Image super-resolution by estimating the enhancement weight of self example and external missing patches

Image super-resolution (SR) is the process of generating a high-resolution (HR) image using one or more low-resolution (LR) inputs. Many SR methods have been proposed, but generating the small-scale structure of an SR image remains a challenging task. We hence propose a single-image SR algorithm that combines the benefits of both internal and external SR methods. First, we estimate the enhancement weights of each LR-HR image patch pair. Next, we multiply each patch by the estimated enhancement weight to generate an initial SR patch. We then employ a method to recover the missing information from the high-resolution patches and create that missing information to generate a final SR image. We then employ iterative back-projection to further enhance visual quality. The method is compared qualitatively and quantitatively with several state-of-the-art methods, and the experimental results indicate that the proposed framework provides high contrast and better visual quality, particularly for non-smooth texture areas.     

多尺度密集残差网络的单幅图像超分辨率重建

目的 近几年应用在单幅图像超分辨率重建上的深度学习算法都是使用单种尺度的卷积核提取低分辨率图像的特征信息,这样很容易造成细节信息的遗漏。另外,为了获得更好的图像超分辨率重建效果,网络模型也不断被加深,伴随而来的梯度消失问题会使得训练时间延长,难度加大。针对当前存在的超分辨率重建中的问题,本文结合GoogleNet思想、残差网络思想和密集型卷积网络思想,提出一种多尺度密集残差网络模型。方法 本文使用3种不同尺度卷积核对输入的低分辨率图像进行卷积处理,采集不同卷积核下的底层特征,这样可以较多地提取低分辨率图像中的细节信息,有利于图像恢复。再将采集的特征信息输入残差块中,每个残差块都包含了多个由卷积层和激活层构成的特征提取单元。另外,每个特征提取单元的输出都会通过短路径连接到下一个特征提取单元。短路径连接可以有效地缓解梯度消失现象,加强特征传播,促进特征再利用。接下来,融合3种卷积核提取的特征信息,经过降维处理后与3×3像素的卷积核提取的特征信息相加形成全局残差学习。最后经过重建层,得到清晰的高分辨率图像。整个训练过程中,一幅输入的低分辨率图像对应着一幅高分辨率图像标签,这种端到端的学习方法使得训练更加迅速。结果 本文使用两个客观评价标准PSNR（peak signal-to-noise ratio）和SSIM（structural similarity index）对实验的效果图进行测试,并与其他主流的方法进行对比。最终的结果显示,本文算法在Set5等多个测试数据集中的表现相比于插值法和SRCNN算法,在放大3倍时效果提升约3.4 dB和1.1 dB,在放大4倍时提升约3.5 dB和1.4 dB。结论 实验数据以及效果图证明本文算法能够较好地恢复低分辨率图像的边缘和纹理信息。     

Large-scale evaluation of splicing localization algorithms for web images

With the proliferation of smartphones and social media, journalistic practices are increasingly dependent on information and images contributed by local bystanders through Internet-based applications and platforms. Verifying the images produced by these sources is integral to forming accurate news reports, given that there is very little or no control over the type of user-contributed content, and hence, images found on the Web are always likely to be the result of image tampering. In particular, image splicing, i.e. the process of taking an area from one image and placing it in another is a typical such tampering practice, often used with the goal of misinforming or manipulating Internet users. Currently, the localization of splicing traces in images found on the Web is a challenging task. In this work, we present the first, to our knowledge, exhaustive evaluation of today’s state-of-the-art algorithms for splicing localization, that is, algorithms attempting to detect which pixels in an image have been tampered with as the result of such a forgery. As our aim is the application of splicing localization on images found on the Web and social media environments, we evaluate a large number of algorithms aimed at this problem on datasets that match this use case, while also evaluating algorithm robustness in the face of image degradation due to JPEG recompressions. We then extend our evaluations to a large dataset we formed by collecting real-world forgeries that have circulated the Web during the past years. We review the performance of the implemented algorithms and attempt to draw broader conclusions with respect to the robustness of splicing localization algorithms for application in Web environments, their current weaknesses, and the future of the field. Finally, we openly share the framework and the corresponding algorithm implementations to allow for further evaluations and experimentation.     

