Geometric distortion-resilient image hashing scheme and its applications on copy detection and authentication

Media hashing is an alternative approach to many applications previously accomplished with watermarking. The major disadvantage of the existing media hashing technologies is their limited resistance to geometric attacks. In this paper, a novel geometric distortion-invariant image hashing scheme, which can be employed to perform copy detection and content authentication of digital images, is proposed. Our major contributions are threefold: (i) a mesh-based robust hashing function is proposed; (ii) a sophisticated hash database for error-resilient and fast matching is constructed; and (iii) the application scalability of our scheme for content copy tracing and authentication is studied. In addition, we further investigate several media hashing issues, including robustness and discrimination, error analysis, and complexity, with respect to the proposed image hashing system. Exhaustive experimental results obtained from benchmark attacks confirm the excellent performance of the proposed method.     

用于图像认证和窜改检测的稳健图像摘要*

本文提出了一种用于图像认证和篡改检测的稳健图像摘要。该算法利用Zernike矩的幅度和修正后的相位旋转不变性生成图像摘要（图像Hash）。图像Hash之间的相似性用汉明距离度量。仿真结果表明该方法对大部分内容不变的图像操作具有鲁棒性。不同图像对Hash之间的汉明距离远大于阈值,因此该方法可用于图像认证。同时该方法还可以检测图像篡改并可以定位图像篡改的位置。     

融合特征关联性的深度哈希图像表示方法

针对深度描述子无法提供图像特征之间关联性的问题进行了研究,提出了一种融合特征关联性的深度哈希图像表示方法,这种方法将深度描述子之间的关系融入到图像内容的描述中,用于提高图像检索性能。首先,通过预训练网络生成图像的特征映射,并在此基础上提取出深度特征描述子。然后,将深度特征描述子映射为深度视觉词,从而用于深度视觉词的频繁项集发现。接下来将离散值的深度视觉词图像表示和哈希值的频繁项集图像表示连接生成图像表示。最后,算法通过图像类内、类间的相似性关系构造优化,得到最优的阈值,用于将图像表示变为哈希值。实验中,将提出的方法与一些优秀的图像表示方法在holiday、Oxford和Paris图像集的图像检索任务中进行了性能比对,用于证明此方法的有效性。     

Robust image hashing using progressive feature selection for tampering detection

The main problem addressed in this paper is the robust tamper detection of the image received in a transmission under various content-preserving attacks. To this aim the progressive feature point selection method is proposed to extract the feature points of high robustness; with which, the local feature and color feature are then generated for each feature point. Afterwards, the robust image hashing construction method is proposed by using the location-context information of the features. The constructed hash is attached to the image before transmission and it can be used for analyzing at destination to filter out the geometric transformations occurred in the received image. After image restoration, the similarity of the global hashes between the source image and restored image is calculated to determine whether the received image has the same contents as the trusted one or has been maliciously tampered. When the received image being judged as a tampered image, the hashes calculated with the proposed Horizontal Location-Context Hashing (HLCH) and Vertical Location-Context Hashing (VLCH) methods will be used to locate the tampered regions. Experimental results on different images with tampering of arbitrary size and location demonstrate that our image authentication and tampering localization scheme are superior to the state-of-the-art methods under various attacks.     

运用边缘方向直方图进行图像拷贝检测

基于内容的图像拷贝检测已经成为图像版权保护的重要手段之一。提出一种基于图像边缘直方图的拷贝检测方法：采用小波模极大值算法提取源图像和待认证图像边缘,然后结合边缘点梯度主方向重新计算其区域点的梯度方向并生成直方图,最后根据源图像与待认证图像所得边缘直方图间的相似性检测待认证图像是否为源图像的拷贝。实验表明：这种图像拷贝检测方法可有效检测出拷贝图像,并可容忍由缩放、旋转、噪声等操作引起的失真。     

面向感知哈希的图像数据集

目的 感知图像哈希又称图像摘要或是图像指纹,是一种有效的图像认证技术,近年来受到了广泛的关注。该技术通过将图像的感知鲁棒特征转化为固定长度的哈希序列,来实现图像版权认证。然而,该领域始终缺乏一个比较通用的数据集,已有数据集所使用的图像内容保留操作和真实场景差异较大,使得训练得到的神经网络架构在应对复杂的图像编辑操作时效果显著下降。方法 针对感知图像哈希任务,面向实际图像内容认证场景构建了一个新的数据集。首先,将现实中常见的图像内容保留操作进行总结和分类,设计了48种单一、复合的图像内容保留操作来生成感知相似图像;然后,根据感知图像哈希的定义,选择与待认证图像语义相似但是感知内容不同的图像作为感知不相似图像,增加了该数据集的辨别难度;最终建立了一个包含116 400幅图像的感知哈希图像数据集。结果 由于本文提出的数据集使用的图像内容保留操作更加复杂,不相似图像也更加难以辨别,使得在该数据集上训练得到的深度神经网络具有较好的泛化能力,即这些神经网络即使不进行重新训练或是微调,也可以在其他数据集上取得较好的认证性能。同时,在该数据集上训练得到的神经网络在不同数据集上性能差别较小,体现了本文数据集具有较好的稳定性。结论 设计了一个针对感知哈希的图像数据集,大量的对比实验表明了该数据集的有效性,该工作可对感知图像哈希领域的发展起到促进作用。     

