一种改进的POCS算法的超分辨率图像重建

图像超分辨率是指从一组模糊的低分辨率图像重建一帧清晰的高分辨率图像的过程.从经典的基于凸集投影POCS(projection onto convex set)的超分辨率图像重建算法出发,分析重建后高分辨率图像边缘模糊的成因,提出了一种基于保留边缘信息的POCS超分辨率图像重建算法.实验结果表明该方法能够明显地提高重建图像的质量.     

一种改善被动毫米波重建图像质量的方法

凸集投影(POCS)算法是一种广泛使用的超分辨率图像重建方法.针对常规POCS算法收敛速度慢、存在边缘震荡效应的问题,论文结合被动毫米波图像降质模型,提出了一种用于被动毫米波图像超分辨率重建方法.该方法有效利用图像的边缘信息,根据不同的区域选择相应的松弛算子,同时建立边缘约束集来保证边缘图像的尖锐性.实验结果表明.在有效消除边缘震荡效应的同时提高了收敛速度,适用于被动毫米波图像的超分辨率处理.     

基于边缘保持的航拍图像凸集投影超分辨率重建算法

针对传统的凸集投影(POCS)超分辨率图像重建算法经常出现的边缘模糊问题,在传统POCS算法原理基础上,文章使用基于梯度插值的算法生成POCS重建的初始估计图像,然后对中心点为边缘像素的空间点扩散函数(PSF)进行修正,使其系数沿梯度大的方向减小,梯度小的方向保持不变。利用改进的POCS算法进行了超分辨率图像重建实验,结果表明,图像峰值信噪比由原来的27.29dB达到28.12dB。该方法有效地保持了边缘特性,改善了超分辨率图像重建质量。     

一种改善图像边缘效果的POCS超分辨率重建方法

利用图像超分辨率重建(SRR)技术可以在现有成像系统基础上提高图像空间分辨力。凸集投影(POCS)是超分辨率重建的主流方法之一。对POCS算法进行了改进,具体改进体现在两个方面:(1)用可控核回归插值图像作为POCS重建的初始估计以提高初始估计图像的质量;(2)将POCS重建中使用的点扩散函数(PSF)由高斯核改为可控核以减少重建图像的边缘振荡效应。对所提出的算法进行了仿真,实验结果显示采用本文方法重建图像的边缘效果有了明显的改善。     

使用改进的POCS算法的超分辨率图像复原

超分辨率复原技术的基本思想就是采用信号处理的方法,在改善图像质量的同时,重建成像系统截至频率外的信息。POCS（凸集投影）算法是一种广泛应用于图像超分辨率复原的方法。针对传统的POCS算法的边缘振荡效应,在分析其产生的原因．造成的影响的基础上,采用改进的POCS算法,以减少边缘振荡。采用基于小波变换模极大值的改进POCS算法进行图像超分辨率复原。实验结果表明,该方法有效的较少了复原图像的边缘振荡效应,是一种有效的图像超分辨率复原方法。     

一种改进的单帧图像超分辨率重建的高效算法

根据图像的降质模型,基于凸集投影（POCS）原理,结合降质图像模型,提出一种使用中值滤波初值处理的高效POCS单帧图像的超分辨率重建方法。计算机仿真结果表明,和双线性内插、经典POCS方法比较,改进后的该方法重建图像信噪比平均提高2．1dB和1．1dB。     

基于改进点扩散函数的遥感图像超分辨率重建

为了提高遥感图像空间域重建质量, 采用改进凸集投影(POCS)算法的点扩散函数, 提出了一种改进的POCS超分辨率重建算法。首先给出POCS算法基本原理以及具体实现步骤, 在此基础上对算法做出改进, 即对待重建的高分辨初始帧进行边缘检测, 对检测到的边缘像素点应用改进的点扩散函数(PSF), 使边缘处像素点对应的PSF水平方向与垂直方向系数依据边缘斜率变化而设置不同的权重; 最后分别采用两组数据集对改进POCS算法的有效性进行验证。结果表明, 该改进的POCS算法有效地提高了图像重建的效果, 两组实验平均绝对误差效果分别提升了0.79%和0.26%, 达到了提高图像重建质量的目的。该算法具有较好的实际应用价值。     

无源毫米波成像改进POCS超分辨率算法

在无源毫米波成像中,因为受天线孔径大小的限制而导致获取的图像分辨率低,所以必须采取有效后处理措施增强分辨率.提出了一种改进的POCS超分辨率算法,该算法结合了Wiener滤波器复原算法和凸集投影(POCS)算法的优点,使用Wiener滤波复原算法恢复图像通带内的低频分量,运用POCS算法作为主迭代过程实现频谱外推,同时保证低频分量不被破坏.实验结果表明,该算法增强了图像的分辨率,改善了收敛速度,减少了计算量,有利于无源毫米波成像超分辨率的实时实现.     

