LogGAN: a Log-level Generative Adversarial Network for Anomaly Detection using Permutation Event Modeling

Information Systems Frontiers - System logs that trace system states and record valuable events comprise a significant component of any computer system in our daily life. Each log contains...     

基于自动编码器的深度伪造图像检测方法

基于深度学习的图像伪造方法生成的图像肉眼难辨,一旦该技术被滥用于制作虚假图像和视频,可能会对国家政治、经济、文化造成严重的负面影响,也可能会对社会生活和个人隐私构成威胁。针对上述问题,提出了一种基于自动编码器的深度伪造Deepfake图像检测方法。首先,借助高斯滤波对图像进行预处理,提取高频信息作为模型输入;然后,利用自动编码器对图像进行特征提取,并在编码器中添加注意力机制模块以获取更好的分类效果;最后,通过消融实验证明,采用所提的预处理方法和添加注意力机制模块有助于伪造图像检测。实验结果表明,与ResNet50、Xception以及InceptionV3相比,所提方法在数据集样本量较小且包含的场景丰富时,可以有效检测多种生成方法所伪造的图像,其平均准确率可达97.10%,明显优于对比方法,且其泛化性能也明显优于对比方法。     

解剖特征引导软式内镜机器人自主介入策略研究

以上消化道介入为例,在前期研制的YunSRobot内镜机器人平台上开展软式内镜自主介入策略研究.基于Faster-RCNN算法和数学形态学图像处理方法从镜下图像中提取消化道解剖特征,并设计自主化镜体操控和弯曲定向策略实现内镜在上消化道中的自主介入.在高仿真度的上消化道模型中开展了对照实验,结果表明本文方法的介入成功率为100％.与非专业人员主从介入相比,平均耗时从262.01 s降低到197 s,介入力的最大值从11.8 N降到9.6 N.与专业人员主从介入相比,介入力最大值基本一致,仅多耗时34.33 s.本文提出的自主化介入策略可以拓展到呼吸道、尿道等其他腔道的软式内镜介入.     

Modeling and Control of Hybrid-driven Gliding Motion for an Underwater Gliding Snake-like Robot

International Journal of Control, Automation and Systems - An underwater gliding snake-like robot (UGSR) combines the advantages of an underwater glider (UG) and an underwater snake-like robot...     

一种融合极限学习机和改进烟花算法的混合式目标跟踪技术

针对背景杂波干扰、目标外观形变导致的跟踪精度差的问题,论文提出了一种基于融合极限学习机和改进烟花算法的混合式目标跟踪技术(OSELM-FWA).该技术在提取图片特征信息时,为了剔除冗余信息、保留边缘轮廓处的颜色信息,提出梯度稀疏矩阵作为特征信息提取工具来提取图片的颜色特征信息;同时为了提高运动模型的预测精度,提出了一种改进烟花算法来生成候选目标样本,该算法不同于传统烟花算法采用固定不变的常量作为爆炸半径参数,而是在其基础上加入目标的运动速率量来调整烟花半径参量,而且为了控制烟花的多样性,该算法还根据烟花置信度方差的变化情况动态调整烟花的变异概率,将烟花的多样性维持在最优范围内;另外为了及时地更新目标外观模型,利用目标的置信度变化率来调整更新样本数目,随后运用在线极限学习机增量的学习相应数量的目标信息.最后,论文将该跟踪技术在标准数据集OTB上进行对比实验,通过实验得到该跟踪技术在处理复杂环境下的目标跟踪问题时,不仅兼顾跟踪的实时性而且拥有较高跟踪精度.     