Perceptual hashing for SAR image segmentation

Image segmentation is an important step in the implementation of the interpretation of synthetic aperture radar (SAR) image due to speckle. This article proposes a SAR image segmentation method based on perceptual hashing. The new algorithm is divided into two phases. The first phase is to obtain initial regions with multi-thresholding based on histogram after reducing the speckle noise. The initial regions are used as input data. And the next phase is to merge regions according to the similarity between regions. In this phase, to segment SAR image effectively, the proposed hashing algorithm is used to obtain hash value and similarity between regions, which preserve the texture features of SAR images. In addition, we can obtain a smooth segmentation result by reducing the redundant information with principal component analysis. Furthermore, morphological methods are used to eliminate the uneven background in the segmentation results. These improvements make our algorithm more effective to segment the images with high speed. The experimental results of four real and one synthetic SAR images verify the efficiency of our algorithm.     

Piecewise supervised deep hashing for image retrieval

In this paper, we propose a novel hash code generation method based on convolutional neural network (CNN), called the piecewise supervised deep hashing (PSDH) method to directly use a latent layer data and the output layer result of the classification network to generate a two-segment hash code for every input image. The first part of the hash code is the class information hash code, and the second part is the feature message hash code. The method we proposed is a point-wise approach and it is easy to implement and works very well for image retrieval. In particular, it performs excellently in the search of pictures with similar features. The more similar the images are in terms of color and geometric information and so on, the better it will rank above the search results. Compared with the hashing method proposed so far, we keep the whole hashing code search method, and put forward a piecewise hashing code search method. Experiments on three public datasets demonstrate the superior performance of PSDH over several state-of-art methods.

     

自监督深度离散哈希图像检索

目的 基于深度学习的图像哈希检索是图像检索领域的热点研究问题。现有的深度哈希方法忽略了深度图像特征在深度哈希函数训练中的指导作用,并且由于采用松弛优化,不能有效处理二进制量化误差较大导致的生成次优哈希码的问题。对此,提出一种自监督的深度离散哈希方法（self-supervised deep discrete hashing,SSDDH）。方法 利用卷积神经网络提取的深度特征矩阵和图像标签矩阵,计算得到二进制哈希码并作为自监督信息指导深度哈希函数的训练。构造成对损失函数,同时保持连续哈希码之间相似性以及连续哈希码与二进制哈希码之间的相似性,并利用离散优化算法求解得到哈希码,有效降低二进制量化误差。结果 将本文方法在3个公共数据集上进行测试,并与其他哈希算法进行实验对比。在CIFAR-10、NUS-WIDE（web image dataset from National University of Singapore）和Flickr数据集上,本文方法的检索精度均为最高,本文方法的准确率比次优算法DPSH（deep pairwise-supervised hashing）分别高3%、3%和1%。结论 本文提出的基于自监督的深度离散哈希的图像检索方法能有效利用深度特征信息和图像标签信息,并指导深度哈希函数的训练,且能有效减少二进制量化误差。实验结果表明,SSDDH在平均准确率上优于其他同类算法,可以有效完成图像检索任务。     

基于尺度不变特征变换的快速图像拷贝检测

针对传统基于尺度不变特征变换（SIFT）特征的图像拷贝检测算法特征提取速度慢、匹配效率不高的问题,提出了一种基于SIFT特征点位置分布与方向分布特征的快速图像拷贝检测算法。首先,提取SIFT特征点二维位置信息,通过计算各个特征点与图像中心点的距离、角度,分块统计各区间的特征点数量,依据数量关系量化生成二值哈希序列,构成一级鲁棒特征;然后,根据特征点一维方向分布特征分块统计各方向子区间特征点数量,依据数量关系构成二级图像特征;最后,拷贝检测时采用级联式过滤框架作出是否为拷贝的判断。仿真实验结果表明,与传统SIFT以128维特征描述子为基础构建哈希序列的图像拷贝检测算法相比,所提算法在保证鲁棒性与独特性不降低的同时,特征提取时间缩短为原来的1/20,匹配时间也缩短了1/2以上,可满足在线拷贝检测的需求。     

基于统计特征的鲁棒图像拷贝检测技术

抗几何攻击的拷贝检测技术是图像拷贝检测领域中的一个重要问题,现有的图像拷贝检测方法中,多数对类似噪声的图像失真比较鲁棒,而对于几何类失真则比较脆弱.因此,提出一种鲁棒图像拷贝检测技术,利用图像高斯过滤后的低频部分中的统计特征(直方图形状和均值)进行拷贝检测.实验结果表明该方法能较好地抵抗不改变内容的几何失真和图像处理攻击.     

