Physical Sciences and Mathematics Commons^™

A System For The Detection Of Adversarial Attacks In Computer Vision Via Performance Metrics, Sarah Reynolds

Doctoral Dissertations and Master's Theses

Adversarial attacks, or attacks committed by an adversary to hijack a system, are prevalent in the deep learning tasks of computer vision and are one of the greatest threats to these models' safe and accurate use. These attacks force the trained model to misclassify an image, using pixel-level changes undetectable to the human eye. Various defenses against these attacks exist and are detailed in this work. The work of previous researchers has established that when adversarial attacks occur, different node patterns in a Deep Neural Network (DNN) are activated within the model. Additionally, it is known that CPU and GPU …

Deep Convolutional Neural Network-Based System For Fish Classification, Ahmad Al Smadi, Atif Mehmood, Ahed Abugabah, Eiad Almekhlafi, Ahmad Mohammad Al-Smadi

All Works

In computer vision, image classification is one of the potential image processing tasks. Nowadays, fish classification is a wide considered issue within the areas of machine learning and image segmentation. Moreover, it has been extended to a variety of domains, such as marketing strategies. This paper presents an effective fish classification method based on convolutional neural networks (CNNs). The experiments were conducted on the new dataset of Bangladesh’s indigenous fish species with three kinds of splitting: 80-20%, 75-25%, and 70-30%. We provide a comprehensive comparison of several popular optimizers of CNN. In total, we perform a comparative analysis of 5 …

Gps-Denied Navigation Using Synthetic Aperture Radar Images And Neural Networks, Teresa White

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Unmanned aerial vehicles (UAV) often rely on GPS for navigation. GPS signals, however, are very low in power and easily jammed or otherwise disrupted. This paper presents a method for determining the navigation errors present at the beginning of a GPS-denied period utilizing data from a synthetic aperture radar (SAR) system. This is accomplished by comparing an online-generated SAR image with a reference image obtained a priori. The distortions relative to the reference image are learned and exploited with a convolutional neural network to recover the initial navigational errors, which can be used to recover the true flight trajectory throughout …

Convolutional Neural Networks For Deflate Data Encoding Classification Of High Entropy File Fragments, Nehal Ameen

University of New Orleans Theses and Dissertations

Data reconstruction is significantly improved in terms of speed and accuracy by reliable data encoding fragment classification. To date, work on this problem has been successful with file structures of low entropy that contain sparse data, such as large tables or logs. Classifying compressed, encrypted, and random data that exhibit high entropy is an inherently difficult problem that requires more advanced classification approaches. We explore the ability of convolutional neural networks and word embeddings to classify deflate data encoding of high entropy file fragments after establishing ground truth using controlled datasets. Our model is designed to either successfully classify file …

Keystroke Dynamics For User Authentication With Fixed And Free Text, Jianwei Li

Master's Projects

YouTube videos often include captivating descriptions and intriguing thumbnails designed to increase the number of views, and thereby increase the revenue for the person who posted the video. This creates an incentive for people to post clickbait videos, in which the content might deviate significantly from the title, description, or thumbnail. In effect, users are tricked into clicking on clickbait videos. In this research, we consider the challenging problem of detecting clickbait YouTube videos. We experiment with multiple state of the art machine learning techniques and a variety of textual features.

Keystroke Dynamics Based On Machine Learning, Han-Chih Chang

Master's Projects

The development of active and passive biometric authentication and identification technology plays an increasingly important role in cybersecurity. Biometrics that utilize features derived from keystroke dynamics have been studied in this context. Keystroke dynamics can be used to analyze the way that a user types by monitoring various keyboard inputs. Previous work has considered the feasibility of user authentication and classification based on keystroke features. In this research, we analyze a wide variety of machine learning and deep learning models based on keystroke-derived features, we optimize the resulting models, and we compare our results to those obtained in related research. …

Malware Classification Using Lstms, Dennis Dang

Master's Projects

Signature and anomaly based detection have long been quintessential techniques used in malware detection. However, these techniques have become increasingly ineffective as malware becomes more complex. Researchers have therefore turned to deep learning to construct better performing models. In this project, we create four different long-short term memory (LSTM) models and train each model to classify malware by family type. Our data consists of opcodes extracted from malware executables. We employ techniques used in natural language processing (NLP) such as word embedding and bidirection LSTMs (biLSTM). We also use convolutional neural networks (CNN). We found that our model consisting of …

Image-Based Malware Classification With Convolutional Neural Networks And Extreme Learning Machines, Mugdha Jain