超分辨率重构

利用低分辨率的图像序列来估计高分辨率图像的方法称为超分辨率图像重建,逐步成为当前科研热点。本文通过POCS视频图像重建算法为例,阐述了超分辨率的概念应用场合及基本策略和分类,并对超分辨率的重构方法和前景进行了展望。     

超分辨率重构

利用低分辨率的图像序列来估计高分辨率图像的方法称为超分辨率图像重建,逐步成为当前科研热点.本文通过POCS视频图像重建算法为例,阐述了超分辨率的概念、应用场合及基本策略和分类,并对超分辨率的重构方法和前景进行了展望.     

超分辨率图像重构边缘振荡的高效去除算法

针对POCS超分辨率图像重构算法在放大倍数加大时存在边缘振荡效应加剧和计算复杂性上升的问题，提出了一个能同时解决这两个问题的新算法，它将放大过程分步进行，并在每步引入边界自适应约束。实验结果表明，该算法在有效消除边缘振荡效应的同时，极大地提高了超分辨率图像重构的速度。     

Theory of projection onto the narrow quantization constraint setand its application

Since the postprocessing of coded images using a priori information depends on the constraints imposed on the coded images, it is important to utilize constraints that are best suited to postprocessing techniques. Among the constraint sets, the quantization constraint set (QCS) is commonly used in the iterative algorithms that are especially based on the theory of projections onto convex sets (POCS). The converged image in the iteration is usually a boundary point of the QCS. But, we can easily conjecture that the possible location of the original image is inside the QCS. In order to obtain an image inside the QCS, we proposed a new convex constraint set, a subset of the QCS called narrow QCS (NQCS) as a substitute for the QCS. In order to demonstrate that the NQCS works better than the QCS on natural images, we present mathematical analysis with examples and simulations by reformulating the iterative algorithm of the constrained minimization problem or of the POCS using the probability theory. Since the initial image of the iteration is the centroid of the QCS, we reach a conclusion that the first iteration is enough to recover the coded image, which implies no need of any theories that guarantee the convergences.     

Optimum equalization of multicarrier systems: a unified geometricapproach

This paper presents a new iterative equalization algorithm that maximizes the capacity for discrete multitone (DMT) systems. The research modifies a previously proposed criterion and applies an appropriate transformation to map the objective function and the constraint set into a canonical region. The resulting constraint set exhibits an identifiable geometric characteristic. Using the gradient projection method in conjunction with projection onto convex sets (POCS) provides us with an iterative search algorithm that facilitates the gradient descent method. We also generalize the approach to two important subclasses of equalizers, namely linear phase and unit tap filters. We also derive a fundamental limit on the performance of the proposed approach. In comparison with the previous methods, the proposed equalization algorithm is less computationally complex and more geometrically intuitive. Simulation experiments confirm the validity of the proposed method for equalization of DMT systems     

改进边缘质量和运动估计的超分辨率图像重构

凸集投影算法（POCS）是一种广泛使用的超分辨率图像重构方法。针对传统的POCS超分辨率图像重构算法出现的边缘模糊及匹配时的局限性问题,首先利用二阶梯度检测出像素周围0°、45°、90°、135°四个边缘。在构造参考帧时采用基于梯度的插值算法,沿边缘方向进行线性插值,沿非边缘方向进行基于一阶梯度的带权插值。在运动估计时,采用SURF匹配算法,提高匹配的鲁棒性和实时性。在修正参考帧时,分别定义中心在四个边缘方向的点扩散函数（PSF）。利用完全参考图像质量评价和无参考图像质量评价分别对仿真实验和实物实验进行了评价,评价结果表明提出的算法较传统POCS算法有明显的改善。      

Study on infrared image super-resolution reconstruction based on an improved POCS algorithm

Aiming at the disadvantages of the traditional projection onto convex sets of blurry edges and lack of image details, this paper proposes an improved projection onto convex sets (POCS) method to enhance the quality of image super-resolution reconstruction (SRR). In traditional POCS method, bilinear interpolation easily blurs the image. In order to improve the initial estimation of high-resolution image (HRI) during reconstruction of POCS algorithm, the initial estimation of HRI is obtained through iterative curvature-based interpolation (ICBI) instead of bilinear interpolation. Compared with the traditional POCS algorithm, the experimental results in subjective evaluation and objective evaluation demonstrate the effectiveness of the proposed method. The visual effect is improved significantly and image detail information is preserved better.     

Removal of compression artifacts using projections onto convex setsand line process modeling

We present a new image recovery algorithm to remove, in addition to blocking, ringing artifacts from compressed images and video. This new algorithm is based on the theory of projections onto convex sets (POCS). A new family of directional smoothness constraint sets is defined based on line processes modeling of the image edge structure. The definition of these smoothness sets also takes into account the fact that the visibility of compression artifacts in an image is spatially varying. To overcome the numerical difficulty in computing the projections onto these sets, a divide-and-conquer (DAC) strategy is introduced. According to this strategy, new smoothness sets are derived such that their projections are easier to compute. The effectiveness of the proposed algorithm is demonstrated through numerical experiments using Motion Picture Expert Group based (MPEG-based) coders-decoders (codecs).