Biblio

Convolutional neural network is a variant of deep neural network. It is widely used to extract image features and used in image classification, image generation and style transfer. In the style transfer task, we can extract the content from one picture through the convolutional neural network, extract the style from another picture, and generate a new picture by the combination. In this paper, we show the general steps of image style transfer based on convolutional neural networks through a specific example, and discuss the future possible applications.

Universal style transfer methods typically leverage rich representations from deep Convolutional Neural Network (CNN) models (e.g., VGG-19) pre-trained on large collections of images. Despite the effectiveness, its application is heavily constrained by the large model size to handle ultra-resolution images given limited memory. In this work, we present a new knowledge distillation method (named Collaborative Distillation) for encoder-decoder based neural style transfer to reduce the convolutional filters. The main idea is underpinned by a finding that the encoder-decoder pairs construct an exclusive collaborative relationship, which is regarded as a new kind of knowledge for style transfer models. Moreover, to overcome the feature size mismatch when applying collaborative distillation, a linear embedding loss is introduced to drive the student network to learn a linear embedding of the teacher's features. Extensive experiments show the effectiveness of our method when applied to different universal style transfer approaches (WCT and AdaIN), even if the model size is reduced by 15.5 times. Especially, on WCT with the compressed models, we achieve ultra-resolution (over 40 megapixels) universal style transfer on a 12GB GPU for the first time. Further experiments on optimization-based stylization scheme show the generality of our algorithm on different stylization paradigms. Our code and trained models are available at https://github.com/mingsun-tse/collaborative-distillation.

Advertisement sharing in vehicular network through vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication is a fascinating in-vehicle service for advertisers and the users due to multiple reasons. It enable advertisers to promote their product or services in the region of their interest. Also the users get to receive more relevant ads. Usually, users tend to contribute in dissemination of ads if their privacy is preserved and if some incentive is provided. Recent researches have focused on enabling both of the parameters for the users by developing fair incentive mechanism which preserves privacy by using Zero-Knowledge Proof of Knowledge (ZKPoK) (Ming et al., 2019). However, the anonymity provided by ZKPoK can introduce internal attacker scenarios in the network due to which authenticated users can disseminate fake ads in the network without payment. As the existing scheme uses certificate-less cryptography, due to which malicious users cannot be removed from the network. In order to resolve these challenges, we employed conditional anonymity and introduced Monitoring Authority (MA) in the system. In our proposed scheme, the pseudonyms are assigned to the vehicles while their real identities are stored in Certification Authority (CA) in encrypted form. The pseudonyms are updated after a pre-defined time threshold to prevent behavioural privacy leakage. We performed security and performance analysis to show the efficiency of our proposed system.

Rapidly growing shared information for threat intelligence not only helps security analysts reduce time on tracking attacks, but also bring possibilities to research on adversaries' thinking and decisions, which is important for the further analysis of attackers' habits and preferences. In this paper, we analyze current models and frameworks used in threat intelligence that suited to different modeling goals, and propose a three-layer model (Goal, Behavior, Capability) to study the statistical characteristics of APT groups. Based on the proposed model, we construct a knowledge network composed of adversary behaviors, and introduce a similarity measure approach to capture similarity degree by considering different semantic links between groups. After calculating similarity degrees, we take advantage of Girvan-Newman algorithm to discover community groups, clustering result shows that community structures and boundaries do exist by analyzing the behavior of APT groups.

This paper studies the synthesis of controllers for cyber-physical systems (CPSs) that are required to carry out complex tasks that are time-sensitive, in the presence of an adversary. The task is specified as a formula in metric interval temporal logic (MITL). The adversary is assumed to have the ability to tamper with the control input to the CPS and also manipulate timing information perceived by the CPS. In order to model the interaction between the CPS and the adversary, and also the effect of these two classes of attacks, we define an entity called a durational stochastic game (DSG). DSGs probabilistically capture transitions between states in the environment, and also the time taken for these transitions. With the policy of the defender represented as a finite state controller (FSC), we present a value-iteration based algorithm that computes an FSC that maximizes the probability of satisfying the MITL specification under the two classes of attacks. A numerical case-study on a signalized traffic network is presented to illustrate our results.

