基于LogGabor纹理特征分析的机场ROI的自动分割

提出了一种基于2D LogGabor纹理特征分析的机场ROI(Region of Interest)自动分割方法。首先将原始图像分块并做傅里叶变换;然后采用基于上下文纹理信息的LogGabor算法,提取图像块的尺度特征;再通过直线拟合,利用尺度特征分割出图像的机场ROI。实验证明文中方法具有较好的检测效果。     

一种改进的图像拼接检测算法

通过对图像拼接技术特点的分析,提出一种基于图像纹理特征分析和马尔科夫模型的改进的拼接图像检测算法。该算法计算图像DCT域上的马尔科夫转移概率矩阵,同时对图像进行纹理分析,得到两类特征共178维。为评估该检测算法的性能,提出了一个具体实现方案,提取了图片数据集的特征,使用支持向量机（Support Vector Machine,SVM）对特征数据进行训练与分类。实验表明,该方法取得了较好的分类效果。     

基于内容的图像检索系统的实现

将小波变换有效地运用于对图像的颜色和纹理特征的提取，并提出了一种将小波变换与图像分块结合起来提取颜色特征的方法，开发出一个综合运用颜色、纹理特征及相关反馈技术进行图像检索的系统。实验结果表明，该系统具有明显的优越性和通用性。     

一种抗混淆的恶意代码变种识别系统

恶意代码变种是当前恶意代码防范的重点和难点.混淆技术是恶意代码产生变种的主要技术,恶意代码通过混淆技术改变代码特征,在短时间内产生大量变种,躲避现有基于代码特征的恶意代码防范方法,对信息系统造成巨大威胁.本文提出一种抗混淆的恶意代码变种识别方法,采用可回溯的动态污点分析方法,配合触发条件处理引擎,对恶意代码及其变种进行...     

基于模糊识别恶意代码检测技术的研究

在恶意代码检测的过程中,假设恶意代码隐藏的比较深,很难对恶意代码特征进行完整、准确的提取.利用传统算法进行恶意代码检测,恶意代码的分布情况都是未知的,没有充分考虑到不同类别代码特征之间的差异性,降低了恶意代码检测的准确性.为此,提出基于模糊识别的恶意代码检测方法.根据支持向量机相关理论,提取恶意代码特征,并将上述特征作为恶意代码识别的依据.建立模糊识别辨别树,计算识别对象属于恶意代码的概率,实现恶意代码的检测.实验结果表明,利用改进算法进行恶意代码检测,能够极大提高检测的准确性.     

指纹图像特征点提取算法研究

为了将指纹纹理中的有效信息提取出来,通过指纹图像预处理和特征点提取两个步骤,提出了一种指纹图像的特征点提取算法。图像预处理主要包含图像增强、二值化、细化,而特征点提取则采用模板搜索提取法。最终通过MATLAB GUI系统将采集的指纹图像进行处理,在用户界面内显示图像处理结果以及特征点提取结果,说明该设计很好地实现了算法目的,并且具有很好的稳定性和扩展性。     

一种基于综合行为特征的恶意代码识别方法

基于行为的分析方法是恶意代码检测技术的发展方向,但现有的以孤立行为特征为依据的恶意代码识别方法误报率较高,本文提出了一种基于代码综合行为特征的恶意代码检测方法-IBC-DA.该算法通过改造的攻击树模型描述恶意代码执行过程中各相关主体间的关系,在此基础上计算得到的恶意性权值能够更加准确地反映代码执行过程对系统的影响.实验表明,利用本文算法进行病毒检测具有较低漏报率和误报率,并对未知恶意代码的防范具有积极意义.     

综合纹理和亮度的夜间场景图像来源检测方法

针对目前基于模式噪声方法处理夜间场景下图像来 源检测准确效果较差问题,提出一种综合纹理和 亮度的夜间场景图像来源检测方法。首先基于夜间场景图片会存在同张图片不同区域会有不 同纹理和亮度 从而会影响提取模式噪声质量不同这一理论依据,将图片根据纹理和亮度分成若干大小相等 的块;然后对 不同块按纹理平缓亮度良好、纹理复杂亮度差、纹理平缓亮度差、温度复杂亮度良好四种情 况处理,其中 对于纹理复杂的区域提取的噪声要抑制纹理干扰、对于亮度条件差的区域提取的模式噪声需 要增强、对于 亮度差且纹理又复杂的区域提取的模式噪声、既要抑制纹理的干扰又要增强模式噪声。实验 结果表明,本 文方法在夜间场景情况对图像检测整 体识别率均在在80%以上,与传 统模式噪声提取算法相比,该算法正确率能够提高4到12个百 分点。     

