基于SIFT图像特征区域的全息水印技术

为了增强水印的不可见性和鲁棒性,提出了一种基于图像特征区域的水印算法。首先利用SIFT(Scale Invariant Feature Transform)算法从载体图像蓝色B通道中提取图像特征点来进行优化和筛选,根据优化后的稳定特征点及其特征尺度确定图像的特征区域,再结合全息技术,对原始水印图像进行双随机相位加密,生成加密全息水印;然后对特征区域进行离散余弦变换(DCT);最后在其中频区域嵌入加密全息水印。在提取水印时无须借助原始图像,是盲水印技术。实验结果表明:该算法重建的水印图像与原始水印图像的NC值高达0.95;水印的嵌入对图像质量影响很小,PSNR值高达55.97,能够抵抗常规信号攻击及缩放、剪切、平移等几何攻击。     

基于非抽样Contourlet变换的彩色图像数字水印算法

研究了小波变换和Contourlet变换的不足,利用非抽样Contourlet变换的多尺度性、多方向性以及平移不变性特征进行彩色图像数字水印.先对图像进行非抽样Contourlet变换得到不同尺度不同方向上的系数.然后为了满足水印的不町见性和稳健性,在低频系数和高频系数中采用不同的嵌入强度.并依据人类视觉的彩色感知特性,在蓝色分量中嵌入较多水印,在其他颜色分量选取变换系数能量最大的区域嵌入水印.与同样嵌入规则的Contourlet变换算法和平稳小波变换算法相比.该算法具有更好的稳健性,利用该算法得到的水印相似度更高,最高町提高约0.5.     

自嵌入双功能图像水印算法

利用自嵌入技术提出一种同时实现版权保护和内容认证的双功能图像水印算法.在嵌入端,首先将原始图像划分成互不重叠的子块,将各子块每个像素的最低m位置0,然后将最低m位置0后的子块进行奇异值分解,通过提取奇异值范数的最高位奇偶性产生原始鲁棒特征水印,然后再将原始鲁棒特征水印嵌入各子块每个像素的最低m位得到含水印图像.检测端从攻击的含水印图像提取鲁棒特征水印的过程与嵌入端产生原始鲁棒特征水印的过程类似,并且通过计算提取的鲁棒特征水印与原始鲁棒特征水印之间的归一化相关度进行版权鉴别,通过判断提取的鲁棒特征水印与攻击图像各子块每个像素的最低m位的一致性实现篡改检测进行内容认证.理论分析和实验结果都表明算法具有非常好的不可见性.实验结果还表明,算法不仅在抵抗添加噪音、剪切、JPEG压缩、平滑、重采样和几何攻击如随机删除行、向右偏移列、向下偏移行表现出很强的鲁棒性,而且能够精确定位出篡改位置和区分篡改类型.因此,算法具有版权保护和内容认证双重功能.     

一种基于差分检测的自适应图像水印新方法

提出了一种新的图像水印实现方法。采用差分法进行水印强度的嵌入和检测,利用一种自适应算法调节水印强度,能使水印在透明性得到保证的前提下尽量地提高水印强度。使用密钥计算水印的嵌入位置,增加水印的安全性。在水印检测过程中,利用一种带优先级的膨胀算法辅助水印的复原。以二值图像作为水印信息,灰度图像作为水印的载体,以另一幅灰度图片作为密钥,通过实验成功地实现了水印的嵌入和检测。实验证明,该方法可以很好地抵御几何裁剪攻击、量化攻击、JPEG压缩攻击和高斯噪声攻击,具有一定的实用价值。     

结合Retinex和HVS的自适应全息水印算法

提出一种水印算法的新思路,可以自适应地确定水印嵌入的区域和强度。基于HVS的人眼掩蔽效应对图像亮度和纹理敏感度的量化指标进行计算,图像先后两次分块。依据评分选择最佳嵌入区域后对其进行单尺度Retinex分解,将其分解为用于评价亮度信息的照度分量和评价纹理信息的反射分量。根据评价结果,设置不同的水印嵌入强度分别嵌入图像小波域中频的各分块中。提取的水印通过数学形态学处理以提高其可辨识度。嵌入水印的不可见性得到最大化实现(PSNR=40.0),鲁棒性较佳,能抵御信号处理和几何攻击。能根据图像特征自适应分析水印的最佳嵌入位置,在保证隐蔽性的同时嵌入尽可能多的水印信息;水印的安全性也得以保障,是一种应用面广、适应性强的数字水印算法。     