面向信息SNP选择的聚类算法

SNP数据在人类遗传病诊断与治疗中存在重要作用,但SNP原始数据存在大量冗余,因此需要选择出信息量大的SNP,完成SNP数据的降维.针对常用聚类算法应用到信息SNP选择时未考虑单个SNP与SNP子集之间相似度的问题,采用一种新的相似度度量方法,提出了一种改进的聚类算法K-MIGS,并将其应用到SNP选择中.K-MIGS算法解决了传统K-means不能挖掘出SNP位点与SNP子集之间的强相关性问题,并在医院提供的临床数据实验中表明,K-MIGS具有更高的非信息SNP子集重构度.最后使用支持向量机、决策树和神经网络对构造的SNP子集进行分类实验,对比K-means、特征加权K-means、ReliefF和MCMR,结果表明K-MIGS分类准确率和F1指标上提升了10％和15％,充分说明K-MIGS在信息SNP选择中具有更好的效果.     

融入类贡献抑制因子的灰度级模糊C均值图像分割

基于灰度级模糊C均值图像分割算法具有分割速度快的优势。由于无损检测图像中背景类和目标类差异较大,该算法不能有效地将目标分割出来,故提出改进的基于灰度级的模糊C均值算法。构建了一种与类大小反向相关的类贡献抑制因子表达式,将之融入目标函数后能够降低较大类对目标函数的贡献,这可避免较小类的聚类中心受较大类的影响而靠近较大类的聚类中心。最小化新的目标函数可得新形式的隶属度和聚类中心表征形式。采用类大小差异较大的无损检测图像进行试验,结果显示本文算法得到的分割图像视觉效果良好,而且指标G_mean也更高,进一步提升了基于灰度级模糊C均值算法适应能力。     

The mass,fake news,and cognition security

The widespread fake news in social networks is posing threats to social stability, economic development, and political democracy, etc. Numerous studies have explored the effective detection approaches of online fake news, while few works study the intrinsic propagation and cognition mechanisms of fake news. Since the development of cognitive science paves a promising way for the prevention of fake news, we present a new research area called Cognition Security (CogSec), which studies the potential impacts of fake news on human cognition, ranging from misperception, untrusted knowledge acquisition, targeted opinion/attitude formation, to biased decision making, and investigates the effective ways for fake news debunking. CogSec is a multidisciplinary research field that leverages the knowledge from social science, psychology, cognition science, neuroscience, AI and computer science. We first propose related definitions to characterize CogSec and review the literature history. We further investigate the key research challenges and techniques of CogSec, including humancontent cognition mechanism, social influence and opinion diffusion, fake news detection, and malicious bot detection. Finally, we summarize the open issues and future research directions, such as the cognition mechanism of fake news, influence maximization of fact-checking information, early detection of fake news, fast refutation of fake news, and so on.     

Reaction mechanism and microstructure evolution of reaction sintered <Emphasis Type="Italic">h</Emphasis>-BN

Hexagonal boron nitride ceramic (h-BN) based on the nitridation of B powders was obtained by reaction sintering method. The effects of sintering temperature on the mechanical properties and microstructure of the resultant products were investigated and the reaction mechanism was discussed. Results showed that the reaction between B and N₂ occurred vigorously at temperatures ranging from 1 000 °C to 1 300 °C, which resulted in the generation of t-BN. When the temperature exceeded 1 450 °C, transformation from t-BN to h-BN began to occur. As the sintering temperature increased, the spherical particles of t-BN gradually transformed into fine sheet particles of h-BN. These particles subsequently displayed a compact arrangement to achieve a more uniform microstructure, thereby increasing the strength.     

禁飞区无人机预警算法研究

针对大量在空中无秩序飞行的无人机有可能会闯入飞机场等禁飞区的情况,为了避免发生空中交通安全事故,提出了一种无人机禁飞区预警算法。首先,该算法经过坐标变换将描述无人机位置点的GPS坐标转换成对应的平面坐标;接着,采用改进的最小二乘曲线拟合算法预测出无人机的飞行轨迹;然后,通过计算预测的飞行轨迹曲线在当前点的切线是否会与描述禁飞区的电子围栏相交,来判断无人机是否会进入禁飞区。同时,所有的无人机都会安装上飞行数据记录模块,来实时地为该算法提供无人机的飞行状态信息。最后,通过MATLAB仿真实验验证了该预警算法的可行性和有效性,表明该算法可以对禁飞区周围的无人机进行预警。