More Like this

1. Development of a Facial Feature Based Image Steganography Technology

   https://doi.org/10.1109/CSCI49370.2019.00126  

   Marella, Pranay; Straub, Jeremy; Bernard, Benjamin (December 2019, Proceedings of the 2019 International Conference on Computational Science and Computational Intelligence (CSCI))

   A new image steganography method is proposed for data hiding. This method uses least significant bit (LSB) insertion to hide a message in one of the facial features of a given image. The proposed technique chooses an image of a face from a dataset of 8-bit color images of head poses and performs facial recognition on the image to extract the Cartesian coordinates of the eyes, mouth, and nose. A facial feature is chosen at random and each bit of the binary representation of the message is hidden at the least significant bit in the pixels of the chosen facial feature. © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. 

   more » « less
2. Use of Bash History Novelty Detection for Identification of Similar Source Attack Generation

   Hance, Jack; Straub, Jeremy (January 2021, Proceedings of the 2020 IEEE 19th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom))

   null (Ed.)

   When a cyberattack occurs, tracking the attack back to an actual individual person can be problematic. Even identifying the workstation or device that it originated from does not necessarily identify the attacker, as the attacking device could, itself, be compromised. A system to determine whether activities that occur are from the same user or not facilitates forensic analysis as well as the detection of concurrent attacks from different devices by the same user. This paper proposes a system for identifying attackers based on behaviors expressed via their use of the Bash command line interface, the most common shell on Linux distributions. Prior systems were limited by issues such as requiring labeled user data which is difficult to acquire or not being specific enough to monitor individual persons. The approach proposed herein does not require labeled data and is specific enough to target individual users. The proposed system analyzes the level of variance between commands used and calculates an anomaly score for each given command. It uses these anomaly scores to compare Bash history sets together to identify if they were created by the same user. © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. 

   more » « less
3. Isolated and Ensemble Audio Preprocessing Methods for Detecting Adversarial Examples against Automatic Speech Recognition

   Rajaratnam, Krishan; Shah, Kunal; Kalita, Jugal (November 2018, Conference on Computational Linguistics and Speech Processing (ROCLING))

   An adversarial attack is an exploitative process in which minute alterations are made to natural inputs, causing the inputs to be misclassified by neural models. In the field of speech recognition, this has become an issue of increasing significance. Although adversarial attacks were originally introduced in computer vision, they have since infiltrated the realm of speech recognition. In 2017, a genetic attack was shown to be quite potent against the Speech Commands Model. Limited-vocabulary speech classifiers, such as the Speech Commands Model, are used in a variety of applications, particularly in telephony; as such, adversarial examples produced by this attack pose as a major security threat. This paper explores various methods of detecting these adversarial examples with combinations of audio preprocessing. One particular combined defense incorporating compressions, speech coding, filtering, and audio panning was shown to be quite effective against the attack on the Speech Commands Model, detecting audio adversarial examples with 93.5% precision and 91.2% recall. 

   more » « less
4. Manually Classified Real and Fake News Articles

   https://doi.org/10.1109/CSCI49370.2019.00262  

   Snell, Nicholas; Fleck, William; Traylor, Terry; Straub, Jeremy (December 2019, Proceedings of the 2019 International Conference on Computational Science and Computational Intelligence (CSCI))

   News articles that are written with an intent to deliberately deceive or manipulate readers are inherently problematic. These so-called 'fake news' articles are believed to have contributed to election manipulation and even resulted in severe injury and death, by actions that they have triggered. Identifying intentionally deceptive and manipulative news article and alerting human readers is key to mitigating the damage that they can produce. The dataset presented in this paper includes manually identified and classified news stories that can be used for the training and testing of classification systems that identify legitimate versus fake and manipulative news stories. © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. 

   more » « less
5. Identifying UAV Swarm Command Methods and Individual Craft Roles Using Only Passive Sensing

   https://doi.org/10.1109/CSCI49370.2019.00199  

   Meredith, Jonathan; Straub, Jeremy; Bernard, Benjamin (December 2019, Proceedings of the 2019 International Conference on Computational Science and Computational Intelligence (CSCI))

   Anti-drone technologies that attack drone clusters or swarms autonomous command technologies may need to identify the type of command system being utilized and the various roles of particular UAVs within the system. This paper presents a set of algorithms to identify what swarm command method is being used and the role of particular drones within a swarm or cluster of UAVs utilizing only passive sensing techniques (which cannot be detected). A testing configuration for validating the algorithms is also discussed. © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. 

   more » « less