基于QR码和Schur分解的双彩色盲水印算法

针对大多数盲水印算法不能抗旋转攻击及鲁棒性差的问题,提出一种将彩色QR码作为水印嵌入彩色宿主图像的强鲁棒性盲水印算法。算法基于Arnold变换对彩色QR码分通道置乱加密并归一化,将水印信息嵌入到彩色载体图像对应通道经小波变换和分块Schur分解后酉矩阵的系数差中。实验结果表明该算法具有较强的鲁棒性,能够抵抗任意角度的旋转攻击,且对裁切、JPEG压缩、高斯噪声、椒盐噪声、中值滤波、对比度增强等攻击具有较强的抵抗能力。算法在提取水印过程中不需要任何原载体图像相关信息,属于盲水印算法,在数字产品版权保护中具有较大的实际实用价值。     

基于视觉显著图的数字水印算法

为了更加有效地解决数字水印不可见性的问题,提出了一种基于视觉显著图的水印方法。对Itti视觉计算注意模进行改进,提取图像视觉显著图;利用二级哈尔小波变换方法,根据图像视觉显著图,将水印信息嵌入到图像的低频域内;采用多组SIFT特征点匹配进行精确几何校正处理,提取水印信息。实验结果表明,含水印图像的PSNR达到80dB以上,提取水印NC值都在0.8以上。因此,数字水印方法具有极好的不可感知性,可以有效抵抗噪声、滤波和JPEG压缩等常规图像处理攻击,尤其是对任意角度的旋转攻击、缩放、平移等几何攻击具有极强的鲁棒性。     

基于Radon变换和四元数实矩阵的彩色图像零水印算法

为了提高空间域彩色图像零水印算法的水印嵌入和检测精度以及抵抗旋转几何变换的能力,提出了一种基于Radon变换和四元数实矩阵表示的抗旋转攻击空间域零水印算法。首先将原始彩色图像使用四元数实矩阵来表征,并计算实数字矩阵的Radon变换不变矩;然后利用少量低阶Radon变换不变矩来设计和构建零水印信息;在水印检测前,通过Radon变换几何校正算法对可能遭受旋转攻击的待检测图像进行旋转角度的校正,然后进行水印的提取。实验结果表明,该方法可以获得良好的图像视觉效果,对于旋转几何攻击具有很强的鲁棒性,同时对滤波、JPEG压缩和剪切攻击也具有一定的鲁棒性。     

一种新的光电成像末制导景象匹配方法

针对光电成像末制导景象匹配中图像存在尺度、旋转、灰度和3D视角差异的问题,构造了一种基于均匀模式的特征描述符,并基于该特征描述符和最稳定极值区域(MSER)提出了一种新的景象匹配算法。算法首先提取基准图像和实时图像的MSER特征,MSER特征具有尺度和仿射不变性,基于新的特征描述符分别对该特征进行旋转和灰度不变性特征描述,然后根据欧氏距离比值准则提取两图像间匹配的MSER特征对,根据随机抽样一致性(RANSAC)算法估计两图像的外极几何关系,实现末制导景象匹配。仿真结果表明,该算法能够在光电成像末制导过程中实现稳定的景象匹配,其稳健性优于传统方法。     

基于四元数实矩阵和对数极坐标的彩色图像零水印算法

为了提高空间域彩色图像零水印算法的水印嵌入、检测的精度,以及抵抗旋转几何变换的能力,提出一种基于对数极坐标和四元数实矩表示的抗旋转攻击空间域零水印算法。将原始彩色图像使用四元数实矩阵来表征,计算实数字矩阵的Hu不变矩;利用7个Hu不变矩来设计和构建零水印信息;在水印检测前,通过对数极坐标校正算法,对可能遭受旋转攻击的待检测图像进行旋转角度的遍历,而后进行水印的提取。实验获得了良好的图象视觉效果,对于旋转几何攻击具有很强的鲁棒性,同时对于滤波,JPEG压缩,剪切攻击也具有一定的鲁棒性。     

Part-Aware Mask-Guided Attention for Thorax Disease Classification

Thorax disease classification is a challenging task due to complex pathologies and subtle texture changes, etc. It has been extensively studied for years largely because of its wide application in computer-aided diagnosis. Most existing methods directly learn global feature representations from whole Chest X-ray (CXR) images, without considering in depth the richer visual cues lying around informative local regions. Thus, these methods often produce sub-optimal thorax disease classification performance because they ignore the very informative pathological changes around organs. In this paper, we propose a novel Part-Aware Mask-Guided Attention Network (PMGAN) that learns complementary global and local feature representations from all-organ region and multiple single-organ regions simultaneously for thorax disease classification. Specifically, multiple innovative soft attention modules are designed to progressively guide feature learning toward the global informative regions of whole CXR image. A mask-guided attention module is designed to further search for informative regions and visual cues within the all-organ or single-organ images, where attention is elegantly regularized by automatically generated organ masks and without introducing computation during the inference stage. In addition, a multi-task learning strategy is designed, which effectively maximizes the learning of complementary local and global representations. The proposed PMGAN has been evaluated on the ChestX-ray14 dataset and the experimental results demonstrate its superior thorax disease classification performance against the state-of-the-art methods.     