基于局部二元模式的快速红外人脸识别系统

红外人脸成像具有对光照、人脸皮肤、表情、姿态等因素变化不敏感的特点,可以在一定程度上弥补这些因素对可见光人脸识别影响的不足。为了充分提取红外人的局部鉴别特征,文中提出了一个基于局部二元模式的快速红外人脸识别系统。该系统首先通过thermoVision A40型红外热像仪获分辨率为320240的红外人脸图像,并通过人脸检测和归一化方法提取大小为6080的标准红外人脸图像。其次,基于人脸图像的对称性,将红外人脸图像分块。通过局部二元模式直方图提取每一分块所包含的纹理模式特征。最后,采用Kruskal-Wallis（KW）特征选择算法,进一步抽取对识别有贡献的局部纹理特征用于分类识别。实验结果表明:提出的热红外人脸系统识别率明显优于基于主成分分析（PCA）和线性鉴别分析（LDA）的传统红外人脸识别系统,可以达到98.6%的识别率。与此同时,提出的红外人脸识别系统识别速度也快于传统基于PCA和LDA系统,可以广泛应用于实时人脸识别中。     

基于分块颜色直方图和GWLBP的图像检索算法

为了提高多特征融合图像检索的效果,本文提出了一种基于分块颜色直方图和GWLBP的图像检索算法。算法采用K-means均值聚类对RGB颜色空间进行颜色聚类,再将4×4均匀分块图像分成9个子块,提取每个子块的颜色体积直方图,并赋予不同权值计算颜色特征;利用Gabor滤波器组对输入图像进行不同分辨率和方向滤波,然后将不同方向上局部滤波器输出结果与全局滤波器输出结果的平均值进行比较,并进行二值化,据此提出3种不同的GWLBP算子来提取纹理特征。最后对图像的颜色和纹理特征高斯归一化,采用加权平均来融合颜色和纹理的特征距离。通过实验仿真可知,与其他3种算法相比,本算法对正常和有旋转倾向的图像都有较高的查全率和查准率。     

Malware variants detection based on ensemble learning

Application programming interface (API) is a procedure call interface to operation system resource. API-based behavior features can capture the malicious behaviors of malware variants. However, existing malware detection approaches have a deal of complex operations on constructing and matching. Furthermore, graph matching is adopted in many approaches, which is a nondeterministic polynominal (NP)-complete problem because of computational complexity. To address these problems, a novel approach is proposed to detect malware variants. Firstly, the API of the malware are divided by their functions and parameters. Then, the classified behavior graph (CBG) is constructed from the API call sequences. Finally, the signature based on CBGs for each malware family is generated. Besides, the malware variants are classified by ensemble learning algorithm. Experiments on 1 220 malware samples show that the true positive rate (TPR) is up to 89.0% with the low false positive rate (FPR) 3.7% by ensemble learning.     

Mobile malware traffic detection approach based on value-derivative GRU

For the dramatic increase in the number and variety of mobile malware had created enormous challenge for information security of mobile network users,a value-derivative GRU-based mobile malware traffic detection approach was proposed in order to solve the problem that it was difficult for a RNN-based mobile malware traffic detection approach to capture the dynamic changes and critical information of abnormal network traffic.The low-order and high-order dynamic change information of the malicious network traffic could be described by the value-derivative GRU approach at the same time by introducing the concept of “accumulated state change”.In addition,a pooling layer could ensure that the algorithm can capture key information of malicious traffic.Finally,simulation were performed to verify the effect of accumulated state changes,hidden layers,and pooling layers on the performance of the value-derivative GRU algorithm.Experiments show that the mobile malware traffic detection approach based on value-derivative GRU has high detection accuracy.     

一种基于深度学习的强对抗性Android恶意代码检测方法

针对现有Android恶意代码检测方法容易被绕过的问题,提出了一种强对抗性的Android恶意代码检测方法.首先设计实现了动静态分析相结合的移动应用行为分析方法,该方法能够破除多种反分析技术的干扰,稳定可靠地提取移动应用的权限信息、防护信息和行为信息.然后,从上述信息中提取出能够抵御模拟攻击的能力特征和行为特征,并利用一个基于长短时记忆网络（Long Short-Term Memory,LSTM）的神经网络模型实现恶意代码检测.最后通过实验证明了本文所提出方法的可靠性和先进性.     

基于权限频繁模式挖掘算法的Android恶意应用检测方法

Android应用所申请的各个权限可以有效反映出应用程序的行为模式,而一个恶意行为的产生需要多个权限的配合,所以通过挖掘权限之间的关联性可以有效检测未知的恶意应用。以往研究者大多关注单一权限的统计特性,很少研究权限之间关联性的统计特性。因此,为有效检测Android平台未知的恶意应用,提出了一种基于权限频繁模式挖掘算法的Android恶意应用检测方法,设计了能够挖掘权限之间关联性的权限频繁模式挖掘算法—PApriori。基于该算法对49个恶意应用家族进行权限频繁模式发现,得到极大频繁权限项集,从而构造出权限关系特征库来检测未知的恶意应用。最后,通过实验验证了该方法的有效性和正确性,实验结果表明所提出的方法与其他相关工作对比效果更优。     

