered sequences, the postfiltered version was judged to be better in 63 out of 75 instances.     

Global point tracking based panoramic image stabilization system

A novel image stabilization system is presented,which consists of a global feature point tracking based motion estimation,a Kalman filtering based motion smoothing and an image mosaic based panoramic compensation.The global motion is estimated using feature point matching and iteration with the least-square method.Then,the Kalman filter is applied to smooth the original motion vectors to effectively alleviate unwanted camera vibrations and follow the intentional camera scan.Lastly,the loss information of im...     

一种结合遗传算法和钻石搜索的多模式快速运动估计方法

为了解决视频编码中运动矢量搜索精度与速度的矛盾,本文提出了一种基于遗传算法(GA)和钻石搜索(DS)的多模式快速运动估计方法——MMS算法.它以图像序列的时空预测矢量作为图像活动剧烈程度的判据,自适应选择搜索模式.针对平缓运动类型使用快速的DS搜索模式,针对剧烈运动类型使用GA/DS联合搜索模式.与现有的次优解快速算法相比,MMS有效地解决了在大运动矢量情况下编码器性能下降的问题,可以从整体上提升编码器的性能,接近理想的全搜索法的结果;与其它直接利用GA进行全局优化的方法相比,MMS利用DS配合GA实现加速收敛.此外,通过引入多模式处理的概念,在保证搜索精度的同时,充分发挥了次优解算法的效率,整体编码速度与DS等快速算法的结果十分接近.这一方法为有效地解决运动估计中的矛盾问题提供了一个新的处理框架.实验结果验证了算法的性能.     

Video denoising, deblocking, and enhancement through separable 4-d nonlocal spatiotemporal transforms

We propose a powerful video filtering algorithm that exploits temporal and spatial redundancy characterizing natural video sequences. The algorithm implements the paradigm of nonlocal grouping and collaborative filtering, where a higher dimensional transform-domain representation of the observations is leveraged to enforce sparsity, and thus regularize the data: 3-D spatiotemporal volumes are constructed by tracking blocks along trajectories defined by the motion vectors. Mutually similar volumes are then grouped together by stacking them along an additional fourth dimension, thus producing a 4-D structure, termed group, where different types of data correlation exist along the different dimensions: local correlation along the two dimensions of the blocks, temporal correlation along the motion trajectories, and nonlocal spatial correlation (i.e., self-similarity) along the fourth dimension of the group. Collaborative filtering is then realized by transforming each group through a decorrelating 4-D separable transform and then by shrinkage and inverse transformation. In this way, the collaborative filtering provides estimates for each volume stacked in the group, which are then returned and adaptively aggregated to their original positions in the video. The proposed filtering procedure addresses several video processing applications, such as denoising, deblocking, and enhancement of both grayscale and color data. Experimental results prove the effectiveness of our method in terms of both subjective and objective visual quality, and show that it outperforms the state of the art in video denoising.     

压缩视频码流中运动矢量的提取

MPEG-2标准中，码流中的运动矢量代表了编码时相邻图像间的运动信息。提出一种针对MPEG-2中不同运动补偿方法的运动矢量两步法提取方法：首先选择性提取解码视频码流中的运动矢量，然后其进行滤波系数调整。分析结果表明经过简单滤波后的运动矢量符合显示时相邻图像间的运动信息。提取出的运动矢量经简单处理后可以直接应用于帧率增强。生成的图像保持了运动连贯性，显示效果良好，方法复杂度低。     

适于微小型飞行器平台的稳像算法

为解决微小型飞行器由于机械振动、气流扰动等原因引起的图像高频抖动,设计了一种适用于该平台的稳像算法。采用基于生物视觉的匹配方法估计帧间运动矢量,建立了图像参数传递的数学模型;结合微小型飞行器的运动特点,提出了带约束的交互式多模型卡尔曼滤波方法(CIMMKF),针对绝对帧位移曲线和旋转角度滤波,引入硬约束条件减小模型不准确性产生的误差,再通过软约束平滑硬约束带来的局部跳变求得图像合适的校正量。最后,给出了一种新颖的微小型飞行器平台稳像算法性能的评估方法。实验结果表明,该稳像算法能够适应飞行器多种状态的交替改变,有效减小滤波延迟,去除高频抖动,保留主动运动,使稳定后图像质量满足观察要求,具有图像信息保留程度高、速度快的特点,尤其适用于微小型飞行器实时视频稳定。     