Air-gapped networks are isolated from the Internet, since they store and process sensitive information. It has been shown that attackers can exfiltrate data from air-gapped networks by sending acoustic signals generated by computer speakers, however this type of covert channel relies on the existence of loudspeakers in the air-gapped environment. In this paper, we present CD-LEAK - a novel acoustic covert channel that works in constrained environments where loudspeakers are not available to the attacker. Malware installed on a compromised computer can maliciously generate acoustic signals via the optical CD/DVD drives. Binary information can then be modulated over the acoustic signals and be picked up by a nearby Internet connected receiver (e.g., a workstation, hidden microphone, smartphone, laptop, etc.). We examine CD/DVD drives and discuss their acoustical characteristics. We also present signal generation and detection, and data modulation and demodulation algorithms. Based on our proposed method, we developed a transmitter and receiver for PCs and smartphones, and provide the design and implementation details. We examine the channel and evaluate it on various optical drives. We also provide a set of countermeasures against this threat - which has been overlooked.

This paper explores using chaos-based cryptography for transmitting multimedia data, mainly speech and voice messages, over public communication channels, such as the internet. The secret message to be transmitted is first converted into a one-dimensional time series, that can be cast in a digital/binary format. The main feature of the proposed technique is mapping the two levels of every corresponding bit of the time series into different multiple chaotic orbits, using a simple encryption function. This one-to-many mapping robustifies the encryption technique and makes it resilient to crypto-analysis methods that rely on associating the energy level of the signal into two binary levels, using return map attacks. A chaotic nonautonomous Duffing oscillator is chosen to implement the suggested technique, using three different parameters that are assumed unknown at the receiver side. Synchronization between the transmitter and the receiver and reconstructing the secret message, at the receiver side, is done using a Lyapunov-based adaptive technique. Achieving stable operation, tuning the required control gains, as well as effective utilization of the bandwidth of the public communication channel are investigated. Two different case studies are presented; the first one deals with text that can be expressed as 8-bit ASCII code, while the second one corresponds to an analog acoustic signal that corresponds to the voice associated with pronouncing a short sentence. Advantages and limitation of the proposed technique are highlighted, while suggesting extensions to other multimedia signals, along with their required additional computational effort.

In this paper, the two methods for ciphering are presented and compared. The aim is to reveal the suitability of chaotic neural network approach to ciphering compared to AES cipher. The durations in seconds of both methods are presented and the two methods are compared. The results show, that the chaotic neural network is fast, suitable for ciphering of short plaintexts. AES ciphering is suitable for longer plaintexts or images and is also more reliable.

In this work, an asymmetric cryptography method for information security was developed, inspired by the fact that the human body generates chaotic signals, and these signals can be used to create sequences of random numbers. Encryption circuit was implemented in a Reconfigurable Hardware (FPGA). To encode and decode an image, the chaotic synchronization between two dynamic systems, such as Hopfield neural networks (HNNs), was used to simulate chaotic signals. The notion of Homotopy, an argument of topological nature, was used for the synchronization. The results show efficiency when compared to state of the art, in terms of image correlation, histogram analysis and hardware implementation.

Research has shown that cryptographic APIs are hard to use. Consequently, developers resort to using code examples available in online information sources that are often not secure. We have developed a web platform, named CryptoExplorer, stocked with numerous real-world secure and insecure examples that developers can explore to learn how to use cryptographic APIs properly. This platform currently provides 3 263 secure uses, and 5 897 insecure uses of Java Cryptography Architecture mined from 2 324 Java projects on GitHub. A preliminary study shows that CryptoExplorer provides developers with secure crypto API use examples instantly, developers can save time compared to searching on the internet for such examples, and they learn to avoid using certain algorithms in APIs by studying misused API examples. We have a pipeline to regularly mine more projects, and, on request, we offer our dataset to researchers.

Most modern cloud and web services are programmatically accessed through REST APIs. This paper discusses how an attacker might compromise a service by exploiting vulnerabilities in its REST API. We introduce four security rules that capture desirable properties of REST APIs and services. We then show how a stateful REST API fuzzer can be extended with active property checkers that automatically test and detect violations of these rules. We discuss how to implement such checkers in a modular and efficient way. Using these checkers, we found new bugs in several deployed production Azure and Office365 cloud services, and we discuss their security implications. All these bugs have been fixed.

New attacks on existing algorithms have revealed various weaknesses. To overcome these weaknesses, researchers are searching for new ones. In this study, a new image encryption algorithm has been proposed to realize this aim. The most obvious component of the proposed image encryption algorithm is the cryptographic substitution box structure used to provide the confusion requirement. The original aspect of the substitution box structure according to similar studies in the literature is that the design was realized by using Lotka-Volterra chaotic system. The analysis results showed that both the substitution box structure and the image encryption algorithm provide the necessary criteria for data security.