贪心非对称深度有监督哈希图像检索方法

近年来,深度有监督哈希检索方法已成功应用于众多图像检索系统中。但现有方法仍然存在一些不足：一是大部分深度哈希学习方法都采用对称策略来训练网络,但该策略训练通常比较耗时,难以用于大规模哈希学习过程;二是哈希学习过程中存在离散优化问题,现有方法将该问题进行松弛,但难以保证得到最优解。为解决上述问题,提出了一种贪心非对称深度有监督哈希图像检索方法,该方法将贪心算法和非对称策略的优势充分结合,进一步提高了哈希检索性能。在两个常用数据集上与17种先进方法进行比较。在CIFAR-10数据集上48 bit条件下,与性能最好的方法相比mAP提高1.3%;在NUS-WIDE数据集上所有bit下,mAP平均提高2.3%。在两个数据集上的实验结果表明,该方法可以进一步提高哈希检索性能。     

Supervised deep hashing for image content security

Multimedia Tools and Applications - Due to the fast growth of image data on the web, it is necessary to ensure the content security of uploaded images. One of the fundamental problems behind this...     

Robust and secure image hashing

Image hash functions find extensive applications in content authentication, database search, and watermarking. This paper develops a novel algorithm for generating an image hash based on Fourier transform features and controlled randomization. We formulate the robustness of image hashing as a hypothesis testing problem and evaluate the performance under various image processing operations. We show that the proposed hash function is resilient to content-preserving modifications, such as moderate geometric and filtering distortions. We introduce a general framework to study and evaluate the security of image hashing systems. Under this new framework, we model the hash values as random variables and quantify its uncertainty in terms of differential entropy. Using this security framework, we analyze the security of the proposed schemes and several existing representative methods for image hashing. We then examine the security versus robustness tradeoff and show that the proposed hashing methods can provide excellent security and robustness.     

Hierarchical deep semantic hashing for fast image retrieval

Multimedia Tools and Applications - Content-Based large-scale image retrieval has recently attracted considerable attention because of the explosive increase of online images. Inspired by recent...     

编码选择哈希算法解决大规模图像检索问题

哈希算法已被广泛应用于解决大规模图像检索的问题. 在已有的哈希算法中, 无监督哈希算法因为不需要数据库中图片的语义信息而被广泛应用. 平移不变核局部敏感哈希(SKLSH)算法就是一种较为代表性的无监督哈希算法.该算法随机的产生哈希函数, 并没有考虑所产生的哈希函数的具体检索效果. 因此, SKLSH算法可能产生一些检索效果表现较差的哈希函数. 在本文中, 提出了编码选择哈希算法(BSH). BSH算法根据SKLSH算法产生的哈希函数的具体检索效果来进行挑选. 挑选的标准主要根据哈希函数在3个方面的表现: 相似性符合度, 信息包含量, 和编码独立性. 然后,BSH算法还使用了一种基于贪心的选择方法来找到哈希函数的最优组合. BSH算法和其他代表性的哈希算法在两个真实图像库上进行了检索效果的对比实验. 实验结果表明, 相比于最初的SKLSH算法和其他哈希算法, BSH算法在检索准确度上有着明显的提高.     

基于图像能量的稳健图像哈希算法

为了提高哈希算法的分类性能和运行效率,提出一种基于梯度变化特征和能量特征的图像哈希算法.首先,对输入图像进行预处理操作形成二次图像,利用Sobel算子对二次图像的红色通道、绿色通道、蓝色通道图像进行x轴和y轴的梯度值计算,将各分量图像的梯度值进行相加得到最终的梯度图像;然后将梯度图像幅值的多方向变化信息作为图像的梯度特征,所有图像子块的能量值作为图像的能量特征;最后将图像的梯度特征与能量特征联合起来并置乱得到最终的哈希序列.实验结果表明:所提算法的区别性和鲁棒性可以达到较好的权衡;与最新的以及较好的哈希算法相比,该算法具有最好的ROC曲线和最短的运行时间,平均运行时间为0.0242 s,并且在拷贝检测的对比实验中,所提算法的查全率—查准率曲线最好.     

Deep linear discriminant analysis hashing for image retrieval

Multimedia Tools and Applications - Currently, due to the exponential growth of online images, it is necessary to consider image retrieval among large number of images, which is very time-consuming...     

一种安全鲁棒的图像哈希方法*

互联网的发展使得多媒体的版权保护成为亟待解决的问题,从而使得用于图像内容认证的哈希方法得到广泛的研究和应用。提出了一种新的基于V系统的图像哈希方法,该方法首先将原始图像强化,然后提取V变换域上的系数生成哈希值。实验结果显示,该方法能够抵抗滤波、噪声、25%以下的裁剪和95%以下的JPEG压缩等攻击,有很好的安全性和鲁棒性。