H.264压缩域中mean-shift聚类运动目标分割算法

针对视频监控、检索,提出了一 种在H.264压缩域下运动对象分 割的新算法。根据实际监控应用特点,算法主要利用H.264码流中提 取的运动矢量(MV)、分块尺寸信息对原始的MV场(MVF)进 行可靠性分析以及空间滤波、mean-shift聚类处理,从而得到可信度较高的 MVF场。首先,从H.264码流中获取原 始MV 并进行归一化处理,同时对原始MVF进行两步空间滤波;其次,对不同的块大 小分配相应的权值作为每个样本 的权重系数,将处理之后的MVF作为样本空间,利用mean-shift聚类获取 真实的MVF;最后,根据可靠的MVF 标记运动目标。实验结果表明,本文提出的mean-shift聚类运动目标分割算 法可以获得有效并可靠的分割结果。     

基于MPEG的全局运动估计技术

提出了一种基于MPEG（moving Picture experts group)的全局运动估计技术;利用四参数模型来描述由摄像机运动带来的全局运动,再利用基于块匹配的算法来求解4个参数,进而作相应的运动补偿,由于MPEG视频流中的运动矢量蕴含了最优块匹配的信息,所以不需要对MPEG视频流完全解码,即可进行参数估计,实验证明该算法速度较快,并能通过运动补偿排除由摄像机运动造成的伪运动区域的干扰。     

基于运动判断的动态帧间滤波方法

提出了一种基于运动判断的自适应动态帧间滤波方法，利用不同帧间图像的共有图像信息部分进行累加滤波，同时又可避免因为子像素的位移引入的混叠效应，从而达到最大限度提高图像质量的目的。给出了该滤波方法的噪声方差方程，并利用该方法对采集的视频图像数据进行了处理，讨论了不同阈值下的滤波效果。     

基于双域滤波的三维块匹配视频去噪算法

将双域滤波和三维块匹配算法相结合,充分利用了视频序列的时空频域的相关性,进行视频去噪。将三维块匹配视频去噪算法中小波阈值去噪得到的图像替换原引导图像,抑制了双域滤波算法基本层不平滑的问题。参考残差矩阵的能量谱对短时傅里叶变换系数进行阈值收缩,减少细节层的噪声残留。实验结果表明,本算法有效地解决了双域滤波算法的基本层不平滑、细节层有噪声残留的问题,主观和客观对比也表明,本算法有着较为优秀的去噪效果。     

HOS-based image sequence noise removal.

In this paper, a new spatiotemporal filtering scheme is described for noise reduction in video sequences. For this purpose, the scheme processes each group of three consecutive sequence frames in two steps: 1) estimate motion between frames and 2) use motion vectors to get the final denoised current frame. A family of adaptive spatiotemporal L-filters is applied. A recursive implementation of these filters is used and compared with its nonrecursive counterpart. The motion trajectories are obtained recursively by a region-recursive estimation method. Both motion parameters and filter weights are computed by minimizing the kurtosis of error instead of mean squared error. Using the kurtosis in the algorithms adaptation is appropriate in the presence of mixed and impulsive noises. The filter performance is evaluated by considering different types of video sequences. Simulations show marked improvement in visual quality and SNRI measures cost as well as compared to those reported in literature.     

基于运动矢量多级分析的视频全局运动估计

基于运动矢量场的视频全局运动估计相较于基于像素的估计方法具有较低的计算复杂度,因而广泛应用于视频分割及视频压缩等领域中。然而噪声和前景目标等外点区域的存在,降低了全局运动估计的准确性。为了提高全局运动估计的准确度,该文提出一种基于运动矢量多级分析的全局运动估计算法,该算法根据局部运动与全局运动的运动特性差异自适应地滤除前景目标区域,由邻域矢量间相似性度量检测出纹理平滑周期区域,最后滤除孤立的噪声区域,由滤波得到的内点区域求解全局运动参数。实验结果表明,该方法能有效地滤除外点区域,提高全局运动估计的准确性。     

An illumination invariant algorithm for subpixel accuracy image stabilization and its effect on MPEG-2 video compression

This paper offers a fast and simple near-closed-form solution for the least mean-squared-error (LMSE) estimation of the frame-to-frame global subpixel motion in an unsteady image sequence. The offered near-closed-form solution achieves unlimited subpixel accuracy by always employing a small and fixed number of computations, independent of the desired subpixel accuracy. The algorithm is designed so that it is insensitive to frame-to-frame intensity variations, which is a distinctive feature of the method. Experimental results demonstrate the superiority of the proposed method to the spatio-temporal differentiation and surface fitting algorithms under different illumination conditions. This paper furthermore discusses the effect of the proposed image stabilization algorithm on the performance of MPEG-2 video compression. We report that removal of global motion down to subpixel accuracy from an unsteady video improves MPEG-2 compression performance significantly (by at least 1 dB for all frame types), in spite of the fact that motion vectors are differentially encoded in MPEG-2. This result is supported via experimental results using the proposed subpixel registration algorithm and an analysis of the macroblock coding preferences accepted in MPEG-2.     