With the deep research on coherent optical OFDM offset quadrature amplitude modulation OFDM/OQAM in these years, and the communication system exposed to potential threat from various capable attackers, which prompt people lay emphasis on encryption methods for transmission. Therefore, in this paper, we systematically discuss an encryption project with the main purpose of improving security in coherent optical OFDM/OQAM (CO-OFDM/OQAM) system, and the scheme applied the chaotic constellation scrambling (CCS) which founded on chaotic cross mapping to encrypt transmitted information. Besides, we also systematically discuss the basic principle of the encryption scheme for CO-OFDM/OQAM system. According to numerous studies and analysis on experiment data with caution, such as the performance of entropy, bit error rate (BER). It's conforms that the security of CO-OFDM/OQAM system have been enhanced.

This paper describes a method for classifying onion sites. According to the results of the research, the most spread model of site in onion space is built. To create such a model, a specially trained neural network is used. The classification of neural network is based on five different categories such as using authentication system, corporate email, readable URL, feedback and type of onion-site. The statistics of the most spread types of websites in Dark Net are given.

The global proliferation of affordable photographing devices and readily-available face image and video editing software has caused a remarkable rise in face manipulations, e.g., altering face skin color using FaceApp. Such synthetic manipulations are becoming a very perilous problem, as altered faces not only can fool human experts but also have detrimental consequences on automated face identification systems (AFIS). Thus, it is vital to formulate techniques to improve the robustness of AFIS against digital face manipulations. The most prominent countermeasure is face manipulation detection, which aims at discriminating genuine samples from manipulated ones. Over the years, analysis of microtextural features using local image descriptors has been successfully used in various applications owing to their flexibility, computational simplicity, and performances. Therefore, in this paper, we study the possibility of identifying manipulated faces via local feature descriptors. The comparative experimental investigation of ten local feature descriptors on a new and publicly available DeepfakeTIMIT database is reported.

AI-synthesized face-swapping videos, commonly known as DeepFakes, is an emerging problem threatening the trustworthiness of online information. The need to develop and evaluate DeepFake detection algorithms calls for datasets of DeepFake videos. However, current DeepFake datasets suffer from low visual quality and do not resemble DeepFake videos circulated on the Internet. We present a new large-scale challenging DeepFake video dataset, Celeb-DF, which contains 5,639 high-quality DeepFake videos of celebrities generated using improved synthesis process. We conduct a comprehensive evaluation of DeepFake detection methods and datasets to demonstrate the escalated level of challenges posed by Celeb-DF.

Video streams acquired from thermal cameras are proven to be beneficial in diverse number of fields including military, healthcare, law enforcement, and security. Despite the hype, thermal imaging is increasingly affected by poor resolution, where it has expensive optical sensors and inability to attain optical precision. In recent years, deep learning based super-resolution algorithms are developed to enhance the video frame resolution at high accuracy. This paper presents a comparative analysis of super resolution (SR) techniques based on deep neural networks (DNN) that are applied on thermal video dataset. SRCNN, EDSR, Auto-encoder, and SRGAN are also discussed and investigated. Further the results on benchmark thermal datasets including FLIR, OSU thermal pedestrian database and OSU color thermal database are evaluated and analyzed. Based on the experimental results, it is concluded that, SRGAN has delivered a superior performance on thermal frames when compared to other techniques and improvements, which has the ability to provide state-of-the art performance in real time operations.

Differential privacy is fast becoming the gold standard in enabling statistical analysis of data while protecting the privacy of individuals. However, practical use of differential privacy still lags behind research progress because research prototypes cannot satisfy the scalability requirements of production deployments. To address this challenge, we present Chorus, a framework for building scalable differential privacy mechanisms which is based on cooperation between the mechanism itself and a high-performance production database management system (DBMS). We demonstrate the use of Chorus to build the first highly scalable implementations of complex mechanisms like Weighted PINQ, MWEM, and the matrix mechanism. We report on our experience deploying Chorus at Uber, and evaluate its scalability on real-world queries.

The opioid abuse epidemic represents a major public health threat to global populations. The role social media may play in facilitating illicit drug trade is largely unknown due to limited research. However, it is known that social media use among adults in the US is widespread, there is vast capability for online promotion of illegal drugs with delayed or limited deterrence of such messaging, and further, general commercial sale applications provide safeguards for transactions; however, they do not discriminate between legal and illegal sale transactions. These characteristics of the social media environment present challenges to surveillance which is needed for advancing knowledge of online drug markets and the role they play in the drug abuse and overdose deaths. In this paper, we present a computational framework developed to automatically detect illicit drug ads and communities of vendors.The SVM- and CNNbased methods for detecting illicit drug ads, and a matrix factorization based method for discovering overlapping communities have been extensively validated on the large dataset collected from Google+, Flickr and Tumblr. Pilot test results demonstrate that our computational methods can effectively identify illicit drug ads and detect vendor-community with accuracy. These methods hold promise to advance scientific knowledge surrounding the role social media may play in perpetuating the drug abuse epidemic.