Master's Projects

Research in the field of malware classification often relies on machine learning models that are trained on high level features, such as opcodes, function calls, and control flow graphs. Extracting such features is costly, since disassembly or code execution is generally required. In this research, we conduct experiments to train and evaluate machine learning models for malware classification, based on features that can be obtained without disassembly or execution of code. Specifically, we visualize malware samples as images and employ image analysis techniques. In this context, we focus on two machine learning models, namely, Convolutional Neural Networks (CNN) and Extreme …

Image-Based Localization Of User-Interfaces, Riti Gupta

Master's Projects

Image localization corresponds to translating the text present in the images from one language to other language. The aim of the project is to develop a methodology to translate the text in image captions from English to Hindi by taking context of the images into account. A lot of work has been done in this field [22], but our aim was to explore if the accuracy can be further improved by consideration of the additional information imparted by the images apart from the text. We have explored Deep Learning using neural networks for this project. In particular, Recurrent Neural Networks …

Bee Shadow Recognition In Video Analysis Of Omnidirectional Bee Traffic, Laasya Alavala

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Over a decade ago, beekeepers noticed that the bees were dying or disappearing without any prior health disorder. Colony Collapse Disorder(CCD) has been a major threat to bee colonies around the world which affects vital human crop pollination. Possible instigators of CCD include viral and fungal diseases, decreased genetic diversity, pesticides and a variety of other factors. The interaction among any of these potential facets may be resulting in immunity loss for honey bees and the increased likelihood of collapse. It is essential to rescue honey bees and improve the health of bee colony.

Monitoring the traffic of bees helps …

Traffic Flow Prediction Using Convolutional Neural Network Accelerated By Spark Distributed Cluster, Yihang Tang

Master's Projects

Obtain information from historical data to forecast traffic flow in a city can be difficult because a precision forecasting demands large amount of data and accurate pattern analysis. Meanwhile, it is also meaningful because it provides a detailed and accurate point-to-point prediction for users. In this project, I use CNN (Convolutional Neural Network) to train the model based on the images captured by webcams in New York City. Then I deploy the training process on a Spark distributed Cluster so that the whole training process is accelerated. To efficiently combine CNN and Apache Spark, the prediction model is re-designed and …

Improving Steering Ability Of An Autopilot In A Fully Autonomous Car, Shivanku Mahna

Master's Projects

The world we live in is developing at a really rapid pace and along with it is developing the technology that we use. We have clearly come a long way from calling a car modern because it had a touch screen infotainment system to calling it modern because it drives on its own. The progress has been so rapid that it demands for us to analyze this and try to improvise a small part of this journey. With the same thought in mind, this project focuses on improvising the steering ability of an autonomous car. In order to make more …

Predicting Off-Target Potential Of Crispr-Cas9 Single Guide Rna, Ishita Mathur

Master's Projects

With advancements in the field of genome engineering, researchers have come up with potential ways for site-specific gene editing. One of the methods uses the Clustered Regularly Interspaced Short Palindromic Repeats - CRISPR-Cas technology. It consists of a Cas9 nuclease and a single guide RNA (sgRNA) that cleaves the DNA at the intended target site. However, the target genome could contain multiple potential off-target sites and cleaving an off-target site can have deleterious effects in case of gene editing in humans.

Lab based assays have been developed to test the off-target effects of guide RNAs. However, it is not feasible …

Smartphone Gesture-Based Authentication, Preethi Sundaravaradhan

Master's Projects

In this research, we consider the problem of authentication on a smartphone based on gestures, that is, movements of the phone. Accelerometer data from a number of subjects was collected and we analyze this data using a variety of machine learning techniques, including support vector machines (SVM) and convolutional neural networks (CNN). We analyze both the fraud rate (or false accept rate) and insult rate (or false reject rate) in each case.

Visual Odometry Using Convolutional Neural Networks, Alec Graves, Steffen Lim, Thomas Fagan, Kevin Mcfall Phd.

The Kennesaw Journal of Undergraduate Research

Visual odometry is the process of tracking an agent's motion over time using a visual sensor. The visual odometry problem has only been recently solved using traditional, non-machine learning techniques. Despite the success of neural networks at many related problems such as object recognition, feature detection, and optical flow, visual odometry still has not been solved with a deep learning technique. This paper attempts to implement several Convolutional Neural Networks to solve the visual odometry problem and compare slight variations in data preprocessing. The work presented is a step toward reaching a legitimate neural network solution.