一种应用形态学滤波的视频对象时空分割算法

从视频图像中提取视频对象是基于内容的视频编码中的一项关键技术。本文提出了一种基于帧间运动信息和形态学滤波的视频对象时空分割算法。该算法首先利用分块高阶统计算法和基于最大类间方差的阈值算法得到目标的运动区域检测模板。然后,用基于交变序列重建滤波的分水岭算法得到前景对象的精确边缘。最后,用区域基时空融合方法将运动检测和形态学分割结果结合起来提取出视频对象。实验结果表明,本文算法能避免区域合并有效提取出具有精确边缘的视频对象,主客观分割效果理想。     

Motion information scalability for MC-EZBC

Interframe wavelet video coding algorithm has received much attention lately because of its high coding efficiency and flexible temporal and spatial scalability. It combines the motion-compensated temporal filtering technique together with the wavelet embedded zeroblock coding technique. Blending these two techniques in a nice manner enables it achieving three types of scalability: SNR, temporal and spatial, in one single bit-stream. However, the performance of the interframe wavelet video coding at low bit-rates is less satisfactory because the bit-rate of the non-scalable motion information is too high.In this paper, we extend the framework of Motion Compensated Embedded Zero Block Coding (MC-EZBC) proposed by RPI (Improved MC-EZBC with Quarter-pixel Motion Vectors, ISO/IEC/JTC1 SC29/WG11 doc. No. m8366, Fairfax, May 2002). Our major contribution is splitting the motion information in MC-EZBC into a few layers. At very low bit-rates only the “coarse” motion vectors are transmitted. Therefore, we are able to produce compressed bit streams at a lower bit-rate and the associated picture quality is significantly better than that of the original scheme. The overhead due to motion information partitioning is negligible at higher rates. Hence, the rate-distortion performance at high rates is about the same as that of the original scheme. In addition, a Hilbert curve scan order is proposed to increase the efficiency up to 5% in encoding the differential motion vectors.     

Feature-based block matching algorithm using integral projections

A fast block matching algorithm using integral projections for the estimation of motion vectors is proposed. In the calculation of the cost function, the proposed algorithm using integral projections as features can reduce the motion estimation computations by a factor of 3 by converting the operations of two-dimensional arrays into those of one-dimensional arrays. Also, the proposed method is not sensitive to additive noise. Simulation results based on the sequence of original image and the sequences of noisy image with different standard deviation are presented.<>     

基于运动估计的模糊图像盲复原

针对由于物体在不同方向移动速度不同导致退化的CCD图像空间移变特性的运动模糊,提出一种基于隔行扫描空间移变特性的CCD图像恢复方案。方法主要依据隔行扫描图像分为上下两场的特性,借助插值算法将单幅图像按场拆分为两幅存在时间差的图像,然后引入视频处理中基于图像序列确定运动向量的块匹配算法,估计模糊参数,从而实现应用图像序列算法估计单幅图像点扩散函数,最后利用最小二成方算法进行复原,其中引入了图像定向信息测量操作保持边缘。仿真实验结果表明,本文算法比以往算法具有大的信噪比(SNR)和图像保真度(IMFID),以及小的平均绝对差(ABSDIFF)和均方误差(MSE)。     

A multistage motion vector processing method for motion-compensated frame interpolation.

In this paper, a novel, low-complexity motion vector processing algorithm at the decoder is proposed for motion-compensated frame interpolation or frame rate up-conversion. We address the problems of having broken edges and deformed structures in an interpolated frame by hierarchically refining motion vectors on different block sizes. Our method explicitly considers the reliability of each received motion vector and has the capability of preserving the structure information. This is achieved by analyzing the distribution of residual energies and effectively merging blocks that have unreliable motion vectors. The motion vector reliability information is also used as a prior knowledge in motion vector refinement using a constrained vector median filter to avoid choosing identical unreliable one. We also propose using chrominance information in our method. Experimental results show that the proposed scheme has better visual quality and is also robust, even in video sequences with complex scenes and fast motion.