Search Constraints
« Previous |
411 - 420 of 475
|
Next »
Search Results
- Creator:
- Agarwal, Geetanjali
- Description:
- Digital imaging has experienced tremendous growth in recent decades, and have been used in a growing number of applications. With such increasing popularity and the availability of low-cost image editing software, the integrity of digital image content can no longer be taken for granted. This thesis introduces a new methodology for the forensic
analysis of digital images. It proposes a novel hashing method using scale-invariant feature transform (SIFT) features points and Discrete Wavelet Transform (DWT) approximation coefficients for image authentication. Experimental results show that the proposed method is robust to various content-preserving operations. In addition, the performance of the proposed method is compared to existing methods. The comparison results show that the proposed method performs better than the existing methods. This thesis also mentions about the Amazon Web Services that are being used in detail. Also, the name of this thesis — Aneka means that which have many variations. This thesis also talks about recognizing nearly duplicate/similar images or detecting differents variations of an image present in a database. Aneka is also one of the names of Lord Vishnu.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Zhang, Zhe
- Description:
- The purpose of this research is to explore electroencephalogram (EEG) signal features and evaluate the performance of different prediction models. This paper presents the methodologies of how to set up experiments to collect and visualize EEG data, discover feature correlations and build efficient classification models. Several experiments are conducted to collect raw EEG data from ten healthy subjects. In the experiment, subjects are asked to watch a looping video which shows rock, paper, and scissors repeatedly 90 seconds for each, they are also imagining doing the same thing in their mind. At the same time, a device called Emotiv Epoc placed on their head will record the EEG signals and transmit them to the computer. After data preprocessing and feature extraction, EEG data is fed into several different prediction models including Support Vector Machines, Logistic Regression, K-nearest Neighbors, and Neural Networks. The performance of the classification process due to different methods is presented and compared based on their accuracy, recall, precision, and F-1 score. The best model achieved in this study is Neural Networks with an accuracy of 76.5%, this is more than twice higher than guessing randomly with an accuracy of 33.3%. A GUI is also built as a brain-computer interface powered by the model with the best performance, which can be further developed for medical use or other purposes.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Gentry, Eric Elwood
- Description:
- As our world of digital devices continues to expand, the potential for digital evidence available to law enforcement during case investigation is ever increasing. The growing amount of digital evidence, along with the considerable lack of Digital Forensic Investigators [9] is causing a backlog to form at many of the digital forensics labs around the world. This backlog leads to delays in evidence analysis and reporting, causing investigators and prosecutors to postpone or even drop on-going cases.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Dwarakanath, Suhas
- Description:
- Due to rapid globalization and changing lifestyle, more people are now visiting foreign countries for business and travel. However, important signages like traffic signs, safety signs and informational signs are displayed in only one language (usually the native language). There are many languages in the world and it is impractical to install signages
in all the languages. In this research, by combining computer vision and bluetooth beacons, multilingual digital information is displayed on the user's smartphone. A mobile application is constructed which enables the user's smartphone to listen to nearby digital signages (with bluetooth beacons) and display the respective signage information in user's preferred language. The application also enables the user to capture an image of a printed signage, extract the information in the signage and translate the information to user's preferred language. Optical Character Recognition (OCR) is used to extract the text from the image and Google Translate is used for the language translation. This system was implemented in the university campus and experiments were conducted by to determine the feasibility of using this system on a larger scale. It was found that the system helps the users to understand their surroundings better in their preferred
language.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Lu, Yite
- Description:
- Sentiment Analysis is a popular topic in machine learning, a subfield of computer science. In the past, Sentiment Analysis has been widely adopted in e-commerce and helps organizations analyze customer satisfaction of products and services. More recently, Sentiment Analysis has expanded its applications across government agencies as well, whether it being to analyze potential human threats within social media or political influence in election campaigns. Even more generally, humans are simply curious about how other humans are feeling. Two major approaches to Sentiment Analysis include lexical semantic analysis and machine learning. In this thesis, I will combine different word embedding techniques and use machine learning to analyze sentiments across published tweets. The overall goal is to discover which approach to Sentiment Analysis offers better performance.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Bension, Jack BJ
- Description:
- The Nation Basketball Association (NBA) has embraced the 21st Century by increasing its use of advanced analytics. New and evolving statistics can be used to determine how efficient a player is while he is on the court. However, even though a player is being efficient, his performance may not lead to victories. This paper creates a Decision Tree Classifier model that helps to determine, through game by game statistics, an NBA player’s value to a team’s chance to win. Players tested in the model demonstrate that having a high PER does not always lead to being a great asset for their team. The models created also distinguish what statistics are important for All-Star and Starter level players. The All-Star model favors individually focused, offensive statistics; whereas the Starter model places a higher level of importance on team statistics.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Zhang, Hang
- Description:
- Turing machines(TMs) are mathematical models of computation that define abstract machines. Because of their simplicity and consistency, they are amenable to mathematical analysis. These hypothetical machines are intended to help explore the concept of what it meant to be computable. These models form the foundation of theoretical computer science. A lot of theoretical computer science has been beard on TMs, and so a lot of the primary results are in the language of Turing machines. It’s essential for computer science students to build up a strong foundation of computer science by understanding theoretical models behind this field. Nowadays, Universities provide computer science education usually offers one or two classes to introduce Turing Machine. However, studying Turing Machine without a useful visual tool can be a little bit less intuitive. In this thesis, a Turing Machine simulator was created. Also, a universal Turing machine was proposed to show how a simulator works at a lower level.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Bharaswadkar, Apurva
- Description:
- Central place foraging algorithms for multiple robots are gaining attention due to their performance and efficiency in various applications like planetary surveys, mining, object transportation and manipulation. In foraging tasks, multiple robots search for resources and deposit the collected resources to a particular location called “nest” or “home”. If the resources are deposited at a central single collection point, it becomes a central place foraging task. The performance of central place foraging approaches is reduced due to reactive interrobot collision avoidance. The performance decreases in two cases, first case is when two or more robots collect the resources from the same cluster and go to the central location for deposition and the second case is when the path of one robot going to nest from its search position or vice versa intersects with the path of another robot searching for resources. The approach proposed in this thesis is called Path Planning And Collision Avoidance Algorithm For Clustered Central Place Foraging (PPCA-CCPFA). PPCA-CCPFA concentrates on improving the performance of central place foraging task in terms of reducing the number of inter robot collisions and improving target collection in given time for clustered resource distributions. We compare our approach to the popular Distributed Deterministic Spiral Algorithm (DDSA). The proposed algorithm detects inter robot collision and finds an alternate collision free path for a robot in case 1 and adds a delay time for a robot in case 2. This approach has shown notable increase in the performance of DDSA with a single 8 x 8 resource cluster. This algorithm is tested on a single cluster resource distribution at random locations in the arena for a swarm size of 3 to 15 robots.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands
- Creator:
- Devlin, Christopher R.
- Description:
- Voting and choice aggregation are used widely not just in politics but in business decision making processes and other areas such as competitive bidding procurement. Stakeholders and others who rely on these systems require them to be fast, efficient, and, most importantly, fair. The focus of this thesis is to illustrate the complexities inherent in voting systems. The algorithms intrinsic in several voting systems are made explicit as a way to simplify choices among these systems. The systematic evaluation of the algorithms associated with choice aggregation will provide a groundwork for future research and the implementation of these tools across public and private spheres.
- Resource Type:
- Thesis
- Campus Tesim:
- Channel Islands