With the rapid growth in diversity of PoE-IoT devices and concept of "Edge intelligence", PoE-IoT security and behavior analysis is the major concern. These PoE-IoT devices lack visibility when the entire network infrastructure is taken into account. The IoT devices are prone to have design faults in their security capabilities. The entire network may be put to risk by attacks on vulnerable IoT devices or malware might get introduced into IoT devices even by routine operations such as firmware upgrade. There have been various approaches based on machine learning(ML) to classify PoE-IoT devices based on network traffic characteristics such as Deep Packet Inspection(DPI). In this paper, we propose a novel method for PoE-IoT classification where ML algorithm, Decision Tree is used. In addition to classification, this method provides useful insights to the network deployment, based on the deviations detected. These insights can further be used for shaping policies, troubleshooting and behavior analysis of PoE-IoT devices.

Data security has risen to be one of the most critical concerns of computer professionals. Tighter legal requirements now exist for the purpose of protecting user data from unauthorized uses and for both preserving and erasing/sanitizing data records to meet legal compliance requirements. To meet the data security requirement, many secure (data) deletion techniques have been proposed to deal with the data security concerns from different system layers. This paper surveys the state-of-the-art secure deletion techniques that have been designed to pursue higher efficiency, verifiability, and portability for emerging types of hard disk drives and flash-based solid-state drives. Meanwhile, the pros and cons of implementing secure deletion in different system layers are also discussed, so as to assist in pursuing better secure deletion designs for future storage systems.

The wireless technology has knocked the door of tremendous usage and popularity in the last few years along with a high growth rate for new applications in the networking domain. Mobile Ad hoc Networks (MANETs) is solitary most appealing, alluring and challenging field where in the participating nodes do not require any active, existing and centralized system or rigid infrastructure for execution purpose and thus nodes have the moving capability on arbitrary basis. Radio range nodes directly communicate with each other through the wireless links whereas outside range nodes uses relay principle for communication. Though it is a rigid infrastructure less environment and has high growth rate but security is a major concern and becomes vital part of providing hostile free environment for communication. The MANET imposes several prominent challenges such as limited energy reserve, resource constraints, highly dynamic topology, sharing of wireless medium, energy inefficiency, recharging of the batteries etc. These challenges bound to make MANET more susceptible, more close to attacks and weak unlike the wired line networks. Theresearch paperismainly focused on two aspects, one is computation termination of cluster head algorithm and another is use of finite state machine for attacks identification.

A recurring principle in consideration of the future of systems engineering is continual dynamic adaptation. Context drives change whether it be from potential loss (threats, vulnerabilities) or from potential gain (opportunity-driven). Contextual-awareness has great influence over the future of systems engineering and of systems security. Those contextual environments contain fitness functions that will naturally select compatible approaches and filter out the incompatible, with prejudice. We don't have to guess at what those environmental shaping forces will look like. William Gibson famously tells us why: “The future is already here, it's just not evenly distributed;” and, sometimes difficult to discern. This paper provides archetypes that 1) characterize general systems engineering for products, processes, and operations; 2) characterize the integration of security to systems engineering; and, 3) characterize contextually aware agile-security. This paper is more of a problem statement than a solution. Solution objectives and tactics for guiding the path forward have a broader range of options for subsequent treatment elsewhere. Our purpose here is to offer a short list of necessary considerations for effective contextually aware adaptive system security in the future of systems engineering.

This report provides a calculation example of hydrodynamic and acoustic characteristics of the ship propulsor using numerical modelling with the help of RANS-models and eddy-resolving approaches in the hydrodynamics task, acoustic analogy in the acoustics tasks and harmonic analysis of the propulsor under hydrodynamic loads.

In order to the development of IoT, IETF developed a standard named 6LoWPAN for increase the usage of IPv6 to the tiny and smart objects with low power. Generally, the 6LoWPAN radio link needs end to end (e2e) security for its IPv6 communication process. 6LoWPAN requires light weight variant of security solutions in IPSec. A new security approach of 6LoWPAN at adaptation layer to provide e2e security with light weight IPSec. The existing security protocol IPsec is not suitable for its 6LoWPAN IoT environment because it has heavy restrictions on memory, power, duty cycle, additional overhead transmission. The IPSec had packet overhead problem due to share the secret key between two communicating peers by IKE (Internet Key Exchange) protocol. Hence the existing security protocol IPSec solutions are not suitable for lightweight-based security need in 6LoWPAN IoT. This paper describes 6LowPSec protocol with AES-CCM (Cipher block chaining Message authentication code with Counter mode) cryptographic algorithm with key size of 128 bits with minimum power consumption and duty cycle.