Curvelet-based fingerprint anti-spoofing

Existing perspiration-based liveness detection algorithms need two successive images (captured in certain time interval), hence they are slow and not useful for real-time applications. Liveness detection methods using extra hardware increase the cost of the system. To alleviate these problems, we propose new curvelet-based method which needs only one fingerprint to detect liveness. Wavelets are very effective in representing objects with isolated point singularities, but fail to represent line and curve singularities. Curvelet transform allows representing singularities along curves in a more efficient way than the wavelets. Ridges oriented in different directions in a fingerprint image are curved; hence curvelets are very significant to characterize fingerprint texture. Textural measures based on curvelet energy and co-occurrence signatures are used to characterize fingerprint image. Dimensionalities of feature sets are reduced by running Pudil’s sequential forward floating selection (SFFS) algorithm. Curvelet energy and co-occurrence signatures are independently tested on three different classifiers: AdaBoost.M1, support vector machine and alternating decision tree. Finally, all the aforementioned classifiers are fused using the “Majority Voting Rule” to form an ensemble classifier. A fingerprint database consisting of 185 real, 90 Fun-Doh and 150 Gummy fingerprints is created by using varieties of artificial materials for casts and moulds of spoof fingerprints. Performance of the new liveness detection approach is found very promising, as it needs only one fingerprint and no extra hardware to detect vitality.     

一种恶意软件实时检测模型

启发式扫描检测入侵行为未知的恶意软件,存在误报及漏报问题,且不能有效监控Rootkit。基于＂通过监控某种恶意行为,实现对一类入侵方式未知的恶意软件的实时检测＂的思想,提出了一种实时检测入侵行为未知恶意软件的Petri网模型,给出了性能测量及优化方法。通过在模型指导下建立的恶意软件实时检测系统中采集关键参数,完成了模型性能评价和调整。设计的系统可实时准确地检测具有特征行为的恶意软件。     

Saliency detection in the compressed domain for adaptive image retargeting

Saliency detection plays important roles in many image processing applications, such as regions of interest extraction and image resizing. Existing saliency detection models are built in the uncompressed domain. Since most images over Internet are typically stored in the compressed domain such as joint photographic experts group (JPEG), we propose a novel saliency detection model in the compressed domain in this paper. The intensity, color, and texture features of the image are extracted from discrete cosine transform (DCT) coefficients in the JPEG bit-stream. Saliency value of each DCT block is obtained based on the Hausdorff distance calculation and feature map fusion. Based on the proposed saliency detection model, we further design an adaptive image retargeting algorithm in the compressed domain. The proposed image retargeting algorithm utilizes multioperator operation comprised of the block-based seam carving and the image scaling to resize images. A new definition of texture homogeneity is given to determine the amount of removal block-based seams. Thanks to the directly derived accurate saliency information from the compressed domain, the proposed image retargeting algorithm effectively preserves the visually important regions for images, efficiently removes the less crucial regions, and therefore significantly outperforms the relevant state-of-the-art algorithms, as demonstrated with the in-depth analysis in the extensive experiments.     

基于代码进化的恶意代码沙箱规避检测技术研究

为了对抗恶意代码的沙箱规避行为,提高恶意代码的分析效率,该文提出基于代码进化的恶意代码沙箱规避检测技术。提取恶意代码的静态语义信息和动态运行时信息,利用沙箱规避行为在代码进化过程中所产生的动静态语义上的差异,设计了基于相似度差异的判定算法。在7个实际恶意家族中共检测出240个具有沙箱规避行为的恶意样本,相比于JOE分析系统,准确率提高了12.5%,同时将误报率降低到1%,其验证了该文方法的正确性和有效性。     

Intelligent Vehicle Parking System (IVPS) Using Wireless Sensor Networks

Android smartphones are employed widely due to its flexible programming system with several user-oriented features in daily lives. With the substantial growth rate of smartphone technologies, cyber-attack against such devices has surged at an exponential rate. Majority of the smartphone users grant permission blindly to various arbitrary applications and hence it weakens the efficiency of the authorization mechanism. Numerous approaches were established in effective malware detection, but due to certain limitations like low identification rate, low malware detection rate as well as category detection, the results obtained are ineffective. Therefore, this paper proposes a convolutional neural network based adaptive red fox optimization (CNN-ARFO) approach to detect the malware applications as benign or malware. The proposed approach comprising of three different phases namely the pre-processing phase, feature extraction phase and the detection phase for the effective detection of android malware applications. In the pre-processing phase, the selected dataset utilizes Minmax technique to normalize the features. Then the malicious APK and the collected benign apps are investigated to identify and extract the essential features for the proper functioning of malware in the extraction phase. Finally, the android mobile applications are detected using CNN based ARFO approach. Then the results based on detecting the benign and malicious applications from the android mobiles are demonstrated by evaluating certain parameters like model accuracy rate, model loss rate, accuracy, precision, recall and f-measure. The resulting outcome revealed that the detection accuracy achieved by the proposed approach is 97.29%.