Markov random field model-based edge-directed image interpolation.

This paper presents an edge-directed image interpolation algorithm. In the proposed algorithm, the edge directions are implicitly estimated with a statistical-based approach. In opposite to explicit edge directions, the local edge directions are indicated by length-16 weighting vectors. Implicitly, the weighting vectors are used to formulate geometric regularity (GR) constraint (smoothness along edges and sharpness across edges) and the GR constraint is imposed on the interpolated image through the Markov random field (MRF) model. Furthermore, under the maximum a posteriori-MRF framework, the desired interpolated image corresponds to the minimal energy state of a 2-D random field given the low-resolution image. Simulated annealing methods are used to search for the minimal energy state from the state space. To lower the computational complexity of MRF, a single-pass implementation is designed, which performs nearly as well as the iterative optimization. Simulation results show that the proposed MRF model-based edge-directed interpolation method produces edges with strong geometric regularity. Compared to traditional methods and other edge-directed interpolation methods, the proposed method improves the subjective quality of the interpolated edges while maintaining a high PSNR level.     

基于统计特征的彩色图像快速插值方法

本文首先阐述了基于统计特征的图像插值方法,该方法通过提取待插入像素所在区域的协方差矩阵和协方差向量,得出适应于边缘位置和方向的插值权重.为了把基于统计特征的图像插值方法应用于彩色图像插值领域,本文提出了以下措施以提高计算速度:仅对Y图像估计插值权重,并同时应用到R、G、B三个分量的插值;对边缘像素应用基于统计特征的图像插值方法,而对非边缘像素应用简单的双线性插值,即混合图像插值方法.这些措施提高了计算速度,并保证了图像质量.实验表明了该算法在计算速度和插值图像质量方面的有效性.     

Polyphase back-projection filtering for image resolution enhancement

The method for reconstruction and restoration of super-resolution images from sets of low-resolution images presented is an extension of the algorithm proposed by Irani and Peleg (1991). After estimating the projective transformation parameters between the image sequence frames, the observed data are transformed into a sequence with only quantised sub-pixel translations. The super-resolution reconstruction is an iterative process, in which a high-resolution image is initialised and iteratively improved. The improvement is achieved by back-projecting the errors between the translated low-resolution images and the respective images obtained by simulating the imaging system. The imaging system's point-spread function (PSF) and the back-projection function are first estimated with a resolution higher than that of the super-resolution image. The two functions are then decimated so that two banks of polyphase filters are obtained. The use of the polyphase filters allows exploitation of the input data without any smoothing and/or interpolation operations. The presented experimental results show that the resolution improvement is better than the results obtained with Irani and Peleg's algorithm.     

一种边缘定向平滑图像插值算法

针对传统图像放大算法边缘处理效果较差,自适应图像插值方法存在高计算复杂度的问题,该文提出一种有效增强图像边缘轮廓的插值放大算法。该算法结合边缘定向平滑滤波器和双线性插值的特点,使得图像在平坦和非平坦区域均能取得理想效果。仿真测试结果表明,与基于统计特征的自适应插值算法相比,该文提出算法能显著提高插值速度,平均运行时间降低8.33 s;与双三次插值算法相比,图像峰值信噪比平均增加0.30 dB。     

Selective Data Pruning-Based Compression Using High-Order Edge-Directed Interpolation

This paper proposes a selective data pruning-based compression scheme to improve the rate-distortion relation of compressed images and video sequences. The original frames are pruned to a smaller size before compression. After decoding, they are interpolated back to their original size by an edge-directed interpolation method. The data pruning phase is optimized to obtain the minimal distortion in the interpolation phase. Furthermore, a novel high-order interpolation is proposed to adapt the interpolation to several edge directions in the current frame. This high-order filtering uses more surrounding pixels in the frame than the fourth-order edge-directed method and it is more robust. The algorithm is also considered for multiframe-based interpolation by using spatio-temporally surrounding pixels coming from the previous frame. Simulation results are shown for both image interpolation and coding applications to validate the effectiveness of the proposed methods.      