尺度不变特征与几何特征融合的人耳识别方法

要提高人耳的识别率,关键是特征的提取与表达.尺度不变特征变换(SIFT)技术是局部点特征提取算法,在尺度空间寻找极值点,提取对图像的尺度和旋转变化具有不变性,对光照变化和图像变形具有较强的适应性的特征向量.尝试用SIFT技术来提取外耳图像的结构特征点以形成稳定的特征描述子,为了克服一幅图像中有多个局部描述子相似的问题,在SIFT特征描述子中融入一个耳廓几何特征.最后采用特征向量的欧氏距离作为两幅图像相似性度量标准进行人耳识别.在耳图像库七进行实验.结果表明,该方法不仅可以有效地提取人耳特征,通过少量特征可获得较高的识别率,而且对耳图像刚体变化具有较强的稳健性.     

Non-local means algorithm with adaptive patch size and bandwidth

Non-local means algorithm is an effective denoising method that consists in some kind of averaging process carried on similar patches in a noisy image. Some internal parameters, such as patch size and bandwidth, strongly influence the performance of non-local means, but with the difficulty of tuning. Many solutions for choosing these two parameters, like cross-validation and Steins unbiased risk estimate criterion, are successful but computationally heavy. In this paper, we introduce a new feature metric that is capable of providing a quantitative measure of geometric structures of image in the presence of noise. The proposed region-based non-local means method first classifies a noisy image into several regions. Then, a local window and a local bandwidth value are selected pixel-wisely according to the property of each region and the local value of the new feature metric. Experiments on standard test images show that the proposed method outperforms the original non-local means version by around 1.34 dB and is comparable to or better than the performance of the current state-of-the-art non-local means based denoising algorithms, both visually and quantitatively.     

Structural signature of plastic deformation in metallic glasses

The structure feature of a model CuZr metallic glass during deformation is investigated by molecular dynamics simulations. A spatially heterogeneous irreversible rearrangement is observed in terms of nonaffine displacement. We find that regions with smaller nonaffine displacement have more Voronoi pentagons, while in those with larger nonaffine displacement other types of faces are more populated. We use the degree of local fivefold symmetry (LFFS) as the structural indicator to predict plastic deformation of local structures and find that the plastic events prefer to be initiated in regions with a lower degree of LFFS and propagate toward regions with a higher degree of LFFS.     

Dark energy in systems of galaxies

The precise observational data of the Hubble Space Telescope have been used to study nearby galaxy systems. The main result is the detection of dark energy in groups, clusters, and flows of galaxies on a spatial scale of about 1–10 Mpc. The local density of dark energy in these systems, which is determined by various methods, is close to the global value or even coincides with it. A theoretical model of the nearby Universe has been constructed, which describes the Local Group of galaxies with the flow of dwarf galaxies receding from this system. The key physical parameter of the group-flow system is zero gravity radius, which is the distance at which the gravity of dark matter is compensated by dark-energy antigravity. The model predicts the existence of local regions of space where Einstein antigravity is stronger than Newton gravity. Six such regions have been revealed in the data of the Hubble space telescope. The nearest of these regions is at a distance of 1–3 Mpc from the center of the Milky Way. Antigravity in this region is several times stronger than gravity. Quasiregular flows of receding galaxies, which are accelerated by the dark-energy antigravity, exist in these regions. The model of the nearby Universe at the scale of groups of galaxies (～1 Mpc) can be extended to the scale of clusters (～10 Mpc). The systems of galaxies with accelerated receding flows constitute a new and probably widespread class of metagalactic populations. Strong dynamic effects of local dark energy constitute the main characteristic feature of these systems.     

Dynamics at the interface dividing collective chaotic and synchronized periodic states in a CML

A study is developed focusing the loss of stability of the interface dividing two regions of different spatial patterns on a coupled map lattice using coupling as the parameter guiding the transition. These patterns are constructed over local periodic/chaotic attractors generating regions of synchronized/collective behavior. The discrete feature of the underlying lattice, the anisotropy that stems from such discreteness and its possible change to an isotropic system through coupling with large number of neighbors are also investigated.     

