Rapid homoglyph prediction and detection through visual hitzone mapping and comparison, 2017

Item — Call Number: MU Thesis Gin

Identifier: b7668723

Staff Only

Scope and Contents

From the Collection:

The collection consists of theses written by students enrolled in the Monmouth University graduate Computer Science program. The holdings are primarily bound print documents that were submitted in partial fulfillment of requirements for the Master of Science degree.

Dates

Creation: 2017

Creator

Ginsberg, Avi (Author, Person)
Yu, Cui (Thesis advisor, Person)

Conditions Governing Access

The collection is open for research use. Access is by appointment only.

Access to the collection is confined to the Monmouth University Library and is subject to patron policies approved by the Monmouth University Library.

Collection holdings may not be borrowed through interlibrary loan.

Research appointments are scheduled by the Monmouth University Library Archives Collections Manager (723-923-4526). A minimum of three days advance notice is required to arrange a research appointment for access to the collection.

Patrons must complete a Researcher Registration Form and provide appropriate identification to gain access to the collection holdings. Copies of these documents will be kept on file at the Monmouth University Library.

Extent

1 Items (print book) : 61 pages ; 8.5 x 11.0 inches (28 cm).

Language of Materials

English

Additional Description

Abstract

Character comparison is far from a new concept. For decades, computer science has worked on Optical Character Recognition and handwriting recognition techniques, which are used in a myriad of applications. Much progress has been made in this field, especially with the more contemporary advancements in artificial intelligence. In these cases, we are trying to compare given input to a known set of characters; that is, we want to map image representation of a character to a text it represents. What if we want to do nearly the opposite? What if we want to take a known character and find unknown characters that visually resemble it? This concept has many uses, most notably in computer/internet security and copyright infringement detection. However, minimal research has been conducted on this topic until now.

Enter: Visual HitZone Mapping and Comparison. This thesis aims to provide a collection of algorithms to enable quick and efficient comparison of glyphs and prediction of homoglyphs with adjustable levels of granularity and percentages of similarity. Such a collection is widely versatile as it can be easily fit to a variety of use-cases right "out of the box." The proposed technique of Visual HitZone simulates the human behavior of "visual screening" and allows significant amounts of visual data to be represented in a format that is easily and quickly searchable. Ultimately, the technique allows the alogrithms to effortlessly scale and perform accurately on any set of glyphs from any language. Additonally, the Visual HitZone concept employs a pre-processing time-memory trade-off to vastly decrease search and comparison times.

Partial Contents

Abstract -- Acknowledgements -- Table of contents -- List of figures -- List of tables -- 1. Introduction -- 2. Literature review -- 3. Alogrithms and techniques -- 4. Experimental study and analysis -- 5. Potential applications -- 6. Conclusion -- 7. References.