A wavelet-based interpolation-restoration method for superresolution (wavelet superresolution)

Superresolution produces high-quality, high-resolution images from a set of degraded, low-resolution images where relative frame-to-frame motions provide different looks at the scene. Superresolution translates data temporal bandwith into enhanced spatial resolution. If considered together on a reference grid, given low-resolution data are nonuniformly sampled. However, data from each frame are sampled regularly on a rectangular grid. This special type of nonuniform sampling is called interlaced sampling. We propose a new wavelet-based interpolation-restoration algorithm for superresolution. Our efficient wavelet interpolation technique takes advantage of the regularity and structure inherent in interlaced data, thereby significantly reducing the computational burden. We present one- and two-dimensional superresolution experiments to demonstrate the effectiveness of our algorithm.This work was supported in part by the National Science Foundartion Grant CCR-9984246.     

Real-time artifact-free image upscaling

The problem of creating artifact-free upscaled images appearing sharp and natural to the human observer is probably more interesting and less trivial than it may appear. The solution to the problem, often referred to also as "single-image super-resolution," is related both to the statistical relationship between low-resolution and high-resolution image sampling and to the human perception of image quality. In many practical applications, simple linear or cubic interpolation algorithms are applied for this task, but the results obtained are not really satisfactory, being affected by relevant artifacts like blurring and jaggies. Several methods have been proposed to obtain better results, involving simple heuristics, edge modeling, or statistical learning. The most powerful ones, however, present a high computational complexity and are not suitable for real-time applications, while fast methods, even if edge adaptive, are not able to provide artifacts-free images. In this paper, we describe a new upscaling method (iterative curvature-based interpolation) based on a two-step grid filling and an iterative correction of the interpolated pixels obtained by minimizing an objective function depending on the second-order directional derivatives of the image intensity. We show that the constraints used to derive the function are related with those applied in another well-known interpolation method, providing good results but computationally heavy (i.e., new edge-directed interpolation (NEDI). The high quality of the images enlarged with the new method is demonstrated with objective and subjective tests, while the computation time is reduced of one to two orders of magnitude with respect to NEDI so that we were able, using a graphics processing unit implementation based on the nVidia Compute Unified Device Architecture technology, to obtain real-time performances.     

区域指导的彩色图像插值

本文提出了一种针对彩色图像的区域指导的插值算法.该方法首先从低分辨率图像中计算象素沿梯度方向的1、2阶方向导数,然后把彩色图像划分为常数、线性和非线性区域,并对不同的区域施用不同复杂度的滤波器.实验结果表明,与传统的双线性插值相比,该方法可以明显改进插值图像的质量;重建所需的时间也比单纯使用复杂滤波器的时间减少很多,并且可以实现彩色图像的任意放大.     

Image super-resolution via sparse representation over multiple learned dictionaries based on edge sharpness

A new algorithm for single-image super-resolution based on selective sparse representation over a set of coupled dictionary pairs is proposed. Patch sharpness measure for high- and low-resolution patch pairs defined via the magnitude of the gradient operator is shown to be approximately invariant to the patch resolution. This measure is employed in the training stage for clustering the training patch pairs and in the reconstruction stage for model selection. For each cluster, a pair of low- and high-resolution dictionaries is learned. In the reconstruction stage, the sharpness measure of a low-resolution patch is used to select the cluster it belongs to. The sparse coding coefficients of the patch over the selected low-resolution cluster dictionary are calculated. The underlying high-resolution patch is reconstructed by multiplying the high-resolution cluster dictionary with the calculated coefficients. The performance of the proposed algorithm is tested over a set of natural images. PSNR and SSIM results show that the proposed algorithm is competitive with the state-of-the-art super-resolution algorithms. In particular, it significantly out-performs the state-of-the-art algorithms for images with sharp edges and corners. Visual comparison results also support the quantitative results.     