基于快速轮廓转动力矩特征的激光主动成像目标识别

针对激光主动成像的图像特性,提出一种基于快速轮廓转动力矩的目标识别方法。将转动力矩的概念引入目标识别中,提出的快速轮廓转动力矩特征(FCTF)不仅包含了轮廓的尺寸、位置、规则度以及目标的亮暗等信息,同时对于旋转、尺度缩放等变换具有不变性。采用转动力矩的快速计算方法,提高了识别算法的计算效率。识别算法首先使用最大稳定极值区域(MSER)算法检测出目标特征区域,并将其变换为圆形区域,然后结合快速转动力矩特征算法提取出目标区域的局部不变特征,最后输入训练好的支持向量机分类器进行识别。实验结果表明相比于已有的激光主动成像目标识别方法,所提算法对于旋转、仿射变换均具有更高的识别率,同时单帧平均运算时间为9.68 ms,满足激光主动成像目标识别系统实时性的要求。     

多波段红外图像的海面舰船目标检测

现有的基于单个红外宽波段的海面舰船目标探测系统在面对复杂海天背景、岛岸背景、恶劣天气、亮带干扰或诱饵弹干扰等情况时,系统的探测率、虚警率、探测距离等性能指标均会受到严重的影响;为此,开展了基于多波段红外图像的海面舰船目标检测方法的研究。通过中波红外多波段数据采集系统实际采集107组五个中波红外波段的图像;波段1-5分别为3.7～4.8,3.7～4.1,4.4～4.8,3.7～3.9和4.65～4.75 μm;对多波段图像进行手动标注构建样本数据集,其中,正样本舰船目标298个,负样本非舰船目标353个。对于多波段红外图像,首先进行PCA降维并采用选择性搜索算法生成初始目标候选区域;针对候选区域中存在大量明显的非舰船目标区域的问题,利用积分图像计算候选区域的局部对比度,依据红外舰船目标的几何和灰度特征从初始目标候选区域中筛选出舰船目标可能性大的区域作为舰船目标候选区域。然后对舰船目标候选区域进行拓展以融入局部上下文信息,对于候选区域对应的5波段红外图像,分别提取每个波段图像的稠密SIFT特征,并将128维SIFT特征向量降为64维,融入SIFT特征的空间和波段位置分布信息得到新的特征向量,基于高斯混合模型对候选区域的特征向量集合进行编码融合得到舰船目标候选区域的费舍尔向量表示,最后利用线性SVM分类器识别出舰船目标。对多波段图像进行舰船目标候选区域生成实验,所提出的基于红外舰船目标的几何和灰度特征的约束方法可以有效地克服选择性搜索算法的不足,从初始目标候选区域中快速定位出舰船目标候选区域,对25组多波段图像进行实验,舰船目标候选区域生成的整体耗时为0.353 s,定位舰船目标区域耗时0.005 s。对100个正负样本进行目标识别测试,所提出的目标识别算法融合了目标的多波段图像特征信息,通过引入费舍尔向量挖掘了多波段图像梯度统计特征的深层次信息,算法的识别率达到了0.97,显著高于单波段红外图像的目标识别率。对25组多波段图像进行舰船目标检测实验,所提出的舰船目标检测方法能够在海天背景、岛岸背景以及亮带干扰等不同场景下完成海面舰船目标的检测工作,舰船目标定位准确,舰船目标召回率达到了0.95,每组多波段图像的平均检测耗时为1.33 s。研究结果表明,充分考虑海面舰船目标在红外图像中与局部海洋背景的辐射差异以及有效地融合舰船目标在多个红外波段图像中的辐射特征,可以增强舰船目标的可分性,提高舰船目标的识别率以及检测率,为基于多波段红外图像的海面舰船目标检测提供了新的技术支持。     

Infrared image enhancement through saliency feature analysis based on multi-scale decomposition

To improve contrast between dim target region and background in infrared (IR) long-range surveillance, this paper proposes a fast image enhancement approach using saliency feature extraction based on multi-scale decomposition. Firstly, a smooth based multi-scale decomposition is designed and applied to original infrared image, generating sub-images with various frequency components at different decomposition levels. The dim target regions of sub-images are extracted by a local frequency-tuned based saliency feature detection method, secondly. With saliency maps created by saliency extraction using multi-scale local windows with different sizes, the sub-images are enhanced at different decomposition scales. Finally, the enhanced result is reconstructed by synthesizing the all sub-images with adjustable synthetic weights. Since salient areas are analyzed based on fast multi-scale image decomposition, IR image can be s enhanced with good contrast successfully and rapidly. Compared with other algorithms, the experimental results prove that the proposed method is robust and efficient for IR image enhancement.     

A radiation scalar for numerical relativity

This Letter describes a scalar curvature invariant for general relativity with a certain, distinctive feature. While many such invariants exist, this one vanishes in regions of space-time which can be said unambiguously to contain no gravitational radiation. In more general regions which incontrovertibly support nontrivial radiation fields, it can be used to extract local, coordinate-independent information partially characterizing that radiation. While a clear, physical interpretation is possible only in such radiation zones, a simple algorithm can be given to extend the definition smoothly to generic regions of space-time.