Super-resolution Reconstruction of Remote Sensing Images by Using Empirical Mode Decomposition and Compressed Sensing

Super resolution (SR) of remote sensing images is significant for improving accuracy of target identification and for image fusing.Conventional fusion-based methods inevitably result in distortion of spectral information,a feasible solution to the problem is the single-image based super resolution.In this work,we proposed a single-image based approach to super resolution of multiband remote sensing images.The method combines the EMD (Empirical Mode Decomposition),compressed sensing and PCA to dictionary learning and super resolution reconstruction of remote sensing color image.First,the original image is decomposed into a series of IMFs(Intrinsic Mode Function) according to their frequency component by using EMD,and the super resolution is implemented only on IMF1,which includes high-frequency component;then the K-SVD algorithm is used to learn and obtain overcomplete dictionaries,and the MOP (Orthogonal Matching Pursuit) algorithm is used to reconstruct the IMF1;Finally,the up-scaled IMF1 is combined with other IMFs to acquire the super resolution of original image.For a multiband image reconstruction,a PCA transform is first implemented on multiband image,and the PC1 is adopted for learning to get overcomplete dictionaries,the obtained dictionaries is then used to super-resolution reconstruction of each multi-spectral band.The Geoeye-1 panchromatic and multi-spectral images are used as experimental data to demonstrate the effectiveness of the proposed algorithm.The results show that the proposed method is workable to exhibit the detail within the images.     

Projection-based high accuracy measurement of straight line edges

We present a novel projection-based high accuracy algorithm to determine the parameters of a straight edge in a noisy image. Our algorithm is equivalent to applying a set of very long directional operators over a range of finely quantized angles. This is known to improve both signal-to-noise ratio and localization in the measurement of straight line edges, but was rarely used in practice because of the expense and coefficient restrictions of the operators. The algorithm is implemented by taking a set of projections of the original grey level image, filtering the projections, and analyzing the peaks in projection space to estimate line offset and angle. It also includes a procedure to analyze the peaks in projection space which provides an accuracy better than the quantization in offset and angle parameters. The algorithm was tested on synthetic and a large number of real images and offers very high (subpixel) offset and angular accuracy. The advantages of the algorithm over traditional approaches are improved signal-to-noise ratio and localization accuracy (due to the effective use of very long directional edge operators), no need for expensive edge operators and related hardware, and no need for sophisticated thresholding of the gradient image for finding edges. It is robust in the presence of many types of texture, patterned and bias noise, light intensity, and focus change, and is ideal for use in industrial machine vision, where a large number of parts with straight edges are processed.     

Dynamic SfM: Detecting Scene Changes from Image Pairs

Detecting changes in scenes is important in many scene understanding tasks. In this paper, we pursue this goal simply from a pair of image recordings. Specifically, our goal is to infer what the objects are, how they are structured, and how they moved between the images. The problem is challenging as large changes make point‐level correspondence establishment difficult, which in turn breaks the assumptions of standard Structure‐from‐Motion (SfM). We propose a novel algorithm for dynamic SfM wherein we first generate a pool of potential corresponding points by hypothesizing over possible movements, and then use a continuous optimization formulation to obtain a low complexity solution that best explains the scene recordings, i.e., the input image pairs. We test the algorithm on a variety of examples to recover the multiple object structures and their changes.     

Combining deblurring and denoising for handheld HDR imaging in low light conditions

This paper proposes a probability formulation that unifies both single-image deblurring and multi-image denoising using variational inference. The proposed formulation is based on a theoretical analysis that compares denoising and deblurring in the same probabilistic framework, and supported by a practical approach that deal with general motion that creates HDR images in the presence of spatially varying motion. Based on this formulation, a new algorithm for deblurring a noisy and blurry image pair is presented. Besides, we provide also an approach that combines existing optical flow and image denoising techniques for High Dynamic Range imaging.     

基于躯干检测的单人不良图片识别

目的 互联网中色情图片传播泛滥,对其自动识别与过滤越来越重要,而目前多数不良图片识别方法对类肤色区域较多的正常图像容易产生误检。为此,针对网络上常见的单人色情写真类图片,在总结已有方法不足的基础上提出一种将躯干部位作为感兴趣区域的不良图片识别算法。方法 首先使用基于Poselet(姿态部件)的人体躯干检测方法定位出与色情信息密切相关的躯干区域,然后基于躯干区域提取具有判别力的Fisher向量,最后使用线性支持向量机(SVM)进行分类。然而,由于人体外观变化很大,躯干检测器输出的置信度最大的位置往往较躯干真实的位置有一定的偏移。为了克服这一缺点,提出一种自适应的算法,即根据躯干检测器输出的置信度自适应地选择多个躯干候选区域,并通过集成多个区域的判别结果来得到最终结果。此外,为了训练基于躯干的SVM分类器和验证算法的有效性,本文通过互联网下载的方式收集了一个包含30000幅单人色情写真图片的大规模数据集,并对色情部位进行了标注,标注信息可用于自动生成训练数据。结果 本文提出的基于躯干的自适应分类算法在收集的大规模数据集上达到了91.7%的识别精度,明显高于传统肤色模型的识别结果,尤其是对于如同泳装模特等皮肤裸露较多或类肤色区域较多的图像,本文方法效果尤为显著。结论 文中基于Poselet的躯干检测能够获取与色情信息更相关的信息,因而相比较于传统方法,在较为准确地检测不良图片的同时,有效地降低皮肤裸露较多的正常图像的误检率,达到了实际应用的要求。     