L1范数的图像超分辨率重建改进算法

介绍了超分辨率图像重建的数学模型和基于L1范数的超分辨率重建算法。针对在所观察到的低分辨率图像不足情况下的超分辨率重建,在L1范数重建算法框架下,提出了一种新的代价方程,在其中增加了关于丢失的低分辨率观察信息的保真度项和正则化项。该方法同时对高分辨率图像和丢失的观察信息进行迭代估计,并利用交替最小方法求解。实验结果表明,在获取低分辨率图像较少的情况下,提出的算法能够有效地改进重建的结果。     

Super-resolution reconstruction based on linear interpolation of wavelet coefficients

High resolution image reconstruction is an image process to reconstruct a high resolution image from a set of blurred, degraded and shifted low resolution images. In this paper, the reconstruction problem is treated as a function approximation. We use linear interpolation to build up an algorithm to obtain the relationship between the detail coefficients in wavelet subbands and the set of low resolution images. We use Haar wavelet as an example and establish the connection between the Haar wavelet subband and the low resolution images. Experiments show that we can use just 3 low resolution images to obtain a high resolution image which has better quality than Tikhonov least-squares approach and Chan et al. Algorithm 3 in low noise cases. We also propose an error correction extension for our method which can lead to very good results even in noisy cases. Moreover, our approach is very simple to implement and very efficient.     

一种快速的亚像素图像配准算法

图像超分辨率重建是在现有红外探测器基础上提升空间分辨率的一种有效方法。超分辨率图像重建是利用一组相互之间存在亚像素位移的低分辨率图像构造出一幅高分辨率的图像,快速、高精度估计图像间的位移是其关键技术之一。提出了一种用于超分辨率重建的亚像素配准算法,算法由特征检测、像素级配准和亚像素级配准三个处理过程组成。在特征检测过程,首先采用梯度算子对图像进行边缘检测,然后对边缘点进行角点预检测,排除非角点像素点,之后再进行Harris角点检测,大大减少了计算量;在像素级配准过程,用NCC算法进行像素级配准,用统计方法去除误匹配点对;在亚像素级配准过程,先对像素级匹配点的邻域进行插值放大,再进行亚像素匹配,误匹配点剔除,相对偏移量计算。对提出的算法进行了仿真实验,结果显示本算法的速度较类似算法速度有较大的提高。     

基于聚类和高斯过程回归的超分辨率重建

在样本学习的思想框架下,针对图像超分辨率问题的研究,提出了数据聚类和高斯过程回归相结合的解决方法.使用K-means对数据进行聚类,在各类中利用高斯过程回归对样本库中高低分辨率图像之间的对应关系进行学习.根据得到的学习模型对需要处理的低分辨率图像所对应的高分辨率图像进行预测,有效地利用了高低分辨率图像之间的统计特性.实验结果表明该方法可以较好地改善超分辨率重建效果.     

An improved image interpolation algorithm

The paper sets forth an improved edge-directed image interpolation algorithm with low time complexity which is the combination of Newton’s method and edge-directed method. It first partitions images into homogeneous areas and edge areas by setting a preset threshold value based on the local structure characteristics, and then specified algorithms are assigned to interpolate each classified areas, respectively. In this way, it achieves the goals of real-time interpolation and good subjective quality. The interpolated images have higher peak signal noise ratios (PSNR) and better visual effects using proposed method than that of using other algorithms referred to in this paper. Experimental results show that proposed method is highly performed in image interpolation.

一种自适应协方差的低复杂度图像放大方法

对于图像放大技术而言,重要的就是要权衡到图像质量以及计算复杂度.传统的基于线性或三次样条插值的方法会带来图像模糊和锯齿边缘等失真,为了解决这一问题,人们提出一种基于迭代和学习的算法,但是这种方法带来了很高的计算复杂度.综合以上几点本文提出了一种基于自适应协方差的图像放大方法(adaptive covariance-based edge diffusion,ACED).该方法能很好地权衡图像放大性能和复杂度之间的关系.在这种方法中,提出了一种联合边缘判别准则,并自适应选择扩散模板来估计局部协方差系数,以高效的减少图像放大带来的失真.实验结果表明,所提出的方法在主观质量和客观质量上都有很大的提升,同时也具有较低的计算复杂度.     

