非抽样Contourlet变换去噪滤波器设计的源相机识别

针对源相机识别和小波滤波器在获取残留噪声图像时会引入明显的场景噪声的问题,提出一种利用非抽样Contourlet变换(NSCT)进行模式噪声提取的新方案。首先根据源相机识别的过程,讨论小波滤波器在提取模式噪声上的不足,接着重点讨论设计基于NSCT滤波器进行模式噪声的提取。实验表明NSCT滤波器不仅使场景噪声得到明显的抑制,而且与小波滤波器相比,对来自三种不同相机的照片的平均识别率提高了近3.667%。     

Helena’s Many Daughters: More Mitogenome Diversity behind the Most Common West Eurasian mtDNA Control Region Haplotype in an Extended Italian Population Sample

The high number of matching haplotypes of the most common mitochondrial (mt)DNA lineages are considered to be the greatest limitation for forensic applications. This study investigates the potential to solve this constraint by massively parallel sequencing a large number of mitogenomes that share the most common West Eurasian mtDNA control region (CR) haplotype motif (263G 315.1C 16519C). We augmented a pilot study on 29 to a total of 216 Italian mitogenomes that represents the largest set of the most common CR haplotype compiled from a single country. The extended population sample confirmed and extended the huge coding region diversity behind the most common CR motif. Complete mitogenome sequencing allowed for the detection of 163 distinct haplotypes, raising the power of discrimination from 0 (CR) to 99.6% (mitogenome). The mtDNAs were clustered into 61 named clades of haplogroup H and did not reveal phylogeographic trends within Italy. Rapid individualization approaches for investigative purposes are limited to the most frequent H clades of the dataset, viz. H1, H3, and H7.     

计算机网络犯罪中电子数据的取证研究

近些年,计算机技术飞速发展,计算机网络犯罪也在暗流涌动,比如曾经备受关注的网络诈骗徐玉玉案,其对社会治安、网络空间安全等都造成了极其不良的影响.文章介绍计算机网络犯罪的概念和特点;其次,介绍计算机网络犯罪中的电子数据基本概念,阐述侦查人员在对案件进行电子数据取证时遵循的基本原则以及一些特殊要求;然后,以路由器为例介绍对...     

面向入侵的取证系统框架

在分析常见入侵攻击的基础上抽象出入侵过程的一般模式,提出针对入侵攻击的取证系统应满足的特征。提出了入侵取证模型,并基于这一取证模型在操作系统内核层实现了取证系统原型KIFS(kernel intrusionforensic system)。在对实际入侵的取证实验中,根据KIFS得到的证据,成功记录并重构了一个针对FreeBSD系统漏洞的本地提升权限攻击的完整过程。     

一种智能手机电源管理方案的设计与实现

叙述了一种以嵌入式Linux为操作系统的智能手机电源管理方案的设计原则、设计框架和实现要点.该方案在嵌入式Linux的基础上,针对智能手机软件环境复杂、对电源持续工作要求高的特点,设计和实现了从内核支持到应用程序的完整的电源管理软件框架.该方案与嵌入式图形运行环境的结合,使得电源管理的状态更容易监测、行为更容易控制;对硬件设计的讨论则体现了软硬件协同设计的思想.     

Is real screen time a determinant of problematic smartphone and social network use among young people?

Smartphone use by adolescents is increasingly attracting the interest of social scientists, parents, and educational stakeholders (teachers, educators, educational policy makers), generally in a negative context. There are, however, many myths surrounding the issue, resulting from the reproduction of stereotypes about the psychosocial functioning of adolescents in cyberspace, as well as inadequately constructed research tools that measure how new media are used. This text is an attempt to show the phenomenon of problematic Internet use through the prism of screen time measured using real data from smartphones. The research was conducted in first half of 2022 among adolescents aged 13–19 in Bosnia and Herzegovina (N = 1185, mean age = 15.47 with standard deviation = 1.84). The research was implemented using a triangulation of tools such as the Smartphone Addiction Scale (SAS Scale), Software Installed on the Smartphone (SIS Scale), The Bergen Social Media Addiction Scale (BSMAS), and the Screen Time Scale (STS). From the data collected, it was noted that: 1) Screen time correlates weakly with problematic smartphone use and social networking; 2)The average period of smartphone use in the study group is 3 h49 minutes per day, of which 37 min are related to web browsing, while 2 h22 minutes are related to social networking; 3) A quarter of adolescents use a smartphone more than 5.5 h per day, of which more than 2 h15 minutes is screen time related to social media; 4) Approximately half of the respondents use smartphones slightly more than 2 h per day; 5) Girls use smartphones for longer than boys; 6) Place of residence (city or countryside) does not affect screen time; 7) Screen time is related to a medium degree to the type of software installed on the smartphone; 8) There is a need to refine the measurement methodology of screen time and to abandon self-declaration in this area.     

Recaptured screen image identification based on vision transformer

Due to the copyright issues often involved in the recapture of LCD screen content, recaptured screen image identification has received lots of concerns in image source forensics. This paper analyzes the characteristics of convolutional neural network (CNN) and vision transformer (ViT) in extracting features and proposes a cascaded network structure that combines local-feature and global-feature extraction modules to detect the recaptured screen image from original images with or without demoiréing operation. We first extract the local features of the input images with five convolutional layers and feed the local features into the ViT to enhance the local perception capability of the ViT module, and further extract the global features of the input images. Through thorough experiments, our method achieves a detection accuracy rate of 0.9691 in our generated dataset and 0.9940 in the existing mixture dataset, both showing the best performance among the compared methods.     

Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling

Age estimation using forensics odontology is an important process in identifying victims in criminal or mass disaster cases. Traditionally, this process is done manually by human expert. However, the speed and accuracy may vary depending on the expertise level of the human expert and other human factors such as level of fatigue and attentiveness. To improve the recognition speed and consistency, researchers have proposed automated age estimation using deep learning techniques such as Convolutional Neural Network (CNN). CNN requires many training images to obtain high percentage of recognition accuracy. Unfortunately, it is very difficult to get large number of samples of dental images for training the CNN due to the need to comply to privacy acts. A promising solution to this problem is a technique called Generative Adversarial Network (GAN). GAN is a technique that can generate synthetic images that has similar statistics as the training set. A variation of GAN called Conditional GAN (CGAN) enables the generation of the synthetic images to be controlled more precisely such that only the specified type of images will be generated. This paper proposes a CGAN for generating new dental images to increase the number of images available for training a CNN model to perform age estimation. We also propose a pseudo-labelling technique to label the generated images with proper age and gender. We used the combination of real and generated images to train Dental Age and Sex Net (DASNET), which is a CNN model for dental age estimation. Based on the experiment conducted, the accuracy, coefficient of determination (R2) and Absolute Error (AE) of DASNET have improved to 87%, 0.85 and 1.18 years respectively as opposed to 74%, 0.72 and 3.45 years when DASNET is trained using real, but smaller number of images.