基于新型切换Lorenz混沌系统的图像加密算法研究

论文提出了新型的切换Lorenz混沌加密系统,在传统的单次图像加密的基础上,通过采用“置乱-扩散-置乱-扩散”的两次图像加密过程,实现了图像的更高水平加密。论文以Lena图像的加密过程为例,采用Matlab软件编写了相关程序,实现了新型的切换Lorenz混沌加密系统;在此基础上,通过直方图分析、信息熵值分析等验证,表明了论文实现的图像加密算法具有较高密匙空间和信息熵值,混乱效果较好,安全性更高,在未来具有更好的应用前景。     

深度学习的单幅图像超分辨率重建方法综述

图像超分辨率重建即使用特定算法将同一场景中的低分辨率模糊图像恢复成高分辨率图像。近年来,随着深度学习的蓬勃发展,该技术在很多领域都得到了广泛的应用,在图像超分辨率重建领域中基于深度学习的方法被研究的越来越多。为了掌握当前基于深度学习的图像超分辨率重建算法的发展状况和研究趋势,对目前图像超分辨率的流行算法进行综述。主要从现有单幅图像超分辨算法的网络模型结构、尺度放大方法和损失函数三个方面进行详细论述,分析各类方法的缺陷和益处,同时通过实验对比分析不同网络模型、不同损失函数在主流数据集上的重建效果,最后展望基于深度学习的单幅图像超分辨重建算法未来的发展方向。     

基于字典学习和非局部相似的超分辨率重建

为提高单帧降质图像的分辨率,提出了一种基于字典学习和非局部相似性的超分辨率重建算法。该算法主要将高分辨率图像减去利用迭代反投影重建结果得到差值图像,再利用K-奇异值分解(K-SVD)算法和联合字典生成的思想形成的字典训练方法,训练差值图像块和低分辨率图像块得到对应的高、低分辨率字典用于超分辨重建。此外,引入非局部相似性的正则项约束以提高重建图像的质量。实验结果表明,所提算法重建得到的图像在主观视觉效果和客观评价上优于基于例子学习的超分辨率算法。     

双粒度协议下基于hint的启发式缓存替换算法

合作式缓存技术是提高机群文件系统性能的关键技术之一.s2fs(scalable single-image file system)是一个单一映像机群文件系统原型,它利用双粒度协议实现了符合严格UNIX语义的合作式缓存.该文为s2fs设计了基于hint的启发式缓存替换算法,并为其建立了性能分析模型.分析结果表明,同现有的合作式缓存替换算法N-chance相比,启发式算法几乎在所有情况下都有效地降低了I/O的响应时间.     

Fast pattern recognition using normalized grey-scale correlation in a pyramid image representation

The ability to quickly locate one or more instances of a model in a grey scale image is of importance to industry. The recognition/localization must be fast and accurate. In this paper we present an algorithm which incorporates normalized correlation into a pyramid image representation structure to perform fast recognition and localization. The algorithm employs an estimate of the gradient of the correlation surface to perform a steepest descent search. Test results are given detailing search time by target size, effect of rotation and scale changes on performance, and accuracy of the subpixel localization algorithm used in the algorithm. Finally, results are given for searches on real images with perspective distortion and the addition of Gaussian noise.     

DWT域图像水印的精确嵌入与盲检测的实现

文章提出了一种用于版权认证的基于小波变换的盲图像水印算法,该算法首先从低频子带中选择N个大系数作为嵌入区间,用它们的的低5位表示水印信息;在嵌入过程中,该算法根据小波变换所具有的良好的空域局部化特性,采用了分块舍入技术,实现了水印信息的精确嵌入;在检测过程中,文章提出了一种新颖的差分检测方法,替代了传统的位置检测方法,既简化了检测过程,又增强了系统的鲁棒性。实验结果证明,该算法具有嵌入信息量大,水印透明性好,对各种常见的图像处理操作、有损压缩、裁剪和几何变换抵抗性强等特点。     

基于熵编码的JPEG压缩域脆弱图像水印算法

为了解决JPEG图像水印算法在嵌入容量、不可感知性、实时性和窜改定位的问题,提出了一种基于熵编码的JPEG压缩域脆弱图像水印算法。首先选择JPEG熵编码阶段进行水印嵌入,这样就可避免正向和逆向的DCT和量化运算,保证算法在嵌入阶段对图像的改变量较小且具有较好的实时性;接着根据熵编码过程中某些比特不会参与哈夫曼编码这一原则,在其中选择符合条件的系数嵌入水印,将水印信息嵌入到这些系数的最低有效位中,以规避修改哈夫曼编码系数后带来的编码误差,进一步加强了算法的透明性。实验表明,该算法不仅具有较大的嵌入容量和很好的不可感知性,而且脆弱性较高且能够对恶意窜改进行准确定位。     