基于压缩卷积神经网络的图像超分辨率算法

为了有效提高深度图像的分辨率,文中借鉴经典SqueezeNet网络结构,提出一种基于Fire Module的卷积神经网络模型。该算法实现了直接从低分辨率图像到高分辨率图像的映射和转化,其中Fire Module作为网络的非线性映射模块,在减少参数的同时可学习图像的深层特征。为了避免插值预处理,在网络的输出层引入反卷积层,实现3倍上采样和高分辨率图像的输出。实验表明,采用该基于Fire Module的卷积神经网络模型的反卷积算法得到的超分辨率图像细节更加丰富,客观指标PSNR值和SSIM值的评价也明显优于其他算法。     

松弛耦合非负矩阵分解的低分辨率人脸识别算法

针对实际监控场景中经常遇到的人脸图像分辨率较低的问题,本文提出了一种利用耦合非负矩阵分解并保持系数松弛的低分辨率人脸识别算法（Relaxed Coupled Nonnegative Matrix Factorization,后文简称RCNMF）。首先,对高低分辨率人脸图像进行非负矩阵矩阵分解（nonnegative matrix factorization,后文简称NMF）,在分解的同时保持组合系数近似一致,从而得到高低分辨率图像的基矩阵。然后,通过低分辨率图像的基矩阵提取训练和测试样本的特征。最后进行识别。实验结果验证了与其他几种基于耦合映射的低分辨率人脸识别方法相比,RCNMF算法的识别性能更好。同时通过实验验证了RCNMF算法的收敛性。      

Super-Resolution Based on Fast Registration and Maximum a Posteriori Reconstruction

In this paper, we propose a maximum a posteriori framework for the super-resolution problem, i.e., reconstructing high-resolution images from shifted, rotated, low-resolution degraded observations. The main contributions of this work are two; first, the use of a new locally adaptive edge preserving prior for the super-resolution problem. Second an efficient two-step reconstruction methodology that includes first an initial registration using only the low-resolution degraded observations. This is followed by a fast iterative algorithm implemented in the discrete Fourier transform domain in which the restoration, interpolation and the registration subtasks of this problem are preformed simultaneously. We present examples with both synthetic and real data that demonstrate the advantages of the proposed framework.     

基于配准的肺4D-CT图像超分辨率重建

肺4D-CT数据在肺癌治疗中有重要意义.但肺4D-CT数据纵向(Z方向)分辨率低,为显示正确比例图像需进行插值运算,由此带来图像的模糊.本文提出了一种基于Active Demons配准的超分辨率重建技术来提高肺4D-CT图像分辨率.我们将不同相位同一位置的低分辨率图像视为不同"帧"图像.首先采用Active Demons配准方法得到不同"帧"图像之间的运动估计;而后采用凸集投影(Projection Onto Convex Set,POCS)超分辨率算法重建高分辨率肺图像.实验结果表明,与三次样条插值和反投影方法相比较,我们的方法能得到更清晰的肺图像,明显增强图像结构.     

Super-resolution based on fast registration and maximum a posteriori reconstruction.

In this paper, we propose a maximum a posteriori ramework for the super-resolution problem, i.e., reconstructing high-resolution images from shifted, rotated, low-resolution degraded observations. The main contributions of this work are two; first, the use of a new locally adaptive edge preserving prior for the super-resolution problem. Second an efficient two-step reconstruction methodology that includes first an initial registration using only the low-resolution degraded observations. This is followed by a fast iterative algorithm implemented in the discrete Fourier transform domain in which the restoration, interpolation and the registration subtasks of this problem are preformed simultaneously. We present examples with both synthetic and real data that demonstrate the advantages of the proposed framework.