多尺度密集时序卷积网络的单幅图像去雨方法

雨滴会降低户外拍摄图像质量,影响图像视觉效果及后续图像分析工作。针对目前去雨算法存在颜色失真、去雨过度化等问题,为了提高计算机视觉算法在中、大雨天气下的准确性,提出多尺度DenseTimeNet(密集时间序列卷积神经网络)的单幅图像去雨方法。该网络由多个尺度DenseTimeNetBlock(密集时序卷积网络密集块)组成,通过卷积下采样技术得到不同尺度下雨线特征信息与降低图像维度后利用时域卷积寻找的时间维度特征信息。在不同维度下学习雨景图和无雨图之间的映射关系,网络主体由密集卷积块和残差网络组成,可加速算法收敛速度,更深度学习图像纹理特征,使特征信息在网络结构进行深度传播,可以更好地复原残损图像。在不同方向,不同大小的雨滴图像上对所提方法进行验证,实验结果表明,该方法相较于现有算法,图像去雨效果良好。     

大形变微分同胚图像配准快速算法

本文提出一种研究大形变图像配准算法. 大形变使得图像信息和拓扑结构有较大的改变, 目前该方面的研究仍然是一个难点. 基于严密数学理论的微分同胚Demons算法是图像配准的著名算法, 为解决大形变配准问题提供了重要基础. 基于对微分同胚Demons算法的研究结合流形学习的思想提出一种大形变图像配准的新算法(MRL算法). 新算法通过挖掘图像的局部和全局流形信息改进微分同胚Demons 速度场的更新, 更好地保持图像的拓扑结构. 对比实验结果表明, 本文所提出的算法能够快速高精度地实现大形变图像的配准.     

基于图像融合与YOLOv3的铝型材表面缺陷检测

我国是工业铝型材制造大国，铝型材生产质量检测意义重大。针对传统的人工目测等方式检测效率低下，稳定性相对较弱；单一YOLOv3方法特征提取不突出，检测精度有限等问题，提出一种基于图像融合与YOLOv3的铝型材表面缺陷检测方法。首先利用图像增强、空域滤波的方法对原始图像进行预处理得到处理图像；然后借鉴SLAM中特征提取与匹配的思想对原始图像和处理图像进行特征提取与匹配；之后进行图像融合得到最终的处理后图像；再通过K-means算法聚类和调参优化，最后利用单阶段物体检测模型YOLOv3对铝型材表面缺陷进行检测。通过一个end-to-end的全卷积神经网络完成从原始图像的输入到Bounding box和box中物体类别与置信度的输出。实验结果表明，此图像融合与YOLOv3的方法对表面缺陷分类检出的平均成功率为98.33%,比单一YOLOv3方法提高了3.75个百分点；验证集mAP值为88.81%，提高了4.18个百分点，具有更强的特征提取能力和泛化能力，能精确检测表面缺陷，进行分类和定位。     

Air‐ground Matching: Appearance‐based GPS‐denied Urban Localization of Micro Aerial Vehicles

In this paper, we address the problem of globally localizing and tracking the pose of a camera‐equipped micro aerial vehicle (MAV) flying in urban streets at low altitudes without GPS. An image‐based global positioning system is introduced to localize the MAV with respect to the surrounding buildings. We propose a novel air‐ground image‐matching algorithm to search the airborne image of the MAV within a ground‐level, geotagged image database. Based on the detected matching image features, we infer the global position of the MAV by back‐projecting the corresponding image points onto a cadastral three‐dimensional city model. Furthermore, we describe an algorithm to track the position of the flying vehicle over several frames and to correct the accumulated drift of the visual odometry whenever a good match is detected between the airborne and the ground‐level images. The proposed approach is tested on a 2 km trajectory with a small quadrocopter flying in the streets of Zurich. Our vision‐based global localization can robustly handle extreme changes in viewpoint, illumination, perceptual aliasing, and over‐season variations, thus outperforming conventional visual place‐recognition approaches. The dataset is made publicly available to the research community. To the best of our knowledge, this is the first work that studies and demonstrates global localization and position tracking of a drone in urban streets with a single onboard camera.     

基于马鞍地形的无线传感器网络节点定位算法

无线传感器网络三维DV-Hop定位算法直接应用于山区地形会存在较大的节点定位误差。对此本文提出了基于山区马鞍地形的节点定位算法NLA-ST。该算法首先利用双曲抛物面函数拟合山区马鞍地形,然后应用三维DV-Hop定位算法计算节点初始位置,并将该初始位置垂直投影到马鞍地形表面。对投影节点坐标的非线性方程组进行了推导,并采用数值解法实现对节点最终坐标的计算。仿真实验表明,NLA-ST定位算法受马鞍地形参数影响较小且具有较高定位精度,能够有效满足实际应用需求。